Professor Hung Nguyen

Tran, C., Hoang, T.D., Ha, Q.P. & Nguyen, H.T. 2012, 'Decentralised Model Predictive Control of Time-Varying Splitting Parallel Systems' in Mohammadpour, Javad; Scherer, Carsten W. (eds), Control of Linear Parameter Varying Systems with Applications, Springer-Verlag Berlin / Heidelberg, Berlin/Heidelberg, pp. 217-251.
View description>>

This chapter is devoted to the development of a decentralised model predictive control (MPC) strategy for splitting parallel systems that have timevarying and unknown splitting ratios. The large-scale system in consideration consists of several dynamically-coupled modular subsystems. Each subsystem is regulated by a dedicated multivariable controller employing the open-loop MPC algorithms in conjunction with stability constraints. The connection topology of the large-scale systems includes serial, parallel and recirculated configurations. The solution to splitting parallel systems in this chapter is not only an alternative to the hybrid approach for duty-standby modes, but also a novel approach that accommodates the concurrent operations of splitting parallel systems. The effectiveness of this approach rests on the newly introduced asymptotically positive real constraint (APRC) which prescribes an approaching characteristic towards a positive real property of the system under control. The asymptotic attribute of APRC smooths out all significant wind-up actions in the control trajectories. The APRCs are developed into a one-time-step quadratic constraint on the local control vectors, which plays the role of a stability constraint for the decentralised MPC. The recursive feasibility is assured by characterizing the APRC with dynamicmultiplier matrices. Numerical simulations for two typical modular systems in an alumina refinery are provided to illustrate the theoretical results.

Tran, C., Tuan, H.D., Ha, Q.P. & Nguyen, H.T. 2012, 'Decentralized Model Predictive Control of Time-varying Splitting Parallel Systems' in Mohammadpour, Javad; Scherer, Carsten W. (eds), Control of Linear Parameter Varying Systems with Applications, Springer, Houston, Stuttgart, pp. 217-251.
View/Download from: UTSePress | Publisher's site
View description>>

This chapter is devoted to the development of a decentralised model predictive control (MPC) strategy for splitting parallel systems that have timevarying and unknown splitting ratios. The large-scale system in consideration consists of several dynamically-coupled modular subsystems. Each subsystem is regulated by a dedicated multivariable controller employing the open-loop MPC algorithms in conjunction with stability constraints. The connection topology of the large-scale systems includes serial, parallel and recirculated configurations. The solution to splitting parallel systems in this chapter is not only an alternative to the hybrid approach for duty-standby modes, but also a novel approach that accommodates the concurrent operations of splitting parallel systems. The effectiveness of this approach rests on the newly introduced asymptotically positive real constraint (APRC) which prescribes an approaching characteristic towards a positive real property of the system under control. The asymptotic attribute of APRC smooths out all significant wind-up actions in the control trajectories. The APRCs are developed into a one-time-step quadratic constraint on the local control vectors, which plays the role of a stability constraint for the decentralised MPC. The recursive feasibility is assured by characterizing the APRC with dynamicmultiplier matrices. Numerical simulations for two typical modular systems in an alumina refinery are provided to illustrate the theoretical results.

Khushaba, R.N., Al-Ani, A., Al-Jumaily, A. & Nguyen, H.T. 2008, 'A Hybrid Nonlinear-Discriminant Analysis Feature Projection Technique' in Wobcke, Wayne; Zhang, Mengjie (eds), Lecture Notes In Computer Science Vol 5360: AI 2008 Advances in Artificial Intelligence, Springer, Germany, pp. 544-550.
View/Download from: UTSePress | Publisher's site
View description>>

Feature set dimensionality reduction via Discriminant Analysis (DA) is one of the most sought after approaches in many applications. In this paper, a novel nonlinear DA technique is presented based on a hybrid of Artificial Neural Networks (ANN) and the Uncorrelated Linear Discriminant Analysis (ULDA). Although dimensionality reduction via ULDA can present a set of statistically uncorrelated features, but similar to the existing DA+s it assumes that the original data set is linearly separable, which is not the case with most real world problems. In order to overcome this problem, a one layer feed-forward ANN trained with a Differential Evolution (DE) optimization technique is combined with ULDA to implement a nonlinear feature projection technique. This combination acts as nonlinear discriminant analysis. The proposed approach is validated on a Brain Computer Interface (BCI) problem and compared with other techniques.

Kulatunga, A.K., Skinner, B., Liu, D. & Nguyen, H.T. 2007, 'Distributed simultaneous task allocation and motion coordination of autonomous vehicles using a parallel computing cluster' in Tzyh-Jong Tarn, Shan-Ben Chen, Changjiu Zhou (eds), Robotic Welding, Intelligence and Automation, Springer, Heidelberg, pp. 409-420.
View/Download from: UTSePress | Publisher's site
View description>>

Task allocation and motion coordination are the main factors that should be consi-dered in the coordination of multiple autonomous vehicles in material handling systems. Presently, these factors are handled in different stages, leading to a reduction in optimality and efficiency of the overall coordination. However, if these issues are solved simultaneously we can gain near optimal results. But, the simultaneous approach contains additional algorithmic complexities which increase computation time in the simulation environment. This work aims to reduce the computation time by adopting a parallel and distributed computation strategy for Simultaneous Task Allocation and Motion Coordination (STAMC). In the simulation experiments, each cluster node executes the motion coordination algorithm for each autonomous vehicle. This arrangement enables parallel computation of the expensive STAMC algorithm. Parallel and distributed computation is performed directly within the interpretive MATLAB environment. Results show the parallel and distributed approach provides sub-linear speedup compared to a single centralised computing node.

Craig, A.R., Tran, Y.H., Wijesuriya, N. & Nguyen, H.T. 2012, 'Regional brain wave activity changes associated with fatigue', Psychophysiology, vol. 49, pp. 574-582.
View/Download from: UTSePress | Publisher's site
View description>>

Assessing brain wave activity is a viable strategy for monitoring fatigue when performing tasks such as driving, and numerous studies have been conducted in this area. However, results of a systematic review on changes in brain wave activity associated with fatigue have revealed equivocal findings. This study investigated brain wave activity associated with fatigue in 48 nonprofessional healthy drivers as they participated in a simulated driving task until they fatigued. The results showed that as a person fatigues, slow wave activity increased over the entire cortex, in theta and in alpha 1 and 2 bands, while no significant changes were found in delta wave activity. Substantial increases also occurred in fast wave activity, though mostly in frontal sites. The results suggest that as a person fatigues, the brain loses capacity and slows its activity, and that attempts to maintain vigilance levels lead to increased beta activity.

Ling, S.S. & Nguyen, H.T. 2012, 'Natural occurrence of nocturnal hypoglycemia detection using hybrid particle swarm optimized fuzzy reasoning model', Artificial Intelligence in Medicine, vol. 55, no. 3, pp. 177-184.
View/Download from: Publisher's site
View description>>

Introduction: Low blood glucose (hypoglycemia) is a common and serious side effect of insulin therapy in patients with diabetes. This paper will make a contribution to knowledge in the modeling and design of a non-invasive hypoglycemia monitor for patients with type 1 diabetes mellitus (T1DM) using a fuzzy-reasoning system. Methods: Based on the heart rate and the corrected QT interval of the electrocardiogram (ECG) signal, we have developed a hybrid particle-swarm-optimization-based fuzzy-reasoning model to recognize the presence of hypoglycemic episodes. To optimize the fuzzy rules and the fuzzy-membership functions, a hybrid particle-swarm-optimization with wavelet mutation operation is investigated. Conclusion: We have investigated the detection for the natural occurrence of nocturnal hypoglycemic episodes in T1DM using a hybrid particle-swarm-optimization-based fuzzy-reasoning model with physiological parameters. In this study, no restricted environment (e.g. patient's dietary requirements) is required. Furthermore, the sampling time is between 5 and 10 min. To conclude, we have shown that the testing performances of the proposed algorithm for detection of advanced hypoglycemic and hypoglycemic episodes for T1DM patients are satisfactory.

Ling, S.S., San, P., Nguyen, H.T. & Leung, F.H. 2012, 'Non-invasive nocturnal hypoglycemia detection for insulin-dependent diabetes mellitus using genetic fuzzy logic method', International Journal of Computational Intelligence and Applications, vol. 11, no. 4, pp. 1250025-1-1250025-17.
View/Download from: Publisher's site
View description>>

Hypoglycemia, or low blood glucose, is the most common complication experienced by Type 1 diabetes mellitus (T1DM) patients. It is dangerous and can result in unconsciousness, seizures and even death. The most common physiological parameter to be e-«ected from hypoglycemic reaction are heart rate (HR) and correct QT interval (QTc) of the electrocardiogram (ECG) signal. Based on physiological parameters, a genetic algorithm based fuzzy reasoning model is developed to recognize the presence of hypoglycemia. To optimize the parameters of the fuzzy model in the membership functions and fuzzy rules, a genetic algorithm is used. A validation strategy based adjustable fitness is introduced in order to prevent the phenomenon of overtraining (over-»tting). For this study, 15 children with 569 sampling data points with Type 1 diabetes volunteered for an overnight study. The e-«ectiveness of the proposed algorithm is found to be satisfactory by giving better sensitivity and speci-»city compared with other existing methods for hypoglycemia detection.

Nuryani, N., Ling, S.S. & Nguyen, H.T. 2012, 'Electrocardiographic signals and swarm-based support vector machine for hypoglycemia detection', Annals Of Biomedical Engineering, vol. 40, no. 4, pp. 934-945.
View/Download from: UTSePress | Publisher's site
View description>>

Cardiac arrhythmia relating to hypoglycemia is suggested as a cause of death in diabetic patients. This article introduces electrocardiographic (ECG) parameters for artificially induced hypoglycemia detection. In addition, a hybrid technique of swarm-based support vector machine (SVM) is introduced for hypoglycemia detection using the ECG parameters as inputs. In this technique, a particle swarm optimization (PSO) is proposed to optimize the SVM to detect hypoglycemia. In an experiment using medical data of patients with Type 1 diabetes, the introduced ECG parameters show significant contributions to the performance of the hypoglycemia detection and the proposed detection technique performs well in terms of sensitivity and specificity.

Su, S.W., Anderson, B., Chen, W. & Nguyen, H.T. 2012, 'Multi-realization Of Nonlinear Systems', Automatica, vol. 48, no. 7, pp. 1455-1461.
View/Download from: Publisher's site
View description>>

The system multi-realization problem is to find a state-variable realization for a set of systems, sharing as many parameters as possible. A multi-realization can be used to efficiently implement a multi-controller architecture for multiple model adaptive control. We extend the linear multi-realization problem to nonlinear systems. The problem of minimal multi-realization of a set of MIMO systems is introduced and solved for static feedback linearizable systems.

Wong, M., He, X.S., Nguyen, H.T. & Yeh, W. 2012, 'Mass Classification in Digitized Mammograms Using Texture Features and Artificial Neural Network', Lecture Notes in Computer Science, vol. 7667, pp. 151-158.
View/Download from: Publisher's site
View description>>

A technique is proposed to classify regions of interests (ROIs) of digitized mammograms into mass and non-mass regions using texture features and artificial neural network (ANN). Fifty ROIs were extracted from the MIAS MiniMammographic Database, with 25 ROIs containing masses and 25 ROIs containing normal breast tissue only. Twelve texture features were derived from the gray level co-occurrence matrix (GLCM) of each region. The sequential forward selection technique was used to select four significant features from the twelve features. These significant features were used in the ANN to classify the ROI into either mass or non-mass region. By using leave-one-out method on the 50 images using the four significant features, classification accuracy of 86% was achieved for ANN. The test result using the four significant features is better than the full set of twelve features. The proposed method is compared with some existing works and promising results are obtained

Yuwono, M., Moulton, B.D., Su, S.W., Celler, B.G. & Nguyen, H.T. 2012, 'Unsupervised machine-learning method for improving the performance of ambulatory fall-detection systems', Biomedical Engineering Online, vol. 11, p. art9.
View/Download from: UTSePress | Publisher's site
View description>>

Background: Falls can cause trauma, disability and death among older people. Ambulatory accelerometer devices are currently capable of detecting falls in a controlled environment. However, research suggests that most current approaches can tend to have insufficient sensitivity and specificity in non-laboratory environments, in part because impacts can be experienced as part of ordinary daily living activities. Method: We used a waist-worn wireless tri-axial accelerometer combined with digital signal processing, clustering and neural network classifiers. The method includes the application of Discrete Wavelet Transform, Regrouping Particle Swarm Optimization, Gaussian Distribution of Clustered Knowledge and an ensemble of classifiers including a multilayer perceptron and Augmented Radial Basis Function (ARBF) neural networks. Results: Preliminary testing with 8 healthy individuals in a home environment yields 98.6% sensitivity to falls and 99.6% specificity for routine Activities of Daily Living (ADL) data. Single ARB and MLP classifiers were compared with a combined classifier. The combined classifier offers the greatest sensitivity, with a slight reduction in specificity for routine ADL and an increased specificity for exercise activities. In preliminary tests, the approach achieves 100% sensitivity on in-group falls, 97.65% on out-group falls, 99.33% specificity on routine ADL, and 96.59% specificity on exercise ADL. Conclusion: The pre-processing and feature-extraction steps appear to simplify the signal while successfully extracting the essential features that are required to characterize a fall. The results suggest this combination of classifiers can perform better than MLP alone. Preliminary testing suggests these methods may be useful for researchers who are attempting to improve the performance of ambulatory fall detection systems.

Chan, K., Ling, S.S., Dillon, T.S. & Nguyen, H.T. 2011, 'Diagnosis Of Hypoglycemic Episodes Using A Neural Network Based Rule Discovery System', Expert Systems With Applications, vol. 38, no. 8, pp. 9799-9808.
View/Download from: UTSePress | Publisher's site
View description>>

Hypoglycemia or low blood glucose is dangerous and can result in unconsciousness, seizures and even death for Type 1 diabetes mellitus (T1DM) patients. Based on the T1DM patients' physiological parameters, corrected QT interval of the electrocardiogram (ECG) signal, change of heart rate, and the change of corrected QT interval, we have developed a neural network based rule discovery system with hybridizing the approaches of neural networks and genetic algorithm to identify the presences of hypoglycemic episodes for TIDM patients. The proposed neural network based rule discovery system is built and is validated by using the real T1DM patients' data sets collected from Department of Health, Government of Western Australia. Experimental results show that the proposed neural network based rule discovery system can achieve more accurate results on both trained and unseen T1DM patients' data sets compared with those developed based on the commonly used classification methods for medical diagnosis, statistical regression, fuzzy regression and genetic programming. Apart from the achievement of these better results, the proposed neural network based rule discovery system can provide explicit information in the form of production rules which compensate for the deficiency of traditional neural network method which do not provide a clear understanding of how they work in prediction as they are in an implicit black-box structure. This explicit information provided by the product rules can convince medical doctors to use the neural networks to perform diagnosis of hypoglycemia on T1DM patients.

Ling, S.S. & Nguyen, H.T. 2011, 'Genetic algorithm based multiple regression with fuzzy inference system for detection of nocturnal hypoglycemic episodes', IEEE Transactions on Information Technology in Biomedicine, vol. 15, no. 2, pp. 308-315.
View/Download from: Publisher's site
View description>>

Hypoglycemia or low blood glucose is dangerous and can result in unconsciousness, seizures, and even death. It is a common and serious side effect of insulin therapy in patients with diabetes. Hypoglycemic monitor is a noninvasivemonitor that measures some physiological parameters continuously to provide detection of hypoglycemic episodes in type 1 diabetes mellitus patients (T1DM). Based on heart rate (HR), corrected QT interval of the ECG signal, change of HR, and the change of corrected QT interval, we develop a genetic algorithm (GA)-based multiple regression with fuzzy inference system (FIS) to classify the presence of hypoglycemic episodes. GA is used to find the optimal fuzzy rules and membership functions of FIS and the model parameters of regression method. From a clinical study of 16 children with T1DM, natural occurrence of nocturnal hypoglycemic episodes is associated with HRs and corrected QT intervals. The overall data were organized into a training set (eight patients) and a testing set (another eight patients) randomly selected. The results show that the proposed algorithm performs a good sensitivity with an acceptable specificity.

Ling, S.S., Jiang, F., Nguyen, H.T. & Chan, K.Y. 2011, 'Hybrid Fuzzy Logic-Based Particle Swarm Optimization For Flow Shop Scheduling Problem', International Journal of Computational Intelligence and Applications, vol. 10, no. 3, pp. 335-356.
View/Download from: UTSePress | Publisher's site
View description>>

This paper, proposes a hybrid fuzzy logic-based particle swarm optimization (PSO) with cross-mutated operation method for the minimization of makespan in permutation flow shop scheduling problem. This problem is a typical non-deterministic polynomial-time (NP) hard combinatorial optimization problem. In the proposed hybrid PSO, fuzzy inference system is applied to determine the inertia weight of PSO and the control parameter of the proposed cross-mutated operation by using human knowledge. By introducing the fuzzy system, the inertia weight becomes adaptive. The cross-mutated operation effectively forces the solution to escape the local optimum. To make PSO suitable for solving flow shop scheduling problem, a sequence-order system based on the roulette wheel mechanism is proposed to convert the continuous position values of particles to job permutations. Meanwhile, a new local search technique namely swap-based local search for scheduling problem is designed and incorporated into the hybrid PSO. Finally, a suite of flow shop benchmark functions are employed to evaluate the performance of the proposed PSO for flow shop scheduling problems. Experimental results show empirically that the proposed method outperforms the existing hybrid PSO methods significantly.

