Heart sound classification using Gaussian mixture model

Abstract

Background: This article represents a new method of classifying the heart sound status using the loudness features from the heart sound. Materials and methods: The method includes the following 3 main steps. First, the heart sound, which is usually found noisy, is heavily filtered by a 6th-order Chebyshev-I filter. The heart sound is then segmented using the event synchronous method to separate the sounds into the first heart sound, the systole and the second heart sound, the diastole. In the second step, the heart sound features namely maximum loudness index and minimum loudness index are obtained from the spectrogram of the sound by taking the row means. As a third step, the heart sound is classified using the Gaussian mixture model approach to categorize the sounds. Results: This method has been tested on a very large database of heart sounds consisting of over 3000 heart sounds recordings with a success rate of 97.77%. Conclusion: Only 2 features are used in this method namely, minimum loudness index and maximum loudness index. Classification of sounds using these 2 features yields high accuracy even under noisy conditions and is comparable to any state-of-the-art technique.

Abnormal Heart Sound Recognition using SVM and LSTM Models in Real-time Mode

“…In a nutshell, we can notice that the proposed approach achieves better accuracy than Mei et al (2021), Shervegar andBhat (2018), andBao et al (2022) approaches using a single SVM. The results in Figure 145 shows that the proposed approach achieves an accuracy of 96.0% for two SVMs and 98.1% for a single SVM compared to 97.77%, 94.07 %, and 92.23% for Shervegar and Bhat (2018), Bao et al (2022), andAl-Mei et al (2021) respectively, with a decrease of about 1.77 for two SVMs and an enhancement of about 0.33% for a single SVM.…”

“…In a nutshell, we can notice that the proposed approach achieves better accuracy than Mei et al (2021), Shervegar andBhat (2018), andBao et al (2022) approaches using a single SVM. The results in Figure 145 shows that the proposed approach achieves an accuracy of 96.0% for two SVMs and 98.1% for a single SVM compared to 97.77%, 94.07 %, and 92.23% for Shervegar and Bhat (2018), Bao et al (2022), andAl-Mei et al (2021) respectively, with a decrease of about 1.77 for two SVMs and an enhancement of about 0.33% for a single SVM. By applying classical approaches for heart sound recognition, the proposed method can detect six common abnormal sounds, and to the best of our knowledge this is the first work that combines the PASCAL and the GitHub datasets' abnormalities by achieving an overall accuracy of 98.6% using the LSTM classifier, and 99.3% using the SVM classifier.…”

A Comprehensive Survey on Heart Sound Analysis in the Deep Learning Era

“…A 6-th order Butterworth filter was designed with a cutoff frequency of 50-950 Hz [47], [48] and 30-900 Hz in [49]. Additionally, several other filters were also used for denoising heart sound, such as a Savitzky-Golay filter [50], [51], Chebyshev low-pass filter [52], [53], and Notch filter [54]. Spectrum-based denoising.…”

“…Loudness-based segmentation. Loudness has proven its potential to segment heart sounds [53], [65]. Specifically, spectrograms extracted from heart sounds are firstly converted into the Bark scale and smoothed with a Hanning window.…”

“…The Naïve Bayes Classifier [41], [91], [102], [103] was widely used for heart sound classification due to its advantage of being easy-touse. Gaussian Mixture Models (GMMs) [53], [95] were used to estimate the data distribution by optimising the weights of Gaussian mixture components and mean and variance in each component. A Gaussian mixture-based HMM [38] was employed for heart sound classification considering the four sequential heart states, i. e., S1, systole, S2, and diastole.…”