Phonocardiography (PCG)-based early detection of heart anomalies using deep learning

Hasan, Awatif; Bahri, Zouhir

doi:10.1049/icp.2023.0663

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, Reed et al [20] provided a computeraided diagnosis method based on a Coifman fourth-order wavelet kernel and a seven-level wavelet decomposition that could be applied to five distinct clinical scenarios. In this paper, we provide an extension to our preliminary study [21] with detailed illustrations for the extensive testing and methods applied. The purpose of this work is to provide comparative research in order to integrate some of the major methodologies mentioned above in a system that is reasonably accurate, easy to use, and cost-effective.…”

Section: Introductionmentioning

confidence: 99%

Comparative Study on Heart Anomalies Early Detection Using Phonocardiography (PCG) Signals

Hasan,

Bahri

2023

IJCDS

View full text Add to dashboard Cite

In this paper, a Phonocardiography (PCG)-based comparative study for cardiovascular anomalies' early detection system is proposed. Some of the main signal processing and deep learning methods applied in the literature to PCG signals are contrasted to differentiate normal heartbeats from abnormal ones and classify five of the most common murmurs, the "lub" and "dub" of "lub-dub" are formed by combining the typical heartbeat sound S1 and S2. The results of this comparative study show an average of 92.14 % for heart anomaly detection and 71.02 % for classification rates. This is achieved by using Deep Neural Network (DNN) classification with Hyperbolic Tangent (tanh) activation function, a 5-layer with 100 neurons in each layer. The Discrete Wavelet Transform (DWT) was found to be the best denoising algorithm and the Heart Sound Envelogram (HSE) was the best segmentation method for the PCG signal. Mel Frequency Cepstral Coefficients (MFCC) Features outperformed their Time and Frequency Domain counterparts. This work proved to be useful in the framework of intelligent and preventative health care systems, offering a convenient early warning home-care tool that should help to direct potentially ill individuals to cardiologists for more precise diagnoses.

show abstract

Section: Introductionmentioning

confidence: 99%