Nguyen, T., Su, S.W. & Nguyen, H.T. 2011, 'Robust neuro-sliding mode multivariable control strategy for powered wheelchairs', IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 19, no. 1, pp. 105-111.
View/Download from: Publisher's site
View description>>

This paper proposes an advanced robust multivariable control strategy for a powered wheelchair system. The new control strategy is based on a combination of the systematic triangularization technique and the robust neuro-sliding mode control approach. This strategy effectively copes with parameter uncertainties and external disturbances in real-time in order to achieve robustness and optimal performance of a multivariable system. This novel strategy reduces coupling effects on a multivariable system, eliminates chattering phenomena, and avoids the plant Jacobian calculation problem. Furthermore, the strategy can also achieve fast and global convergence using less computation. The effectiveness of the new multivariable control strategy is verified in real-time implementation on a powered wheelchair system. The obtained results confirm that robustness and desired performance of the overall system are guaranteed, even under parameter uncertainty and external disturbance effects.

Tran, C., Hoang, T.D., Ha, Q.P. & Nguyen, H.T. 2011, 'Stabilising agent design for the control of interconnected systems', International Journal of Control, vol. 84, no. 6, pp. 1140-1156.
View/Download from: UTSePress | Publisher's site
View description>>

This article presents a new control design strategy for stabilising large-scale interconnected systems operating in semi-automatic control modes. The large-scale system is modelled by subsystems connected to each other in an arbitrary configuration. Each subsystem is regulated by a dedicated multivariable controller that also allows for a manual control mode. The notion of asymptotically positive realness constraint (APRC) is introduced and applied for deriving the interconnection stabilisability condition in the time domain. The interactions between subsystems are taken into consideration in the stability condition. The APRC is subsequently employed in the so-called stabilising agent to accommodate the closed-loop control and man-in-the-loop coexistence. The multipliers of the APRC quadratic supply rate are updated on-the-fly to ensure that the constraint satisfaction of stabilising agents is recursively feasible. The stabilising agents are developed independently from the control law under the same auspice controller. Due to this independence, operational errors from the manual control adjustments, that may destabilise the control systems, can be avoided. The decentralised agents render stabilising bounds for the manipulated variables in the automatic control mode, and at the same time, provide warning signals and manipulation guidance for the operators to prevent possible plant-wide destabilisation in the manual control mode. Our main results are illustrated through numerical simulations for an industrial modular system.

Boord, P.R., Craig, A.R., Tran, Y.H. & Nguyen, H.T. 2010, 'Discrimination of left and right leg motor imagery for brain-computer interfaces', Medical & Biological Engineering & Computing, vol. 48, no. 4, pp. 343-350.
View/Download from: UTSePress | Publisher's site
View description>>

This article reports on a study to identify electroencephalography (EEG) signals with potential to provide new BCI channels through mental motor imagery (MMI). Leg motion was assessed to see if left and right leg MMI could be discriminated in the EEG. The study also explored simultaneous observation of leg movement as a means to enhance MMI evoked EEG signals. The results demonstrate that MMI of the left and right leg produce a contralateral preponderance of EEG alpha band desynchronization, which can be spatially discriminated. This suggests that lower extremity MMI could provide signals for additional BCI channels. The study also shows that movement imitation enhances alpha band desynchronization during MMI, and might provide a useful aid in the identification and training of BCI signals.

Su, S.W., Chen, W., Liu, D., Fang, Y., Kuang, W., Yu, X., Guo, T., Celler, B.G. & Nguyen, H.T. 2010, 'Dynamic Modelling of Heart Rate Response Under Different Exercise Intensity', The Open Medical Informatics Journal, vol. 4, pp. 81-85.
View/Download from: UTSePress | Publisher's site
View description>>

Heart rate is one of the major indications of human cardiovascular response to exercises. This study investigates human heart rate response dynamics to moderate exercise. A healthy male subject has been asked to walk on a motorised treadmill under a predefined exercise protocol. ECG, body movements, and oxygen saturation (SpO2) have been reliably monitored and recorded by using non-invasive portable sensors. To reduce heart rate variation caused by the influence of various internal or external factors, the designed step response protocol has been repeated three times. Experimental results show that both steady state gain and time constant of heart rate response are not invariant when walking speed is faster than 3 miles/hour, and time constant of offset exercise is noticeably longer than that of onset exercise.

Nguyen, H.T. & Su, S.W. 2009, 'Conditions for triangular decoupling control', International Journal of Control, vol. 82, no. 9, pp. 1-7.
View/Download from: UTSePress | Publisher's site
View description>>

The main purpose of this article is to explore the relationship of two existing conditions for the triangular decoupling problem. The first one is the triangular-diagonal-dominance condition proposed by Hung and Anderson. The second one is the stable coprime factorisation-described condition proposed by Gomez and Goodwin, which has been proven as a necessary and sufficient condition for the triangular decoupling problem. This article proves that the two conditions are actually equivalent. It also provides easy-to-use criteria for assessment of the solvability of the triangular decoupling problem.

Su, S.W., Celler, B.G., Savkin, A.V., Nguyen, H.T., Cheng, T.M., Guo, Y. & Wang, L. 2009, 'Transient and steady state estimation of human oxygen uptake based on noninvasive portable sensor measurements', Medical & Biological Engineering & Computing, vol. 47, no. 10, pp. 1111-1117.
View/Download from: UTSePress | Publisher's site
View description>>

The main motivation of this study is to establish an ambulatory cardio-respiratory analysis system for the monitoring and evaluation of exercise and regular daily physical activity. We explored the estimation of oxygen uptake by using noninvasive portable sensors. These sensors are easy to use but may suffer from malfunctions under free living environments. A promising solution is to combine sensors with different measuring mechanisms to improve both reliability and accuracy of the estimation results. For this purpose, we selected a wireless heart rate sensor and a tri-axial accelerometer to form a complementary sensor platform. We analyzed the relationship between oxygen uptake measured by gas analysis and data collected from the simple portable sensors using multivariable nonlinear modeling approaches. It was observed that the resulting nonlinear multivariable model could not only achieve a better estimate compared with single input single output models, but also had greater potential to improve reliability.

Thornton-benko, E., Nguyen, H.T., Hung, W. & Thornton, B.S. 2009, 'Improved observer dependent perception of weak edges when scanning an image in real time indicated by introducing 1/f noise into the primary visual cortex V1. Theory and experimental support', Australian Physical And Engineering Sciences in Me..., vol. 32, no. 3, pp. 136-149.
View/Download from: UTSePress | Publisher's site
View description>>

We present results of a new process for generating 1/f type noise sequences and introducing the noise in the primary visual cortex which then enables improved perception of weak edges when an observer is scanning a complex image in real time to detect detail such as in mammogram reading sessions. It can be explained by an adaptation of information theory for functional rather than previous task-based methods for formulating processes for edge formation in early vision. This is enabled from a two "species" classification of the interaction of opposing on-centre and off-centre neuron processes. We show that non-stationary stochastic resonances predicted by theory can occur with 1/f noise in the primary visual cortex VI and suggest that signalling exchanges between VI and the lateral geniculate nucleus (LGN) of the thalamus can initiate neural activity for saccadic action (and observer attention) for weak edge perception. Improvements predicted by our theory were shown from 600 observations by two groups of observers of limited experience and an experienced radiologist for reference (but not for diagnosis). They scanned and rated the definition of microcalcification in clusters separately rated by the experienced radiologist. The results and supporting theory showed dependence on the observer's attention and orderly scanning. Using a compact simplified equipment configuration the methodology has important clinical applications for conjunction searches of features and for detection of objects in poor light conditions for vehicles.

Du, H., Zhang, N. & Nguyen, H.T. 2008, 'Mixed H-2/H-infinity control of tall buildings with reduced-order modelling technique', Structural Control & Health Monitoring, vol. 15, no. 1, pp. 64-89.
View/Download from: UTSePress | Publisher's site
View description>>

In this paper, a reduced-order technique based on the dynamic condensation method is applied to obtain a reduced-order model of an experimental tall building which has 20 floors and is 2.5 m high. The experimental model is designed to imitate a practical

Nguyen, H.T. 2008, 'Intelligent Technologies for Real-time Biomedical Engineering Applications', International Journal of Automation and Control, vol. 1, no. 2/3, pp. 274-285.
View/Download from: UTSePress | Publisher's site
View description>>

Intelligent technologies are essential for many biomedical engineering applications in order to cope with a wide variety of patient conditions or user disability. The development of advanced optimisation training algorithms such as adaptive optimal Bayesian neural networks is particularly useful when only limited training data are available. Two specific biomedical engineering applications will be presented. The first application concerns the development of a non-invasive monitor for real-time detection of hypoglycaemic episodes in Type 1 diabetes mellitus patients (T1DM). The second application relates to the development of real-time hands-free wheelchair control systems using head movement to provide mobility independence for severely disabled people.

Nguyen, H.T., Nguyen Thanh, S., Taylor, P.W. & Middleton, J. 2007, 'Head Direction Command Classification using an Adaptive Optimal Bayesian Neural Network', International Journal of Factory Automation, Robotics and Soft Computing, vol. 1, no. 3, pp. 98-103.
View/Download from: UTSePress
View description>>

Mobility has become very important for our quality of life. Head movement is a natural form of pointing and can be used to directly replace the joystick for severely disabled people. In this paper, we describe the development of an optimal Bayesian neural network for the classification of head direction commands in a hands-free wheelchair control system as it allows strong generalisation during the training phase and does not require a validation data set. Experimental results show that with limited training data, an adaptive optimal Bayesian neural network can be developed to classify head direction commands by disabled users with a high sensitivity and specificity of 93.75% and 97.92% respectively.

Tran, T., Ha, Q.P. & Nguyen, H.T. 2007, 'Robust Non-Overshoot Time Responses using Cascade Sliding Mode-PID Control', Journal of Advanced Computational Intelligence and Intelligent Informatics, vol. 11, no. 10, pp. 1224-1231.
View/Download from: UTSePress
View description>>

Overshoot is a serious problem in automatic control systems. This paper presents a new method for elimination of the step response overshoot in a conventional PID-controlled system and enhancement of its robustness by cascading a sliding mode controller in the outer loop. The idea is first to use the cascade control principle to model the under-damped system under PID control with a second-order system. Then, by making use of the sliding mode control outer loop, a robust, reduced-order response can be obtained to suppress the control overshoot. The proposed approach can also deal with time delay systems. Its validity is verified through simulation for some dynamic systems subject to highly nonlinear uncertainties, where overshoot remains an issue.

Smith, P.J., Vigneswaran, S., Ngo, H.H., Nguyen, H.T. & Aim, R.B. 2006, 'A New Approach to backwash initiation in Membrance Systems', Journal of Membrane Science, vol. 278, no. 1-2, pp. 381-389.
View/Download from: UTSePress | Publisher's site

Smith, P.J., Vigneswaran, S., Ngo, H.H., Nguyen, H.T. & Ben-Aim, R. 2006, 'Application of an automation system and a supervisory control and data acquisition (SCADA) system for the optimal operation of a membrane adsorption hybrid system', Water Science And Technology, vol. 53, no. 4-5, pp. 179-184.
View/Download from: UTSePress
View description>>

The application of automation and supervisory control and data acquisition (SCADA) systems to municipal water and wastewater treatment plants is rapidly increasing. However, the application of these systems is less frequent in the research and developmen

Smith, P.J., Shon, H., Vigneswaran, S., Ngo, H.H. & Nguyen, H.T. 2006, 'Productivity enhancement in a cross-flow ultrafiltration membrane system through automated de-clogging operations', Journal Of Membrane Science, vol. 280, no. 1-2, pp. 82-88.
View/Download from: UTSePress | Publisher's site
View description>>

A membrane system only has a limited operational lifetime, whereby it becomes so severely fouled that continued operation must be stopped. In the cross-flow configuration of membrane filtration of wastewater, both increased cross-flow velocities and decr

Smith, P.J., Vigneswaran, S., Ngo, H.H., Ben-Aim, R. & Nguyen, H.T. 2005, 'Design of a generic control system for optimising back flush durations in a submerged membrane hybrid reactor', Journal Of Membrane Science, vol. 255, no. 38749, pp. 99-106.
View/Download from: UTSePress | Publisher's site
View description>>

Organic fouling on the membrane can be minimised through powdered activated carbon (PAC) usage in the submerged membrane reactor to adsorb dissolved organic matter and reduce direct organic loading on the membrane. However, fouling cannot be totally alle

Boord, P.R., Barriskill, A.B., Craig, A.R. & Nguyen, H.T. 2004, 'Brain-Computer Interface - FES Integration: Towards a Hands-free Neuroprosthesis Command System', Neuromodulation, vol. 7, no. 4, pp. 267-276.
View/Download from: UTSePress | Publisher's site
View description>>

This paper presents a critical review of brain-computer interfaces (BCIs) and their potential for neuroprosthetic applications. Summaries are provided for the command interface requirements of hand grasp, multijoint, and lower extremity neuroprostheses,and the characteristics of various BCIs are discussed in relation to these requirements. The current limitations of BCIs and areas of research that need to be addressed to enhance BCI - FES inetgration.

Smith, P.J., Vigneswaran, S., Ngo, H.H., Ben-Aim, R. & Nguyen, H.T. 2004, 'Investigation of Membrane De-Clogging Techniques in the Submerged Filtration Absorption Hybrid System (SMFAHS)', Fluid/Particle Separation Journal, vol. 16, no. 2, pp. 165-173.
View/Download from: UTSePress

Chai, R., Ling, S.S., Hunter, G. & Nguyen, H.T. 2012, 'Mental non-motor imagery tasks classifications of brain computer interface for wheelchair commands using genetic algorithm-based neural network', 2012 IEEE World Congress on Computational Intelligence (WCCI 2012) / 2012 International Joint Conference on Neural Networks (IJCNN 2012), Brisbane, Australia, June 2012 in Proceedings of the 2012 International Joint Conference on Neural Networks (IJCNN 2012), ed Hussein A. Abbass, IEEE, USA, pp. 978-984.
View/Download from: Publisher's site
View description>>

A genetic algorithm (GA)-based neural network classification in the application of brain computer interface (BCI) for controlling a wheelchair is presented in this paper. This study uses an electroencephalography (EEG) as a non-invasive BCI approach to discriminate three non-motor imagery mental tasks for disabled individuals who may have difficulty in using BCI based motor imagery tasks. The three tasks classification is mapped into three wheelchair movements: left, right and forward and the relevant combination mental tasks used in this study are mental arithmetic, letter composing, Rubik's cube rolling, visual counting, ringtone imagery and spatial navigation. The results show the proposed system provides good classification performance after selecting the most effective of three discriminative tasks across combination of the different non-motor imagery mental tasks for the five subjects tested. The average classification accuracy is between 76% and 85 %, with information transfer rates varies from 0.5 to 0.8 bits per trial.

Chai, R., Ling, S.S., Hunter, G. & Nguyen, H.T. 2012, 'Mental Task Classifications Using Prefrontal Cortex Electroencephalograph Signals', the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2012), San Diego, CA, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2012), ed Michael C.K. Khoo, IEEE, USA, pp. 1831-1834.
View description>>

For an electroencephalograph (EEG)-based brain computer interface (BCI) application, the use of gel on the hair area of the scalp is needed for low impedance electrical contact. This causes the set up procedure to be time consuming and inconvenient for a practical BCI system. Moreover, studies of other cortical areas are useful for BCI development. As a more convenient alternative, this paper presents the EEG based-BCI using the prefrontal cortex non-hair area to classify mental tasks at three electrodes position: Fp1, Fpz and Fp2. The relevant mental tasks used are mental arithmetic, ringtone, finger tapping and words composition with additional tasks which are baseline and eyes closed. The feature extraction is based on the Hilbert Huang Transform (HHT) energy method and the classification algorithm is based on an artificial neural network (ANN) with genetic algorithm (GA) optimization. The results show that the dominant alpha wave during eyes closed can still clearly be detected in the prefrontal cortex. The classification accuracy for five subjects, mental tasks vs. baseline task resulted in average accuracy is 73% and the average accuracy for pairs of mental task combinations is 72%.

