Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/17115
DC FieldValue
dc.titleBrain signal processing and neurological therapy
dc.contributor.authorPAN YAOZHANG
dc.date.accessioned2010-05-13T19:33:06Z
dc.date.available2010-05-13T19:33:06Z
dc.date.issued2009-08-24
dc.identifier.citationPAN YAOZHANG (2009-08-24). Brain signal processing and neurological therapy. ScholarBank@NUS Repository.
dc.identifier.urihttp://scholarbank.nus.edu.sg/handle/10635/17115
dc.description.abstractIn this thesis, neurological therapies which consist of advanced engineering technologies such as brain imaging, signal processing, pattern recognition, intelligent control, and advanced robotics are presented for motivating future development of neurological therapies. By investigating the characteristics of these neurological disorders, pattern recognition based brain signal processing approaches, and multimodal human robot interaction (HRI) based advanced robotics are presented for neurological therapies of these neurological disorders. The first application is the detection and prevention of epilepsy. For detection of epileptic seizures, a new electroencephalography (EEG)-based brain state identification method is presented. Several statistical features which are specifically suited for detection of epileptic spike waves are derived and support vector machine (SVM) is used to classify the low dimensional features. It is illustrated by experimental evaluation that the proposed method is a promising way for automatic seizure detection. Once epileptic states are identified from normal states of epilepsy patients, the problem of controlling the synaptic plasticity to constrain bursting activity in epileptic seizures can be addressed by a direct drug injection or electrical stimulation of related brain region. With a good understanding of dynamical changes in the brain during seizures onset and the mechanisms that cause these changes, a model based control is designed to develop close-loop stimulation system for brain states restoration in epileptic seizures onset. Numerical simulations are carried out to illustrate the effectiveness of the proposed controls. Another important application is stroke rehabilitation. Recently, brain computer interface (BCI)-based robotic rehabilitation is introduced which directly translates brain signals that involve motor or mental imagery into commands for controlling the robot and bypasses the normal motor output neural pathways. In this work, a human-friendly interactive robot is developed as a visual and motion feedback for BCI system to help the patients to be more cognitively engaged in rehabilitative training process. For the BCI system, a feature fusion of common spatial pattern (CSP) and autoregressive (AR) spectral analysis is proposed to extract features from EEG signal with left hand movement imagination or right hand movement imagination for further classification of these two brain states. Quadratic discriminant analysis (QDA) is utilized as classifier for the combined feature vectors. The feature fusion method is proved to outperform each of the single-feature extraction algorithms in motor imagery BCI system through both off-line and real-time experiments. Finally, social therapy of autism is studied based on some well-developed hypothesis of cognitive and social science. An interactive robot, RoBear, is developed with multimodal HRI to help autistic children become more socially engaged. Under the multimodal HRI framework we proposed in this study, RoBear is able to identify the face and voice, and sensitive to the emotional change of the human working with it. Scale invariant neighborhood linear embedding (SINLE) is proposed for sound source recognition motivated by neighborhood linear embedding (NLE) and scale adaptation of humanb s perception. Weighted locally linear embedding (WLLE) motivated by weighted distance measurement and locally linear embedding (LLE) is proposed for feature extraction of face images to obtain more compact and low-dimensional representations. WLLE is demonstrated to outperform several well-known face recognition algorithms through extensive experiments. During the interaction between child and robot, the robot will elicit physical and psychological states of the child, followed by therapy of management according to social norms.
dc.language.isoen
dc.subjectBrain Signal Processing, Neurological Therapy, Pattern Recognition, Human Robot Interaction
dc.typeThesis
dc.contributor.departmentELECTRICAL & COMPUTER ENGINEERING
dc.contributor.supervisorGE SHUZHI
dc.contributor.supervisorABDULLAH AL MAMUN
dc.description.degreePh.D
dc.description.degreeconferredDOCTOR OF PHILOSOPHY
dc.identifier.isiutNOT_IN_WOS
Appears in Collections:Ph.D Theses (Open)

Show simple item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Pan Yaozhang _PhD thesis_final.pdf2.42 MBAdobe PDF

OPEN

NoneView/Download

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.