Self-Supervised Multimodal Biosignal Processing for Early Detection of Cardiac Arrhythmias Using Wearable
##plugins.themes.bootstrap3.article.sidebar##
##plugins.themes.bootstrap3.article.main##
Abstract
This study proposes a multimodal self-supervised framework for early detection of cardiac arrhythmias based on wearables combining 1-lead ECG, PPG, and IMU. The core method includes contrastive pretraining + masked reconstruction on synchronized windows and adaptive fusion weighted by Signal Quality Index (SQI) and aleatoric uncertainty, complemented by domain adaptation for invariant
representation across devices and populations. The unlabeled corpus for pretraining contains 2,400 hours of free-living data from 820
participants (three different devices), while fine-tuning and clinical testing used 1,100 hours of labeled data (n=210; paroxysmal AF,
PVC/PAC, SVT, episodic brady/tachycardia). In subject-wise testing, the model achieved Se 92.8%, Sp 97.1%, F1 90.3%, AUROC 0.972 for AF; F1 83.6% for PVC/PAC; and Se 88.9% for SVT. At episode-level evaluation (≥30 s), AF sensitivity was 94.6% with false alarms per hour
(FPh) of 0.28 and a median time-to-detection of 22 s. Robustness increased at high activity (ECE 0.032, NLL −27%), leave-device-out
generalization remained strong (AUROC 0.957), and the on-device implementation met resource limits (~68 ms/window on an edge-class MCU, <2.3 MB memory). These results demonstrate that signal quality/uncertainty-aware multimodal SSL can suppress false alarms without sacrificing sensitivity, enabling reliable and label-efficient home monitoring for wearable-based arrhythmia screening.
##plugins.themes.bootstrap3.article.details##
[2] Lubitz, S. A., et al. (2022). Detection of Atrial Fibrillation in a Large Population Using Wearable Devices: The Fitbit Heart Study. Circulation, 146(19), 1415–1424. doi:10.1161/CIRCULATIONAHA.122.060291.
[3] Pereira, T., et al. (2020). Photoplethysmography-based atrial fibrillation detection: a review. npj Digital Medicine, 3, 3. doi:10.1038/s41746-019-0207-9.
[4] Orphanidou, C., et al. (2015). Signal-quality indices for the electrocardiogram and photoplethysmogram: derivation and applications to wireless monitoring. IEEE Journal of Biomedical and Health Informatics, 19(3), 832–838. doi:10.1109/JBHI.2014.2338351.
[5] Rahman, S., et al. (2022). Robustness of electrocardiogram signal quality indices. Journal of The Royal Society Interface, 19(190), 20220012. doi:10.1098/rsif.2022.0012.
[6] harlton, P. H., et al. (2023). The 2023 wearable photoplethysmography roadmap. Physiological Measurement, 44(11), 111001. doi:10.1088/1361-6579/acead2.
[7] Mehari, T., et al. (2022). Self-supervised representation learning from 12-lead ECG data. Computers in Biology and Medicine, 141, 105114. doi:10.1016/j.compbiomed.2021.105114.
[8] Kiyasseh, D., Zhu, T., & Clifton, D. (2021). CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patients. Proceedings of ICML (PMLR), 139, 5606–5618. (PMLR; prosiding ICML umumnya terindeks Scopus)
[9] Yang, S., et al. (2024). Masked self-supervised ECG representation learning via time–frequency masking and reconstruction. Neural Computing and Applications. doi:10.1007/s00521-024-09486-4.
[10] Sarkar, P., & Etemad, A. (2022). Self-supervised ECG Representation Learning for Emotion Recognition. IEEE Transactions on Affective Computing, 13(3), 1541–1554. doi:10.1109/TAFFC.2020.3014842. (Contoh kuat SSL pada ECG; metodologi relevan).
[11] Inui, T., et al. (2020). Use of a Smart Watch for Early Detection of Paroxysmal Atrial Fibrillation. JMIR Cardio, 4(1), e14857. doi:10.2196/14857.
[12] Xu, H., et al. (2021). Assessing Electrocardiogram and Respiratory Signal Quality in Wearable Recordings: An Unsupervised Isolation-Forest Approach. JMIR mHealth and uHealth, 9(8), e25415. doi:10.2196/25415.
[13] Niu, L., et al. (2020). A Deep-Learning Approach to ECG Classification Based on Adversarial Domain Adaptation. Healthcare (Basel), 8(4), 437. doi:10.3390/healthcare8040437
[14] Gliner, V., et al. (2023). Using domain adaptation for classification of healthy and abnormal ECG in mobile-captured images. Scientific Reports, 13, 15463. doi:10.1038/s41598-023-40693-6
[15] Kim, K. B., & Baek, H. J. (2023). Photoplethysmography in Wearable Devices: A Comprehensive Review of Technological Advances, Current Challenges, and Future Directions. Electronics, 12(13), 2923. doi:10.3390/electronics12132923.