Showing papers by "Choo Min Lim published in 2011"

Application of Higher Order Spectra to Identify Epileptic EEG

Abstract: Epilepsy is characterized by the spontaneous and seemingly unforeseeable occurrence of seizures, during which the perception or behavior of patients is disturbed. An automatic system that detects seizure onsets would allow patients or the people near them to take appropriate precautions, and could provide more insight into this phenomenon. Various methods have been proposed to predict the onset of seizures based on EEG recordings. The use of nonlinear features motivated by the higher order spectra (HOS) has been reported to be a promising approach to differentiate between normal, background (pre-ictal) and epileptic EEG signals. In this work, we made a comparative study of the performance of Gaussian mixture model (GMM) and Support Vector Machine (SVM) classifiers using the features derived from HOS and from the power spectrum. Results show that the selected HOS based features achieve 93.11% classification accuracy compared to 88.78% with features derived from the power spectrum for a GMM classifier. The SVM classifier achieves an improvement from 86.89% with features based on the power spectrum to 92.56% with features based on the bispectrum.

Human Cardiovascular Model and Applications

Abstract: Cardiovascular diseases (CVDs) can be known as a class of diseases which affect different parts of the cardiovascular system such as the heart or blood vessels. Hemodynamic signals are an important tool used by doctors to diagnose the type of CVD occurred in a patient. Diagnosing the correct type of CVD in a patient early will allow the patient to have the suitable medical treatment. Some examples of CVDs include coronary heart disease, cerebrovascular disease and peripheral arterial disease. A human cardiovascular model is developed in order to simulate different hemodynamic signals of the cardiovascular system. The hemodynamic signals include the blood pressures, flow rates and volumes in various part of the cardiovascular system. This paper presents a model which is able to simulate hemodynamic signals and they are able to represent the human arterial blood pressure accurately. Hence this model can also be used to simulate hypertensive patients in order to design control systems for regulation of blood pressure. Signal verification has been performed and the stability of the model is being investigated. Applications of the human cardiovascular model are also presented.

FES Controller Design Based on Threshold Control Theory for Single Joint Movement

Abstract: The threshold control theory is applied to build a physiology-based FES controller. The proposed control strategy is tested in simulation study with an integrated musculoskeletal model. The results suggest that the controller based on the threshold control theory alone can realize the task with small feedback delays. However, it is not capable to produce the movement fast enough to match the desired trajectory. Thus two extensions to improve the performance are proposed. Firstly, the feedback on actual movement velocity is replaced by the differences between the actual and desired velocities. Secondly, the feedback of joint angular acceleration is introduced to calculate the dynamic threshold. Simulation results suggest that such extensions can 1) improve the movement speed; 2) reduce the response time; and 3) reduce influence of external perturbation and feedback delays.

An adaptive PI algorithm for regulation of blood pressure of hypertension patients

Abstract: It is known that postsurgical hypertension is common in cardiac patients and untreated hypertension patients and may result in complications. To prevent this side effect, regulation of the blood pressure for this kind of patient is necessary. The aim of an automatic drug control system is to quickly reduce the oscillatory change in mean blood pressure through infusion of sodium nitroprusside (SNP). In this paper, a new adaptive PI algorithm updating variations in time-delay and sensitivity of the patient is proposed and its effectiveness is analysed in detail. Simulation under clinical conditions is also carried out and the results show that this control system can handle the changes in patient's dynamics.