{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T22:08:15Z","timestamp":1790633295104,"version":"4.1.0"},"reference-count":26,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T00:00:00Z","timestamp":1785283200000},"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. Digit. Health"],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Wearable electrocardiography (ECG) and smartphone technology create new opportunities for arrhythmia monitoring outside clinical environments. However, deployable mobile cardiac systems must address more than just classification accuracy. They must reliably acquire physiological signals under real-world conditions, protect the privacy of sensitive cardiac data, adapt to user- and sensor-specific variability, and minimize false alarms in daily use. We present Cardio-FL, a smartphone-based digital health prototype for privacy-preserving, adaptive arrhythmia monitoring.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>The system integrates wearable ECG acquisition with activity-aware on-device classification, camera-based photoplethysmography (PPG) pulse-regularity assessment, and federated on-device learning. ECG signals are streamed from a Movesense wearable sensor to the developed Android application via Bluetooth Low Energy, preprocessed locally, and converted into compact representations composed of derivative ECG samples and normalized RR-interval features. The deployed neural network follows an encoder-classifier-decoder architecture with only 5,575 parameters. The classifier distinguishes between normal beats, supraventricular ectopic beats (SVEB), and ventricular ectopic beats (VEB). At the same time, the encoder and decoder are further adapted on-device using unlabeled ECG segments and a reconstruction-based federated learning strategy. To improve signal representations without increasing runtime complexity, the model is pretrained prior to deployment. The encoder is first pretrained using masked reconstruction on the PTB-XL dataset and subsequently fine-tuned together with the classifier on the MIT-BIH beat-classification task before deployment. The developed Android application incorporates continuous ECG acquisition and preprocessing, on-device inference, and Flower-based federated encoder-decoder adaptation. Both inference and training are gated by user activity, while PPG-based pulse-regularity assessment is also performed after ECG alerts as a post-alert support workflow rather than as a clinically validated diagnostic confirmation step.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>In offline evaluation, masked pretraining combined with validation-selected SVEB-aware calibration improves the deployed model performance from 92.80% to 94.15% accuracy and from 69.24% to 74.53% macro-F1 compared with MIT-BIH-only training. SVEB F1 improved from 35.86% to 46.11%, while VEB F1 increased from 75.44% to 80.31%.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The results demonstrate the feasibility of integrating wearable ECG acquisition, compact on-device inference, activity-aware operation, smartphone-camera PPG assessment, and reconstruction-based federated adaptation within a single Android application. Cardio-FL provides a practical framework for privacy-preserving and adaptive mobile arrhythmia monitoring. We further identify the validation steps required for real-world clinical integration.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fdgth.2026.1900115","type":"journal-article","created":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T04:33:38Z","timestamp":1785299618000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Privacy-preserving adaptive arrhythmia monitoring with wearable ECG, smartphone PPG, and federated on-device learning"],"prefix":"10.3389","volume":"8","author":[{"given":"Aneta","family":"Kartali","sequence":"first","affiliation":[{"name":"Faculty of Computer and Information Science, University of Ljubljana","place":["Ljubljana, Slovenia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marija","family":"Koshtrevska","sequence":"additional","affiliation":[{"name":"Faculty of Computer and Information Science, University of Ljubljana","place":["Ljubljana, Slovenia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stevan","family":"Joki\u0107","sequence":"additional","affiliation":[{"name":"Faculty of Information Technology, Alfa BK University","place":["Belgrade, Serbia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nenad","family":"Gligori\u0107","sequence":"additional","affiliation":[{"name":"Zentrix Lab LLC","place":["Pan\u010devo, Serbia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Octavian M.","family":"Machidon","sequence":"additional","affiliation":[{"name":"Faculty of Computer and Information Science, University of Ljubljana","place":["Ljubljana, Slovenia"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,29]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1365","DOI":"10.3390\/s23031365","article-title":"A tiny matched filter-based CNN for inter-patient ECG classification and arrhythmia detection at the edge","volume":"23","author":"Farag","year":"2023","journal-title":"Sensors"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1823","DOI":"10.1038\/s41591-022-01979-w","article-title":"Smartphone-based screening for atrial fibrillation: a pragmatic randomized clinical