A Cross-Platform Smartphone Auscultation SDK and Optimized Filters for Severe Aortic Stenosis Detection
Files
Date
2025-01-07
Contributor
Advisor
Department
Instructor
Depositor
Speaker
Researcher
Consultant
Interviewer
Narrator
Transcriber
Annotator
Journal Title
Journal ISSN
Volume Title
Publisher
Volume
Number/Issue
Starting Page
3559
Ending Page
Alternative Title
Abstract
Initial studies suggest that valve replacement may also benefit asymptomatic patients with severe aortic stenosis, who don't typically seek medical attention and thus require screening. As echocardiography, the current gold standard, is time-intensive and hence costly, a more convenient and broadly accessible alternative would be desirable. We present a cross-platform smartphone auscultation software development kit (SDK) for Android and iOS that uses the built-in microphone to record heart sounds. Our initial exploration shows that such recordings can detect 89% of severe aortic stenosis patients, compared to 95% for a digital stethoscope. In addition, we tackle the issue of smartphone audio quality as an image-to-image translation problem between spectrograms of smartphone and stethoscope recordings. Both CycleGAN and CUT are able to significantly decrease background noise, bringing the perceptual quality quantitatively and qualitatively closer to that of a digital stethoscope.
Description
Keywords
Personal Health Management with Digital Solutions, audio processing, digital health, machine learning, smartphone auscultation, unpaired image-to-image translation
Citation
Extent
10
Format
Geographic Location
Time Period
Related To
Proceedings of the 58th Hawaii International Conference on System Sciences
Related To (URI)
Table of Contents
Rights
Attribution-NonCommercial-NoDerivatives 4.0 International
Rights Holder
Local Contexts
Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.