{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T17:08:39Z","timestamp":1785604119396,"version":"3.56.0"},"reference-count":27,"publisher":"Public Library of Science (PLoS)","issue":"11","license":[{"start":{"date-parts":[[2024,11,25]],"date-time":"2024-11-25T00:00:00Z","timestamp":1732492800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000265","name":"Medical Research Council","doi-asserted-by":"publisher","award":["MR\/S036644\/1"],"award-info":[{"award-number":["MR\/S036644\/1"]}],"id":[{"id":"10.13039\/501100000265","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>The detection of heart disease using a stethoscope requires significant skill and time, making it expensive and impractical for widespread screening in low-resource environments. Machine learning analysis of heart sound recordings can improve upon the accessibility and accuracy of diagnoses, but existing approaches require further validation on larger and more representative clinical datasets. For many previous algorithms, segmenting the signal into its individual sound components is a key first step. However, segmentation algorithms often struggle to find S1 or S2 sounds in the presence of strong murmurs or noise that significantly alter or mask the expected sound. Segmentation errors then propagate to the subsequent disease classifier steps. We propose a novel recurrent neural network and hidden semi-Markov model (HSMM) algorithm that can both segment the signal and detect a heart murmur, removing the need for a two-stage algorithm. This algorithm formed the \u2018CUED_Acoustics\u2019 entry to the 2022 George B. Moody PhysioNet challenge, where it won the first prize in both the challenge tasks. The algorithm\u2019s performance exceeded that of many end-to-end deep learning approaches that struggled to generalise to new test data. As our approach both segments the heart sound and detects a murmur, it can provide interpretable predictions for a clinician. The model also estimates the signal quality of the recording, which may be useful for a screening environment where non-experts are using a stethoscope. These properties make the algorithm a promising tool for screening of abnormal heart murmurs.<\/jats:p>","DOI":"10.1371\/journal.pdig.0000436","type":"journal-article","created":{"date-parts":[[2024,11,25]],"date-time":"2024-11-25T13:26:39Z","timestamp":1732541199000},"page":"e0000436","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":10,"title":["A recurrent neural network and parallel hidden Markov model algorithm to segment and detect heart murmurs in phonocardiograms"],"prefix":"10.1371","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3361-0457","authenticated-orcid":true,"given":"Andrew","family":"McDonald","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mark J. F.","family":"Gales","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anurag","family":"Agarwal","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2024,11,25]]},"reference":[{"key":"pdig.0000436.ref001","unstructured":"World Health Organisation. Cardiovascular diseases fact sheet; 2023. Available from: https:\/\/www.who.int\/health-topics\/cardiovascular-diseases."},{"issue":"7","key":"pdig.0000436.ref002","doi-asserted-by":"crossref","first-page":"883","DOI":"10.4244\/EIJY16M06_02","article-title":"Aortic valve stenosis: what do people know? A heart valve disease awareness survey of over 8, 800 people aged 60 or over","volume":"12","author":"L Gaede","year":"2016","journal-title":"EuroIntervention"},{"issue":"12","key":"pdig.0000436.ref003","doi-asserted-by":"crossref","first-page":"853","DOI":"10.1038\/s41569-021-00570-z","article-title":"Global epidemiology of valvular heart disease","volume":"18","author":"S Coffey","year":"2021","journal-title":"Nature Reviews Cardiology"},{"issue":"8","key":"pdig.0000436.ref004","doi-asserted-by":"crossref","first-page":"713","DOI":"10.1056\/NEJMoa1603693","article-title":"Global, Regional, and National Burden of Rheumatic Heart Disease, 1990\u20132015","volume":"377","author":"DA Watkins","year":"2017","journal-title":"New England Journal of Medicine"},{"key":"pdig.0000436.ref005","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-59259-338-5","volume-title":"Essentials of Bedside Cardiology: With a Complete Course in Heart Sounds and Murmurs on CD","author":"J Constant","year":"2003","edition":"2"},{"issue":"22","key":"pdig.0000436.ref006","doi-asserted-by":"crossref","first-page":"1832","DOI":"10.1136\/heartjnl-2018-313082","article-title":"Cardiac auscultation poorly predicts the presence of valvular heart disease in asymptomatic primary care patients","volume":"104","author":"SKM Gardezi","year":"2018","journal-title":"Heart"},{"issue":"8","key":"pdig.0000436.ref007","doi-asserted-by":"crossref","first-page":"E10","DOI":"10.1088\/1361-6579\/aa7ec8","article-title":"Recent advances in heart sound analysis","volume":"38","author":"GD Clifford","year":"2017","journal-title":"Physiological Measurement"},{"issue":"9","key":"pdig.0000436.ref008","doi-asserted-by":"crossref","first-page":"e0000324","DOI":"10.1371\/journal.pdig.0000324","article-title":"Heart murmur detection from phonocardiogram recordings: The George B. Moody PhysioNet Challenge 2022","volume":"2","author":"MA Reyna","year":"2023","journal-title":"PLOS Digital Health"},{"key":"pdig.0000436.ref009","doi-asserted-by":"crossref","unstructured":"McDonald A, Gales M, Agarwal A. Detection of Heart Murmurs in Phonocardiograms with Parallel Hidden Semi-Markov Models. In: 2022 Computing in Cardiology Conference (CinC). Computing in Cardiology; 2022. Available from: