Chee Ming Ting

Spectral Estimation of Nonstationary EEG Using Particle Filtering With Application to Event-Related Desynchronization (ERD)

This paper proposes non-Gaussian models for parametric spectral estimation with application to event-related desynchronization (ERD) estimation of nonstationary EEG. Existing approaches for time-varying spectral estimation use time-varying autoregressive (TVAR) state-space models with Gaussian state noise. The parameter estimation is solved by a conventional Kalman filtering. This study uses non-Gaussian state noise to model autoregressive (AR) parameter variation with estimation by a Monte Carlo particle filter (PF). Use of non-Gaussian noise such as heavy-tailed distribution is motivated by its ability to track abrupt and smooth AR parameter changes, which are inadequately modeled by Gaussian models. Thus, more accurate spectral estimates and better ERD tracking can be obtained. This study further proposes a non-Gaussian state space formulation of time-varying autoregressive moving average (TVARMA) models to improve the spectral estimation. Simulation on TVAR process with abrupt parameter variation shows superior tracking performance of non-Gaussian models. Evaluation on motor-imagery EEG data shows that the non-Gaussian models provide more accurate detection of abrupt changes in alpha rhythm ERD. Among the proposed non-Gaussian models, TVARMA shows better spectral representations while maintaining reasonable good ERD tracking performance.

ECG based personal identification using extended Kalman filter

This paper proposes a new approach for electrocardiogram (ECG) based personal identification based on extended Kalman filtering (EKF) framework. The framework uses nonlinear ECG dynamic models formulated to represent noisy ECG signal. The advantage of the models is the ability to capture distinct ECG features used for biometric recognition such as temporal and amplitude distances between PQRST points. Moreover the inherent modeling of additive noise provides robust recognition. Log-likelihood scoring is proposed for classification. The algorithm is evaluated on identification task on 13 subjects of MIT-BIH Arrhythmia Database using single lead data. Identification rate of 87.50% is achieved on 30s test recordings of normal beat. Experimental results using artificial additive white noise show that the model is robust to noise for SNR level above 20dB.

A Unified Estimation Framework for State-Related Changes in Effective Brain Connectivity

Objective: This paper addresses the critical problem of estimating time-evolving effective brain connectivity. Current approaches based on sliding window analysis or time-varying coefficient models do not simultaneously capture both slow and abrupt changes in causal interactions between different brain regions. Methods: To overcome these limitations, we develop a unified framework based on a switching vector autoregressive (SVAR) model. Here, the dynamic connectivity regimes are uniquely characterized by distinct vector autoregressive (VAR) processes and allowed to switch between quasi-stationary brain states. The state evolution and the associated directed dependencies are defined by a Markov process and the SVAR parameters. We develop a three-stage estimation algorithm for the SVAR model: 1) feature extraction using time-varying VAR (TV-VAR) coefficients, 2) preliminary regime identification via clustering of the TV-VAR coefficients, 3) refined regime segmentation by Kalman smoothing and parameter estimation via expectation-maximization algorithm under a state-space formulation, using initial estimates from the previous two stages. Results: The proposed framework is adaptive to state-related changes and gives reliable estimates of effective connectivity. Simulation results show that our method provides accurate regime change-point detection and connectivity estimates. In real applications to brain signals, the approach was able to capture directed connectivity state changes in functional magnetic resonance imaging data linked with changes in stimulus conditions, and in epileptic electroencephalograms, differentiating ictal from nonictal periods. Conclusion: The proposed framework accurately identifies state-dependent changes in brain network and provides estimates of connectivity strength and directionality. Significance: The proposed approach is useful in neuroscience studies that investigate the dynamics of underlying brain states.

Face identification and verification using PCA and LDA

Algorithms based on PCA (Principal Components Analysis) and LDA (Linear Discriminant Analysis) are among the most popular appearance-based approaches in face recognition. PCA is recognized as an optimal method to perform dimension reduction, yet being claimed as lacking discrimination ability. LDA once proposed to obtain better classification by using class information. Disputes over the comparison of PCA and LDA have motivated us to study their performance. In this paper, we describe both of these statistical subspace methods and evaluated them using The Database of Faces which comprises 40 subjects with 10 images each. Both identification and verification results have revealed the superiority of LDA over PCA for this medium-sized database.

Malay Sign Language Gesture Recognition system

This paper describes hardware design and sensors setting and configuration for Malay sign language gesture recognition systems. A set of sensors consists of accelerometers and flexure sensors has been setup to capture the movement or gesture of shoulder, elbow, wrist, palm and fingers. This project is based on the need of developing an electronic device that can translate the sign language into speech (sound) in order to enable communication to take place between the mute and deaf community with the common public. Hence, the main objective of this project is to develop a system that can convert the sign language into speech so that deaf people are able to communicate efficiently with normal people. This Human-Computer Interaction system is able to recognize 25 common words signing in Bahasa Isyarat Malaysia (BIM) by using the Hidden Markov Models (HMM) method. Both hands are involved to perform the BIM with all the sensors connect wirelessly to a PC with a Bluetooth module. This project aims to capture the hand gestures which involve multiple axis of movement. Altogether 24 sensors have been setup in different hand locations to capture hand and wrist movement in different directions.

