PhysioVec: A Multi-stage Deep-Learning Framework for Searching Online Health Information with Breath Sound

Conference Publication ResearchOnline@JCU
Huang, Yi;Song, Insu
Abstract

The COVID-19 outbreak presents a major challenge in diagnosing and monitoring respiratory diseases. IoT has the potential to address the challenges by remotely providing patients with rich information about respiratory health. However, current IoT-based health monitoring systems do not provide users with sufficient information to access the rich information in Health Social Network (HSN). We developed PhysioVec, a framework for searching HSN using breath sounds. PhysioVec consists of three components: Local Recurrent Transformer (LRT), a Multivariate radial-basis Logistic Interpreter (MLI), and an existing sentence embedding module. LRT combines local attention and recurrent Transformer to reduce overfitting and improve performance in the segmentation of breathing sounds. Physiological information detected from breathing sounds is used to search for relevant health information. PhysioVec achieved 100%., 59.8%., 92.2%., and 100% precision in the top one search results for breath sound with the common cold, influenza, pneumonia, and bronchitis, respectively. Our proposed framework allows users to search HSN for useful information just by recording their breathing sounds on mobile phones.

Journal

N/A

Publication Name

5th International Conference on Big Data and Artificial Intelligence, BDAI 2022

Volume

N/A

ISBN/ISSN

9781665470810

Edition

N/A

Issue

N/A

Pages Count

8

Location

Fuzhou, China

Publisher

Institute of Electrical and Electronics Engineers

Publisher Url

N/A

Publisher Location

Piscataway, NJ, USA

Publish Date

N/A

Url

N/A

Date

N/A

EISSN

N/A

DOI

10.1109/BDAI56143.2022.9862712