trial","volume":"28","author":"Rizas","year":"2022","journal-title":"Nat Med"},{"key":"B3","author":"Silva","year":""},{"key":"B4","doi-asserted-by":"publisher","first-page":"1535335","DOI":"10.3389\/fdgth.2025.1535335","article-title":"Integrated portable ECG monitoring system with CNN classification for early arrhythmia detection","volume":"7","author":"Panwar","year":"2025","journal-title":"Front Digit Health"},{"key":"B5","doi-asserted-by":"publisher","first-page":"44622","DOI":"10.1109\/JSEN.2025.3625425","article-title":"A wearable single-arm electrocardiograph system: design and experimental evaluation","volume":"25","author":"Marcucci","year":"2025","journal-title":"IEEE Sens J"},{"key":"B6","doi-asserted-by":"publisher","first-page":"e12770","DOI":"10.2196\/12770","article-title":"Deep learning approaches to detect atrial fibrillation using photoplethysmographic signals: algorithms development study","volume":"7","author":"Kwon","year":"2019","journal-title":"JMIR Mhealth Uhealth"},{"key":"B7","author":"Voisin","year":""},{"key":"B8","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1038\/s41746-019-0207-9","article-title":"Photoplethysmography based atrial fibrillation detection: a review","volume":"3","author":"Pereira","year":"2020","journal-title":"npj Digital Med"},{"key":"B9","author":"McMahan","year":""},{"key":"B10","author":"Beutel","year":""},{"key":"B11","doi-asserted-by":"publisher","first-page":"21038","DOI":"10.1038\/s41598-024-71366-7","article-title":"Fed-CL: an atrial fibrillation prediction system using ECG signals employing federated learning mechanism","volume":"14","author":"Alreshidi","year":"2024","journal-title":"Sci Rep"},{"key":"B12","doi-asserted-by":"publisher","first-page":"e0000793","DOI":"10.1371\/journal.pdig.0000793","article-title":"Cross-device federated unsupervised learning for the detection of anomalies in single-lead electrocardiogram signals","volume":"4","author":"Kapsecker","year":"2025","journal-title":"PLOS Digit Health"},{"key":"B13","doi-asserted-by":"publisher","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":"Goldberger","year":"2000","journal-title":"Circulation"},{"key":"B14","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1109\/51.932724","article-title":"The impact of the MIT-BIH arrhythmia database","volume":"20","author":"Moody","year":"2001","journal-title":"IEEE Eng Med Biol Mag"},{"key":"B15","doi-asserted-by":"publisher","first-page":"154","DOI":"10.1038\/s41597-020-0495-6","article-title":"PTB-XL, a large publicly available electrocardiography dataset","volume":"7","author":"Wagner","year":"2020","journal-title":"Sci Data"},{"key":"B16","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1109\/TBME.1985.325532","article-title":"A real-time QRS detection algorithm","author":"Pan","year":"1985","journal-title":"IEEE Trans Biomed Eng"},{"key":"B17","doi-asserted-by":"publisher","first-page":"e0334607","DOI":"10.1371\/journal.pone.0334607","article-title":"Hybrid machine learning models for enhanced arrhythmia detection from ECG signals using autoencoder and convolution features","volume":"20","author":"Biswas","year":"2025","journal-title":"PLoS One"},{"key":"B18","doi-asserted-by":"publisher","first-page":"57119","DOI":"10.1109\/ACCESS.2025.3555518","article-title":"Winograd transform-based fast detection of heart disease using ECG signals and chest X-ray images","volume":"13","author":"Anand","year":"2025","journal-title":"IEEE Access"},{"key":"B19","doi-asserted-by":"publisher","first-page":"100617","DOI":"10.1016\/j.biosx.2025.100617","article-title":"Enhanced cardiovascular diagnostics using wearable ECG and bioimpedance monitoring with LightGBM classifier","volume":"24","author":"Jain","year":"2025","journal-title":"Biosens Bioelectron: X"},{"key":"B20","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1007\/s13042-025-02822-7","article-title":"Automated human emotion recognition from EEG signals using chaotic local binary pattern and ensemble learning","volume":"17","author":"Chhabra","year":"2026","journal-title":"Int J Mach Learn Cybern"},{"key":"B21","doi-asserted-by":"publisher","first-page":"988","DOI":"10.3390\/electronics15050988","article-title":"Large-scale real-world smartphone photoplethysmography datasets for vascular assessment","volume":"15","author":"Joki\u0107","year":"2026","journal-title":"Electronics"},{"key":"B22","author":"Guo","year":""},{"key":"B23","author":"Li","year":""},{"key":"B24","doi-asserted-by":"crossref","DOI":"10.1007\/11787006_1","article-title":"Differential privacy","volume-title":"Automata, Languages and Programming","author":"Dwork","year":"2006"},{"key":"B25","year":""},{"key":"B26","doi-asserted-by":"publisher","first-page":"21434","DOI":"10.1364\/OE.16.021434","article-title":"Remote plethysmographic imaging using ambient light","volume":"16","author":"Verkruysse","year":"2008","journal-title":"Opt Express"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1900115\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T04:33:41Z","timestamp":1785299621000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1900115\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,29]]},"references-count":26,"alternative-id":["10.3389\/fdgth.2026.1900115"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1900115","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,29]]},"article-number":"1900115"}}