Estimating Effective Connectivity from fMRI Data Using Factor-based Subspace Autoregressive Models

We consider the problem of identifying large-scale effective connectivity of brain networks from fMRI data. Standard vector autoregressive (VAR) models fail to estimate reliably networks with large number of nodes. We propose a new method based on factor modeling for reliable and efficient high-dimensional VAR analysis of large networks. We develop a subspace VAR (SVAR) model from a factor model (FM), where observations are driven by a lower-dimensional subspace of common latent factors with an AR dynamics. We consider two variants of principal components (PC) methods that provide consistent estimates for the FM hence the implied SVAR model, even of large dimensions. Information criterion is used to select the optimal subspace dimension. We established asymptotic normality and convergence rates for the estimated SVAR coefficients matrix. Evaluation on simulated resting-state fMRI shows that the SVAR models are more robust and produce better connectivity estimates than the classical model for a moderately-large network analysis. Results on real data by varying the subspace dimensions identify strong connections in the default mode network and reveal hierarchical connectivity of resting-state networks with distinct functional relevance.

ESTIMATING DYNAMIC CORTICAL CONNECTIVITY FROM MOTOR IMAGERY EEG USING KALMAN SMOOTHER & EM ALGORITHM

This paper considers identifying effective cortical connectivity from scalp EEG. Recent studies use time-varying multivariate autoregressive (TV-MAR) models to better describe the changing connectivity between cortical regions where the TV coefficients are estimated by Kalman filter (KF) within a state-space framework. We extend this approach by incorporating Kalman smoothing (KS) to improve the KF estimates, and the expectation-maximization (EM) algorithm to infer the unknown model parameters from EEG. We also consider solving the volume conduction problem by modeling the induced instantaneous correlations using a full noise covariate. Simulation results show the superiority of KS in tracking the coefficient changes. We apply two derived frequency domain measures i.e. TV partial directed coherence (TV-PDC) and TV directed transfer function (TV-DTF), to investigate dynamic causal interactions between motor areas in discriminating motor imagery (MI) of left and right hand. Event-related changes of information flows around beta-band, in a unidirectional way between left and right hemispheres are observed during MI. A difference in inter-hemispheric connectivity patterns is found between left and right-hand movements, implying potential usage for BCI.

PCA, LDA and neural network for face identification

Algorithms based on Principal Component Analysis (PCA) and subspace Linear Discriminant Analysis (LDA) are popular in face recognition. PCA is used to perform dimension reduction on human face data and LDA creates another subspace to improve discriminant of PCA features. In this paper, we propose Artificial Neural Networks (ANN) as an alternative to replace Euclidean distances in classification of human face features extracted by PCA and LDA. ANN is well recognized by its robustness and good learning ability. The algorithms were evaluated using the Database of Faces which comprises 40 subjects and with a total size of 400 images. Experimental results show that ANN reasonably improves the performance of PCA and LDA method. LDA-NN achieves an average recognition accuracy of 95.8%.

Text independent Speaker Identification using Gaussian mixture model

This paper describes text-independent (TI) Speaker Identification (ID) using Gaussian mixture models (GMM). The use of GMM approach is motivated by that the individual Gaussian components of a GMM are shown to represent some general speaker-dependent spectral shapes that are effective for speaker identity modeling. For speaker model training, a fast re-estimation algorithm based on highest likelihood mixture clustering is introduced. In this work, the GMM is evaluated on TI Speaker ID task via series of experiments (model convergence, effect of feature set, number of Gaussian components, and training utterance length on identification rate). The database consisted of Malay clean sentence speech database uttered by 10 speakers (3 female and 7 male). Each speaker provides the same 40 sentences utterances (average length- 3.5s) with different text. The sentences for testing were different from those for training. The GMM achieved 98.4% identification rate using 5 training sentences. The model training based on highest likelihood clustering is shown to perform comparably to conventional expectation-maximization training but consumes much shorter computational time.

Modeling Effective Connectivity in High-Dimensional Cortical Source Signals

To study the effective connectivity among sources in a densely voxelated (high-dimensional) cortical surface, we develop the source-space factor VAR model. The first step in our procedure is to estimate cortical activity from multichannel electroencephalograms (EEG) using anatomically constrained brain imaging methods. Following parcellation of the cortical surface into disjoint regions of interest (ROIs), latent factors within each ROI are computed using principal component analysis. These factors are ROI-specific low-rank approximations (or representations) which allow for efficient estimation of connectivity in the high-dimensional cortical source space. The second step is to model effective connectivity between ROIs by fitting a VAR model jointly on all the latent processes. Measures of cortical connectivity, in particular partial directed coherence, are formulated using the VAR parameters. We illustrate the proposed model to investigate connectivity and interactions between cortical ROIs during rest.