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

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.

Phonocardiogram Segmentation with Tiny Computing

“…A wide range of feature extraction techniques for PCG segmentation have been explored in related works. A Fourier synchro-squeezed transform (FSST) is an accurate technique that has been used in conjunction with a Bidirectional Long Short-Term Memory (BiLSTM) neural network [4]. This work adopted a similar approach utilising a series of short-time Fourier transforms (STFTs) for data pre-processing due to their simplicity and the availability of open-source implementations of STFT optimized for STM32 MCUs.…”

“…PCG segmentation techniques studied in previous works are diverse and include algorithmic techniques, statistical models and deep learning models. The latter utilised convolutional neural networks (CNNs) [2], long short-term memory networks (LSTMs) [3] and temporal convolutional networks (TCNs) [4].…”

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

“…The determination of the abnormal part of the heart sound can only be achieved when the state of the heart sound is determined, which can be done by heart sound segmentation. The common methods for heart sound segmentation mainly includes ECG signal-based segmentation methods [4][5][6], envelope-based segmentation methods [7][8][9][10][11], feature-based segmentation methods [12][13][14][15], machine learning-based segmentation methods [16][17][18][19][20], and Hidden Markov Model (HMM)-based segmentation methods [16,[21][22][23][24]. In earlier times, there has been only some achievements made in segmentation methods based on ECG signals and envelopes.…”