Automatic heart sound classification from segmented/unsegmented phonocardiogram signals using time and frequency features

Abstract

Objective: Heart abnormality detection using heart sound signals (phonocardiogram (PCG)) has been an active research area for the last few decades. In this paper, automatic heart sound classification using segmented and unsegmented PCG signals is presented. Approach: In this paper: (i) we perform an in-depth analysis of various time and frequency domain features, followed by experimental determination of effective feature subsets for improved classification performance; (ii) both segmented and unsegmented PCG signals are studied and important results concerning the respective feature subsets and their classification performances are reported; and (iii) different classification algorithms, including the support vector machine, kth nearest neighbor, decision tree, ensemble classifier, artificial neural network and long short-term memory network (LSTMs), are employed to evaluate the performance of the proposed feature subsets and their comparison with other established features and methods is presented. Main results: It is observed that LSTM performs better on mel-frequency cepstral coefficient (MFCC) features extracted from unsegmented PCG data, with an area under curve (AUC) score of 91.39%, however, the MFCC features do not show a consistent performance with other classifiers (the second highest AUC score is 62.08% with the decision tree classifier). In contrast, in the case of time-frequency features from segmented data, the performance of all the classifiers is appreciable with AUC scores over 70%. In particular, the conventional machine learning techniques shows consistency in achieving over 80% in AUC scores. Significance: The results of this study highlight the importance of time and frequency domain features. Thus it is necessary to employ both the time and frequency features of segmented PCG signals to achieve improved classification.

Deep Learning Methods for Heart Sounds Classification: A Systematic Review

“…Heart sound signals are a kind of sequential data with a strong temporal correlation and can thus be suitably processed by RNNs. Indeed, they have been proven to be very effective and are commonly used for heart sounds classification [46][47][48][49][50].…”

“…However, compared with other non-deep learning methods, it eliminates the tedious process of manual feature extraction. Khan et al [47] used LSTM in combination with the MFCC features of the unsegmented data and achieved the best area under the curve score of 91.39% in the PhysioNet/CinC Challenge, performing better than other algorithms such as SVM, KNN, decision tree, and ANN for various time and frequency-domain features. Similarly, Raza et al [49] used the LSTM model to classify three kinds of heart sounds-namely, normal, murmur, and extrasystole-in PASCAL heart sound dataset B.…”

“…It should be noted that the utilized datasets were balanced, with each class of heart sound containing 200 recordings. Most deep learning-based methods do not utilize a segmentation algorithm to identify S1, S2, systole, and diastole heart sounds, such as [25][26][27][28]30,31,[33][34][35][36][37][38]44,45,[47][48][49][50][51]54,55]. The methods are nevertheless very efficient for automatic heart sounds classification.…”

Research on Segmentation and Classification of Heart Sound Signals Based on Deep Learning

“…They compared the classification results of segmented and unsegmented heart sound signals and concluded that using segmented heart sound signals can contributes to better classification. However, in the experiment, they used the improved empirical wavelet transformation and standardized Shannon average energy to preprocess and automatically segment the signals to identify the systolic and diastolic interval of the signal, instead of the segmentation of the four states [35]. An example of a normal heart sound includes two heart sound cycles, and each cycle consists of the following heart sound components: S1, sys, S2, and dia.…”

Rheumatic Heart Disease Screening Based on Phonocardiogram

“…In most cases, the incorrect localization of the FHS was partially blamed for the poor classification performance. Based on this, researchers proposed a classification approach without the requirement of segmentation [ 32 , 33 , 34 , 35 , 36 ]. This paper proposes a classification system for heart sounds that does not require segmentation.…”