{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T06:25:27Z","timestamp":1787034327287,"version":"build-2736575974"},"reference-count":42,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T00:00:00Z","timestamp":1787011200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med. Technol."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Cardiovascular diseases (CVDs) represent a major global health threat, making early detection crucial. While cardiac auscultation is cost-effective, its reliance on clinical expertise leads to significant variability in diagnostic accuracy. Existing automated heart sound classification methods suffer from inadequate feature representation and limited model performance due to constraints imposed by the randomness, nonstationarity, and complex acoustic environment of phonocardiogram (PCG) signals.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>In this paper, an improved dual-branch CNN-BiLSTM network with hybrid cepstral fusion, named HSD-Net, is proposed for automated heart sound classification. To overcome feature representation limitations, we introduce a hybrid cepstral fusion feature set that combines Mel-Frequency Cepstral Coefficients (MFCC) and Inverse Mel-Frequency Cepstral Coefficients (IMFCC) with their first-order delta coefficients, capturing complementary low-frequency and high-frequency acoustic information. This feature set is processed by a dedicated dual-branch architecture: a convolutional neural network (CNN) branch extracts localized spectrotemporal patterns, while a bidirectional Long Short-Term Memory (BiLSTM) branch models long-range temporal dependencies. A Squeeze-and-Excitation (SE) attention mechanism is integrated to adaptively recalibrate feature importance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Extensive evaluations on two public datasets demonstrate that the proposed HSD-Net achieves superior performance compared with existing methods. On the PhysioNet\/Computing in Cardiology Challenge 2016 dataset, the model achieved an accuracy of 96.25%, a sensitivity of 95.85%, and a specificity of 96.50%. On the additional five-class Yaseen dataset, the model achieved an average accuracy of 99.32% in valvular heart disease classification. Ablation studies quantitatively confirmed that both the hybrid features and the dual-branch architecture contribute significantly to the overall performance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The proposed HSD-Net provides an effective framework for automated heart sound analysis. This work offers a reliable technical foundation for developing clinical decision-support systems, with significant potential for application in early screening and telemedicine.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmedt.2026.1883548","type":"journal-article","created":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T05:38:09Z","timestamp":1787031489000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["HSD-Net: a dual-branch CNN-BiLSTM network with hybrid cepstral fusion for heart sound classification"],"prefix":"10.3389","volume":"8","author":[{"given":"Caijian","family":"Hua","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Sichuan University of Science and Engineering","place":["Yibin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ye","family":"Tian","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Sichuan University of Science and Engineering","place":["Yibin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liuying","family":"Li","sequence":"additional","affiliation":[{"name":"Traditional Chinese Medicine Department, Zigong First People\u2019s Hospital","place":["Zigong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xia","family":"Zhou","sequence":"additional","affiliation":[{"name":"Traditional Chinese Medicine Department, Zigong First People\u2019s Hospital","place":["Zigong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,18]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"125928","DOI":"10.1016\/j.eswa.2024.125928","article-title":"Explainable AI assisted heart disease diagnosis through effective feature engineering and stacked ensemble learning","volume":"265","author":"Ghose","year":"2025","journal-title":"Expert Syst Appl"},{"key":"B2","doi-asserted-by":"publisher","first-page":"102289","DOI":"10.1016\/j.artmed.2022.102289","article-title":"Machine learning-based heart disease diagnosis: a systematic literature review","volume":"128","author":"Ahsan","year":"2022","journal-title":"Artif Intell Med"},{"key":"B3","author":"Strunic","year":""},{"key":"B4","first-page":"11","article-title":"Factors influencing cardiac auscultation proficiency in physician trainees","volume":"46","author":"Lam","year":"2005","journal-title":"Singapore Med J"},{"key":"B5","doi-asserted-by":"publisher","first-page":"822","DOI":"10.1109\/TBME.2015.2475278","article-title":"Logistic regression-hsmm-based heart sound segmentation","volume":"63","author":"Springer","year":"2015","journal-title":"IEEE