{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,7]],"date-time":"2026-11-07T18:45:00Z","timestamp":1794077100857,"version":"4.8.0"},"reference-count":37,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2024,3,19]],"date-time":"2024-03-19T00:00:00Z","timestamp":1710806400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>Cardiovascular diseases (CVDs) account for a high fatality rate worldwide. Heart murmurs can be detected from phonocardiograms (PCGs) and may indicate CVDs. Still, they are often overlooked as their detection and correct clinical interpretation require expert skills. In this work, we aim to predict the presence of murmurs and clinical outcomes from multiple PCG recordings employing an explainable multitask model.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Approach<\/jats:title>\n                    <jats:p>Our approach consists of a two-stage multitask model. In the first stage, we predict the murmur presence in single PCGs using a multiple instance learning (MIL) framework. MIL also allows us to derive sample-wise classifications (i.e. murmur locations) while only needing one annotation per recording (\u201cweak label\u201d) during training. In the second stage, we fuse explainable hand-crafted features with features from a pooling-based artificial neural network (PANN) derived from the MIL framework. Finally, we predict the presence of murmurs and the clinical outcome for a single patient based on multiple recordings using a simple feed-forward neural network.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Main results<\/jats:title>\n                    <jats:p>We show qualitatively and quantitatively that the MIL approach yields useful features and can be used to detect murmurs on multiple time instances and may thus guide a practitioner through PCGs. We analyze the second stage of the model in terms of murmur classification and clinical outcome. We achieved a weighted accuracy of 0.714 and an outcome cost of 13612 when using the PANN model and demographic features on the CirCor dataset (hidden test set of the George B. Moody PhysioNet challenge 2022, team \u201cHeart2Beat\u201d, rank 12 \/ 40).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec004\">\n                    <jats:title>Significance<\/jats:title>\n                    <jats:p>To the best of our knowledge, we are the first to demonstrate the usefulness of MIL in PCG classification. Also, we showcase how the explainability of the model can be analyzed quantitatively, thus avoiding confirmation bias inherent to many post-hoc methods. Finally, our overall results demonstrate the merit of employing MIL combined with handcrafted features for the generation of explainable features as well as for a competitive classification performance.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pdig.0000461","type":"journal-article","created":{"date-parts":[[2024,3,19]],"date-time":"2024-03-19T13:24:22Z","timestamp":1710854662000},"page":"e0000461","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":11,"title":["Multiple instance learning framework can facilitate explainability in murmur detection"],"prefix":"10.1371","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6053-6558","authenticated-orcid":true,"given":"Maurice","family":"Rohr","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Benedikt","family":"M\u00fcller","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8446-4635","authenticated-orcid":true,"given":"Sebastian","family":"Dill","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0522-5877","authenticated-orcid":true,"given":"G\u00f6khan","family":"G\u00fcney","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7948-8181","authenticated-orcid":true,"given":"Christoph","family":"Hoog Antink","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2024,3,19]]},"reference":[{"issue":"25","key":"pdig.0000461.ref001","doi-asserted-by":"crossref","first-page":"2980","DOI":"10.1016\/j.jacc.2020.11.021","article-title":"The Global Burden of Cardiovascular Diseases and Risks: A Compass for Global Action","volume":"76","author":"GA Roth","year":"2020","journal-title":"Journal of the American College of Cardiology"},{"issue":"9","key":"pdig.0000461.ref002","doi-asserted-by":"crossref","first-page":"1","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"},{"issue":"23","key":"pdig.0000461.ref003","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"},{"issue":"3","key":"pdig.0000461.ref004","first-page":"250","article-title":"Heart Murmurs in Children: Evaluation and Management","volume":"105","author":"B Ford","year":"2022","journal-title":"American Family Physician"},{"issue":"6","key":"pdig.0000461.ref005","doi-asserted-by":"crossref","first-page":"610","DOI":"10.1001\/archinte.166.6.610","article-title":"Competency in Cardiac Examination Skills in Medical Students, Trainees, Physicians, and Maculty: A Multicenter Study","volume":"166","author":"JM Vukanovic-Criley","year":"2006","journal-title":"Archives of Internal Medicine"},{"key":"pdig.0000461.ref006","unstructured":"Overdiek M, Petersen T. Was Deutschland \u00fcber Algorithmen und K\u00fcnstliche Intelligenz wei\u00df und denkt. Ergebnisse einer repr\u00e4sentativen Bev\u00f6lverungsumfrage. Bertelsmann Stiftung; 2022. Available from: https:\/\/www.bertelsmann-stiftung.de\/fileadmin\/files\/BSt\/Publikationen\/GrauePublikationen\/DG_Was_Deutschland_ueber_Algorithmen_KI.pdf."