Chai, R., Hunter, G., Ling, S.S. & Nguyen, H.T. 2012, 'Real-Time Microcontroller based Brain Computer Interface for Mental Task Classifications using Wireless EEG Signals from Two Channels', The Ninth IASTED International Conference on Biomedical Engineering, Innsbruck, Austria, February 2012 in Proceedings of the Ninth IASTED International Conference on Biomedical Engineering, ed C. Hellmich, ACTA Press, Canada, pp. 336-342.
View/Download from: Publisher's site
View description>>

A brain computer interface (BCI) using electroencephalography (EEG) to measure brain activities could provide severely disabled people with alternative means of control and communication. In a practical system, portability, low power and real-time operation are the keys requirements. This could be accomplished by using an embedded microcontroller based system. The main contribution of this paper shows the development of a real-time BCI prototype system to classify groups of mental tasks based on such a system. The relevant mental tasks used are mental arithmetic, figure rotation, letter composing, visual counting and eyes closed action. Moreover, the system uses a separate two channels only wireless EEG measurement module with the active positions at parietal and occipital lobes. The result shows the wireless EEG module has a good performance with a CMRR of more than 95dB. In addition, the size of the module is small (36x36 mm2) and current consumption is low enough to operate off a 3V coin cell battery. The mental tasks were classified using a feed-forward back-propagation artificial neural network (ANN) trained with the Levenberg-Marquardt algorithm. An accuracy of around 70% was achieved with bit rate at around 0.4 bits/trial for six subjects tested to select between three separate mental tasks.

Chai, R., Ling, S.S., Hunter, G. & Nguyen, H.T. 2012, 'Toward Fewer EEG Channels and Better Feature Extractor of Non-Motor Imagery Mental Tasks Classification for a Wheelchair Thought Controller', 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2012), San Diego, CA, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2012), ed Michael C.K. Khoo, IEEE, USA, pp. 5266-5269.
View description>>

This paper presents a non-motor imagery tasks classification electroencephalography (EEG) based brain computer interface (BCI) for wheelchair control. It uses only two EEG channels and a better feature extractor to improve the portability and accuracy in the practical system. In addition, two different features extraction methods, power spectral density (PSD) and Hilbert Huang Transform (HHT) energy are compared to find a better method with improved classification accuracy using a Genetic Algorithm (GA) based neural network classifier. The results from five subjects show that using the original eight channels with three tasks, accuracy between 76% and 85% is achieved. With only two channels in combination with the best chosen task using a PSD feature extractor, the accuracy is reduced to between 65% and 79%. However, the HHT based method provides an improved accuracy between 70% and 84% for the classification of three discriminative tasks using two EEG channels.

Chan, K.Y., Ling, S.S., Nguyen, H.T. & Jiang, F. 2012, 'A hypoglycemic episodes diagnosis system based on neural networks for Type 1 diabetes mellitus', CEC 2012, Australia, June 2012 in IEEE Congress on Evolutionary Computation, ed Hussein A. Abbass, IEEE, US, pp. 2046-2051.
View/Download from: Publisher's site
View description>>

Hypoglycemia (or low blood glucose) is dangerous for Type 1 diabetes mellitus (T1DM) patients, as this can cause unconsciousness or even death. However, it is impossible to monitor the hypoglycemia by measuring patients+ blood glucose levels all the time, especially at night. In this paper, a hypoglycemic episode diagnosis system is proposed to determine T1DM patients+ blood glucose levels based on these patients+ physiological parameters which can be measured online. It can be used not only to diagnose hypoglycemic episodes in T1DM patients, but also to generate a set of rules, which describe the domains of physiological parameters that lead to hypoglycemic episodes. The hypoglycemic episode diagnosis system addresses the limitations of the traditional neural network approaches which cannot generate implicit information. The performance of the proposed hypoglycemic episode diagnosis system is evaluated by using real T1DM patients+ data sets collected from the Department of Health, Government of Western Australia, Australia. Results show that satisfactory diagnosis accuracy can be obtained. Also, explicit knowledge can be produced such that the deficiency of traditional neural networks can be overcome. A clear understanding of how they perform diagnosis can be indicated.

Handojoseno, A.M., Shine, J.M., Nguyen, T., Tran, Y.H., Lewis, S. & Nguyen, H.T. 2012, 'The detection of freezing of gait in Parkinson's Disease patients using EEG signals based on wavelet decomposition', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 69-72.
View/Download from: Publisher's site
View description>>

Freezing of Gait (FOG) is one of the most disabling gait disturbances of ParkinsonÔ++s disease (PD). The experience has often been described as Ô+ feeling like their feet have been glued to the floor while trying to walkÔ+ and as such it is a common cause of falling in PD patients. In this paper, EEG subbands Wavelet Energy and Total Wavelet Entropy were extracted using the multiresolution decomposition of EEG signal based on the Discrete Wavelet Transform and were used to analyze the dynamics in the EEG during freezing. The Back Propagation Neural Network classifier has the ability to identify the onset of freezing of PD patients during walking using these features with average values of accuracy, sensitivity and specificity are around 75 %. This results have proved the feasibility of utilized EEG in future treatment of FOG.

Ling, S.S., Nuryani, N. & Nguyen, H.T. 2012, 'Hybrid Particle Swarm - based Fuzzy Support Vector Machine for Hypoglycaemia Detection', Australia, June 2012 in IEEE International Conference on Fuzzy Systems, ed Hussein A. Abbass, IEEE, US, pp. 450-455.
View/Download from: Publisher's site
View description>>

Severe hypoglycemia is potentially life-threatening. This article introduces a novel hypoglycemia detection strategy using a hybrid particle swarm - based fuzzy support vector machine (SFisSvm) technique. The inputs of this system are six electrocardiographic (ECG) parameters. The system parameters of SFisSvm are optimized using a particle swarm optimization method. The proposed hypoglycemia detector system is a combination of two subsystems, namely, fuzzy inference system (FIS) and support vector machine (SVM). Two most significant inputs, heart rate and RTpc are fed to FIS, and its output is used for input of the SVM. The other ECG parameters and the output of FIS are fed to SVM and, then, are classified to indicate the presence of hypoglycemia. In this study, three and five membership functions are investigated for FIS. Furthermore, radial basis function (RBF), sigmoid and linear kernel functions are employed for mapping the inputs to high dimensional space in SVM. Performances of SFisSvm with different kernel functions are compared. As conclusion, the performance of SFisSvm is found with 75.19%, 83.71% and 79.33% in terms of sensitivity, specificity and geometric mean.

Ling, S.S., Nguyen, H.T., Leung, F.H., Chan, K.Y. & Jiang, F. 2012, 'Intelligent fuzzy particle swarm optimization with cross-mutated operation', IEEE Congress on Evolutionary Computation (CEC), Australia, June 2012 in IEEE Congress on Evolutionary Computation, ed Hussein A. Abbass, IEEE, US, pp. 3009-3016.
View/Download from: Publisher's site
View description>>

This paper presents a novel fuzzy particle swarm optimization with cross-mutated operation (FPSOCM), where a fuzzy logic is applied to determine the inertia weight of PSO and the control parameter of the proposed cross-mutated operation based on human knowledge. By introducing the fuzzy system, the value of the inertia weight of PSO becomes adaptive. The new cross-mutated operation effectively drives the solution to escape from local optima. To illustrate the performance of the FPSOCM, a suite of benchmark test functions are employed. Experimental results show the proposed FPSOCM method performs better than some existing hybrid PSO methods in terms of solution quality and solution reliability (standard deviation upon many trials). Moreover, an industrial application of economic load dispatch is given to show that the FPSOCM method performs statistically more significant than the existing hybrid PSO methods

Nguyen, V., Nguyen, B.L., Su, S.W. & Nguyen, H.T. 2012, 'Development of a Bayesian neural network to perform obstacle avoidance for an intelligent wheelchair', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 1884-1887.
View/Download from: Publisher's site
View description>>

This paper presents an extension of a real-time obstacle avoidance algorithm for our laser-based intelligent wheelchair, to provide independent mobility for people with physical, cognitive, and/or perceptual impairments. The laser range finder URG-04LX mounted on the front of the wheelchair collects immediate environment information, and then the raw laser data are directly used to control the wheelchair in real-time without any modification. The central control role is an obstacle avoidance algorithm which is a neural network trained under supervision of Bayesian framework, to optimize its structure and weight values. The experiment results demonstrated that this new approach provides safety, smoothness for autonomous tasks and significantly improves the performance of the system in difficult tasks such as door passing.

Nguyen, B.L., Nguyen, V., Ling, S.S. & Nguyen, H.T. 2012, 'A particle swarm optimization-based neural network for detecting nocturnal hypoglycemia using electroencephalograph (EEG) signals', Brisbane, Australia, June 2012 in 2012 IEEE World Congress on Computational Intelligence, ed Hussein A. Abbass, IEEE, Piscataway, USA, pp. 2730-2735.
View/Download from: Publisher's site
View description>>

For patients with Type 1 Diabetes Mellitus (T1DM), hypoglycemia or the state of low blood glucose level is a very common but dangerous complication. Hypoglycemia episodes can lead to a large number of serious symptoms and effects, including unconsciousness, coma and even death. The variety of hypoglycemia symptoms is originated from the inadequate supply of glucose to the brain. By analyzing electroencephalography (EEG) signals from five T1DM patients during an overnight study, we find that under hypoglycemia, both centroid theta frequency and centroid alpha frequency change significantly against non-hypoglycemia conditions. Furthermore, a neural network is developed to detect hypoglycemia using the mentioned two EEG features. A standard particle swarm optimization strategy is applied to optimize the parameters of this neural network. By using the proposed method, we obtain the classification performance of 82% sensitivity and 63% specificity. The results demonstrate that hypoglycemia episodes can be detected non-invasively and effectively from EEG signals.

Nguyen, B.L., Nguyen, V., Ling, S.S. & Nguyen, H.T. 2012, 'An adaptive strategy of classification for detecting hypoglycemia using only two EEG channels', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 3515-3518.
View/Download from: Publisher's site
View description>>

Hypoglycemia is the most common but highly feared side effect of the insulin therapy for patients with Type 1 Diabetes Mellitus (T1DM). Severe episodes of hypoglycemia can lead to unconsciousness, coma, and even death. The variety of hypoglycemic symptoms arises from the activation of the autonomous central nervous system and from reduced cerebral glucose consumption. In this study, electroencephalography (EEG) signals from five T1DM patients during an overnight clamp study were measured and analyzed. By applying a method of feature extraction using Fast Fourier Transform (FFT) and classification using neural networks, we establish that hypoglycemia can be detected non-invasively using EEG signals from only two channels. This paper demonstrates that a significant advantage can be achieved by implementing adaptive training. By adapting the classifier to a previously unseen person, the classification results can be improved from 60% sensitivity and 54% specificity to 75% sensitivity and 67% specificity.

Nguyen, J., Nguyen, T., Tran, Y.H., Su, S.W., Craig, A.R. & Nguyen, H.T. 2012, 'Real-time performance of a hands-free semi-autonomous wheelchair system using a combination of stereoscopic and spherical vision', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 3069-3072.
View/Download from: Publisher's site
View description>>

This paper is concerned with the operational performance of a semi-autonomous wheelchair system named TIM (Thought-controlled Intelligent Machine), which uses cameras in a system configuration modeled on the vision system of a horse. This new camera configuration utilizes stereoscopic vision for 3-Dimensional (3D) depth perception and mapping ahead of the wheelchair, combined with a spherical camera system for 360-degrees of monocular vision. The unique combination allows for static components of an unknown environment to be mapped and any surrounding dynamic obstacles to be detected, during real-time autonomous navigation, minimizing blind-spots and preventing accidental collisions with people or obstacles. Combining this vision system with a shared control strategy provides intelligent assistive guidance during wheelchair navigation, and can accompany any hands-free wheelchair control technology for people with severe physical disability. Testing of this system in crowded dynamic environments has displayed the feasibility and real-time performance of this system when assisting handsfree control technologies, in this case being a proof-of-concept brain-computer interface (BCI).

Nguyen, L., Su, S.W. & Nguyen, H.T. 2012, 'Identification of hypoglycemia and hyperglycemia in Type 1 diabetic patients using ECG parameters', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 2716-2719.
View/Download from: Publisher's site
View description>>

Hypoglycemia and Hyperglycemia are both serious diseases related to diabetes mellitus. Among Type 1 Diabetic patients, there are who experience both hypoglycemic and hyperglycemic events. The aim of this study was to identify of hypoglycemia and hyperglycemia based on ECG changes in this population. An ECG Acquisition and Analysis System based on LabVIEW software has been developed for collecting ECG signals and extracting features with abnormal changes. ECG parameters included Heart rate (HR), corrected QT interval (QTc), PR interval, corrected RT interval (RTc) and corrected TpTe interval (TpTeC). The results indicated that hypoglycemic and hyperglycemic states produce significant inverse changes on those ECG parameters.

San, P., Ling, S.S. & Nguyen, H.T. 2012, 'Hybrid Particle Swarm Optimization Based Normalized Radial Basis Function Neural Network For Hypoglycemia Detection', WCCI 2012 IEEE World Congress on Computational Intelligence, Brisbane, Australia, June 2012 in International Joint Conference on Neural Networks, ed Hussein A. Abbass, IEEE, US, pp. 2718-2723.
View/Download from: Publisher's site
View description>>

In this study, a normalized radial basis function neural network (NRBFNN) is presented for detection of hypoglycemia episodes by using physiological parameters of electrocardiogram (ECG) signal. ypoglycemia is a common and serious side effect of insulin therapy in patients with Type 1 diabetes. Based on heart rate (HR) and corrected QT interval (QTc) of electrocardiogram (ECG) signal, a hybrid particle swarm optimization based normalized RBFNN is developed for recognization of hypoglycemia episodes. A global learning algorithm called hybrid particle swarm optimization with wavelet mutation (HPSOWM) is used to optimize the parameters of NRBFNN. From a clinical study of 15 children with Type 1 diabetes, natural occurrence of nocturnal hypoglycemic episodes associated with increased heart rates and corrected QT interval are studied. The overall data are organized into a training set (5 patients), validation set (5 patients) and testing set (5 patients) randomly selected. Using the optimized NRBFNN, the testing performance for detection of hypoglycemic episodes are satisfactory with 76.74% of sensitivity and 51.82% of specificity.

San, P., Ling, S.S. & Nguyen, H.T. 2012, 'Intelligent detection of hypoglycemic episodes in children with Type 1 diabetes using adaptive neural-fuzzy inference system', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 6325-6328.
View/Download from: Publisher's site
View description>>

Hypoglycemia, or low blood glucose, is the most common complication experienced by Type 1 diabetes mel- litus (T1DM) patients. It is dangerous and can result in unconsciousness, seizures and even death. The most common physiological parameter to be effected from hypoglycemic reaction are heart rate (HR) and correct QT interval (QTc) of the electrocardiogram (ECG) signal. Based on physiological parameters, an intelligent diagnostics system, using the hybrid approach of adaptive neural fuzzy inference system (ANFIS), is developed to recognize the presence of hypoglycemia. The proposed ANFIS is characterized by adaptive neural network capabilities and the fuzzy inference system. To optimize the membership functions and adaptive network parameters, a global learning optimization algorithm called hybrid particle swarm optimization with wavelet mutation (HPSOWM) is used. For clinical study, 15 children with Type 1 diabetes volunteered for an overnight study. All the real data sets are collected from the Department of Health, Government of Western Australia. Several experiments were conducted with 5 patients each, for a training set (184 data points), a validation set (192 data points) and a testing set (153 data points), which are randomly selected. The effectiveness of the proposed detection method is found to be satisfactory by giving better sensitivity, 79.09% and acceptable specificity, 51.82%.

San, P., Ling, S.S. & Nguyen, H.T. 2012, 'Optimized variable translation wavelet neural network and its application on hypoglycemia detection system', 7th IEEE Conference on Industrial Electronics and Applications, Singapore, July 2012 in 7th IEEE Conference on Industrial Electronics and Applications, ed Wenxiang XIE, IEEE, Singapore, pp. 547-551.
View/Download from: Publisher's site
View description>>

An hybrid particle swarm optimization based optimized variable translation wavelet neural network (VTWNN) is proposed for detection of hypoglycemic episodes in patients with Type 1 diabetes mellitus (T1DM). Due to excellent performance in capturing nonstationary signal and nonlinear function modeling of VTWNN, it is used as a suitable classifier in hypoglycemia detection system. A global training algorithm called hybrid particle swarm optimization with wavelet mutation (HPSOWM) operation is investigated for parameters optimization of proposed VTWNN detection system. In this clinical study, 15 children with Type 1 diabetes were observed overnight. All the real data sets collected from Department of Heath, Government of Western Australia. Several experiments are performed over a randomly selected training set 5 patients (184 data points), validation set 5 patients (192 data points) and testing set 5 patients (153 data points) respectively. Using variable translation wavelet neural network (VTWNN), the value of testing sensitivity and specificity are 79.07 % and 50.00 %. The results show that the proposed detection system performs well in terms of good sensitivity and acceptable specificity.

