{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,19]],"date-time":"2026-09-19T16:50:53Z","timestamp":1789836653916,"version":"4.0.1"},"reference-count":45,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2025,7,11]],"date-time":"2025-07-11T00:00:00Z","timestamp":1752192000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100015624","name":"University of Jeddah","doi-asserted-by":"crossref","award":["UJ-20-005-DR"],"award-info":[{"award-number":["UJ-20-005-DR"]}],"id":[{"id":"10.13039\/501100015624","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>This paper presents a novel privacy-preserving architecture, a fusion of Federated Learning with Personalized Models and Differential Privacy (FLPMDP), for diagnosing arrhythmia from 12-lead electrocardiogram (ECG) signals. The architecture supports collaborative training in decentralized healthcare institutions without exposing sensitive patient information. By employing gated recurrent units (GRUs) for temporal sequence modeling along with feature fusion techniques and local differential privacy enforcement, FLPMDP ensures robust classification performance with data confidentiality. The architecture is evaluated on four experimental setups and demonstrates significant performance gain over centralized and federated baseline models. An empirical experiment on a large ECG dataset of 10,646 recordings indicates that the FLPMDP approach achieves an average accuracy of 93.71%. The FLPMDP approach yields F1-scores of 0.98, 0.93, 0.88, and 0.89 for sinus bradycardia (SB), atrial fibrillation (AFIB), supraventricular tachycardia (GSVT), and sinus rhythm (SR), respectively. Additionally, FLPMDP recorded a specificity up to 0.98, with a Kappa score of 0.8971 and a Matthews Correlation Coefficient of 0.9042, indicating high diagnostic accuracy and model strength. Comparative analysis against state-of-the-art methods\u2014such as CNN, ResNet, and attention-based RNNs\u2014indicate that FLPMDP consistently outperforms current models in accuracy, sensitivity, and robustness when facing non-IID data conditions. In the context of this research, federated learning is highly pertinent to modern healthcare, enabling secure and collaborative model training across institutions while complying with data privacy. The proposed FLPMDP framework offers a scalable and privacy-compliant solution for real-time arrhythmia detection, marking a step forward in deploying trustworthy artificial intelligence for decentralized medical diagnostics.<\/jats:p>","DOI":"10.1371\/journal.pone.0327108","type":"journal-article","created":{"date-parts":[[2025,7,11]],"date-time":"2025-07-11T17:41:38Z","timestamp":1752255698000},"page":"e0327108","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":9,"title":["Fusion of Personalized Federated Learning (PFL) with Differential Privacy (DP) Learning for Diagnosis of Arrhythmia Disease"],"prefix":"10.1371","volume":"20","author":[{"given":"Syed Mohsin","family":"Bokhari","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sarmad","family":"Sohaib","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8430-206X","authenticated-orcid":true,"given":"Muhammad","family":"Shafi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2025,7,11]]},"reference":[{"key":"pone.0327108.ref001","volume-title":"World Health Statistics 2020: Monitoring Health for the SDGs, Sustainable Development Goals","author":"World Health Organization","year":"2020"},{"key":"pone.0327108.ref002","doi-asserted-by":"crossref","unstructured":"Bokhari SM, Shafi M, Noor F, Sohaib S, Alqahtany S, Donnelly M. Assessing ECG interpretation expertise in medical practitioners through eye movement data and neuromorphic models. IEEE Access. 