},{"key":"pdig.0000461.ref007","unstructured":"Bundes\u00e4rztekammer. Stellungnahme der Bundes\u00e4rztekammer zum Entwurf eines Gesetzes f\u00fcr ein Zukunftsprogramm Krankenh\u00e4user (Krankenhauszukunftsgesetz \u2013 KHZG); 2020. Available from: https:\/\/www.bundesaerztekammer.de\/fileadmin\/user_upload\/_old-files\/downloads\/pdf-Ordner\/Stellungnahmen\/SN_BAEK_KHZG_GE_07092020_final.pdf."},{"issue":"11","key":"pdig.0000461.ref008","doi-asserted-by":"crossref","first-page":"e745","DOI":"10.1016\/S2589-7500(21)00208-9","article-title":"The False Hope of Current Approaches to Explainable Artificial Intelligence in Health Care","volume":"3","author":"M Ghassemi","year":"2021","journal-title":"The Lancet Digital Health"},{"issue":"6","key":"pdig.0000461.ref009","doi-asserted-by":"crossref","first-page":"2162","DOI":"10.1109\/JBHI.2020.3027910","article-title":"A Robust Interpretable Deep Learning Classifier for Heart Anomaly Detection without Segmentation","volume":"25","author":"T Dissanayake","year":"2020","journal-title":"IEEE Journal of Biomedical and Health Informatics"},{"key":"pdig.0000461.ref010","doi-asserted-by":"crossref","unstructured":"Ren Z, Qian K, Dong F, Dai Z, Nejdl W, Yamamoto Y, et al. Deep attention-based neural networks for explainable heart sound classification. Machine Learning with Applications. 2022; p. 100322.","DOI":"10.1016\/j.mlwa.2022.100322"},{"key":"pdig.0000461.ref011","unstructured":"Holzinger A, Biemann C, Pattichis CS, Kell DB. What Do We Need to Build Explainable AI Systems for the Medical Domain? arXiv preprint arXiv:171209923. 2017."},{"issue":"8","key":"pdig.0000461.ref012","doi-asserted-by":"crossref","first-page":"1616","DOI":"10.1088\/1361-6579\/aa7840","article-title":"Detection of Pathological Heart Sounds","volume":"38","author":"M Abdollahpur","year":"2017","journal-title":"Physiological Measurement"},{"key":"pdig.0000461.ref013","doi-asserted-by":"crossref","unstructured":"Potes C, Parvaneh S, Rahman A, Conroy B. Ensemble of Feature-based and Deep Learning-based Classifiers for Detection of Abnormal Heart Sounds. In: 2016 Computing in Cardiology Conference (CinC). IEEE; 2016. p. 621\u2013624.","DOI":"10.22489\/CinC.2016.182-399"},{"key":"pdig.0000461.ref014","doi-asserted-by":"crossref","unstructured":"Goda MA, Hajas P. Morphological Determination of Pathological PCG Signals by Time and Frequency Domain Analysis. In: 2016 Computing in Cardiology Conference (CinC). IEEE; 2016. p. 1133\u20131136.","DOI":"10.22489\/CinC.2016.324-249"},{"key":"pdig.0000461.ref015","unstructured":"Ortiz JJG, Phoo CP, Wiens J. Heart Sound Classification Based on Temporal Alignment Techniques. In: 2016 Computing in Cardiology Conference (CinC). IEEE; 2016. p. 589\u2013592."},{"issue":"8","key":"pdig.0000461.ref016","doi-asserted-by":"crossref","first-page":"1631","DOI":"10.1088\/1361-6579\/aa7982","article-title":"Ensemble Methods with Outliers for Phonocardiogram Classification","volume":"38","author":"MN Homsi","year":"2017","journal-title":"Physiological Measurement"},{"issue":"9","key":"pdig.0000461.ref017","doi-asserted-by":"crossref","first-page":"095003","DOI":"10.1088\/1361-6579\/ac1d59","article-title":"Deep Learning Based Classification of Unsegmented Phonocardiogram Spectrograms Leveraging Transfer Learning","volume":"42","author":"KN Khan","year":"2021","journal-title":"Physiological Measurement"},{"key":"pdig.0000461.ref018","doi-asserted-by":"crossref","unstructured":"Rubin J, Abreu R, Ganguli A, Nelaturi S, Matei I, Sricharan K. Classifying Heart Sound Recordings Using Deep Convolutional Neural Networks and Mel-Frequency Cepstral Coefficients. In: 2016 Computing in Cardiology Conference (CinC). IEEE; 2016. p. 813\u2013816.","DOI":"10.22489\/CinC.2016.236-175"},{"issue":"8","key":"pdig.0000461.ref019","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"},{"key":"pdig.0000461.ref020","doi-asserted-by":"crossref","unstructured":"Summerton S, Wood D, Murphy D, Redfern O, Benatan M, Kaisti M, et al. Two-stage Classification for Detecting Murmurs from Phonocardiograms Using Deep and Expert. In: 2022 Computing in Cardiology (CinC). vol. 49. IEEE; 2022. p. 1\u20134.","DOI":"10.22489\/CinC.2022.322"},{"key":"pdig.0000461.ref021","unstructured":"Xu Y, Bao X, Lam HK, Kamavuako EN. Hierarchical Multi-Scale Convolutional Network for Murmurs Detection on PCG Signals. In: 2022 Computing in Cardiology (CinC). vol. 49. IEEE; 2022. p. 1\u20134."