Thuraisingham, R., Tran, Y.H., Craig, A.R. & Nguyen, H.T. 2012, 'Frequency analysis of eyes open and eyes closed EEG signals using the Hilbert-Huang transform', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 2865-2868.
View/Download from: Publisher's site
View description>>

Frequency analysis based on the Hilbert-Huang transform (HHT) is examined as an alternative to Fourier spectral analysis in the study of EEG signals. This method overcomes the need for the EEG signal to be linear and stationary, assumptions necessary for the application of Fourier spectral analysis. The HHT method comprises two components: empirical mode decomposition (EMD) of the signal into intrinsic mode functions (IMFÔ++s); and the Hilbert transform of the IMFÔ++s. This technique is applied here in the study of consecutive eyes open (EO), eyes closed (EC) EEG signals of able bodied and spinal cord injured participants. The study found that in this EO, EC pair the instantaneous frequencies in the EO state were higher compared to the EC state. The Hilbert weighted frequency, a measure of the mean of the instantaneous frequencies present in an IMF, is used here to detect these changes from EO to the EC state in an EEG signal. Although there was a good detection of this change with information obtained from just one IMF (94% in able-bodied persons and 84% in SCI persons), almost 100% success in detecting between group differences was achieved using all the IMF's. This result has implications for assistive technology that rely on EEG changes in EO and EC states.

Tran, Y.H., Thuraisingham, R., Craig, A.R. & Nguyen, H.T. 2012, 'Stationary and variability in eyes open and eyes closed EEG signals from able-bodied and spinal cord injured persons', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 2861-2864.
View/Download from: Publisher's site
View description>>

This paper examines the assumption of stationarity used in EEG brain activity analyses, despite EEG data often being non-stationary. Transformations necessary to obtain stationary data from measured non-stationary EEG data and methods to assess non-stationarity are illustrated using eyes open (EO) and eyes closed (EC) data. The study shows that even short time EEG records of 10s duration exhibit nonstationary behavior. Examination of the change in variance when going from the EO to the EC state for both able bodied and spinal cord injured participants show that the difference in variance is consistently positive and statistically significant only when stationary data is used. This has implications for brain computer interfaces that utilizes changes in EO and EC EEG signals.

Truong, B.C., Hoang, T.D., Ha, K. & Nguyen, H.T. 2012, 'Global optimization for human skin investigation in TeraHertz', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 5474-5477.
View/Download from: Publisher's site
View description>>

In this paper, the electromagnetic interaction between human skin and terahertz radiation is investigated through the double Debye parametersÔ++ extraction algorithm. The changes of skin content are contrasted at the frequencies below one terahertz(THz) but the recent approaches could provide only a rough estimation. We propose an global optimization based identification, which results in globally accurate estimators in the frequency range up to two THz, and thus supports the validity of Debye model for Terahertz waveÔ++s propagation and reflection in skin. Simulation results confirm our prominent methodology.

Yuwono, M., Su, S.W., Moulton, B.D. & Nguyen, H.T. 2012, 'Gait cycle spectrogram analysis using a torso-attached inertial sensor', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 6539-6542.
View/Download from: Publisher's site
View description>>

Measurement of gait parameters can provide important information about a person's health and safety. Automatic analysis of gait using kinematic sensors is a newly emerging area of research. We describe a new way to detect walking, and measure gait cadence, by using time-frequency signal processing together with spectrogram analysis of signals from a chest-worn inertial measurement unit (IMU). A pilot study of 11 participants suggests that this method is able to distinguish between walk and non-walk activities with up to 88.70% sensitivity and 97.70% specificity. Limitations of the method include instability associated with manual fine-tuning of local and global threshold levels.

Yuwono, M., Su, S.W., Moulton, B.D. & Nguyen, H.T. 2012, 'Gait episode identification based on wavelet feature clustering of spectrogram images', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 2949-2952.
View/Download from: Publisher's site
View description>>

Automatic analysis of gait using kinematic sensors is a newly emerging area of research. We describe a new way to detect walking, and measure gait cadence, by using time-frequency signal processing together with spectrogram analysis of signals from a chest-worn inertial measurement unit (IMU). A pilot study of 11 participants suggests that this method is able to distinguish between walk and non-walk activities with up to 88.70% sensitivity and 97.70% specificity. Limitations of the method include instability associated with manual fine-tuning of local and global threshold levels.

Yuwono, M., Su, S.W., Moulton, B.D. & Nguyen, H.T. 2012, 'Optimization strategies for rapid centroid estimation', 34th Annual Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, California, USA, August 2012 in Proceedings of the 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Michael C.K. Khoo, IEEE, USA, pp. 6212-6215.
View/Download from: Publisher's site
View description>>

Particle swarm algorithm has been extensively utilized as a tool to solve optimization problems. Recently proposed particle swarm-¦based clustering algorithm called the Rapid Centroid Estimation (RCE) is a lightweight alteration to Particle Swarm Clustering (PSC). The RCE in its standard form is shown to be superior to conventional PSC algorithm. We have observed some limitations in RCE including the possibility to stagnate at a local minimum combination and the restriction in swarm size. We propose strategies to optimize RCE further by introducing RCE+ and swarm RCE+. Five benchmark datasets from UCI machine learning database are used to test the performance of these new strategies. In Glass dataset swarm RCE+ is able to achieve highest purity centroid combinations with less iteration (90.3%-¦1.1% in 9-¦5 iterations) followed by RCE+ (89%-¦3.5% in 65-¦62 iterations) and RCE (87%-¦5.9% in 54-¦44). Similar quality is also reflected in other benchmark datasets including Iris, Wine, Breast Cancer, and Diabetes.

Craig, A.R., Tran, Y.H., Wijesuriya, N.S., Thuraisingham, R. & Nguyen, H.T. 2011, 'Switching rate changes associated with mental fatigue for assistive technologies', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, Massachusetts, USA, August 2011 in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Technical Committee, IEEE, Boston, Massachusetts, USA, pp. 3071-3074.
View/Download from: UTSePress
View description>>

This paper presents research that investigated the effects of mental fatigue on brain activity associated with eyes open and eyes closed conditions. The changes associated with electroencephalography (EEG) alpha wave activity (8-13Hz) during eye closure has previously been shown to be an effective strategy for switching and activating devices as an environmental control system (ECS) designed for people with severe disability like spinal cord injury (SCI). The results showed that switching times did increase due to fatigue, however, these increases were not large (around 1 second longer to switch) and this difference was not significant. When baselines were readjusted taking into account the change in alpha wave activity due to the fatigue, switching reduced to times typically seen when the person was alert. Error rates were similar between the alert and fatigue sates. Implications of these results for a hands-free ECS are discussed.

Kitoko, V., Nguyen, T., Nguyen, J., Tran, Y.H. & Nguyen, H.T. 2011, 'Performance of dry electrode with bristle in recording EEG rhythms across brain state changes', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, USA, August 2011 in Engineering in Medicine and Biology Society,EMBC, 2011 Annual International Conference of the IEEE, ed Technical Committee, IEEE, Boston, USA, pp. 59-62.
View/Download from: UTSePress | Publisher's site
View description>>

In this paper we evaluate the physiological performance of a silver-silver chloride dry electrode with bristle (B-Electrode) in recording EEG data. For this purpose, we compare the performance of the bristle electrode in recording EEG data with the standard wet gold-plated cup electrode (G-Electrode) using two different brain state change tasks including resting condition with eyes-closed and performing mathematical task with eyes-open. Using a 2 channel recording device, eyes-closed command data were collected from each of 6 participants for a period of 20sec and the same procedure was applied for the mathematical calculation task. These data were used for statistical and classification analyse. Although, B-electrode has shown a slightly higher performance compared with G-electrode in both tasks, but analyse did not reveal any significant differences between both electrodes in all six subjects tested.

Ling, S.S., Nguyen, H.T. & Leung, F.H. 2011, 'Hypoglycemia detection using fuzzy inference system with genetic algorithm', IEEE International Conference on Fuzzy Systems, Taipei, Taiwan, June 2011 in IEEE International Conference on Fuzzy Systems 2011, ed Chin-Teng Lin, IEEE, USA, pp. 2225-2231.
View/Download from: Publisher's site
View description>>

Abstract+In this paper, we develope a genetic algorithm based fuzzy inference system to recognize hypoglycemic episodes based on heart rate and corrected QT interval of the telectrocardiogram (ECG) signal. Genetic algorithm is introduced to optimize the membership functions and fuzzy rules. A practical experiment based on data from 15 children with T1DM is studied. All the data sets are collected from the Department of Health, Government of Western Australia. To prevent the phenomenon of overtraining (over-fitting), a validation strategy that may adjust the fitness function is proposed. Thus, the data are organized into a training set, a validation set, and a testing set randomly selected. The classification results in term of sensitivity, specificity, and receiver operating characteristic (ROC) analysis show that the proposed classification method performs well.

Ling, S.S., Jiang, F., Chan, K.Y. & Nguyen, H.T. 2011, 'Permutation flow shop scheduling: fuzzy particle swarm optimization approach', IEEE International Conference On Fuzzy Systems, Taipei, Taiwam, June 2011 in IEEE International Conference on Fuzzy Systems 2011, ed Chin Teng Lin, IEEE, USA, pp. 572-578.
View description>>

Abstract+A fuzzy particle swarm optimization (PSO) for the minimization of makespan in permutation flow shop scheduling problem is presented in this paper. In the proposed fuzzy PSO, the inertia weight of PSO and the control parameter of the crossmutated operation are determined by a set of fuzzy rules. To escape the local optimum, cross-mutated operation is introduced. In order to make PSO suitable for solving permutation flow shop scheduling problem, a roulette wheel mechanism is proposed to convert the continuous position values of particles to job permutations. Meanwhile, a swap-based local search for scheduling problem is designed for the local exploration on a discrete job permutation space. Flow shop benchmark functions are employed to evaluate the performance of the fuzzy PSO for flow shop scheduling problems and the results indicate that the algorithm performs better compared with existing hybrid PSO algorithms.

Nguyen, V., Su, S.W. & Nguyen, H.T. 2011, 'Development of a Bayesian recursive algorithm to find freespaces for an intelligent wheelchair', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, Massachusetts, USA, August 2011 in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed N/A, IEEE, United States, pp. 7250-7253.
View/Download from: UTSePress
View description>>

This paper introduces a new shared control strategy for an intelligent wheelchair using a Bayesian recursive algorithm. Using the local environment information gathered by a laser range finder sensor and commands acquired through a user interface, a Bayesian recursive algorithm has been developed to find the most appropriate free-space, which corresponds to the highest posterior probability value. Then, an autonomous navigation algorithm will assist to manoeuvre the wheelchair in the chosen freespace. Experiment results demonstrate that the new method provides excellent performance with great flexibility and fast response.

Nguyen, B.L., Ling, S.S., Jones, T.W. & Nguyen, H.T. 2011, 'Identification of hypoglycemic states for patients with T1DM using various parameters derived from EEG signals', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, Massachusetts, USA, August 2011 in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Technical Committee, IEEE, Boston, Massachusetts, USA, pp. 2760-2763.
View/Download from: UTSePress
View description>>

For patients with Type 1 Diabetes Mellitus (T1DM), hypoglycemia is a very common but dangerous complication which can lead to unconsciousness, coma and even death. The variety of hypoglycemia symptoms is originated from the inadequate supply of glucose to the brain. In this study, we explore the connection between hypoglycemic episodes and the electrical activity of neurons within the brain or electroencephalogram (EEG) signals. By analyzing EEG signals from a clinical study of five children with T1DM, associated with hypoglycemia at night, we find that some EEG parameters change significantly under hypoglycemia condition. Based on these parameters, a method of detecting hypoglycemic episodes using EEG signals with a feed-forward multi-layer neural network is proposed. In our application, the classification results are 72% sensitivity and 55% specificity when the EEG signals are acquired from 2 electrodes C3 and O2. Furthermore, signals from different channels are also analyzed to observe the contributions of each channel to the performance of hypoglycemia classification.

Nguyen, J., Tran, Y.H., Su, S.W. & Nguyen, H.T. 2011, 'Semi-autonomous wheelchair developed using a unique camera system configuration biologically inspired by equine vision', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, Massachusetts, USA, August 2011 in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Technical Committee, IEEE, Boston, Massachusetts, USA, pp. 5762-5765.
View/Download from: UTSePress
View description>>

This paper is concerned with the design and development of a semi-autonomous wheelchair system using cameras in a system configuration modeled on the vision system of a horse. This new camera configuration utilizes stereoscopic vision for 3-Dimensional (3D) depth perception and mapping ahead of the wheelchair, combined with a spherical camera system for 360-degrees of monocular vision. This unique combination allows for static components of an unknown environment to be mapped and any surrounding dynamic obstacles to be detected, during real-time autonomous navigation, minimizing blind-spots and preventing accidental collisions with people or obstacles. This novel vision system combined with shared control strategies provides intelligent assistive guidance during wheelchair navigation and can accompany any hands-free wheelchair control technology. Leading up to experimental trials with patients at the Royal Rehabilitation Centre (RRC) in Ryde, results have displayed the effectiveness of this system to assist the user in navigating safely within the RRC whilst avoiding potential collisions.

Nguyen, T., Nguyen, H.T., Su, S.W. & Celler, B.G. 2011, 'Robust online adaptive neural network control for the regulation of treadmill exercises', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, USA, August 2011 in Engineering in Medicine and Biology Society,EMBC, 2011 Annual International Conference of the IEEE, ed Technical Committee, IEEE, USA, pp. 1005-1008.
View/Download from: UTSePress | Publisher's site
View description>>

The paper proposes a robust online adaptive neural network control scheme for an automated treadmill system. The proposed control scheme is based on Feedback-Error Learning Approach (FELA), by using which the plant Jacobian calculation problem is avoided. Modification of the learning algorithm is proposed to solve the overtraining issue, guaranteeing to system stability and system convergence. As an adaptive neural network controller can adapt itself to deal with system uncertainties and external disturbances, this scheme is very suitable for treadmill exercise regulation when the model of the exerciser is unknown or inaccurate. In this study, exercise intensity (measured by heart rate) is regulated by simultaneously manipulating both treadmill speed and gradient in order to achieve fast tracking for which a single input multi output (SIMO) adaptive neural network controller has been designed. Real-time experiment result confirms that robust performance for nonlinear multivariable system under model uncertainties and unknown external disturbances can indeed be achieved.

Nuryani, N., Ling, S.S. & Nguyen, H.T. 2011, 'Ventricular Repolarization Variability for Hypoglycemia Detection', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, Massachusetts, USA, August 2011 in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Technical Committee, IEEE, Boston, Massachusetts, USA, pp. 7961-7964.
View/Download from: UTSePress
View description>>

Hypoglycemia is the most acute and common complication of Type 1 diabetes and is a limiting factor in a glycemic management of diabetes. In this paper, two main contributions are presented; firstly, ventricular repolarization variabilities are introduced for hypoglycemia detection, and secondly, a swarm-based support vector machine (SVM) algorithm with the inputs of the repolarization variabilities is developed to detect hypoglycemia. By using the algorithm and including several repolarization variabilities as inputs, the best hypoglycemia detection performance is found with sensitivity and specificity of 82.14% and 60.19%, respectively.

San, P., Ling, S.S. & Nguyen, H.T. 2011, 'Block-based neural network for hypoglycemia detection', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, Massachusetts, USA, August 2011 in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Technical Committee, IEEE, Boston, Massachusetts, USA, pp. 5666-5669.
View/Download from: UTSePress
View description>>

In this paper, evolvable block based neural network (BBNN) is presented for detection of hypoglycemia episodes. The structure of BBNN consists of a two-dimensional (2D) array of fundamental blocks with four variable inputoutput nodes and weight connections. Depending on the structure settings, each block can have one of four different internal configurations. To provide early detection of hypoglycemia episodes, the physiological parameters such as heart rate (HR) and corrected QT interval (QTc) of electrocardiogram (ECG) signal are used as the inputs of BBNN. The overall structure and weights of BBNN are optimized by an evolutionary algorithm called hybrid particle swarm optimization with wavelet mutation (HPSOWM). The optimized structures and weights of BBNN are capable to compensate large variations of ECG patterns caused by individual and temporal difference since a fixed structure classifiers are easy to fail to trace ECG signals with large variations. The ECG data of 15 patients are organized into a training set, a testing set and a validation set, each of which has randomly selected 5 patients. The simulation results shows that the proposed algorithm, BBNN with HPSOWM can successfully detect the hypoglycemic episodes in T1DM in term of testing sensitivity (76.74%) and test specificity (50.91%).

Tran, C., Ha, Q.P. & Nguyen, H.T. 2011, 'Semi-automatic control of modular systems with intermittent data losses', IEEE Conference on Automation Science and Engineering, Trieste Italy, August 2011 in Proceedings of the 2011 IEEE Conference on Automation Science and Engineering, ed Maria Pia, IEEE, Piscataway, NJ 08854, pp. 625-630.
View description>>

This paper presents a control procedure of distributed stabilising agents for dynamically-coupled systems operating in the imperfect data environment of a mesh device network. A multivariable controller is applied to each single modular subsystem, which also allows for a manual control mode. To deal with the device network, intermittent data losses are compensated for on-the-fly using the incrementally accumulative quadratic constraint (AQC). The incrementally AQC is employed in the procedure of stabilising agents to accommodate the coexistence of closed-loop control and man-inthe- loop regulation. These agents render stabilising bounds for the manipulated variables in the automatic control mode, and at the same time, provide warning signals and manipulation guidance for the operators to prevent possible plant-wide destabilisation in the semi-automatic control mode. Taking the control constraints into consideration, the feasibility of AQCbased stabilising bounds is guaranteed for the consecutive datalost periods of device networks. The innovative aspect of the proposed approach rests on the stability condition developed from the input and output evolution prescribed in the controller AQC and the system dissipativity, as well as the method of remedying data losses right after the incidents. Simulation results are provided for the model predictive control of an industrial modular system in the mineral processing industry.