2025.","DOI":"10.1109\/ACCESS.2025.3528253"},{"issue":"18","key":"pone.0327108.ref003","doi-asserted-by":"crossref","first-page":"2516","DOI":"10.1161\/CIRCULATIONAHA.108.821306","article-title":"Atrial fibrillation and heart failure: treatment considerations for a dual epidemic","volume":"119","author":"E Anter","year":"2009","journal-title":"Circulation"},{"issue":"1","key":"pone.0327108.ref004","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1093\/ehjci\/jeab213","article-title":"The impact of atrial fibrillation and stroke risk factors on left atrial blood flow characteristics","volume":"23","author":"M Spartera","year":"2021","journal-title":"Eur Heart J Cardiovasc Imaging"},{"issue":"3","key":"pone.0327108.ref005","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1007\/s13239-021-00527-w","article-title":"Left atrial remodeling mechanisms associated with atrial fibrillation","volume":"12","author":"D Qiu","year":"2021","journal-title":"Cardiovasc Eng Technol"},{"key":"pone.0327108.ref006","doi-asserted-by":"crossref","first-page":"219","DOI":"10.1109\/RBME.2020.2976507","article-title":"A review on the state of the art in atrial fibrillation detection enabled by machine learning","volume":"14","author":"A Rizwan","year":"2021","journal-title":"IEEE Rev Biomed Eng"},{"issue":"6","key":"pone.0327108.ref007","doi-asserted-by":"crossref","first-page":"871","DOI":"10.1016\/j.jelectrocard.2016.07.033","article-title":"Automated detection of atrial fibrillation using R-R intervals and multivariate-based classification","volume":"49","author":"A Kennedy","year":"2016","journal-title":"J Electrocardiol"},{"key":"pone.0327108.ref008","first-page":"453402","article-title":"A novel automatic detection system for ECG arrhythmias using maximum margin clustering with immune evolutionary algorithm","volume":"2013","author":"B Zhu","year":"2013","journal-title":"Comput Math Methods Med"},{"key":"pone.0327108.ref009","doi-asserted-by":"crossref","first-page":"132","DOI":"10.1016\/j.compbiomed.2015.03.005","article-title":"Automatic detection of atrial fibrillation using stationary wavelet transform and support vector machine","volume":"60","author":"S Asgari","year":"2015","journal-title":"Comput Biol Med"},{"issue":"10201","key":"pone.0327108.ref010","doi-asserted-by":"crossref","first-page":"861","DOI":"10.1016\/S0140-6736(19)31721-0","article-title":"An artificial intelligence-enabled ECG algorithm for the identification of patients with atrial fibrillation during sinus rhythm: a retrospective analysis of outcome prediction","volume":"394","author":"ZI Attia","year":"2019","journal-title":"Lancet"},{"issue":"1","key":"pone.0327108.ref011","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1038\/s41591-018-0268-3","article-title":"Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network","volume":"25","author":"AY Hannun","year":"2019","journal-title":"Nat Med"},{"key":"pone.0327108.ref012","doi-asserted-by":"crossref","unstructured":"Lin CC, Yang CM. Heartbeat classification using normalized RR intervals and wavelet features. In: 2014 International Symposium on Computer, Consumer and Control. 2014. p. 650\u20133.","DOI":"10.1109\/IS3C.2014.175"},{"key":"pone.0327108.ref013","doi-asserted-by":"crossref","unstructured":"Zhang M, Wang Y, Luo T. Federated learning for arrhythmia detection of non-IID ECG. In: Proceedings of the 2020 IEEE 6th International Conference Computer and Communications (ICCC). 2020. p. 1176\u201380.","DOI":"10.1109\/ICCC51575.2020.9344971"},{"issue":"1","key":"pone.0327108.ref014","doi-asserted-by":"crossref","first-page":"30","DOI":"10.1038\/s41591-018-0307-0","article-title":"The practical implementation of artificial intelligence technologies in medicine","volume":"25","author":"J He","year":"2019","journal-title":"Nat Med"},{"key":"pone.0327108.ref015","doi-asserted-by":"crossref","first-page":"1144","DOI":"10.1186\/1471-2458-14-1144","article-title":"A systematic review of barriers to data sharing in public health","volume":"14","author":"WG van Panhuis","year":"2014","journal-title":"BMC Public Health"},{"issue":"1","key":"pone.0327108.ref016","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41467-019-10933-3","article-title":"Estimating the success of re-identifications in incomplete datasets using generative models","volume":"10","author":"L Rocher","year":"2019","journal-title":"Nat Commun"},{"issue":"2117","key":"pone.0327108.ref017","first-page":"1291","article-title":"Multivariate empirical mode decomposition","volume":"466","author":"N Rehman","year":"2010","journal-title":"Proc Roy SocA"},{"key":"pone.0327108.ref018","doi-asserted-by":"crossref","first-page":"375","DOI":"10.3389\/fneur.2020.00375","article-title":"Deep convolutional neural network-based Epileptic Electroencephalogram (EEG) signal classification","volume":"11","author":"Y Gao","year":"2020","journal-title":"Front Neurol"},{"issue":"13","key":"pone.0327108.ref019","doi-asserted-by":"crossref","first-page":"3551","DOI":"10.1109\/TSP.2017.2690524","article-title":"Tensor decomposition for signal processing and machine learning","volume":"65","author":"ND Sidiropoulos","year":"2017","journal-title":"IEEE Trans Signal Process"},{"issue":"12","key":"pone.0327108.ref020","doi-asserted-by":"crossref","first-page":"13902","DOI":"10.1109\/TCYB.2021.3121312","article-title":"An accurate GRU-based power time-series prediction approach with selective state updating and stochastic optimization","volume":"52","author":"W Zheng","year":"2022","journal-title":"IEEE Trans Cybern"},{"issue":"3","key":"pone.0327108.ref021","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1049\/htl2.12045","article-title":"Automatic cardiac arrhythmias classification using CNN and attention-based RNN network","volume":"10","author":"J Sun","year":"2023","journal-title":"Healthc Technol Lett"},{"key":"pone.0327108.ref022","doi-asserted-by":"crossref","unstructured":"Truex S, Baracaldo N, Anwar A, Steinke T, Ludwig H, Zhang R, et al. A hybrid approach to privacy-preserving federated learning. In: Proceedings of the 12th ACM Workshop on Artificial Intelligence and Security. 2019. p. 1\u201311.","DOI":"10.1145\/3338501.3357370"},{"issue":"3","key":"pone.0327108.ref023","doi-asserted-by":"crossref","first-page":"485","DOI":"10.3390\/e25030485","article-title":"PLDP-FL: federated learning with personalized local differential privacy","volume":"25","author":"X Shen","year":"2023","journal-title":"Entropy (Basel)"},{"issue":"2","key":"pone.0327108.ref024","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3298981","article-title":"Federated machine learning: concept and applications","volume":"10","author":"Q Yang","year":"2019","journal-title":"ACM Trans Intell Syst Technol"},{"key":"pone.0327108.ref025","unstructured":"PhysioNet. ECG arrhythmia dataset. 2020. https:\/\/physionet.org\/content\/ecg-arrhythmia\/1.0.0\/"},{"issue":"368","key":"pone.0327108.ref026","doi-asserted-by":"crossref","first-page":"829","DOI":"10.1080\/01621459.1979.10481038","article-title":"Robust locally weighted regression and smoothing scatterplots","volume":"74","author":"WS Cleveland","year":"1979","journal-title":"J Am Stat Assoc"},{"issue":"403","key":"pone.0327108.ref027","doi-asserted-by":"crossref","first-page":"596","DOI":"10.1080\/01621459.1988.10478639","article-title":"Locally weighted regression: an approach to regression analysis by local fitting","volume":"83","author":"WS Cleveland","year":"1988","journal-title":"J Am Stat Assoc"},{"issue":"2","key":"pone.0327108.ref028","doi-asserted-by":"crossref","first-page":"490","DOI":"10.1137\/040616024","article-title":"A review of image denoising algorithms, with a new one","volume":"4","author":"A Buades","year":"2005","journal-title":"Multiscale Model Simul"},{"issue":"2","key":"pone.0327108.ref029","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1109\/72.279181","article-title":"Learning long-term dependencies with gradient descent is difficult","volume":"5","author":"Y Bengio","year":"1994","journal-title":"IEEE Trans Neural Netw"},{"key":"pone.0327108.ref030","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1007\/978-3-642-24797-2_4","article-title":"Long short-term memory","volume-title":"Supervised sequence labelling with recurrent neural networks","author":"A Graves","year":"2012"},{"key":"pone.0327108.ref031","doi-asserted-by":"crossref","first-page":"108245","DOI":"10.1016\/j.measurement.2020.108245","article-title":"DGRU based human activity recognition using channel state information","volume":"167","author":"SM Bokhari","year":"2021","journal-title":"Measurement"},{"key":"pone.0327108.ref032","doi-asserted-by":"crossref","unstructured":"Cho K, Van Merri\u00ebnboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv preprint 2014. https:\/\/arxiv.org\/abs\/1406.1078","DOI":"10.3115\/v1\/D14-1179"},{"issue":"10","key":"pone.0327108.ref033","doi-asserted-by":"crossref","first-page":"9530","DOI":"10.1109\/JIOT.2020.2991416","article-title":"Personalized federated learning with differential privacy","volume":"7","author":"R Hu","year":"2020","journal-title":"IEEE Internet Things J"},{"key":"pone.0327108.ref034","doi-asserted-by":"crossref","unstructured":"Zhao Z, Fang H, Relton SD, Yan R, Liu Y, Li Z, et al. Adaptive lead weighted ResNet trained with different duration signals for classifying 12-lead ECGs. In: 2020 Computing in Cardiology. 