Classification of heart sounds based on quality assessment and wavelet scattering transform

Mei, Na; Wang, Hongxia; Zhang, Yatao; Liu, Feifei; Jiang, Xinge; Wei, Shoushui

doi:10.1016/j.compbiomed.2021.104814

Cited by 36 publications

(24 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, it is crucial to process the original heart sound signal through feature engineering before feeding it into the neural network for training. There are several commonly used feature extraction methods in heart sound classification tasks, including discrete wavelet transform coefficients (DWT) Mei et al (2021) , and Mel frequency cepstral coefficients (MFCC) Yang and Hsieh (2016) . In this paper, the MFCC-based first and second-order difference coefficients are used as the input tensor of the neural network.…”

Section: Introductionmentioning

confidence: 99%

Heart sound classification based on improved mel-frequency spectral coefficients and deep residual learning

Zhang

Wang

et al. 2022

Front. Physiol.

View full text Add to dashboard Cite

Heart sound classification plays a critical role in the early diagnosis of cardiovascular diseases. Although there have been many advances in heart sound classification in the last few years, most of them are still based on conventional segmented features and shallow structure-based classifiers. Therefore, we propose a new heart sound classification method based on improved mel-frequency cepstrum coefficient features and deep residual learning. Firstly, the heart sound signal is preprocessed, and its improved features are computed. Then, these features are used as input features of the neural network. The pathological information in the heart sound signal is further extracted by the deep residual network. Finally, the heart sound signal is classified into different categories according to the features learned by the neural network. This paper presents comprehensive analyses of different network parameters and network connection strategies. The proposed method achieves an accuracy of 94.43% on the dataset in this paper.

show abstract

Section: Introductionmentioning

confidence: 99%

Heart sound classification based on improved mel-frequency spectral coefficients and deep residual learning

Zhang

Wang

et al. 2022

Front. Physiol.

View full text Add to dashboard Cite

show abstract

“…In the next year, Mei [14] introduced a CVD audio classification based on wavelet scattering transform (WST). The dataset used is taken from Computing in Cardiology (CinC) [21] The proposed method was structured as Quality Assessment, Wavelet Scattering, Audio Classification, and Voting.…”

Section: Related Workmentioning

confidence: 99%

“…However, a manual investigation and monitoring of foetal heartbeat are time-consuming. For these reasons, artificial intelligence (AI) approaches surge to the surface for tackling such issues [13,14] Many works have been introduced using machine learning [15][16][17][18]. Their works were based on a single modal feature like MFCC.…”

Section: Introductionmentioning

confidence: 99%

Arrhythmia Foetus Heartbeat Detection Using Optimized Neural Network Based on Phonocardiograph Ensemble Feature and Principal Component Analysis

2023

IJIES

View full text Add to dashboard Cite

High-risk maternal health condition is alarming, especially in developing countries. Intensive monitoring is mandatory to prevent such issue. However, the long-term invasive method to pregnant women harms both the baby and the mother. In this research, we proposed a cost-efficient non-invasive foetal heartbeat classification based on a phonocardiograph with feature assembly. Since the high number of features and computationally expensive, we cut the size to half by utilizing Principal Component Analysis. Furthermore, data balancing using SMOTE is incorporated to improve classification performance. We proposed a method based on a neural network and optimized it using Random Search optimization. Eventually, the proposed method gained the top position in all data balancing compared to other machine learning algorithms, with 91.7 % for both accuracy and Area Under Curve with a score at 91.6 %.

show abstract

“…These preprocessing methods provided an (Anden and Mallat, 2014). The wavelet scattering method has been wildly used for acoustic scene classification (Li et al, 2019), speech recognition (Fousek et al, 2015;Joy et al, 2020), and heart sound classification (Mei et al, 2021), which yielded efficient representations for audio processing. However, wavelet scattering currently was seldom used in ECG analysis and application.…”

Section: Introductionmentioning

confidence: 99%

“…First-order scattering coefficients characterize persistent phenomena such as tendency and envelope, while second-order scattering coefficients characterize transient phenomena such as shock signals and amplitude modulation ( Anden and Mallat, 2014 ). The wavelet scattering method has been wildly used for acoustic scene classification ( Li et al, 2019 ), speech recognition ( Fousek et al, 2015 ; Joy et al, 2020 ), and heart sound classification ( Mei et al, 2021 ), which yielded efficient representations for audio processing. However, wavelet scattering currently was seldom used in ECG analysis and application.…”

Section: Introductionmentioning

confidence: 99%

Wearable Electrocardiogram Quality Assessment Using Wavelet Scattering and LSTM

et al. 2022

Self Cite

View full text Add to dashboard Cite

As the fast development of wearable devices and Internet of things technologies, real-time monitoring of ECG signals is quite critical for cardiovascular diseases. However, dynamic ECG signals recorded in free-living conditions suffered from extremely serious noise pollution. Presently, most algorithms for ECG signal evaluation were designed to divide signals into acceptable and unacceptable. Such classifications were not enough for real-time cardiovascular disease monitoring. In the study, a wearable ECG quality database with 50,085 recordings was built, including A/B/C (or high quality/medium quality/low quality) three quality grades (A: high quality signals can be used for CVD detection; B: slight contaminated signals can be used for heart rate extracting; C: heavily polluted signals need to be abandoned). A new SQA classification method based on a three-layer wavelet scattering network and transfer learning LSTM was proposed in this study, which can extract more systematic and comprehensive characteristics by analyzing the signals thoroughly and deeply. Experimental results (mACC = 98.56%, mF1 = 98.55%, SeA = 97.90%, SeB = 98.16%, SeC = 99.60%, +PA = 98.52%, +PB = 97.60%, +PC = 99.54%, F1A = 98.20%, F1B = 97.90%, F1C = 99.60%) and real data validations proved that this proposed method showed the high accuracy, robustness, and computationally efficiency. It has the ability to evaluate the long-term dynamic ECG signal quality. It is advantageous to promoting cardiovascular disease monitoring by removing contaminating signals and selecting high-quality signal segments for further analysis.

show abstract

Classification of heart sounds based on quality assessment and wavelet scattering transform

Cited by 36 publications

References 32 publications

Heart sound classification based on improved mel-frequency spectral coefficients and deep residual learning

Heart sound classification based on improved mel-frequency spectral coefficients and deep residual learning

Arrhythmia Foetus Heartbeat Detection Using Optimized Neural Network Based on Phonocardiograph Ensemble Feature and Principal Component Analysis

Wearable Electrocardiogram Quality Assessment Using Wavelet Scattering and LSTM

Contact Info

Product

Resources

About