HeAR -- Health Acoustic Representations
Health acoustic sounds such as coughs and breaths are known to contain useful health signals with significant potential for monitoring health and disease, yet are underexplored in the medical machine learning community. The existing deep learning systems for health acoustics are often narrowly train...
Gespeichert in:
Hauptverfasser: | , , , , , , , , , , , , , , , , , |
---|---|
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | |
container_volume | |
creator | Baur, Sebastien Nabulsi, Zaid Weng, Wei-Hung Garrison, Jake Blankemeier, Louis Fishman, Sam Chen, Christina Kakarmath, Sujay Maimbolwa, Minyoi Sanjase, Nsala Shuma, Brian Matias, Yossi Corrado, Greg S Patel, Shwetak Shetty, Shravya Prabhakara, Shruthi Muyoyeta, Monde Ardila, Diego |
description | Health acoustic sounds such as coughs and breaths are known to contain useful
health signals with significant potential for monitoring health and disease,
yet are underexplored in the medical machine learning community. The existing
deep learning systems for health acoustics are often narrowly trained and
evaluated on a single task, which is limited by data and may hinder
generalization to other tasks. To mitigate these gaps, we develop HeAR, a
scalable self-supervised learning-based deep learning system using masked
autoencoders trained on a large dataset of 313 million two-second long audio
clips. Through linear probes, we establish HeAR as a state-of-the-art health
audio embedding model on a benchmark of 33 health acoustic tasks across 6
datasets. By introducing this work, we hope to enable and accelerate further
health acoustics research. |
doi_str_mv | 10.48550/arxiv.2403.02522 |
format | Article |
fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2403_02522</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2403_02522</sourcerecordid><originalsourceid>FETCH-LOGICAL-a672-d29146866df0eba86d448c271522c1183378d885a8c0a9fd1c7abe31044b7d823</originalsourceid><addsrcrecordid>eNotzr0OgjAYheEuDka9ACfZnIr9o_0YiVExITEh7uSjLZFExQAavXsVnc72noeQOWehgihiK2yf9SMUismQiUiIMVmmPskDSoPU47k_BYlt7l1f2yD3t9Z3_tpjXzfXbkpGFZ47P_vvhBy3m-M6pdlht18nGUVtBHUi5kqD1q5ivkTQTimwwvDPl-UcpDTgACIEyzCuHLcGSy85U6o0DoSckMUvO0iLW1tfsH0VX3ExiOUbwsY4_A</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>HeAR -- Health Acoustic Representations</title><source>arXiv.org</source><creator>Baur, Sebastien ; Nabulsi, Zaid ; Weng, Wei-Hung ; Garrison, Jake ; Blankemeier, Louis ; Fishman, Sam ; Chen, Christina ; Kakarmath, Sujay ; Maimbolwa, Minyoi ; Sanjase, Nsala ; Shuma, Brian ; Matias, Yossi ; Corrado, Greg S ; Patel, Shwetak ; Shetty, Shravya ; Prabhakara, Shruthi ; Muyoyeta, Monde ; Ardila, Diego</creator><creatorcontrib>Baur, Sebastien ; Nabulsi, Zaid ; Weng, Wei-Hung ; Garrison, Jake ; Blankemeier, Louis ; Fishman, Sam ; Chen, Christina ; Kakarmath, Sujay ; Maimbolwa, Minyoi ; Sanjase, Nsala ; Shuma, Brian ; Matias, Yossi ; Corrado, Greg S ; Patel, Shwetak ; Shetty, Shravya ; Prabhakara, Shruthi ; Muyoyeta, Monde ; Ardila, Diego</creatorcontrib><description>Health acoustic sounds such as coughs and breaths are known to contain useful
health signals with significant potential for monitoring health and disease,
yet are underexplored in the medical machine learning community. The existing
deep learning systems for health acoustics are often narrowly trained and
evaluated on a single task, which is limited by data and may hinder
generalization to other tasks. To mitigate these gaps, we develop HeAR, a
