Paperdoi:10.1109/ICASSP55912.2026.11463615
Feasibility of Ectopic Beat Detection and Count Estimation from Smartwatch-Based Photoplethysmography
ICASSP 2026
TL;DR
Pipeline on free-living smartwatch PPG for ectopic-beat (EB) detection and per-segment count estimation, evaluated in three scenarios. Best F1 0.884 for EB vs sinus rhythm; count MAE as low as 0.042 per segment. AF is the main confounder.
- Form factor: wrist
Abstract
Ectopic beats (EBs) are common cardiac irregularities associated with risks of atrial fibrillation (AF) and sudden cardiac death, highlighting the need for reliable long-term monitoring. Electrocardiogram (ECG)-based methods remain the clinical gold standard, but their limited monitoring duration and patient burden restrict long-term daily use. In contrast, smartwatch photoplethysmography (PPG) enables passive, long-term data collection and offers a promising alternative for daily monitoring. In this study, we developed a pipeline using free-living smartwatch PPG data for ectopic beat detection and count estimation. Performance was evaluated under three experimental scenarios reflecting different application contexts. The pipeline achieved an F1 score of up to 0.884 in distinguishing ectopic beats from sinus rhythm and a mean absolute error as low as 0.042 per segment, while AF emerged as a major confounding factor. These findings demonstrate the feasibility of smartwatch-based PPG for ectopic beat monitoring in free-living conditions, providing a foundation for future long-term, passive monitoring solutions.
Verbatim from OpenAlex (opens in a new tab) · Abstract via OpenAlex (publisher-deposited); © the authors/publisher
Results on ChatPPG leaderboards
All leaderboardsNo results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.
Tasks & signals
Tasks
Signals
Keywords
How it works
Cite this paper
APA
Lu, B., Dong, Q., Desai, S., Zhou, H., Tanade, C., Lee, S., & Kang, C. (2026). Feasibility of Ectopic Beat Detection and Count Estimation from Smartwatch-Based Photoplethysmography. ICASSP 2026. https://doi.org/10.1109/ICASSP55912.2026.11463615
BibTeX
@inproceedings{lu2026feasibility,
title = {{Feasibility of Ectopic Beat Detection and Count Estimation from Smartwatch-Based Photoplethysmography}},
author = {Lu, Baiying and Dong, Quan and Desai, Sharanya and Zhou, Hao and Tanade, Cyrus and Lee, Simon and Kang, Chaoyi},
year = {2026},
booktitle = {ICASSP 2026},
doi = {10.1109/ICASSP55912.2026.11463615}
}Built from the hub record. Check it against the publisher before submitting.
Added to the ChatPPG research hub