},{"key":"pdig.0000461.ref022","doi-asserted-by":"crossref","unstructured":"Lu H, Yip JB, Steigleder T, Grie\u00dfhammer S, Sai Jitin Jami NV, Eskofier B, et al. A Lightweight Robust Approach for Automatic Heart Murmurs and Clinical Outcomes Classification from Phonocardiogram Recordings. In: 2022 Computing in Cardiology (CinC). vol. 49. IEEE; 2022. p. 1\u20134.","DOI":"10.22489\/CinC.2022.165"},{"key":"pdig.0000461.ref023","unstructured":"Maron O, Lozano-P\u00e9rez T. A Framework for Multiple-Instance Learning. Advances in Neural Information Processing Systems. 1997;10."},{"key":"pdig.0000461.ref024","doi-asserted-by":"crossref","unstructured":"Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. In: International Conference on Medical Image Computing and Computer-assisted Intervention. Springer; 2015. p. 234\u2013241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"issue":"6","key":"pdig.0000461.ref025","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":"2022","journal-title":"IEEE Journal of Biomedical and Health Informatics"},{"key":"pdig.0000461.ref026","unstructured":"Oliveira J, Renna F, Costa PD, Nogueira M, Oliveira C, Elola A, et al. The CirCor DigiScope Phonocardiogram Dataset (version 1.0.3). PhysioNet; 2022."},{"key":"pdig.0000461.ref027","unstructured":"Judge R, Mangrulkar R. Heart Sound and Murmur Library; 2015."},{"issue":"8","key":"pdig.0000461.ref028","doi-asserted-by":"crossref","first-page":"1685","DOI":"10.1088\/1361-6579\/aa7620","article-title":"Heart Sounds Analysis Using Probability Assessment","volume":"38","author":"F Plesinger","year":"2017","journal-title":"Physiological measurement"},{"key":"pdig.0000461.ref029","doi-asserted-by":"crossref","unstructured":"Springer DB, Brennan T, Zuhlke LJ, Abdelrahman HY, Ntusi N, Clifford GD, et al. Signal Quality Classification of Mobile Phone-recorded Phonocardiogram Signals. In: 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE; 2014. p. 1335\u20131339.","DOI":"10.1109\/ICASSP.2014.6853814"},{"key":"pdig.0000461.ref030","doi-asserted-by":"crossref","unstructured":"Rohr M, M\u00fcller B, Dill S, G\u00fcney G, Hoog Antink C. Two-Stage Multitask-Lerner for PCG Murmur Location Detection. In: 2022 Computing in Cardiology (CinC). vol. 49. IEEE; 2022. p. 1\u20134.","DOI":"10.22489\/CinC.2022.138"},{"issue":"11","key":"pdig.0000461.ref031","doi-asserted-by":"crossref","first-page":"2180","DOI":"10.1109\/TASLP.2018.2858559","article-title":"Adaptive Pooling Operators for Weakly Labeled Sound Event Detection","volume":"26","author":"B McFee","year":"2018","journal-title":"IEEE\/ACM Transactions on Audio, Speech, and Language Processing"},{"issue":"4","key":"pdig.0000461.ref032","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"},{"issue":"4","key":"pdig.0000461.ref033","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.0000461.ref034","doi-asserted-by":"crossref","unstructured":"McFee B, Raffel C, Liang D, Ellis DPW, McVicar M, Battenberg E, et al. librosa: Audio and Music Signal Analysis in Python. In: Kathryn Huff, James Bergstra, editors. Proceedings of the 14th Python in Science Conference; 2015. p. 18\u201324.","DOI":"10.25080\/Majora-7b98e3ed-003"},{"key":"pdig.0000461.ref035","doi-asserted-by":"crossref","unstructured":"Chen T, Guestrin C. Xgboost: A Scalable Tree Boosting System. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2016. p. 785\u2013794.","DOI":"10.1145\/2939672.2939785"},{"key":"pdig.0000461.ref036","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 (CinC). vol. 49. IEEE; 2022. p. 1\u20134.","DOI":"10.22489\/CinC.2022.020"},{"key":"pdig.0000461.ref037","doi-asserted-by":"crossref","unstructured":"Liu Y, Chen H, Wang Y, Zhang P. Power Pooling: An Adaptive Pooling Function for Weakly Labelled Sound Event Detection. In: 2021 International Joint Conference on Neural Networks (IJCNN). IEEE; 2021. p. 1\u20137.","DOI":"10.1109\/IJCNN52387.2021.9533332"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0000461","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,19]],"date-time":"2024-03-19T13:24:42Z","timestamp":1710854682000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0000461"}},"subtitle":[],"editor":[{"given":"Dukyong","family":"Yoon","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2024,3,19]]},"references-count":37,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2024,3,19]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0000461","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2022.12.08.22283240","asserted-by":"object"}]},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,3,19]]}}}