Mohammed Bahoura

Wavelet speech enhancement based on the Teager energy operator

We propose a new speech enhancement method based on the time adaption of wavelet thresholds. The time dependence is introduced by approximating the Teager energy of the wavelets coefficients. This technique does not require an explicit estimation of the noise level or of the a priori knowledge of the SNR, which is usually needed in most of the popular enhancement methods. Performance of the proposed method is evaluated on speech recorded in real conditions and with artificial noise.

DSP implementation of wavelet transform for real time ECG wave forms detection and heart rate analysis

An algorithm based on wavelet transform (WTs) suitable for real time implementation has been developed in order to detect ECG characteristics. In particular, QRS complexes, P and T waves may be distinguished from noise, baseline drift or artefacts. This algorithm is implemented in a DSP (SPROC-1400) with a 50 MHz frequency clock. The performance of this algorithm is discussed, its accuracy is evaluated and a comparison is made with a similar algorithm implemented in C language. For the standard MIT/BIH arrhythmia database, this algorithm correctly detects 99.7% of the QRS complexes.

Pattern recognition methods applied to respiratory sounds classification into normal and wheeze classes

In this paper, we present the pattern recognition methods proposed to classify respiratory sounds into normal and wheeze classes. We evaluate and compare the feature extraction techniques based on Fourier transform, linear predictive coding, wavelet transform and Mel-frequency cepstral coefficients (MFCC) in combination with the classification methods based on vector quantization, Gaussian mixture models (GMM) and artificial neural networks, using receiver operating characteristic curves. We propose the use of an optimized threshold to discriminate the wheezing class from the normal one. Also, post-processing filter is employed to considerably improve the classification accuracy. Experimental results show that our approach based on MFCC coefficients combined to GMM is well adapted to classify respiratory sounds in normal and wheeze classes. McNemars test demonstrated significant difference between results obtained by the presented classifiers (p<0.05).

An integrated automated system for crackles extraction and classification

Abstract This paper presents an integrated automated system for crackles recognition. This system comprises three serial modules with following functions: (1) separation of crackles from vesicular sounds using a wavelet packet filter (WPST–NST); (2) detection of crackles by fractal dimension (FD); (3) classification of crackles based on Gaussian mixture models (GMM). The WPST–NST filter incorporates a multi-resolution decomposition of the original respiratory signal and an entropy-based best basis selection of the coefficients. Two thresholds are defined, in time and frequency domains respectively, to separate the crackles from the respiratory sounds. Then, a denoising filter is applied to the discontinuous output of WPST–NST and a crackle-peak-detector (CPD) localizes the individual crackles by means of their fractal dimension. After that, three feature parameters, including the Gaussian bandwidth (GBW), the peak frequency (PF) and the maximal deflection width (MDW), of the crackles are extracted. Finally, crackles are classified into fine crackles (FC) and coarse crackles (CC) using Gaussian mixture models.

Wavelet speech enhancement based on time-scale adaptation

We propose a new speech enhancement method based on time and scale adaptation of wavelet thresholds. The time dependency is introduced by approximating the Teager energy of the wavelet coefficients, while the scale dependency is introduced by extending the principle of level dependent threshold to wavelet packet thresholding. This technique does not require an explicit estimation of the noise level or of the a priori knowledge of the SNR, as is usually needed in most of the popular enhancement methods. Performance of the proposed method is evaluated on speech recorded in real conditions (plane, sawmill, tank, subway, babble, car, exhibition hall, restaurant, street, airport, and train station) and artificially added noise. MEL-scale decomposition based on wavelet packets is also compared to the common wavelet packet scale. Comparison in terms of signal-to-noise ratio (SNR) is reported for time adaptation and time-scale adaptation of the wavelet coefficients thresholds. Visual inspection of spectrograms and listening experiments are also used to support the results. Hidden Markov Models speech recognition experiments are conducted on the AURORA-2 database and show that the proposed method improves the speech recognition rates for low SNRs.

Respiratory sounds classification using cepstral analysis and Gaussian mixture models

The cepstral analysis is proposed with Gaussian mixture models (GMM) method to classify respiratory sounds in two categories: normal and wheezing. The sound signal is divided in overlapped segments, which are characterized by a reduced dimension feature vectors using Mel-frequency cepstral coefficients (MFCC) or subband based cepstral parameters (SBC). The proposed schema is compared with other classifiers: vector quantization (VQ) and multi-layer perceptron (MLP) neural networks. A post processing is proposed to improve the classification results.

Separation of Crackles From Vesicular Sounds Using Wavelet Packet Transform

Crackles are discontinuous adventitious lung sounds with an explosive and transient character, and occur frequently in pulmonary diseases. They are the most important parameter for the diagnosis. This paper presents a new filter for automatic separation of the crackles from vesicular sounds. The proposed filter is based on the wavelet packet transform and applies two thresholds, which are defined in time-domain and frequency-domain respectively, to wavelet coefficients to achieve the separation task. This filter is more accurate and efficient compared to its rivals. Experimental results are given in detail and demonstrate its excellent performance

FPGA-implementation of a sequential adaptive noise canceller using Xilinx System Generator

This paper presents a sequential architecture of a pipelined LMS-based adaptive noise cancellation to remove the power-line interference (50/60 Hz) from electrocardiogram (ECG). This architecture is implemented on on FPGA using XUP Virtex-II Pro development board and Xilinx System Generator (XSG). The proposed architecture was evaluated using real ECG signals from the MIT-BIH database. Hardware requirement of this adaptive noise canceller is presented for various filter lengths.

New parameters for respiratory sound classification

In this paper, a new approach based on cepstral analysis is proposed to classify respiratory sounds. The sound signal is divided into segments, which are characterized by a reduced number of cepstral coefficients. Those segments are then classified as whether containing wheezes or normal respiratory sounds, by using the vector quantization (VQ) method. This approach is tested and compared to other kind of features extraction like the autoregressive representation and the wavelet transform.

Respiratory sounds classification using Gaussian mixture models

The Gaussian mixture models (GMM) method is proposed to classify respiratory sounds in two categories: normal and wheezing. The sound signal is divided in overlapped segments, which are characterized by reduced dimension feature vectors using cepstral or wavelet transforms. The proposed method is compared with other classifiers: vector quantization (VQ) and multilayer perceptron (MLP) neural networks. A post processing is proposed to improve the test results.