It is expected that an automatic detection and classification algorithm for the abnormities of first heart sound (S1) can realize computer artificial intelligence diagnosis of some relative cardiovascular disease. Few studies have focused on the detection and classification of the abnormities of S1 and given out in detail the essential differences between abnormal and normal S1. This work applied Empirical Wavelet Transform (EWT) to decompose S1 and extracted the instantaneous frequency (IF) of mitral component (M1) and tricuspid component (T1) by using Hilbert Transform. Firstly, the heart sound signal is preprocessed following these processes: filtering, resampling, normalization and segmentation. Secondly, S1 is decomposed into several modes based on EWT. First two maximal points with a distance greater than 20Hz in Fourier Spectrum of S1 are selected and the nearest minimal points on both sides of the maximal points are found out as the boundaries for segmentation of the spectrum. S1 is decomposed into 5 modes and every mode’s IF are calculated through Hilbert transformation. At last, a k-mean cluster algorithm is applied to cluster the IF of different modes. TD and $A_{peak\_{}ratio}$ are calculated for decision tree classifier and S1s are divided into three categories: normal S1, S1 with abnormal split and S1with abnormal amplitude change. When the proposed method is applied to detect normal S1, Se=94.6%, Pp=98.6% and Oa=93.3%; When it is applied to detect S1 with abnormal split, Se=92.6%, Pp=96.9% and Oa=90%; When it is applied to detect S1 with abnormal amplitude change, Se=94.4%, Pp=95.7% and Oa=90.6%; Comparison experiments are carried out between the proposed method and HVD method. The results show Oa of the proposed method is higher than HVD method when detecting the three different S1s.

/pdf/detection-and-classification-of-abnormities-of-first-heart-4g146fqean.pdf

Detection and Classification of Abnormities of First Heart Sound Using Empirical Wavelet Transform

Abstrac:Presented a new method of recognizing and sorting of heart sound signal, using heart sound signal analysis and improved back propagation artificial Neural Network. Firstly, the heart sound signal has to be digitally filtered to eliminating the noise. The neural network adopted is improved back propagation three-layer artificial Neural Network, with an optimized BPN6-3-2 network topology. Get the characteristic vector of six component through ARMA model from filtered heart sound signal to act as the input parameters of the input layer. After about 1200 times training of 20 samples, the network can correctly distinguishes coronary heart disease sample from non-coronary heart disease samples. Meanwhile, this network can also be applied to sort the 28 non-trained heart sound signal samples correctly but two samples. The results of experiment show, that the method of recognizing and sorting heart sound signal discussed in this paper, apart from two exceptions, the rest of the samples is fully consistent with the clinic diagnosis. So it is a new, promising, non-invasive method of diagnosing coronary heart disease.

The Research of Non-Invasive Method of Coronary Heart Disease Based on Neural Network and Heart Sound Signals

Damage Identification for Transmission Towers Based on HHT

Studying the features of network traffic is a necessary method in many applications such as managing and monitoring the network, detecting network attack etc. Due to the non-stationary and random characteristics of the traffic signal, Hilbert-Huang Transformation(HHT) is an effective method in signal processing. However, endpoints effect comes subsequently in HHT. In this particular case, the paper introduces a multiple residual error gray forecast method to realize data continuation so as to inhibiting endpoints effect in applying HHT to network traffic signal. Experimental results show that this method can effectively restrain the endpoint divergence phenomenon in applying HHT to network traffic signal, but it is not applicable to deterministic signals.

Applying Multiple Residual Error Gray Forecast to Restrain Endpoints Effect in HHT of Network Traffic

The heart sound signal, as a kind of weak biological signal under the background of strong noise, is easily subject to interference from noise of various sources. De-noising of heart sound signals, therefore, forms the primary basis for achieving non-invasive diagnosis of coronary heart disease. The paper proposes the five-level wavelet decomposition method for heart sound signals using Daubechies 6 wavelet (db6), which has yielded satisfactory results. A bi-spectrum estimation of the de-noised heart sound signals is performed based on the ARMA model, with appropriately and meticulously selected parameters. Experiments conducted on a total of 36 subjects, one half having healthy hearts and the other half afflicted with coronary heart disease, indicate that the db6-based decomposition method is capable of satisfactorily differentiating normal heart sound signals from abnormal ones.

A New Method of Sorting of Heart Sound Signal Based on Wavelet Transform and Parameter Model Method

According to the valvular theory, the vibrations affected by the mitral and tricuspid valves closure in early systole produce the first heart sound (S1). S1 usually includes many frequency components. Analysis of the first heart sound will be helpful to diagnose the clinic heart diseases. In this paper, a new method using the Hilbert-Huang Transform(HHT) was used to decompose S1 into a series of time-frequency atoms. By examining 30 cases of heart sound data, It has being proved the HHT is an efficiency method to study the heart sound, S1 consists of two major high-frequency components, M1 and T1,the abnormal bioprosthetic mitral valve generates higher frequency components than the normal one.

Liang Hong

Papers

Time-frequency analysis of the first heart sound based on the hilbert-huang transform