Temporal Convolutional Network Connected with an Anti-Arrhythmia Hidden Semi-Markov Model for Heart Sound Segmentation

Yibo Yin,Ming Liu,Kainan Ma

doi:10.3390/app10207049

Abstract

Heart sound segmentation (HSS) is a critical step in heart sound processing, where it improves the interpretability of heart sound disease classification algorithms. In this study, we aimed to develop a real-time algorithm for HSS by combining the temporal convolutional network (TCN) and the hidden semi-Markov model (HSMM), and improve the performance of HSMM for heart sounds with arrhythmias. We experimented with TCN and determined the best parameters based on spectral features, envelopes, and one-dimensional CNN. However, the TCN results could contradict the natural fixed order of S1-systolic-S2-diastolic of heart sound, and thereby the Viterbi algorithm based on HSMM was connected to correct the order errors. On this basis, we improved the performance of the Viterbi algorithm when detecting heart sounds with cardiac arrhythmias by changing the distribution and weights of the state duration probabilities. The public PhysioNet Computing in Cardiology Challenge 2016 data set was employed to evaluate the performance of the proposed algorithm. The proposed algorithm achieved an F1 score of 97.02%, and this result was comparable with the current state-of-the-art segmentation algorithms. In addition, the proposed enhanced Viterbi algorithm for HSMM corrected 30 out of 30 arrhythmia errors after checking one by one in the dataset.

Highlights

Cardiovascular diseases are the main noncommunicable diseases and they contribute to more deaths than all other causes combined [1]
The activation function used in the temporal convolutional network (TCN) networks was ReLU [28], and the activation function employed in the final fully-connected layer was Softmax
We proposed an algorithm for segmenting heart sounds using a CNN-based sequence processing architecture, the temporal convolutional networks, which can allow real-time operation with high performance and low computational complexity

Summary

Introduction

Cardiovascular diseases are the main noncommunicable diseases and they contribute to more deaths than all other causes combined [1]. Cardiac auscultation is the most common and cost-effective noninvasive screening method for heart conditions. Auscultation is a difficult technique that requires sufficient experience. Around 20% of medical interns can effectively perform auscultation [2,3]. Automatic algorithms for auscultation can make the subjective experience of the auscultation technique objective and simplify the mastery of auscultation. The segmentation of heart sounds is important for improving algorithms for classifying heart sound diseases [4]. No sounds appear during systole and diastole, but heart

Objectives

Methods

Results

Discussion

Conclusion