Integrating Deep Learning for Heart and Vascular Acoustic Analysis in Cardiovascular Health Assessment

Date

2023-12-07

Author

Özçil, Ceyda

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

332
views

134
downloads

Cite This

Atherosclerosis, a cardiovascular disease, disrupts blood flow due to occlusions. The transformation from laminar into turbulent flow produces an acoustic phenomena known as bruits. In this study, heart sounds recorded by phonocardiography were classified as normal and abnormal using different combinations of feature extraction and classification techniques. An experiment-based model was employed to generate pulsating flow sound at different stenosis levels. Deep learning and feature comparison methodologies were applied to explore the correlation between phonocardiography and vascular sounds. Beyond promising results in heart sound classification, the study demonstrated an apparent relationship between phonocardiography recordings and 50-90\% stenosed vascular sounds. This outcome highlights that coronary artery disease could be detected by utilizing the phonocardiography.

Subject Keywords

Heart Sound Classification, Phonocardiography, Stenosis Detection

URI

https://hdl.handle.net/11511/107744

Collections

Graduate School of Natural and Applied Sciences, Thesis