2020. p. 1\u20134.","DOI":"10.22489\/CinC.2020.112"},{"issue":"1","key":"pone.0327108.ref035","doi-asserted-by":"crossref","first-page":"18738","DOI":"10.1038\/s41598-021-97118-5","article-title":"ECG-based machine-learning algorithms for heartbeat classification","volume":"11","author":"S Aziz","year":"2021","journal-title":"Sci Rep"},{"key":"pone.0327108.ref036","doi-asserted-by":"crossref","first-page":"190","DOI":"10.1016\/j.ins.2017.06.027","article-title":"Application of deep convolutional neural network for automated detection of myocardial infarction using ECG signals","volume":"415","author":"UR Acharya","year":"2017","journal-title":"Inf Sci"},{"key":"pone.0327108.ref037","doi-asserted-by":"crossref","unstructured":"Santos DR, Protani A, Giusti L, Aillet AS, Brutti P, Serio L. Feasibility analysis of federated neural networks for explainable detection of atrial fibrillation. arXiv preprint 2024. https:\/\/arxiv.org\/abs\/2410.19781","DOI":"10.1109\/HealthCom60970.2024.10880809"},{"issue":"2","key":"pone.0327108.ref038","doi-asserted-by":"crossref","first-page":"132","DOI":"10.4258\/hir.2023.29.2.132","article-title":"Standardized database of 12-lead electrocardiograms with a common standard for the promotion of cardiovascular research: KURIAS-ECG","volume":"29","author":"H Yoo","year":"2023","journal-title":"Healthc Inf Res"},{"issue":"3","key":"pone.0327108.ref039","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1049\/htl2.12045","article-title":"Automatic cardiac arrhythmias classification using CNN and attention-based RNN network","volume":"10","author":"J Sun","year":"2023","journal-title":"Healthc Technol Lett"},{"issue":"1","key":"pone.0327108.ref040","doi-asserted-by":"crossref","first-page":"67","DOI":"10.1186\/s12911-024-02464-9","article-title":"Continual learning framework for a multicenter study with an application to electrocardiogram","volume":"24","author":"J Kim","year":"2024","journal-title":"BMC Med Inform Decis Mak"},{"issue":"1","key":"pone.0327108.ref041","doi-asserted-by":"crossref","first-page":"3741","DOI":"10.1038\/s41467-023-39472-8","article-title":"Practical intelligent diagnostic algorithm for wearable 12-lead ECG via self-supervised learning on large-scale dataset","volume":"14","author":"J Lai","year":"2023","journal-title":"Nat Commun"},{"issue":"1","key":"pone.0327108.ref042","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1109\/JPROC.2020.3004555","article-title":"A comprehensive survey on transfer learning","volume":"109","author":"F Zhuang","year":"2021","journal-title":"Proc IEEE"},{"issue":"4","key":"pone.0327108.ref043","first-page":"1912","article-title":"Context-aware federated learning for healthcare monitoring using wearable devices","volume":"22","author":"J Zhang","year":"2023","journal-title":"IEEE Trans Mobile Comput"},{"issue":"6","key":"pone.0327108.ref044","first-page":"1587","article-title":"Contextual deep learning for ECG-based diagnosis using patient-specific data streams","volume":"70","author":"M Chen","year":"2023","journal-title":"IEEE Trans Biomed Eng"},{"issue":"1","key":"pone.0327108.ref045","first-page":"33","article-title":"Enhancing arrhythmia detection with contextual federated learning: a multi-institutional study","volume":"28","author":"A Gupta","year":"2024","journal-title":"IEEE J Biomed Health Inf"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0327108","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,11]],"date-time":"2025-07-11T17:41:48Z","timestamp":1752255708000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0327108"}},"subtitle":[],"editor":[{"given":"Agnese","family":"Sbrollini","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,7,11]]},"references-count":45,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2025,7,11]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0327108","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,11]]}}}