Tran, C., Ha, Q.P., Nguyen, H.T. & Hoang, T.D. 2011, 'Toward Plant-wide Control of Reticulated Systems Arising in Alumina Refinery with Online Stabilisation', World Congress of the International Federation of Automatic Control, Milano Italy, August 2011 in Preprints of the 18th World Congress of the International Federation of Automatic Control (IFAC), ed S. Bittanti, A. Cenedese, S. Zampieri, International Federation of Automatic Control (IFAC), Milano Italy, pp. 10529-10534.
View description>>

This paper presents a novel distributed model predictive control strategy for reticulated systems of the alumina refining process. The plant-wide control is facilitated by the constructive method of online stabilisations that is applicable to the model predictive controllers (MPC) as stability constraints. The plant-wide process is modeled as a large-scale system formed by the subsystems of different unit operations interconnected to each other. The stability condition for the interconnected system is derived from the accumulative quadratic constraint (AQC), which is subsequently developed into receding-horizon stability constraints for MPC. The proposed online stabilisation scheme can be implemented for a department and/or the whole alumina refinery, which consists of four main departments of digestion, clarification, precipitation-filtration, and evaporation. The theoretical results are illustrated by simulations for a typical example of three dynamically-coupled subsystems.

Tran, Y.H., Thuraisingham, R., Craig, A.R., Tomlinson, E., Davis, G.M., Middleton, J. & Nguyen, H.T. 2011, 'Changes in Blood Volume Pulse during Excercise Recovery in Activity-based Therapy for Spinal Cord Injury', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, Massachusetts, USA, August 2011 in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Technical Committee, IEEE, Boston, Massachusetts, USA, pp. 693-696.
View/Download from: UTSePress
View description>>

This paper presents the results of cardiovascular changes that occur during a novel rehabilitation strategy called activity based therapy (ABT). Blood volume pulse (BVP) signals were measured during functional electrical stimulation (FES)-induced cycling in adults with spinal cord injury (SCI) persons and results were compared to a passive cycling task and able-bodied controls performing normal cycling. BVP signals were compared during three conditions, a baseline preexercise condition, 5 minutes after exercise and after 30- minutes rest following exercise. Exercise recovery was evaluated using normalized inner products values in BVP signals. The results showed that FES-induced cycling in SCI participants resulted in a significantly greater peripheral resistance level and longer time to recover from exercise compared with passive cycling and normal cycling in ablebodied controls.

Weng, K., Zhang, Y., Nguyen, T., Haddad, A., Celler, B.G., Su, S.W., Guo, Y. & Nguyen, H.T. 2011, 'Multi-Loop Integral Control By Using Redundant Control Inputs For Passive Fault Tolerant Implementation', International Conference on Measurement and Control Engineering, Pueto Rico, USA, October 2011 in 2011 2nd International Conference on Measurement and Control Engineering (ICMCE 2011), ed Technical Committee, ASME Press, USA, pp. 7-11.
View/Download from: UTSePress | Publisher's site
View description>>

One of the advantages of using more than one actuator and multi-loop control structure for the control of a single output variable is its potential to tolerate system failures. Based on the analysis of the steady state behaviour of the process, this paper presents a theoretical examination of processes with redundant actuators. The concept of Decentralized Integral Controllability (DIC) has been extended to non-square nonlinear processes, and a steady state sufficient condition has been provided for the multi-loop integral control configuration. We illustrate the proposed analysis method by using the example of the regulation of heart rate response for treadmill exercises, in which both treadmill speed and gradient are served as control inputs for the regulation of a single output, heart rate.

Yuwono, M., Handojoseno, A.M. & Nguyen, H.T. 2011, 'Optimization of head movement recognition using augmented radial basis function neural network', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, Massachusetts, USA, August 2011 in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Technical Committee, IEEE, Boston, Massachusetts, USA, pp. 2776-2779.
View/Download from: UTSePress
View description>>

For people with severe spine injury, head movement recognition control has been proven to be one of the most convenient and intuitive ways to control a power wheelchair. While substantial research has been done in this area, the challenge to improve system reliability and accuracy remains due to the diversity in movement tendencies and the presence of movement artifacts. We propose a Neural- Network Configuration which we call Augmented Radial Basis Function Neural-Network (ARBF-NN). This network is constructed as a Radial Basis Function Neural-Network (RBF-NN) with a Multilayer Perceptron (MLP) augmentation layer to negate optimization limitation posed by linear classifiers in conventional RBF-NN. The RBF centroid is optimized through Regrouping Particle Swarm Optimization (RegPSO) seeded with K-Means. The trial results of ARBFNN on Head-movement show a significant improvement on recognition accuracy up to 98.1% in sensitivity.

Chan, K.Y., Ling, S.S., Dillon, T.S. & Nguyen, H.T. 2010, 'Classification of Hypoglycemic Episodes for Type 1 Diabetes Mellitus based on Neural Networks', IEEE Congress on Evolutionary Computation, Barcelona, Spain, July 2010 in IEEE Congress on Evolutionary Computation (CEC) - 2010 IEEE World Congress on Computational Intelligence, ed Pilar Sobrevilla, IEEE, USA, pp. 1444-1448.
View/Download from: UTSePress | Publisher's site
View description>>

Hypoglycemia is dangerous for Type 1 diabetes mellitus (T1DM) patients. Based on the physiological parameters, we have developed a classification unit with hybridizing the approaches of neural networks and genetic algorithm to identify the presences of hypoglycemic episodes for TIDM patients. The proposed classification unit is built and is validated by using the real T1DM patients' data sets collected from Department of Health, Government of Western Australia. Experimental results show that the proposed neural network based classification unit can achieve more accurate results on both trained and unseen T1DM patients' data sets compared with those developed based on the commonly used classification methods for medical diagnosis including statistical regression, fuzzy regression and genetic programming.

Ling, S.S., Nuryani, N. & Nguyen, H.T. 2010, 'Evolved Fuzzy Reasoning Model for Hypoglycaemic Detection', IEEE Engineering in Medicine and Biology Society Annual Conference, Buenos Aires, Argentina, August 2010 in Proceedings of the 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society 'Merging Medical Humanism and Technology', ed Ricardo Armentano, Piscataway, USA, USA, pp. 4662-4665.
View/Download from: UTSePress | Publisher's site
View description>>

Hypoglycaemia is a serious side effect of insulin therapy in patients with diabetes. We measure physiological parameters (heart rate, corrected QT interval of the electrocardiogram (ECG) signal) continuously to provide early detection of hypoglycemic episodes in Type 1 diabetes mellitus (T1DM) patients. Based on the physiological parameters, an evolved fuzzy reasoning model (FRM) to recognize the presence of hypoglycaemic episodes is developed. To optimize the fuzzy rules and the fuzzy membership functions of FRM, an evolutionary algorithm called hybrid particle swarm optimization with wavelet mutation operation is investigated. All data sets are collected from Department of Health, Government of Western Australia for a clinical study. The results show that the proposed algorithm performs well in terms of the clinical sensitivity and specificity.

Ling, S.S., Nguyen, H.T. & Chan, K.Y. 2010, 'Genetic algorithm based fuzzy multiple regression for the nocturnal hypoglycemic classification', IEEE Congress on Evolutionary Computation, Barcelona, Spain, July 2010 in IEEE Congress on Evolutionary Computation (CEC) - 2010 IEEE WORLD Congress on Computational Intelligence, ed Pilar Sobrevilla, IEEE, USA, pp. 2659-2664.
View/Download from: UTSePress | Publisher's site
View description>>

Low blood glucose (Hypoglycaemia) is dangerous and can result in unconsciousness, seizures and even death. It has a common and serious side effect of insulin therapy in patients with diabetes. We measure physiological parameters (heart rate, corrected QT interval of the electrocardiogram (ECG) signal, change of heart rate, and the change of corrected QT interval) continuously to provide detection of hypoglycaemic. Based on these physiological parameters, we have developed a genetic algorithm based multiple regression model to determine the presence of hypoglycaemic episodes. Genetic algorithm is used to determine the optimal parameters of the multiple regression. The overall data were organized into a training set (8 patients) and a testing set (another 8 patient) which are randomly selected. The clinical results show that the proposed algorithm can achieve predictions with good sensitivities and acceptable specificities.

Ling, S.S., Nuryani, N. & Nguyen, H.T. 2010, 'Hypoglycaemia detection for Type 1 diabetic patients based on ECG parameters using fuzzy support vector machine', International Joint Conference on Neural Networks, Spain, July 2010 in International Joint Conference on Neural Networks - 2010 IEEE World Congress on Computational Intelligence, ed Pilar SOBREVILLA, IEEE, USA, pp. 2253-2259.
View/Download from: UTSePress | Publisher's site
View description>>

Nocturnal hypoglycaemia in type 1 diabetic patients can be dangerous in which symptoms may not be apparent while blood glucose level decreases to very low level, and for this reason, an effective detection system for hypoglycaemia is crucial. This research work proposes a detection system for the hypoglycaemia based on the classification of electrocardiographic (ECG) parameters. The classification uses a Fuzzy Support Vector Machine (FSVM) with inputs of heart rate, corrected QT (QT c) interval and corrected TpTe (TpTec) interval. Three types of kernel functions (radial basis function (RBF), exponential radial basis function (ERBF) and polynomial function) are investigated in the classification. Moreover, parameters of the kernel functions are tuned to find the optimum of the classification. The results show that the FSVM classification using RBF kernel function demonstrates better performance than using SVM. However, both classifiers result approximately same performance if ERBF and polynomial kernel functions are used.

Nguyen, H.T. & Jones, T.W. 2010, 'Detection of Nocturnal Hypoglycemic Episodes using EEG Signals', IEEE Engineering in Medicine and Biology Society Annual Conference, Buenos Aires, Argentina, August 2010 in Proceedings of the 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society 'Merging Medical Humanism and Technology', ed Ricardo Armentano, The Printing House, Inc., USA, pp. 4930-4933.
View/Download from: UTSePress | Publisher's site
View description>>

Hypoglycemia (low blood glucose) or the fear of hypoglycemia constitutes a significant barrier to the achievement of good glycemic control in the insulin treated diabetic patients. By measuring physiological responses derived from EEG and analyzing these, we establish that hypoglycemia can be detected non-invasively. From a clinical study of six children with type 1 diabetes (T1D), associated with hypoglycemic episodes at night, their centroid (centre of gravity) alpha frequency reduced significantly (P<;;0.001) and their centroid theta frequency increased significantly (P<;;0.02). The overall data were organized into a training set (3 patients) and a test set (another 3 patients) randomly selected. Using the optimal Bayesian neural network which was derived from the training set with the highest log evidence, the estimated blood glucose profiles produced a significant correlation (P<;;0.005) against measured values in the test set.

Nguyen, J., Su, S.W. & Nguyen, H.T. 2010, 'Spherical Vision Cameras in a Semi-autonomous Wheelchair System', IEEE Engineering in Medicine and Biology Society Annual Conference, Buenos Aires, Argentina, August 2010 in Proceedings of the 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society 'Merging Medical Humanism and Technology', ed Ricardo Armentano, The Printing House, Inc., USA, pp. 4064-4067.
View/Download from: UTSePress | Publisher's site
View description>>

This paper is concerned with the methods developed for extending the capabilities of a spherical vision camera system to allow detection of surrounding objects and whether or not they pose a danger for movement in that direction during autonomous navigation of a power wheelchair. A Point Grey Research (PGR) Ladybug2 spherical vision camera system was attached to the power wheelchair for surrounding vision. The objective is to use this Ladybug2 system to provide information about obstacles all around the wheelchair and aid the automated decision-making process involved during navigation. Through instantaneous neural network classification of individual camera images to determine whether obstacles are present, detection of obstacles have been successfully achieved with accuracies reaching 96%. This assistive technology has the purpose of automated obstacle detection, navigational path planning and decision-making, and collision avoidance during navigation.

Nuryani, N., Ling, S.S. & Nguyen, H.T. 2010, 'Electrocardiographic T-wave Peak-to-end Interval for Hypoglycaemia Detection', IEEE Engineering in Medicine and Biology Society Annual Conference, Buenos Aires, Argentina, August 2010 in Proceedings of the 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Merging Medical Humanism and Technology, ed Ricardo Armentano, The Printing House, Inc., USA, pp. 618-621.
View/Download from: UTSePress | Publisher's site
View description>>

Electrocardiographic T wave peak-to-end interval (TpTe) is one parameter of T wave morphology, which contains indicators for hypoglycaemia. This paper shows the corrected TpTe (TpTec) interval as one of the inputs contributing to detect hypoglycaemia. Support vector machine (SVM) and fuzzy support vector machine (FSVM) utilizing radial basis function (RBF) are used as the classification methods in this paper. By comparing with the classification systems using inputs of corrected QT interval (QTc) and heart rate only, the results indicate that the inclusion of TpTec in combination with QTc and heart rate performs better in the detection of hypoglycaemia in terms of sensitivity, specificity and accuracy.

Su, S.W., Nguyen, H.T. & Ha, Q.P. 2010, 'Laboratory Demonstration for Model Predictive Multivariable Control with a Coupled Drive System', International Conference on Control, Automation, Robotics and Vision, Singapore, December 2010 in Proc. of the 11th International Conference on Control, Automation, Robotics and Vision (ICARCV 2010), ed IEEE Technical Committee, IEEE, Singapore, pp. 762-767.
View/Download from: UTSePress | Publisher's site
View description>>

Teaching multivariable control usually involves a certain level of mathematical sophistication and hence requires some labaratorial exemplification of the material given in formal lectures. This paper reports on a hands-on approach to multivariable control education via the implementation of a model predictive controller on a two-input, two output coupled drive apparatus. This scaled-down system represents many industrial processes while provides an excellent set-up for demonstrating the cross-coupled effects in multi-input multi-output systems. Here, a model predictive controller (MPC) is developed and implemented on the basis of a constrained optimization problem to show control performance via the belt tension and velocity outputs, demonstrate the decoupling capability, and also illustrate such issues as control input saturation, the selection of operating point, reference inputs, and system robustness to external disturbance and varying parameters. The implementation is based on Labview and MATLAB Model Predictive Control Toolbox.

Tran, C., Nguyen, H.T. & Ha, Q.P. 2010, 'Stability of Complex Systems with Mixed Connection Configurations under Shared Control', Int. Conf. Control, Automation, Robotics and Vision, Singapore, December 2010 in Proc. of the 11th. Int. Conf. Control, Automation, Robotics and Vision (ICARCV 2010), ed IEEE Technical Committee, IEEE, Singapore, pp. 512-517.
View/Download from: UTSePress | Publisher's site
View description>>

This paper presents a new stabilizing method for the control of complex systems operating in semi-automatic modes. The complex system is modeled by several spatially-coupled subsystems interconnected in parallel, serial and cycle configurations. Each subsystem is regulated by a dedicated autonomous controller that also allows for a manual control mode. An interconnection stability condition which takes the couplings between subsystems into consideration is derived from the renowned dissipative systems theory. Built upon this stability condition, decentralized stabilizing agents for autonomous controllers are subsequently deployed independently and segregatedly from the control algorithms. Due to this independence, human errors from manmachine interactions, that may destabilize the control systems, can be avoidable; also different types of control algorithms and controllers of subsystems are interoperable with the same stabilizing mechanism. To accomplish such tasks simultaneously, the stabilizing agents render overriding outputs for the automatic controllers, and at the same time, provide instability warning signals and manipulation guidance to the operators to successfully regulate the subsystems in the manual control mode, yet maintain the plant-wide stability. Real-time data of control inputs and plant outputs is exerted under the auspices of controller dissipativity indices and trajectories to stabilize the systems with closed-loop control and man-in-the-loop coexistence. Our main results are illustrated in simulation for a three-unit system.

Tran, Y.H., Craig, A.R., Wijesuriya, N. & Nguyen, H.T. 2010, 'Improving Classification Rates for Use in Fatigue Countermeasure Devices using Brain Activity', IEEE Engineering in Medicine and Biology Society Annual Conference, Buenos Aires, Argentina, August 2010 in Proceedings of the 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society 'Merging Medical Humanism and Technology', ed Ricardo Armentano, The Printing House, Inc., USA, pp. 4460-4463.
View/Download from: UTSePress | Publisher's site
View description>>

Fatigue can be defined as a state that involves psychological and physical tiredness with a range of symptoms such as tired eyes, yawning and increased blink rate. It has major implications for work place and road safety as well as a negative symptom of many acute and chronic illnesses. As such there has been considerable research dedicated to systems or algorithms that can be used to detect and monitor the onset of fatigue. This paper examines using electroencephalography (EEG) signals to classify fatigue and alert states as a function of subjective self-report, driving performance and physiological symptoms. The results show that EEG classification network for fatigue improved from 75% to 80% when these factors are applied, especially when the data is grouped by subjective self-report of fatigue with classification accuracy improving to 84.5%.

