Heart sound classification based on improved mel frequency cepstrum coefficient and integrated decision network method / 生物医学工程学杂志

Heart sound analysis is significant for early diagnosis of congenital heart disease. A novel method of heart sound classification was proposed in this paper, in which the traditional mel frequency cepstral coefficient (MFCC) method was improved by using the Fisher discriminant half raised-sine function (F-HRSF) and an integrated decision network was used as classifier. It does not rely on segmentation of the cardiac cycle. Firstly, the heart sound signals were framed and windowed. Then, the features of heart sounds were extracted by using improved MFCC, in which the F-HRSF was used to weight sub-band components of MFCC according to the Fisher discriminant ratio of each sub-band component and the raised half sine function. Three classification networks, convolutional neural network (CNN), long and short-term memory network (LSTM), and gated recurrent unit (GRU) were combined as integrated decision network. Finally, the two-category classification results were obtained through the majority voting algorithm. An accuracy of 92.15%, sensitivity of 91.43%, specificity of 92.83%, corrected accuracy of 92.01%, and F score of 92.13% were achieved using the novel signal processing techniques. It shows that the algorithm has great potential in early diagnosis of congenital heart disease.

Heart sound classification based on sub-band envelope and convolution neural network / 生物医学工程学杂志

Automatic classification of heart sounds plays an important role in the early diagnosis of congenital heart disease. A kind of heart sound classification algorithms based on sub-band envelope feature and convolution neural network was proposed in this paper, which did not need to segment the heart sounds according to cardiac cycle accurately. Firstly, the heart sound signal was divided into some frames. Then, the frame level heart sound signal was filtered with Gammatone filter bank to obtain the sub-band signals. Next, the sub-band envelope was extracted by Hilbert transform. After that, the sub-band envelope was stacked into a feature map. Finally, type Ⅰ and type Ⅱ convolution neural network were selected as classifier. The result shown that the sub-band envelope feature was better in type Ⅰ than type Ⅱ. The algorithm is tested with 1 000 heart sound samples. The test results show that the overall performance of the algorithm proposed in this paper is significantly improved compared with other similar algorithms, which provides a new method for automatic classification of congenital heart disease, and speeds up the process of automatic classification of heart sounds applied to the actual screening.

Segmentation of heart sound signals based on duration hidden Markov model / 生物医学工程学杂志

Heart sound segmentation is a key step before heart sound classification. It refers to the processing of the acquired heart sound signal that separates the cardiac cycle into systolic and diastolic, etc. To solve the accuracy limitation of heart sound segmentation without relying on electrocardiogram, an algorithm based on the duration hidden Markov model (DHMM) was proposed. Firstly, the heart sound samples were positionally labeled. Then autocorrelation estimation method was used to estimate cardiac cycle duration, and Gaussian mixture distribution was used to model the duration of sample-state. Next, the hidden Markov model (HMM) was optimized in the training set and the DHMM was established. Finally, the Viterbi algorithm was used to track back the state of heart sounds to obtain S

Classification of heart sound signals in congenital heart disease based on convolutional neural network / 生物医学工程学杂志

Cardiac auscultation is the basic way for primary diagnosis and screening of congenital heart disease(CHD). A new classification algorithm of CHD based on convolution neural network was proposed for analysis and classification of CHD heart sounds in this work. The algorithm was based on the clinically collected diagnosed CHD heart sound signal. Firstly the heart sound signal preprocessing algorithm was used to extract and organize the Mel Cepstral Coefficient (MFSC) of the heart sound signal in the one-dimensional time domain and turn it into a two-dimensional feature sample. Secondly, 1 000 feature samples were used to train and optimize the convolutional neural network, and the training results with the accuracy of 0.896 and the loss value of 0.25 were obtained by using the Adam optimizer. Finally, 200 samples were tested with convolution neural network, and the results showed that the accuracy was up to 0.895, the sensitivity was 0.910, and the specificity was 0.880. Compared with other algorithms, the proposed algorithm has improved accuracy and specificity. It proves that the proposed method effectively improves the robustness and accuracy of heart sound classification and is expected to be applied to machine-assisted auscultation.

Research on pretreatment and envelope extraction algorithm of heart sound signal / 生物医学工程学杂志

In this work, a new method of heart sound signal preprocessing is presented. First, the heart sound signals are decomposed by using multilayer wavelet transform. And then double parameters as thresholds are used in processing each layer after decomposition for denoising. Next, reconstruction of heart sound signals could be done after processing last layer. Four methods, i.e. wavelet transform, Hilbert-Huang transform (HHT), mathematical morphology, and normalized average Shannon energy, were used to extract the envelop of the heart sound signals respectively after reconstruction of heart sounds. All methods were improved in this study. We finally in our study chose 30 cases of raw heart sound signals, which were selected randomly from a database comed from The Clinical Medicine Institute of Montreal, and processed them by using the improved methods. The results were satisfactory. It showed that the extracted envelope with the original signal has a high degree of matching, whether it is a low frequency portion or high frequency portion. Most of all information of heart sound has been maintained in the envelope.

Application and comparison of continuous wavelet transform and matching pursuit method in analyzing phonocardiographic signals / 生物医学工程学杂志

Heart sounds are highly valuable to the clinical diagnoses of most cardiovascular diseases, so the analysis of phonocardiographic signals is helpful to diagnosing cardiovascular diseases clinically. Phonocardiographic signals are non-stable, so it is necessary to choose appropriate method in time-frequency analysis. The traditional method such as Fourier Transform is dissatisfactory. Continuous Wavelet Transform (CWT) and Matching (MPM) Pursuit Method are both effective methods. They can be used to extract and cluster the characteristics of the signals. By analysis and comparison, the two methods showed the advantages over traditional methods. Additionally, their respective merits and demerits are indicated.

Application of independent component analysis for separating phonocardiogram signals / 生物医学工程学杂志

Independent component analysis (ICA) is a novel method developed in recent years for Blind Source Separation. In this paper, the phonocardiogram (PCG) was separated into three components by applying ICA. The basic principle of ICA was introduced in this paper. A fast and robust fixed-point algorithm for ICA was used to analyze PCG signals in this study. The experiments showed that ICA could separate the components of heart sounds from PCG signals successfully.

Frequency detection of the first heart sound based on wavelet transformation / 生物医学工程学杂志

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. In this paper, a method using the multi-resolution analysis of wavelet transformation is recommended for detecting the frequency range of S1. First, S1 was decomposed into different levels on frequency. Then the normalized Shannon energy of the different levels was calculated. The level containing the maximum energy is the major components' level of S1. The frequency range of this level is the major frequency range of S1. The frequency range of S1 was successfully detected by the method.