http:\/\/dx.doi.org\/10.22489\/CinC.2022.020.","DOI":"10.22489\/CinC.2022.020"},{"issue":"8","key":"pdig.0000436.ref010","doi-asserted-by":"crossref","first-page":"1645","DOI":"10.1088\/1361-6579\/aa6a3d","article-title":"DropConnected neural networks trained on time-frequency and inter-beat features for classifying heart sounds","volume":"38","author":"E Kay","year":"2017","journal-title":"Physiological Measurement"},{"issue":"12","key":"pdig.0000436.ref011","doi-asserted-by":"crossref","first-page":"2181","DOI":"10.1088\/0967-3334\/37\/12\/2181","article-title":"An open access database for the evaluation of heart sound algorithms","volume":"37","author":"C Liu","year":"2016","journal-title":"Physiological Measurement"},{"key":"pdig.0000436.ref012","unstructured":"Bentley PJ, Nordehn G, Coimbra MT, Mannor S. The PASCAL Classifying Heart Sounds Challenge 2011 (CHSC2011) Results; 2011. Available from: http:\/\/www.peterjbentley.com\/heartchallenge\/index.html."},{"issue":"6","key":"pdig.0000436.ref013","doi-asserted-by":"crossref","first-page":"2524","DOI":"10.1109\/JBHI.2021.3137048","article-title":"The CirCor DigiScope dataset: from murmur detection to murmur classification","volume":"26","author":"J Oliveira","year":"2021","journal-title":"IEEE Journal of Biomedical and Health Informatics"},{"issue":"23","key":"pdig.0000436.ref014","doi-asserted-by":"crossref","first-page":"e215","DOI":"10.1161\/01.CIR.101.23.e215","article-title":"PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals","volume":"101","author":"AL Goldberger","year":"2000","journal-title":"Circulation"},{"key":"pdig.0000436.ref015","unstructured":"Oliveira J, Renna F, Costa P, Nogueira M, Oliveira AC, Elola A, et al. The CirCor DigiScope Phonocardiogram Dataset (version 1.0.3); 2022. Available from: https:\/\/doi.org\/10.13026\/tshs-mw03."},{"issue":"4","key":"pdig.0000436.ref016","first-page":"822","article-title":"Logistic Regression-HSMM-Based Heart Sound Segmentation","volume":"63","author":"DB Springer","year":"2016","journal-title":"IEEE Transactions on Biomedical Engineering"},{"key":"pdig.0000436.ref017","unstructured":"Kay E. Cardiac Acoustics: Understanding and Detecting Heart Murmurs. PhD Thesis, University of Cambridge; 2018."},{"issue":"9","key":"pdig.0000436.ref018","doi-asserted-by":"crossref","first-page":"1964","DOI":"10.1109\/TBME.2018.2843258","article-title":"Heart Sound Segmentation-An Event Detection Approach Using Deep Recurrent Neural Networks","volume":"65","author":"E Messner","year":"2018","journal-title":"IEEE Transactions on Biomedical Engineering"},{"key":"pdig.0000436.ref019","doi-asserted-by":"crossref","unstructured":"McDonald A. A recurrent neural network and parallel hidden Markov model algorithm to detect heart murmurs. GitHub repository; 2023. Available from: https:\/\/github.com\/am2234\/parallel-hsmm-murmur.","DOI":"10.1101\/2023.12.26.23300540"},{"key":"pdig.0000436.ref020","doi-asserted-by":"crossref","first-page":"117194","DOI":"10.1016\/j.jsv.2022.117194","article-title":"Stethoscope acoustics","volume":"539","author":"M Nussbaumer","year":"2022","journal-title":"Journal of Sound and Vibration"},{"key":"pdig.0000436.ref021","doi-asserted-by":"crossref","unstructured":"Cho K, van Merrienboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, et al. Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation; 2014. Available from: https:\/\/arxiv.org\/abs\/1406.1078.","DOI":"10.3115\/v1\/D14-1179"},{"key":"pdig.0000436.ref022","unstructured":"Kingma DP, Ba JL. Adam: a Method for Stochastic Optimization. In: International Conference on Learning Representations 2015. San Diego; 2015. p. 1\u201315."},{"issue":"4","key":"pdig.0000436.ref023","doi-asserted-by":"crossref","first-page":"513","DOI":"10.1088\/0967-3334\/31\/4\/004","article-title":"Segmentation of heart sound recordings by a duration-dependent hidden Markov model","volume":"31","author":"SE Schmidt","year":"2010","journal-title":"Physiological Measurement"},{"key":"pdig.0000436.ref024","unstructured":"Prokhorenkova L, Gusev G, Vorobev A, Dorogush AV, Gulin A. CatBoost: unbiased boosting with categorical features; 2017. Available from: https:\/\/arxiv.org\/abs\/1706.09516."},{"key":"pdig.0000436.ref025","unstructured":"PhysioNet Challenge organisers. George B. Moody PhysioNet Challenge 2022; 2022. Available from: https:\/\/physionetchallenges.org\/2022\/."},{"issue":"13","key":"pdig.0000436.ref026","doi-asserted-by":"crossref","first-page":"811","DOI":"10.3109\/08958378.2014.955932","article-title":"Screening tests: a review with examples","volume":"26","author":"LD Maxim","year":"2014","journal-title":"Inhalation Toxicology"},{"issue":"4","key":"pdig.0000436.ref027","doi-asserted-by":"crossref","first-page":"329","DOI":"10.1001\/jama.2022.10561","article-title":"Rethinking Algorithm Performance Metrics for Artificial Intelligence in Diagnostic Medicine","volume":"328","author":"MA Reyna","year":"2022","journal-title":"JAMA"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0000436","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,11,25]],"date-time":"2024-11-25T13:26:51Z","timestamp":1732541211000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0000436"}},"subtitle":[],"editor":[{"given":"Matthew A","family":"Reyna","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2024,11,25]]},"references-count":27,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2024,11,25]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0000436","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2023.12.26.23300540","asserted-by":"object"}]},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,11,25]]}}}