Trans Biomed Eng"},{"key":"B6","doi-asserted-by":"publisher","first-page":"642","DOI":"10.1109\/JBHI.2018.2841197","article-title":"Adaptive sojourn time HSMM for heart sound segmentation","volume":"23","author":"Oliveira","year":"2019","journal-title":"IEEE J Biomed Health Inform"},{"key":"B7","doi-asserted-by":"publisher","first-page":"104814","DOI":"10.1016\/j.compbiomed.2021.104814","article-title":"Classification of heart sounds based on quality assessment and wavelet scattering transform","volume":"137","author":"Mei","year":"2021","journal-title":"Comput Biol Med"},{"key":"B8","doi-asserted-by":"publisher","first-page":"45","DOI":"10.3390\/bioengineering10010045","article-title":"An optimal approach for heart sound classification using grid search in hyperparameter optimization of machine learning","volume":"10","author":"Fuadah","year":"2023","journal-title":"Bioengineering"},{"key":"B9","doi-asserted-by":"publisher","first-page":"105899","DOI":"10.1016\/j.bspc.2023.105899","article-title":"Abnormal heart sound detection using time-frequency analysis and machine learning techniques","volume":"90","author":"Nia","year":"2024","journal-title":"Biomed Signal Process Control"},{"key":"B10","doi-asserted-by":"publisher","first-page":"105793","DOI":"10.1016\/j.bspc.2023.105793","article-title":"Hilbert domain characterizations of wavelet packets for automated heart sound abnormality detection","volume":"90","author":"Karan","year":"2024","journal-title":"Biomed Signal Process Control"},{"key":"B11","doi-asserted-by":"publisher","first-page":"10315","DOI":"10.1038\/s41598-025-93989-0","article-title":"Multiscale analysis of heart sound signals in the wavelet domain for heart murmur detection","volume":"15","author":"Vimalajeewa","year":"2025","journal-title":"Sci Rep"},{"key":"B12","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1016\/j.neunet.2020.06.015","article-title":"Heart sound classification based on improved MFCC features and convolutional recurrent neural networks","volume":"130","author":"Deng","year":"2020","journal-title":"Neural Netw"},{"key":"B13","doi-asserted-by":"publisher","first-page":"13019","DOI":"10.1109\/JSEN.2024.3373226","article-title":"Heart murmur severity stages classification using multikernel residual CNN","volume":"24","author":"Das","year":"2024","journal-title":"IEEE Sens J"},{"key":"B14","doi-asserted-by":"publisher","first-page":"9393","DOI":"10.1109\/JSEN.2018.2870759","article-title":"Phonocardiographic sensing using deep learning for abnormal heartbeat detection","volume":"18","author":"Latif","year":"2018","journal-title":"IEEE Sens J"},{"key":"B15","doi-asserted-by":"publisher","first-page":"7252","DOI":"10.1109\/TMC.2025.3547842","article-title":"An AI-assisted all-in-one integrated coronary artery disease diagnosis system using a portable heart sound sensor with an on-board executable lightweight model","volume":"24","author":"Zhang","year":"2025","journal-title":"IEEE Trans Mobile Comput"},{"key":"B16","doi-asserted-by":"publisher","first-page":"3956","DOI":"10.3390\/app10113956","article-title":"Classification of heart sounds using convolutional neural network","volume":"10","author":"Li","year":"2020","journal-title":"Appl Sci"},{"key":"B17","author":"Meintjes","year":""},{"key":"B18","doi-asserted-by":"publisher","first-page":"3533","DOI":"10.1007\/s00034-023-02588-9","article-title":"A heart sound signal classification method based on the mixed characteristics of mel cepstrum coefficient and second-order spectrum","volume":"43","author":"Zhang","year":"2024","journal-title":"Circuits Syst Signal Process"},{"key":"B19","doi-asserted-by":"publisher","first-page":"123099","DOI":"10.1109\/ACCESS.2024.3451660","article-title":"A hybrid framework of transformer encoder and residential conventional for cardiovascular disease recognition using heart sounds","volume":"12","author":"Al-Tam","year":"2024","journal-title":"IEEE Access"},{"key":"B20","doi-asserted-by":"publisher","first-page":"290","DOI":"10.3390\/bioengineering12030290","article-title":"Heart sound classification based on multi-scale feature fusion and channel attention module","volume":"12","author":"Li","year":"2025","journal-title":"Bioengineering"},{"key":"B21","doi-asserted-by":"publisher","first-page":"126839","DOI":"10.1016\/j.eswa.2025.126839","article-title":"Multichannel feature fusion network-based technique for heart sound signal classification and recognition","volume":"273","author":"Xiong","year":"2025","journal-title":"Expert Syst Appl"},{"key":"B22","doi-asserted-by":"publisher","first-page":"025702","DOI":"10.1088\/2040-8986\/ada0c4","article-title":"Heart sound recognition using hybrid joint transform correlator","volume":"27","author":"Mohapatra","year":"2025","journal-title":"J Opt"},{"key":"B23","doi-asserted-by":"publisher","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":"Liu","year":"2016","journal-title":"Physiol