scalable self-supervised learning-based deep learning system using masked
autoencoders trained on a large dataset of 313 million two-second long audio
clips. Through linear probes, we establish HeAR as a state-of-the-art health
audio embedding model on a benchmark of 33 health acoustic tasks across 6
datasets. By introducing this work, we hope to enable and accelerate further
health acoustics research.</description><identifier>DOI: 10.48550/arxiv.2403.02522</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Learning</subject><creationdate>2024-03</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2403.02522$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2403.02522$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Baur, Sebastien</creatorcontrib><creatorcontrib>Nabulsi, Zaid</creatorcontrib><creatorcontrib>Weng, Wei-Hung</creatorcontrib><creatorcontrib>Garrison, Jake</creatorcontrib><creatorcontrib>Blankemeier, Louis</creatorcontrib><creatorcontrib>Fishman, Sam</creatorcontrib><creatorcontrib>Chen, Christina</creatorcontrib><creatorcontrib>Kakarmath, Sujay</creatorcontrib><creatorcontrib>Maimbolwa, Minyoi</creatorcontrib><creatorcontrib>Sanjase, Nsala</creatorcontrib><creatorcontrib>Shuma, Brian</creatorcontrib><creatorcontrib>Matias, Yossi</creatorcontrib><creatorcontrib>Corrado, Greg S</creatorcontrib><creatorcontrib>Patel, Shwetak</creatorcontrib><creatorcontrib>Shetty, Shravya</creatorcontrib><creatorcontrib>Prabhakara, Shruthi</creatorcontrib><creatorcontrib>Muyoyeta, Monde</creatorcontrib><creatorcontrib>Ardila, Diego</creatorcontrib><title>HeAR -- Health Acoustic Representations</title><description>Health acoustic sounds such as coughs and breaths are known to contain useful
health signals with significant potential for monitoring health and disease,
yet are underexplored in the medical machine learning community. The existing
deep learning systems for health acoustics are often narrowly trained and
evaluated on a single task, which is limited by data and may hinder
generalization to other tasks. To mitigate these gaps, we develop HeAR, a
scalable self-supervised learning-based deep learning system using masked
autoencoders trained on a large dataset of 313 million two-second long audio
clips. Through linear probes, we establish HeAR as a state-of-the-art health
audio embedding model on a benchmark of 33 health acoustic tasks across 6
datasets. By introducing this work, we hope to enable and accelerate further
health acoustics research.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzr0OgjAYheEuDka9ACfZnIr9o_0YiVExITEh7uSjLZFExQAavXsVnc72noeQOWehgihiK2yf9SMUismQiUiIMVmmPskDSoPU47k_BYlt7l1f2yD3t9Z3_tpjXzfXbkpGFZ47P_vvhBy3m-M6pdlht18nGUVtBHUi5kqD1q5ivkTQTimwwvDPl-UcpDTgACIEyzCuHLcGSy85U6o0DoSckMUvO0iLW1tfsH0VX3ExiOUbwsY4_A</recordid><startdate>20240304</startdate><enddate>20240304</enddate><creator>Baur, Sebastien</creator><creator>Nabulsi, Zaid</creator><creator>Weng, Wei-Hung</creator><creator>Garrison, Jake</creator><creator>Blankemeier, Louis</creator><creator>Fishman, Sam</creator><creator>Chen, Christina</creator><creator>Kakarmath, Sujay</creator><creator>Maimbolwa, Minyoi</creator><creator>Sanjase, Nsala</creator><creator>Shuma, Brian</creator><creator>Matias, Yossi</creator><creator>Corrado, Greg S</creator><creator>Patel, Shwetak</creator><creator>Shetty, Shravya</creator><creator>Prabhakara, Shruthi</creator><creator>Muyoyeta, Monde</creator><creator>Ardila, Diego</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240304</creationdate><title>HeAR -- Health Acoustic