Weng, K., Turk, B., Dolores, L., Nguyen, T., Celler, B.G., Su, S.W. & Nguyen, H.T. 2010, 'Fast tracking of a given heart rate profile in treadmill exercise', IEEE Engineering in Medicine and Biology Society Annual Conference, Buenos Aires, Argentina, August 2010 in Proceedings of the 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society 'Merging Medical Humanism and Technology', ed Ricardo Armentano, The Printing House, Inc., USA, pp. 2569-2572.
View/Download from: UTSePress | Publisher's site
View description>>

This paper investigates the application of a multi-loop PID controller in an automated treadmill exercise machine. The approach is to design a computer-controlled treadmill control system for the regulation of heart rate (HR) during treadmill exercise. A single-input and multiple-output (SIMO) controller was implemented to fast track a given heart rate profile in treadmill exercise. Two separate single-input and single-output (SISO) PID control systems are initially implemented to modify either the treadmill speed or its angle of inclination in order to achieve a desired HR. The purpose of this paper is to apply a SIMO control system by implementing a control algorithm which includes the two PID controllers working simultaneously to track the desired HR profile. The performance of the SIMO and SISO control systems are compared through the closed loop responses recorded during experimentation. This would also help future development of safe treadmill exercise system.

Dalvand, H., Nguyen, H.T. & Ha, Q.P. 2009, 'Design of second-order sliding mode controllers for MR damper-embedded smart structures', International Symposium on Automation and Robotics in Construction, Austin USA, June 2009 in Proceedings of the 26th International Symposium on Automation and Robotics in Construction, ed Caldas, C., O'Brien, W., Chi S., Gong, J., Luo, X., IAARC-University of Texas at Austin, Texas, USA, pp. 332-340.
View/Download from: UTSePress
View description>>

Design of a current controlled system for MR damper-embedded civil structures

Ling, S.S., Nguyen, H.T. & Chan, K.Y. 2009, 'A New Particle Swarm Optimization Algorithm for Neural Network Optimization', International Conference on network and System Security, Gold Coast, Australia, October 2009 in 2nd International Workshop on Data Mining and Artificial Intelligence (DMAI 2009), IEEE International Conference on Network and System Security, ed Wanlei Zhou, IEEE, USA, pp. 516-521.
View/Download from: UTSePress | Publisher's site
View description>>

This paper presents a new particle swarm optimization (PSO) algorithm for tuning parameters (weights) of neural networks. The new PSO algorithm is called fuzzy logic-based particle swarm optimization with cross-mutated operation (FPSOCM), where the fuzzy inference system is applied to determine the inertia weight of PSO and the control parameter of the proposed cross-mutated operation by using human knowledge. By introducing the fuzzy system, the value of the inertia weight becomes variable. The cross-mutated operation is effectively force the solution to escape the local optimum. Tuning parameters (weights) of neural networks is presented using the FPSOCM. Numerical example of neural network is given to illustrate that the performance of the FPSOCM is good for tuning the parameters (weights) of neural networks.

Nguyen, H.T., Ghevondian, N. & Jones, T.W. 2009, 'Real-time detection of nocturnal hypoglycemic episodes using a novel non-invasive hypoglycemia monitor', IEEE Engineering in Medicine and Biology Society Annual Conference, Minneapolis, Minnesota, USA, September 2009 in Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Zhi-Pei Liang, IEEE, USA, pp. 3822-3825.
View/Download from: UTSePress | Publisher's site
View description>>

Hypoglycemia or low blood glucose is a common and serious side effect of insulin therapy in patients with diabetes. Hypoglycemia is unpleasant and can result in unconsciousness, seizures and even death. HypoMon is a realtime non-invasive monitor that measures relevant physiological parameters continuously to provide detection of hypoglycemic episodes in Type 1 diabetes mellitus patients (T1DM). Based on heart rate and corrected QT interval of the ECG signal, we have continued to develop effective algorithms for early detection of nocturnal hypoglycemia. From a clinical study of 24 children with T1DM, associated with natural occurrence of hypoglycemic episodes at night, their heart rates increased (1.021-¦0.264 vs. 1.068-¦0.314, P<0.053) and their corrected QT intervals increased significantly (1.030-¦0.079 vs. 1.052-¦0.078, P<0.002). It is interesting to note that QT interval and heart rate are less correlated when the patients experienced hypoglycemic episodes through natural occurrence compared to when clamp studies were performed. The overall data were organized into a training set (12 patients) and a test set (another 12 patients) randomly selected. Using the optimal Bayesian neural network which was derived from the training set with the highest log evidence, the estimated blood glucose profiles produced a significant correlation (P<0.02) against measured values in the test set.

Nguyen, J., Nguyen, H.T. & Nguyen, H.T. 2009, 'Semi-autonomous wheelchair system using stereoscopic cameras', IEEE Engineering in Medicine and Biology Society Annual Conference, Minneapolis, Minnesota, USA, September 2009 in Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Zhi-Pei Liang, IEEE, USA, pp. 5068-5071.
View/Download from: UTSePress
View description>>

This paper is concerned with the design and development of a semi-autonomouswheelchair system using stereoscopic cameras to assist hands-free control technologies for severely disabled people. The stereoscopic cameras capture an image from both the left and right cameras, which are then processed with a Sum of Absolute Differences (SAD) correlation algorithm to establish correspondence between image features in the different views of the scene. This is used to produce a stereo disparity image containing information about the depth of objects away from the camera in the image. A geometric projection algorithm is then used to generate a 3- Dimensional (3D) point map, placing pixels of the disparity image in 3D space. This is then converted to a 2-Dimensional (2D) depth map allowing objects in the scene to be viewed and a safe travel path for the wheelchair to be planned and followed based on the user's commands. This assistive technology utilising stereoscopic cameras has the purpose of automated obstacle detection, path planning and following, and collision avoidance during navigation. Experimental results obtained in an indoor environment displayed the effectiveness of this assistive technology.

Nguyen, T., Nguyen, H.T. & Su, S.W. 2009, 'Robust Multivariable Strategy and its Applications to a Powered Wheelchair', IEEE Engineering in Medicine and Biology Society Annual Conference, Minneapolis, Minnesota, USA, September 2009 in Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Zhi-Pei Liang, IEEE, Minnesota, USA, pp. 7114-7117.
View/Download from: UTSePress
View description>>

The paper proposes a systematic robust multivariable control strategy based on combination of systematic triangularization technique and robust control strategies. Two design stages are required. In the first design stage, multivariable control problem is reduced into a series of scalar control problems via triangularization technique. For each specific scalar system, two advanced control strategies are proposed and implemented in the second design stage. The first one is based on Model Predictive Control, which is an iterative, finite horizon optimization procedure. The second control strategy is known as Neuro-Sliding Mode Control, which integrates Sliding Mode Control (SMC) and Neural Network Design to achieve both chattering-free and system robustness. Real-time implementation on a powered wheelchair system confirms that robustness and desired performance of a multivariable system under model uncertainties and unknown external disturbances can indeed be achieved by the combination of triangularization technique and Neuro-Sliding Mode Control.

Smith, S., Winchester, D., Jamieson, R. & Nguyen, H.T. 2009, 'Information Systems Security Compliance in e-Government', Pacific Asia Conference on Information Systems, Hyderabad, Inidia, July 2009 in Proceedings of Pacific Asia Conference on Information Systems (PACIS) 2009, ed Mahajan, S., AIS Electronic Library, India, pp. 1-13.
View/Download from: UTSePress
View description>>

The aim of this research paper is the development of a Fuzzy Logic model framed on Activity Theory to predict and benchmark compliance of Government agencies activities, with information systems security (ISS) standard, AS17799 (2006). The ISS standard has 10 main categories and 127 controls for which survey questions were asked in an online process. This project is a longitudinal study that commenced in 2002. The questions for the Fuzzy Logic project were piloted in August 2002, followed by three annual surveys from November 2002. The paper describes the development of an enhanced Fuzzy Logic model using activity Theory. The results from the Fuzzy Logic model helped to focus attention and monitor the progress of agencies that appear unlikely to reach ISS compliance. The main contribution of this study is the simplification of a complex system guided by Activity Theory using a fuzzy logic tool for analysis of a large number of inputs across a large number of agencies. A practical contribution to the New South Wales Government was that the Fuzzy Logic tool removed the complexity in computation, saved time and resources. Our approach using Fuzzy Logic also permits input from expertÔ++s embracing an organisations human capital.

Su, S.W., Nguyen, H.T., Jarman, R., Zhu, J., Lowe, D.B., McLean, P.B., Huang, S., Nguyen, T., Nicholson, R.S. & Weng, K. 2009, 'Model Predictive Control of Gantry Crane with Input Nonlinearity Compensation', International Conference on Control, Automation and Systems Engineering, Penang, Malaysia, February 2009 in International Conference on Control, Automation and Systems Engineering, ed Ardil C, World Academy of Science, Engineering and Technology, Penang, Malaysia, pp. 312-316.
View/Download from: UTSePress
View description>>

This paper proposed a nonlinear model predictive control (MPC) method for the control of gantry crane. One of the main motivations to apply MPC to control gantry crane is based on its ability to handle control constraints for multivariable systems. A pre-compensator is constructed to compensate the input nonlinearity (nonsymmetric dead zone with saturation) by using its inverse function. By well tuning the weighting function matrices, the control system can properly compromise the control between crane position and swing angle. The proposed control algorithm was implemented for the control of gantry crane system in System Control Lab of University of Technology, Sydney (UTS), and achieved desired experimental results.

Su, S.W., Anderson, B., Chen, W. & Nguyen, H.T. 2009, 'Multi-realisation of nonlinear systems', IEEE Conference on Decision and Control, Shanghai, China, December 2009 in Proceedings of the 48th IEEE Conference on Decision and Control, 2009 held jointly with the 2009 28th Chinese Control Conference. CDC/CCC 2009., ed IEEE Technical Committee, IEEE, China, pp. 5901-5905.
View/Download from: UTSePress | Publisher's site
View description>>

The system multi-realization problem is to find a state-variable realization for a set of systems, sharing as many parameters as possible. A multi-realization can be used to efficiently implement a multi-controller architecture for Multiple Model Adaptive Control (MMAC). We extend the linear multi-realization problem to nonlinear systems. The problem of minimal multi-realization of a set of MIMO systems is introduced and solved for feedback linearizable systems.

Thuraisingham, R., Tran, Y.H., Craig, A.R., Wijesuriya, N.S. & Nguyen, H.T. 2009, 'Using microstate intensity for the analysis of spontaneous EEG: tracking changes from alert to fatigue state', IEEE Engineering in Medicine and Biology Society Annual Conference, Minneapolis, Minnesota, USA, September 2009 in Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Zhi-Pei Liang, IEEE, Minneapolis, Minnesota, USA, pp. 4982-4985.
View/Download from: UTSePress
View description>>

Fatigue is a negative symptom of many illnesses and also has major implications for road safety. This paper presents results using a method called microstate segmentation (MSS). It was used to distinguish changes from an alert to a fatigue state. The results show a significant increase in MSS instantaneous amplitude during the fatigue state. Plotting the linear gradient of the nonlinear part of the phase data from the MSS also showed a significant difference (P<0.01) in the gradients of the alert state compared to the fatigue state. The results suggest that MSS can be used in analyzing spontaneous electroencephalography (EEG) signals to detect changes in physiological states. The results have implications for countermeasures used in detecting fatigue.

Tran, Y.H., Thuraisingham, R., Craig, A.R. & Nguyen, H.T. 2009, 'Evaluating the efficacy of an automated procedure for EEG artifact removal', IEEE Engineering in Medicine and Biology Society Annual Conference, Minneapolis, Minnesota, USA, September 2009 in Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Zhi-Pei Liang, IEEE, USA, pp. 376-379.
View/Download from: UTSePress
View description>>

Electroencephalography (EEG) signals are often contaminated with artifacts arising from many sources such as those with ocular and muscular origins. Artifact removal techniques often rely on the experience of the EEG technician to detect these artifact components for removal. This paper presents the results comparing an automated procedure (AT) against visually (VT) choosing artifactual components for removal, using second order blind identification (SOBI) and canonical correlation analyses. The results show that the resulting EEG signal after artifact removal for the AT and VT were comparable using a technique that measures the variance amongst electrodes and spectral energy. The AT technique is objective, faster and easier to use, and shown here to be comparable to the standard technique of visually detecting artifact component

Trieu, T., Willey, K. & Nguyen, H.T. 2009, 'Adaptive shared control strategies based on the Bayesian recursive technique for an intelligent wheelchair', IEEE Engineering in Medicine and Biology Society Annual Conference, Minneapolis, Minnesota, USA, September 2009 in Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Zhi-Pei Liang, IEEE, USA, pp. 7118-7121.
View/Download from: UTSePress
View description>>

In this paper we present an adaptive shared control method for an intelligent wheelchair based on the Bayesian recursive technique to assist a disable user in performing obstacle avoidance tasks. Three autonomous tasks have been developed for different types of environments to improve the performance of the overall system. The system combines local environmental information gathered using a laser range finder sensor with the userÔ++s intentions to select the most suitable autonomous task in different situations. The evidences of these tasks are estimated by the Bayesian recursive technique during movements of the wheelchair. The most appropriate task is chosen to be the with the highest evidence value. Experimental results show significant performance improvements compared to our previously reported shared control methods.

Weng, K., Nguyen, T., Nguyen, H.T. & Su, S.W. 2009, 'Rate estimation for the monitoring of rehabilitation exercises', IEEE Engineering in Medicine and Biology Society Annual Conference, Minneapolis, Minnesota, USA, September 2009 in Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Zhi-Pei Liang, et al, The Printing House, Inc., USA, pp. 6267-6270.
View/Download from: UTSePress | Publisher's site
View description>>

This study investigates the rate estimation problem encountered in rehabilitation exercise monitoring by using noninvasive portable sensors. The purpose of this paper has two main parts. The first part is to find suitable approaches for the rate detection of tri-axial accelerometer (TA) signals and ECG signals respectively. It is found that the integral type approaches (the average magnitude difference function (AMDF) and autocorrelation function (ACF)) are particularly suitable for TA signal pre-processing, while differential type approaches are very efficient for electrocardiographic (ECG) signal pre-processing. The second part is to develop a square wave matching method to detect the rate from the pre-processed signals. Experimental results indicate that the proposed methods can effectively detect pace rate from TA and heart rate from ECG and remove undesirable spikes.

Nguyen, H.T., Nguyen, J. & Nguyen, H.T. 2008, 'Bayesian Recursive Algorithm for Width Estimation of Freespace for a Power Wheelchair using Stereoscopic Cameras', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, Guy; Galiana, H., IEEE, Vancouver, Canada, pp. 4234-4237.
View/Download from: UTSePress | Publisher's site
View description>>

This paper is concerned with the estimation of freespace based on a Bayesian recursive (BR) algorithm for an autonomous wheelchair using stereoscopic cameras by severely disabled people. A stereo disparity map processed from both the left and right camera images is constructed to generate a 3D point map through a geometric projection algorithm. This is then converted to a 2D distance map for the purpose of freespace estimation. The width of freespace is estimated using a BR algorithm based on uncertainty information and control data. Given the probabilities of this width computed, a possible movement decision is then made for the mobile wheelchair. Experimental results obtained in an indoor environment show the effectiveness of this estimation algorithm.

Nguyen, H.T., Ghevondian, N. & Jones, T.W. 2008, 'Detection of Nocturnal Hypoglycemic Episodes (Natural Occurence) in Children with Type 1 Diabetes using an Optimal Bayesian Neural Network Algorithm', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, Guy; Galiana, H., IEEE, Vancouver, Canada, pp. 1311-1314.
View/Download from: UTSePress | Publisher's site
View description>>

Hypoglycemia or low blood glucose is dangerous and can result in unconsciousness, seizures and even death. It is a common and serious side effect of insulin therapy in patients with diabetes. HypoMon is a non-invasive monitor that measures some physiological parameters continuously to provide detection of hypoglycemic episodes in Type 1 diabetes mellitus patients (T1DM). Based on heart rate and corrected QT interval of the ECG signal, we have continued to develop Bayesian neural network detection algorithms to recognize the presence of hypoglycemic episodes. From a clinical study of 16 children with T1DM, natural occurrence of nocturnal hypoglycemic episodes are associated with increased heart rates (1.033-¦0.242 vs. 1.082-¦0.298, P<0.06) and increased corrected QT intervals (1.031-¦0.086 vs. 1.060-¦0.084, P<0.001). The overall data were organized into a training set (8 patients) and a test set (another 8 patients) randomly selected. Using the optimal Bayesian neural network with 10 hidden nodes which was derived from the training set with the highest log evidence, the sensitivity (true positive) value for detection of hypoglycemia in the test set is 89.2%.

Nguyen, T., Nguyen, H.T. & Su, S.W. 2008, 'Neuro-Sliding Mode Multivariable Control of a Powered Wheelchair', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, G; Galiana, H, IEEE, Vancouver, Canada, pp. 3471-3474.
View/Download from: UTSePress | Publisher's site
View description>>

This paper proposes a neuro-sliding mode multivariable control approach for the control of a powered wheelchair system. In the first stage, a systematic decoupling technique is applied to the wheelchair system in order to reduce the multivariable control problem into two independent scalar control problems. Then two Neuro-Sliding Mode Controllers (NSMCs) are designed for these independent subsystems to guarantee system robustness under model uncertainties and unknown external disturbances. Both off-line and on-line trainings are involved in the second stage. Realtime experimental results confirm that robust performance for this multivariable wheelchair control system under model uncertainties and unknown external disturbances can indeed be achieved.

Nguyen, T., Nguyen, H.T. & Su, S.W. 2008, 'Optimal Path-following Control of a Smart Powered Wheelchair', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, G; Galiana, H, IEEE, Vancouver, Canada, pp. 5025-5028.
View/Download from: UTSePress | Publisher's site
View description>>

This paper proposes an optimal path-following control approach for a smart powered wheelchair. LyapunovÔ++s second method is employed to find a stable position tracking control rule. To guarantee robust performance of this wheelchair system even under model uncertainties, an advanced robust tracking is utilised based on the combination of a systematic decoupling technique and a neural network design. A calibration procedure is adopted for the wheelchair system to improve positioning accuracy. After the calibration, the accuracy is improved significantly. Two real-time experimental results obtained from square tracking and door passing tasks confirm the performance of proposed approach.

Su, S.W., Nguyen, H.T. & Ha, Q.P. 2008, 'Integral Controller Design for Nonlinear Systems using Inverse Optimal Control', International Conference on Control, Automation, Robotics and Vision, Hanoi, Vietnam, December 2008 in Proceedings of the 10th International Conference on Control, Automation, Robotics and Vision, ed Wen, C, Research Publishinh Services, Singapore, pp. 2154-2158.
View/Download from: UTSePress | Publisher's site
View description>>

This paper proposes an integral controller design scheme for nonlinear systems based on optimal control and the passivity theorem in order to suppress the effect of external disturbances. The main strategy is to augment an optimal controller with a PI type controller. To guarantee the proposed controller has a desired stability margin, the passivity-based design method is introduced. Here, the inverse optimal control technique is employed to avoid the need of solving a Hamilton- Jacobi equation. An illustrative example is given to show the design procedure and the controller effectiveness.

Su, S.W., Celler, B.G., Savkin, A.V., Nguyen, H.T., Cheng, T.M., Guo, Y. & Wang, L. 2008, 'Portable sensor based dynamic estimation of human oxygen uptake via nonlinear multivariable modeling', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, G; Galiana, H, IEEE, Vancouver, Canada, pp. 2431-2434.
View/Download from: UTSePress | Publisher's site
View description>>

Noninvasive portable sensors are becoming popular in biomedical engineering practice due to its ease of use. This paper investigates the estimation of human oxygen uptake (VO2) of treadmill exercises by using multiple portable sensors (wireless heart rate sensor and triaxial accelerometers). For this purpose, a multivariable Hammerstein model identification method is developed. Well designed PRBS type of exercises protocols are employed to decouple the identification of linear dynamics with that of nonlinearities of Hammerstein systems. The support vector machine regression is applied to model the static nonlinearities. Multivariable ARX modelling approach is used for the identification of dynamic part of the Hammerstein systems. It is observed the obtained nonlinear multivariable model can achieve better estimations compared with single input single output models. The established multivariable model has also the potential to facilitate dynamic estimation of energy expenditure for outdoor exercises, which is the next research step of this study.

Tran, Y.H., Wijesuriya, N.S., Thuraisingham, R., Craig, A.R. & Nguyen, H.T. 2008, 'Increase in Regularity and Decrease in Variability seen in Electroencephalogram (EEG) Signals from Alert to Fatigue suring a Driving Simulated Task', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, G; Galiana, H, IEEE, Vancouver, Canada, pp. 1096-1099.
View/Download from: UTSePress | Publisher's site
View description>>

Driver fatigue is a prevalent problem and a major risk for road safety accounting for approximately 20-40% of all motor vehicle accidents. One strategy to prevent fatigue related accidents is through the use of countermeasure devices. Research on countermeasure devices has focused on methods that detect physiological changes from fatigue, with the fast temporal resolution from brain signals, using the electroencephalogram (EEG) held as a promising technique This paper presents the results of nonlinear analysis using sample entropy and second-order difference plots quantified by central tendency measure (CTM) on alert and fatigue EEG signals from a driving simulated task. Results show that both sample entropy and second-order difference plots significantly increases the regularity and decreases the variability of EEG signals from an alert to a fatigue state.

Trieu, T., Nguyen, H.T. & Willey, K. 2008, 'Advanced Obstacle Avoidance for a Laser-based Wheelchair using Optimised Bayesian Neural Networks', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, Guy; Galiana, H., IEEE, Vancouver, Canada, pp. 3463-3466.
View/Download from: UTSePress | Publisher's site
View description>>

In this paper we present an advanced method of obstacle avoidance for a laser based intelligent wheelchair using optimized Bayesian neural networks. Three neural networks are designed for three separate sub-tasks: passing through a door way, corridor and wall following and general obstacle avoidance. The accurate usable accessible space is determined by including the actual wheelchair dimensions in a real-time map used as inputs to each networks. Data acquisitions are performed separately to collect the patterns required for specified sub-tasks. Bayesian frame work is used to determine the optimal neural network structure in each case. Then these networks are trained under the supervision of Bayesian rule. Experiment results showed that compare to the VFH algorithm our neural networks navigated a smoother path following a near optimum trajectory.

Trieu, T., Nguyen, H.T. & Willey, K. 2008, 'Shared Control Strategies for Obstacle Avoidance Tasks in an Intelligent Wheelchair', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, G; Galiana, H, IEEE, Vancouver, Canada, pp. 3463-3466.
View/Download from: UTSePress | Publisher's site
View description>>

In this paper we present a method of shared control strategy for an intelligent wheelchair to assist a disable user in performing obstacle avoidance tasks. The system detects obstacles in front of the wheelchair using a laser range finder sensor. As the wheelchair moves the information from the laser range finder is combined with data from the encoders mounted in its driving wheels to build a 360-¦ real-time map. The accuracy of the map is improved by eliminating the systematic error that would result from both the uncertainty of effective wheelbase and unequal driving wheel diameters. The usable wheelchair accessible space is determined by including the actual wheelchair dimensions in producing the real-time map. In making a decision the shared control method considers the user's intentions via the head-movement interface, accessible space of the environment and user safety. The experiments show promising results in the intelligent wheelchair system.

Wan, S.H. & Nguyen, H.T. 2008, 'Human Computer Interaction using Hand Gesture', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, G; Galiana, H, IEEE, Vancouver, Canada, pp. 2357-2360.
View/Download from: UTSePress | Publisher's site
View description>>

Hand gesture is a very natural form of human interaction and can be used effectively in human computer interaction (HCI). This project involves the design and implementation of a HCI using a small hand-worn wireless module with a 3-axis accelerometer as the motion sensor. The small stand-alone unit contains an accelerometer and a wireless Zigbee transceiver with microcontroller. To minimize intrusiveness to the user, the module is designed to be small (3cm by 4 cm). A time-delay neural network algorithm is developed to analyze the time series data from the 3-axis accelerometer. Power consumption is reduced by the noncontinuous transmission of data and the use of low-power components, efficient algorithm and sleep mode between sampling for the wireless module. A home control interface is designed so that the user can control home appliances by moving through menus. The results demonstrate the feasibility of controlling home appliances using hand gestures and would present an opportunity for a section of the aging population and disabled people to lead a more independent life.

Wijesuriya, N.S., Tran, Y.H., Thuraisingham, R., Nguyen, H.T. & Craig, A.R. 2008, 'Effects of Mental Fatigue on 8-13 Hz Brain Activity in People with Spinal Cord Injury', IEEE Engineering in Medicine and Biology Society Annual Conference, Vancouver, Canada, August 2008 in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dumont, G; Galiana, H, IEEE, Vancouver, Canada, pp. 5716-5719.
View/Download from: UTSePress | Publisher's site
View description>>

Brain computer interfaces (BCIs) can be implemented into assistive technologies to provide Ô++hands-freeÔ++ control for the severely disabled. BCIs utilise voluntary changes in oneÔ++s brain activity as a control mechanism to control devices in the personÔ++s immediate environment. Performance of BCIs could be adversely affected by negative physiological conditions such as fatigue and altered electrophysiology commonly seen in spinal cord injury (SCI). This study examined the effects of mental fatigue from an increase in cognitive demand on the brain activity of those with SCI. Results show a trend of increased alpha (8-13Hz) activity in able-bodied controls after completing a set of cognitive tasks. Conversely, the SCI group showed a decrease in alpha activity due to mental fatigue. Results suggest that the brain activity of SCI persons are altered in its mechanism to adjust to mental fatigue. These altered brain conditions need to be addressed when using BCIs in clinical populations such as SCI. The findings have implications for the improvement of BCI technology.

Craig, D.A. & Nguyen, H.T. 2007, 'Adaptive EEG thought pattern classifier for advanced wheelchair control', IEEE Engineering in Medicine and Biology Society Annual Conference, Lyon, France, August 2007 in Proceedings of the 29th International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dittmar, A; Clark, J, IEEE, Lyon, France, pp. 2544-2547.
View/Download from: UTSePress | Publisher's site
View description>>

This paper presents a real-time Electroencephalogram (EEG) classification system, with the goal of enhancing the control of a head-movement controlled power wheelchair for patients with chronic Spinal Cord Injury (SCI). Using a 32 channel recording device, mental command data was collected from 10 participants. This data was used to classify three different mental commands, to supplement the five commands already available using head movement control. Of the 32 channels that were recorded only 4 were used in the classification, achieving an average classification rate of 82%. This paper also demonstrates that there is an advantage to be gained by doing adaptive training of the classifier. That is, customizing the classifier to a person previously unseen by the classifier caused their average recognition rates to improve from 52.5% up to 77.5%.

Nguyen, H.T., Ghevondian, N., Nguyen Thanh, S. & Jones, T.W. 2007, 'Detection of hypoglycemic episodes in children with type 1 diabetes using an optimal Bayesian neural network algorithm', IEEE Engineering in Medicine and Biology Society Annual Conference, Lyon, France, August 2007 in Proceedings of the 29th International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dittmar, A; Clark, J, IEEE, Lyon, France, pp. 3140-3143.
View/Download from: UTSePress | Publisher's site
View description>>

Hypoglycemia or low blood glucose is a common and serious side effect of insulin therapy in patients with diabetes. HypoMon is a non-invasive monitor that measures some physiological parameters continuously to provide detection of hypoglycemic episodes in Type 1 diabetes mellitus patients (T1DM). Based on heart rate, corrected QT interval of the ECG signal and skin impedance, a Bayesian neural network detection algorithm has been developed to recognize the presence of hypoglycemic episodes. From a clinical study of 25 children with T1DM, associated with hypoglycemic episodes, their heart rates increased (1.152-¦0.157 vs. 1.035-¦0.108, P<0.0001), their corrected QT intervals increased (1.088-¦0.086 vs. 1.020-¦0.062, P<0.0001) and their skin impedances reduced significantly (0.679-¦0.195 vs. 0.837-¦0.203, P<0.0001). The overall data were organized into a training set (14 cases) and a test set 14 cases) randomly selected. Using an optimal Bayesian neural network with 11 hidden nodes, and an algorithm developed from the training set, a sensitivity of 0.8346 and specificity of 0.6388 were achieved for the test set.

Nguyen, T., Nguyen, J., Pham, M.D. & Nguyen, H.T. 2007, 'Real-time obstacle detection for an autonomous wheelchair using stereoscopic cameras', IEEE Engineering in Medicine and Biology Society Annual Conference, Lyon, France, August 2007 in Proceedings of the 29th International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dittmar, A; Clark, J, IEEE, Lyon, France, pp. 4775-4778.
View/Download from: UTSePress | Publisher's site
View description>>

This paper is concerned with the development of a real-time obstacle avoidance system for an autonomous wheelchair using stereoscopic cameras by severely disabled people. Based on the left and right images captured from stereoscopic cameras mounted on the wheelchair, the optimal disparity is computed using the Sum of Absolute Differences (SAD) correlation method. From this disparity, a 3D depth map is constructed based on a geometric projection algorithm. A 2D map converted from this 3D map can then be employed to provide an effective obstacle avoidance strategy for this wheelchair. Experiment results obtained in a practical environment show the effectiveness of this real-time implementation.

Nguyen, T., Nguyen, H.T. & Su, S.W. 2007, 'Advanced robust tracking control of a powered wheelchair', IEEE Engineering in Medicine and Biology Society Annual Conference, Lyon, France, August 2007 in Proceedings of the 29th International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dittmar, A; Clark, J, IEEE, Lyon, France, pp. 4767-4770.
View/Download from: UTSePress | Publisher's site
View description>>

In this paper, the dynamic multivariable model of the wheelchair system is obtained including the presence of transportation lags. The triangular diagonal dominance (TDD) decoupling technique is applied to reduce this multivariable control problem into two independent scalar control problems. An advanced robust control technique for the wheelchair has been developed based on the combination of a TDD decoupling strategy and neural network controller design. The results obtained from the real-time implementation confirm that robust performance for this multivariable wheelchair control system can indeed be achieved.

Skinner, B., Nguyen, H.T. & Liu, D. 2007, 'Classification of EEG signals using a genetic-based machine learning classifier', IEEE Engineering in Medicine and Biology Society Annual Conference, Lyon, France, August 2007 in Proceedings of the 29th International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dittmar, A; Clark, J, IEEE, Lyon, France, pp. 3120-3123.
View/Download from: UTSePress | Publisher's site
View description>>

This paper investigates the efficacy of the geneticbased learning classifier system XCS, for the classification of noisy, artefact-inclusive human electroencephalogram (EEG) signals represented using large condition strings (108bits). EEG signals from three participants were recorded while they performed four mental tasks designed to elicit hemispheric responses. Autoregressive (AR) models and Fast Fourier Transform (FFT) methods were used to form feature vectors with which mental tasks can be discriminated. XCS achieved a maximum classification accuracy of 99.3% and a best average of 88.9%. The relative classification performance of XCS was then compared against four non-evolutionary classifier systems originating from different learning techniques. The experimental results will be used as part of our larger research effort investigating the feasibility of using EEG signals as an interface to allow paralysed persons to control a powered wheelchair or other devices.

Skinner, B., Nguyen, H.T. & Liu, D. 2007, 'Distributed classifier migration in XCS for classification of electroencephalographic signals', IEEE Congress on Evolutionary Computation, Singapore, September 2007 in Proceedings of the IEEE Congress on Evolutionary Computation, ed Srinivasan, D, IEEE, SIngapore, pp. 2829-2836.
View/Download from: UTSePress | Publisher's site
View description>>

This paper presents an investigation into combining migration strategies inspired by multi-deme parallel genetic algorithms with the XCS learning classifier system to provide parallel and distributed classifier migration. Migrations occur between distributed XCS classifier sub-populations using classifiers ranked according to numerosity, fitness or randomly selected. The influence of the degree-of-connectivity introduced by fully-connected, bi-directional ring and uni-directional ring topologies is examined. Results indicate that classifier migration is an effective method for improving classification accuracy, improving learning speed and reducing final classifier population size, in the single-step classification of noisy, artefact- inclusive human electroencephalographic signals. The experimental results will be used as part of our larger research effort investigating the feasibility of using EEG signals as an interface to allow paralysed persons to control a powered wheelchair or other devices.

Skinner, B., Nguyen, H.T. & Liu, D. 2007, 'Hybrid optimisation method using PGA and SQP algorithm', Symposium on Foundations of Computational Intelligence, Honolulu, Hawaii, USA, April 2007 in Proceedings of the IEEE Symposium on Foundations of Computational Intelligence, ed Fogel, D, IEEE, Hawaii, pp. 73-80.
View/Download from: UTSePress
View description>>

This paper investigates the hybridisation of two very different optimisation methods, namely the Parallel Genetic Algorithm (PGA) and Sequential Quadratic Programming (SQP) algorithm. The different characteristics of genetic-based and traditional quadratic programming-based methods are discussed and to what extent the hybrid method can benefit the solving of optimisation problems with nonlinear complex objective and constraint functions. Experiments show the hybrid method effectively combines the robust and global search property of Parallel Genetic Algorithms with the high convergence velocity of the Sequential Quadratic Programming Algorithm, thereby reducing computation time, maintaining robustness and increasing solution quality.