Meas"},{"key":"B24","doi-asserted-by":"publisher","first-page":"2344","DOI":"10.3390\/app8122344","article-title":"Classification of heart sound signal using multiple features","volume":"8","author":"Yaseen, Son","year":"2018","journal-title":"Appl Sci"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1084420","DOI":"10.3389\/fphys.2022.1084420","article-title":"Heart sound classification based on improved mel-frequency spectral coefficients and deep residual learning","volume":"13","author":"Li","year":"2022","journal-title":"Front Physiol"},{"key":"B26","author":"Trang","year":""},{"key":"B27","doi-asserted-by":"publisher","first-page":"121","DOI":"10.3390\/electronics9010121","article-title":"An automatic diagnosis of arrhythmias using a combination of CNN and LSTM technology","volume":"9","author":"Zheng","year":"2020","journal-title":"Electronics"},{"key":"B28","author":"Hu","year":""},{"key":"B29","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput"},{"key":"B30","author":"Snyder","year":""},{"key":"B31","doi-asserted-by":"publisher","first-page":"115011","DOI":"10.1088\/1361-6579\/aae7fa","article-title":"Supervised threshold-based heart sound classification algorithm","volume":"39","author":"Han","year":"2018","journal-title":"Physiol Meas"},{"key":"B32","doi-asserted-by":"publisher","first-page":"3811","DOI":"10.1007\/s00371-023-03031-5","article-title":"PCTMF-Net: heart sound classification with parallel CNNs-transformer and second-order spectral analysis","volume":"39","author":"Wang","year":"2023","journal-title":"Vis Comput"},{"key":"B33","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1016\/j.neucom.2018.09.101","article-title":"Heart sounds classification using a novel 1-D convolutional neural network with extremely low parameter consumption","volume":"392","author":"Xiao","year":"2020","journal-title":"Neurocomputing"},{"key":"B34","doi-asserted-by":"publisher","first-page":"105940","DOI":"10.1016\/j.cmpb.2021.105940","article-title":"Convolutional and recurrent neural networks for the detection of valvular heart diseases in phonocardiogram recordings","volume":"200","author":"Alkhodari","year":"2021","journal-title":"Comput Methods Programs Biomed"},{"key":"B35","doi-asserted-by":"publisher","first-page":"54","DOI":"10.3390\/info12020054","article-title":"Lightweight end-to-end neural network model for automatic heart sound classification","volume":"12","author":"Li","year":"2021","journal-title":"Information"},{"key":"B36","doi-asserted-by":"publisher","first-page":"11942","DOI":"10.3390\/app132111942","article-title":"Heart sound classification using wavelet analysis approaches and ensemble of deep learning models","volume":"13","author":"Lee","year":"2023","journal-title":"Appl Sci"},{"key":"B37","doi-asserted-by":"publisher","first-page":"103632","DOI":"10.1016\/j.compbiomed.2020.103632","article-title":"Automated detection of heart valve diseases using chirplet transform and multiclass composite classifier with PCG signals","volume":"118","author":"Ghosh","year":"2020","journal-title":"Comput Biol Med"},{"key":"B38","doi-asserted-by":"publisher","first-page":"105750","DOI":"10.1016\/j.cmpb.2020.105750","article-title":"Automatic diagnosis of multiple cardiac diseases from PCG signals using convolutional neural network","volume":"197","author":"Baghel","year":"2020","journal-title":"Comput Methods Programs Biomed"},{"key":"B39","doi-asserted-by":"publisher","first-page":"114022","DOI":"10.1016\/j.measurement.2023.114022","article-title":"Multiclassification for heart sound signals under multiple networks and multi-view feature","volume":"225","author":"Wang","year":"2024","journal-title":"Measurement"},{"key":"B40","doi-asserted-by":"publisher","first-page":"2221","DOI":"10.3390\/electronics12102221","article-title":"Assisting heart valve diseases diagnosis via transformer-based classification of heart sound signals","volume":"12","author":"Yang","year":"2023","journal-title":"Electronics"},{"key":"B41","author":"Bella","year":""},{"key":"B42","doi-asserted-by":"publisher","first-page":"109716","DOI":"10.1016\/j.sigpro.2024.109716","article-title":"Paramps: convolutional neural networks based on tensor decomposition for heart sound signal analysis and cardiovascular disease diagnosis","volume":"227","author":"Duan","year":"2025","journal-title":"Signal Process"}],"container-title":["Frontiers in Medical Technology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmedt.2026.1883548\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T05:38:10Z","timestamp":1787031490000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmedt.2026.1883548\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,18]]},"references-count":42,"alternative-id":["10.3389\/fmedt.2026.1883548"],"URL":"https:\/\/doi.org\/10.3389\/fmedt.2026.1883548","relation":{},"ISSN":["2673-3129"],"issn-type":[{"value":"2673-3129","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,18]]},"article-number":"1883548"}}