Representations</title><author>Baur, Sebastien ; Nabulsi, Zaid ; Weng, Wei-Hung ; Garrison, Jake ; Blankemeier, Louis ; Fishman, Sam ; Chen, Christina ; Kakarmath, Sujay ; Maimbolwa, Minyoi ; Sanjase, Nsala ; Shuma, Brian ; Matias, Yossi ; Corrado, Greg S ; Patel, Shwetak ; Shetty, Shravya ; Prabhakara, Shruthi ; Muyoyeta, Monde ; Ardila, Diego</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a672-d29146866df0eba86d448c271522c1183378d885a8c0a9fd1c7abe31044b7d823</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Baur, Sebastien</creatorcontrib><creatorcontrib>Nabulsi, Zaid</creatorcontrib><creatorcontrib>Weng, Wei-Hung</creatorcontrib><creatorcontrib>Garrison, Jake</creatorcontrib><creatorcontrib>Blankemeier, Louis</creatorcontrib><creatorcontrib>Fishman, Sam</creatorcontrib><creatorcontrib>Chen, Christina</creatorcontrib><creatorcontrib>Kakarmath, Sujay</creatorcontrib><creatorcontrib>Maimbolwa, Minyoi</creatorcontrib><creatorcontrib>Sanjase, Nsala</creatorcontrib><creatorcontrib>Shuma, Brian</creatorcontrib><creatorcontrib>Matias, Yossi</creatorcontrib><creatorcontrib>Corrado, Greg S</creatorcontrib><creatorcontrib>Patel, Shwetak</creatorcontrib><creatorcontrib>Shetty, Shravya</creatorcontrib><creatorcontrib>Prabhakara, Shruthi</creatorcontrib><creatorcontrib>Muyoyeta, Monde</creatorcontrib><creatorcontrib>Ardila, Diego</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Baur, Sebastien</au><au>Nabulsi, Zaid</au><au>Weng, Wei-Hung</au><au>Garrison, Jake</au><au>Blankemeier, Louis</au><au>Fishman, Sam</au><au>Chen, Christina</au><au>Kakarmath, Sujay</au><au>Maimbolwa, Minyoi</au><au>Sanjase, Nsala</au><au>Shuma, Brian</au><au>Matias, Yossi</au><au>Corrado, Greg S</au><au>Patel, Shwetak</au><au>Shetty, Shravya</au><au>Prabhakara, Shruthi</au><au>Muyoyeta, Monde</au><au>Ardila, Diego</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>HeAR -- Health Acoustic Representations</atitle><date>2024-03-04</date><risdate>2024</risdate><abstract>Health acoustic sounds such as coughs and breaths are known to contain useful
health signals with significant potential for monitoring health and disease,
yet are underexplored in the medical machine learning community. The existing
deep learning systems for health acoustics are often narrowly trained and
evaluated on a single task, which is limited by data and may hinder
generalization to other tasks. To mitigate these gaps, we develop HeAR, a
scalable self-supervised learning-based deep learning system using masked
autoencoders trained on a large dataset of 313 million two-second long audio
clips. Through linear probes, we establish HeAR as a state-of-the-art health
audio embedding model on a benchmark of 33 health acoustic tasks across 6
datasets. By introducing this work, we hope to enable and accelerate further
health acoustics research.</abstract><doi>10.48550/arxiv.2403.02522</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | DOI: 10.48550/arxiv.2403.02522 |
ispartof | |
issn | |
language | eng |
recordid | cdi_arxiv_primary_2403_02522 |
source | arXiv.org |
subjects | Computer Science - Artificial Intelligence Computer Science - Learning |
title | HeAR -- Health Acoustic Representations |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T03%3A37%3A36IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=HeAR%20--%20Health%20Acoustic%20Representations&rft.au=Baur,%20Sebastien&rft.date=2024-03-04&rft_id=info:doi/10.48550/arxiv.2403.02522&rft_dat=%3Carxiv_GOX%3E2403_02522%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |