Sheikh Hussain Shaikh Salleh

Analysis and classification of heart sounds and murmurs based on the instantaneous energy and frequency estimations

This paper proposes the use of the instantaneous energy and the frequency estimation in the classification of the heart sounds and murmurs for common heart diseases. It has been known that the present of the heart murmurs in ones heart sound indicates that there is a potential heart problem. Thus, the goal of this work is to develop a technique for detecting and classifying murmurs. Such a technique can be used as part of a heart diagnostic system. The analysis is performed based on a set of 102 data for various heart sounds. To discriminate the various heart sounds, the instantaneous energy and frequency estimation is used to estimate the features of heart sound. The techniques used to estimate the instantaneous frequency are the central finite difference frequency estimation (CFDFE) and zero crossing frequency estimation (ZCFE). From the instantaneous energy and frequency estimate, the energy and frequencies of the heart sounds are defined as the features of the heart sounds that can uniquely discriminate the various heart sounds.

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.

Analysis of heart sounds based on continuous wavelet transform

This paper presents the application of wavelet transform analysis method to the heart sounds signal. The heart sounds is a non-stationery signal, thus it is very important to study the frequency and time information. One of the time-frequency analysis methods is short time Fourier transforms. However, the STFT analysis is limited by the time and frequency resolution. The wavelet transform was introduced to curb the resolution problem in STFT. The wavelet transform is a multi-resolution time-scale analysis that gives high resolution for low frequency components and low resolution for high frequency components. Since majority of heart sounds component lies in low frequency, thus the application of wavelet transform to heart sounds is very suitable. Results in time-frequency representation clearly show that the wavelet transform is capable to distinguish between the normal with a few types of abnormal heart sounds. The murmurs caused by particular heart diseases such as aortic regurgitation, aortic stenosis, mitral regurgitation, mitral stenosis, pulmonary regurgitation and tricuspid regurgitation were clearly shown under continuous wavelet representation.

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.

Application of Genetic Algorithm in unit selection for Malay speech synthesis system

Corpus based speech synthesis can produce high quality synthetic speech due to it high sensitivity to unit context. Large speech database is embedded in synthesis system and search algorithm (unit selection) is needed to search for the optimal unit sequence. Speech feature which served as target cost is estimated from the input text. The acoustic parameters which served as join cost are derived from mel frequency cepstral coefficients (MFCCs) and Euclidean distance. In this paper, a new method which is Genetic Algorithm is proposed to search for optimal unit sequence. Genetic Algorithm (GA) is a population based search algorithm that is based on the biological principles of selection, reproduction, crossover and mutation. It is a stochastic search algorithm for solving optimization problem. The speech unit sequence that has minimum join cost will be synthesized into complete waveform data.

Speech pitch detection using short-time energy

Research on pitch detection for speech has been done and still ongoing since there is not one algorithm found that perfectly detects the pitch. This paper will show the sum of square energy method in detecting the voiced/unvoiced speech since they have different level of energy. The threshold was found through experiments to find the unvoiced energy level from candidates with various words consist of unvoiced phonemes. Pitch detection algorithm was then implemented and the percentage of pitch detected in words was evaluated to test the accuracy of the algorithm proposed in this paper.

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.

Speaker-independent phonation recognition for Malay plosives using neural networks

The paper investigates the use of neural networks in recognizing the phonation of the speech sounds. The proposed method classifies the Malay plosive sounds of adults and children based on phonation in a speaker-independent manner. The proposed method achieves encouraging result with an average accuracy of 98%.

Window width estimation and the application of the windowed Wigner-Ville distribution in the analysis of heart sounds and murmurs

This paper looks at the analysis of heart sounds and murmur using time-frequency signal analysis. The techniques used are the Wigner-Ville distribution (WVD) and windowed Wigner-Ville distribution (WWVD) that belonged to the bilinear class of time-frequency distribution. These techniques developed to provide high-resolution time-frequency representation for time-varying signals. Due the nonlinear operation involved, interference terms are introduced in the time-frequency representation. The signals of interest are modeled as multicomponent signals and the characteristics of the signal in time-lag plane are observed. From the time-lag plane, the interference components are identified, and the appropriate window width is selected in the WWVD to remove the interference. Analysis results show that WWVD produces more accurate time-frequency representation compared to the WVD and the signal-to-interference is used to quantify the improvement.