Tran, Y.H., Thuraisingham, R., Wijesuriya, N.S., Nguyen, H.T. & Craig, A.R. 2007, 'Detecting neural changes during stress and fatigue effectively: a comparison of spectral analysis and sample entropy', IEEE EMBS Neural Engineering Conference, Kohala Coast, Hawaii, USA, May 2007 in Proceedings of the 3rd International IEEE EMBS Conference on Neural Engineering of the IEEE Engineering in Medicine and Biology Society, ed Akay, M, IEEE, Kohala Coast, Hawaii, USA, pp. 350-353.
View/Download from: UTSePress | Publisher's site
View description>>

Brain computer interface (BCI) technology as its name implies, relies upon decoding brain signals into operational commands. Aside from needing effective means of control, successful BCIs need to remain stable in varying physiological conditions. BCIs need to be developed with mechanisms to recognise and respond to physiological states (such as stress and fatigue) that can disrupt user capability. This paper compares a spectral analysis of EEG signals technique with a nonlinear method of sample entropy to detect changes in brain dynamics during moments of stress and fatigue. The results demonstrated few changes in the spectral frequency bands of the EEG during fatigue and stress conditions. However, when the EEG signals were analysed with the nonlinear technique of sample entropy the results indicated a reduction of complexity during moments of fatigue and stress and an increase in complexity during moments of engagement to the task.

Tran, Y.H., Thuraisingham, R., Boord, P.R., Nguyen, H.T. & Craig, A.R. 2007, 'Using fractal analysis to improve switching rates in hands-free environmental control technology for the severely disabled', IEEE EMBS Neural Engineering Conference, Kohala, Hawaii, USA, May 2007 in Proceedings of the 3rd International IEEE EMBS Conference on Neural Engineering of the IEEE Engineering in Medicine and Biology Society, ed Akay, M, IEEE, Kohala Coast, Hawaii, USA, pp. 406-409.
View/Download from: UTSePress | Publisher's site
View description>>

A negative impact on the quality of life of the severely neurologically disordered such as spinal cord injured persons is the loss of the ability to control devices in their immediate environment. Consequently, we have conducted research on technology designed to restore some measure of independence by providing hands free control over these devices by using EEG signals associated with eye closure (EC) and eye opening (EO). In a previous study we demonstrated that the nonlinear technique fractal dimension analysis was a viable alternative to spectral analysis in detecting these signals in the EEG of able bodied persons. This paper explores the efficacy of using fractal dimension to detect EC/EO signals in a spinal cord injured population. The fractal dimension method was found to improve from the standard spectral analysis technique in that there was a significant reduction is the occurrence of false positive and false negative switching. This improved detection of EC/EO in the brain activity of severely disabled people will be utilised in our technology for remote switching of electrical devices.

Trieu, T., Nguyen, H.T. & Willey, K. 2007, 'Obstacle avoidance for power wheelchair using Bayesian neural network', IEEE Engineering in Medicine and Biology Society Annual Conference, Lyon, France, August 2007 in Proceedings of the 29th International Conference of the IEEE Engineering in Medicine and Biology Society, ed Dittmar, A; Clark, J, IEEE, Lyon, France, pp. 4771-4774.
View/Download from: UTSePress | Publisher's site
View description>>

In this paper we present a real-time obstacle avoidance algorithm using a Bayesian neural network for a laser based wheelchair system. The raw laser data is modified to accommodate the wheelchair dimensions, allowing the freespace to be determined accurately in real-time. Data acquisition is performed to collect the patterns required for training the neural network. A Bayesian frame work is applied to determine the optimal neural network structure for the training data. This neural network is trained under the supervision of the Bayesian rule and the obstacle avoidance task is then implemented for the wheelchair system. Initial results suggest this approach provides an effective solution for autonomous tasks, suggesting Bayesian neural networks may be useful for wider assistive technology applications.

Craig, D.A., Nguyen, H.T. & Burchey, H.A. 2006, 'Two Channel EEG thought pattern classifier', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, New York, USA, August 2006 in Proceedings of the 28th IEEE EMBS Annual International Conference, ed N/A, IEEE, New York, USA, pp. 1291-1294.
View/Download from: UTSePress | Publisher's site
View description>>

This paper presents a real-time electro-encephalogram (EEG) identification system with the goal of achieving hands free control. With two EEG electrodes placed on the scalp of the user, EEG signals are amplified and digitised directly using a ProComp+ encoder and transferred to the host computer through the RS232 interface. Using a real-time multilayer neural network, the actual classification for the control of a powered wheelchair has a very fast response. It can detect changes in the user's thought pattern in 1 second. Using only two EEG electrodes at positions O1 and C4 the system can classify three mental commands (forward, left and right) with an accuracy of more than 79 %

Ghevondian, N., Nguyen, H.T., Jones, T.W., Siafarikas, A. & Ratnam, N. 2006, 'Predicting Hypoglycemia Non-invasively in Type 1 Adolescent Diabetes Using the HypoMon', The American Diabetes Association Scientific Session, Washington, USA, June 2006 in The American Diabetes Association's 66th Scientific Session, ed N/A, ADA, USA, pp. 1-1.
View/Download from: UTSePress

King, L.M., Nguyen, H.T. & Lal, S. 2006, 'Early driver fatigue detection from electroncephalography signals using artificial neural networks', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, New York, USA, August 2006 in Proceedings of the 28th IEEE EMBS Annual International Conference, ed N/A, IEEE, New York, USA, pp. 2187-2190.
View/Download from: UTSePress | Publisher's site
View description>>

This paper describes a driver fatigue detection system using an artificial neural network (ANN). Using electroencephalogram (EEG) data sampled from 20 professional truck drivers and 35 non professional drivers, the time domain data are processed into alpha, beta, delta and theta bands and then presented to the neural network to detect the onset of driver fatigue. The neural network uses a training optimization technique called the magnified gradient function (MGF). This technique reduces the time required for training by modifying the standard back propagation (SBP) algorithm. The MGF is shown to classify professional driver fatigue with 81.49% accuracy (80.53% sensitivity, 82.44% specificity) and non-professional driver fatigue with 83.06% accuracy (84.04% sensitivity and 82.08% specificity)

Nguyen, A., Ha, Q.P. & Nguyen, H.T. 2006, 'Virtual-head robot tracking and three-point l-l control for multiple mobile robots', IEEE Workshop on Distributed Intelligent Systems, Prague, CZECH REPUBLIC, June 2006 in Dis 2006: Ieee Workshop On Distributed Intelligent Systems: Collective Intelligence And Its Applications, Proceedings, ed Ceballos, S, IEEE Computer Soc, Los Alamitos, USA, pp. 73-78.
View/Download from: UTSePress
View description>>

In the context of robotic formation control, the commonly-used virtual robot tracking combined with l-l control has limitations in the establishment of a line formation, the possibility of collision between robots, and the singularity cases involved. Thi

Nguyen, H.T., Ghevondian, N. & Jones, T.W. 2006, 'Neural-network detection of hypoglycemic episodes in children with type 1 diabetes using physiological responses', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, New York, USA, August 2006 in Proceedings of the 28th IEEE EMBS Annual International Conference, ed N/A, IEEE, New York, USA, pp. 6053-6056.
View/Download from: UTSePress | Publisher's site
View description>>

The most common and highly feared adverse effect of intensive insulin therapy in patients with diabetes is the increased risk of hypoglycemia. Symptoms of hypoglycemia arise from the activation of the autonomous central nervous systems and from reduced cerebral glucose consumption. HypoMon is a non-invasive monitor that measures some physiological parameters continuously to provide detection of hypoglycemic episodes in Type 1 diabetes mellitus patients (T1DM). Based on heart rate, corrected QT interval of the ECG signal and skin impedance, a neural network detection algorithm has been developed to recognize the presence of hypoglycemic episodes. From a clinical study of 21 children with T1DM, associated with hypoglycemic episodes, their heart rates increased (1.16plusmn0.16 vs. 1.03plusmn0.11, P<0.0001), their corrected QT intervals increased (1.09plusmn0.09 vs. 1.02plusmn0.07, P<0.0001) and their skin impedances reduced significantly (0.66plusmn0.19 vs. 0.82plusmn0.21, P<0.0001). The overall data were obtained and grouped into a training set, a validation set and a test set, each with 7 patients randomly selected. Using a feedforward multi-layer neural network with 9 hidden nodes, and an algorithm developed from the training set and the validation set, a sensitivity of 0.9516 and specificity of 0.4142 were achieved for the test set. A more advanced neural network algorithm will be developed to improve the specificity of test sets in the near future

Nguyen Thanh, S., Nguyen, H.T. & Taylor, P.W. 2006, 'Bayesian Neural Network Classification of Head Movement Direction Using Various Advanced Optimisation Training Algorithms', International Conference on Biomedical Robotics and Biomechatronics, Pisa, Italy, August 2006 in Proceedings of the 1st IEEE/RAS-EMBS 2006 International Conference on Biomedical Robotics and Biomechatronics, ed N/A, IEEE, Pisa, Italy, pp. 1-6.
View/Download from: UTSePress | Publisher's site
View description>>

Head movement is one of the most effective hands-free control modes for powered wheelchairs. It provides the necessary mobility assistance to severely disabled people and can be used to replace the joystick directly. In this paper, we describe the development of Bayesian neural networks for the classification of head movement commands in a hands-free wheelchair control system. Bayesian neural networks allow strong generalisation of head movement classifications during the training phase and do not require a validation data set. Various advanced optimisation training algorithms are explored. Experimental results show that Bayesian neural networks can be developed to classify head movement commands by abled and disabled people accurately with limited training data

Nguyen Thanh, S., Nguyen, H.T., Taylor, P.W. & Middleton, J.W. 2006, 'Improved Head Direction Command Classification Using An Optimised Bayesian Neural Network', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, New York, USA, August 2006 in Proceedings of the 28th IEEE EMBS Annual International Conference, ed N/A, IEEE, New York, USA, pp. 5679-5682.
View/Download from: UTSePress | Publisher's site
View description>>

Assistive technologies have recently emerged to improve the quality of life of severely disabled people by enhancing their independence in daily activities. Since many of those individuals have limited or non-existing control from the neck downward, alternative hands-free input modalities have become very important for these people to access assistive devices. In hands-free control, head movement has been proved to be a very effective user interface as it can provide a comfortable, reliable and natural way to access the device. Recently, neural networks have been shown to be useful not only for real-time pattern recognition but also for creating user-adaptive models. Since multi-layer perceptron neural networks trained using standard back-propagation may cause poor generalisation, the Bayesian technique has been proposed to improve the generalisation and robustness of these networks. This paper describes the use of Bayesian neural networks in developing a hands-free wheelchair control system. The experimental results show that with the optimised architecture, classification Bayesian neural networks can detect head commands of wheelchair users accurately irrespective to their levels of injuries

Craig, D.A. & Nguyen, H.T. 2005, 'Wireless Real-time Head Movement System using a Personal Digital Assistant (PDA) for Control of a Power Wheelchair', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Shanghai, China, September 2005 in Proceedings of the 27th Annual International Conference of the Engineering in Medicine and Biology Society, ed Zhang, Y.T., IEEE, Shanghai, China, pp. 6235-6238.
View/Download from: UTSePress | Publisher's site
View description>>

Loss of mobility can occur for a variety of reasons, such as spinal cord injury or motor neurone disease. The onset of these conditions often brings with it an associated loss of personal independence, which is primarily due to the fact that the sufferer is no longer able to control their mobility. This project aims to address this problem through the creation of a head movement based wheelchair control system. Using a personal digital assistant (PDA) artificial intelligence techniques on an embedded LINUX operating system, a wireless head movement wheelchair control system has been designed and implemented. This system provides relief for sufferers of conditions which inhibit mobility through a method of wheelchair control which offers enhanced ease of use, attractiveness and independence.

King, L.M., Nguyen, H.T. & Taylor, P.W. 2005, 'Hands-free Head-movement Gesture Recognition Using Artificial Neural Networks and the Magnified Gradient Function', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Shanghai, China, September 2005 in Proceedings of the 27th Annual International Conference of the Engineering in Medicine and Biology Society, ed Zhang, Y.T., IEEE, Shanghai, China, pp. 2063-2066.
View/Download from: UTSePress | Publisher's site
View description>>

This paper presents a hands-free head-movement gesture classification system using a neural network employing the magnified gradient function (MGF) algorithm. The MGF increases the rate of convergence by magnifying the first order derivative of the activation function, whilst guaranteeing convergence. The MGF is tested on able-bodied and disabled users to measure its accuracy and performance. It is shown that for able-bodied users, a classification improvement from 98.25% to 99.85% is made, and 92.08% to 97.50% for disabled users

Tran, T., Ha, Q.P. & Nguyen, H.T. 2005, 'A Cascade Sliding Mode-PID Controller for Non-overshoot Time Responses', International Symposium on Intelligent Technologies in Tech'05, Phuket, Thailand, December 2005 in Proceedings of the 6th International Symposium on Intelligent Technologies, ed Santiprahob; Pratit, Daengdej; Jirapun, Faculty of Science and Technology, Assumption University, Bangkok, Thailand, pp. 27-33.
View/Download from: UTSePress

Mitchell, R.A., Nguyen, H.T., Thornton, B.S., Hung, W., Lee, W.B. & Rickard, M.T. 2004, 'Mammogram Object Detection Using Dendronic Image Analysis', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Francisco, USA, September 2004 in Proceedings of the 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed The IEEE Engineering in Medicine and Biology Society, The Institute of Electrical and Electronics Engineers, Piscataway, USA, pp. 1763-1765.
View/Download from: UTSePress | Publisher's site
View description>>

Breast cancer can be treated with better patient outcomes and significantly lower costs if detected early. Using the spatial dendronic structure, image masks can be obtained, showing regions in the mammogram image corresponding to the breast and lead marker. The technique is robust to noise and placement of the breast within the image. The technique not only reduces the size of the region to be analysed, but also provides the dendronic structure of the breast in which stealth-like masses can be found more easily.

Nguyen, H.T., King, L.M. & Knight, G. 2004, 'Real-Time Head Movement System and Embedded Linux Implementation for the Control of Power Wheelchairs', Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Francisco, USA, September 2004 in Proceedings of the 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, ed The IEEE Engineering in Medicine and Biology Society, The Institute of Electrical and Electronics Engineers, Piscataway, New Jersey, USA, pp. 4892-4895.
View/Download from: UTSePress | Publisher's site
View description>>

Mobility has become very important for our quality of life. A loss of mobility due to an injury is usually accompanied by a loss of self-confidence. For many individuals, independent mobility is an important aspect of self-esteem. Head movement is a natural form of pointing and can be used to directly replace the joystick whilst still allowing for similar control. Through the use of embedded LINUX and artificial intelligence, a hands-free head movement wheelchair controller has been designed and implemented successfully. This system provides for severely disabled users an effective power wheelchair control method with improved posture, ease of use and attractiveness

Nguyen, V., Nguyen, H.T. & Ha, Q.P. 2004, 'Sliding Mode Neural Controller for Nonlinear Systems with Higher-Order and Uncertainties', IEEE Conference on Robotics, Automation and Mechatronics, Singapore, December 2004 in Proceedings of the 2004 IEEE Conference on Robotics, Automation and Mechatronics (RAM), ed Unknown, IEEE, Piscataway, USA, pp. 1026-1031.
View/Download from: UTSePress
View description>>

In this paper, we propose a new neural controller architecture which is derived from an adaptive sliding mode control framework for a SISO nonlinear system with higher-order and uncertainties. This neural controller can overcome some disadvantages inherent in sliding mode controllers such as the chattering problem, complex calculation of the equivalent control term and unavailable knowledge of the upper bounds of system uncertainties. Experimental results for a coupled drives CE8 system show that a real-time neural controller has been implemented successfully.

Nguyen Thanh, S., Nguyen, H.T. & Taylor, P.W. 2004, 'Hands-Free Control of Power Wheelchairs Using Bayesian Neural Network Classification', IEEE International Conference on Cybernetics and Intelligent Systems, Singapore, December 2004 in Proceedings of the 2004 IEEE Conference on Cybernetics and Intelligent Systems (CIS2004), ed Unknown, IEEE, Piscataway, USA, pp. 745-749.
View/Download from: UTSePress | Publisher's site
View description>>

This paper describes the formulation and implementation of Bayesian neural networks for head-movement classification in a hands-free wheelchair navigation system. Bayesian neural network training adjusts the weight decay parameters automatically to their near-optimal values that give the best generalisation. Moreover, no separate validation set is used so all available data can be used for training. Experimental results are presented showing that Bayesian neural network can classify the head movement accurately

Skinner, B., Nguyen, H.T. & Liu, D. 2004, 'Performance Study of a Multi-Deme Parallel Genetic Algorithm with Adaptive Mutation', International Conference on Autonomous Robots and Agents, Palmerston North, New Zealand, December 2004 in Proceedings of the 2nd International Conference on Autonomous Robots and Agents (ICARA'04), ed Mukhopadhyay, S; and Sen Gupta, G, Massey University, Palmerston North, New Zealand, pp. 88-94.
View/Download from: UTSePress