Key result
Smartwatch-reconstructed 12-lead ECGs achieve ~81% visual correlation with conventional 12-lead recordings.
Why the study?
Smartwatch cardiac assessment is limited by single-lead ECG acquisition and the inability to directly obtain augmented limb leads needed to evaluate ischemic and arrhythmogenic conditions.
Does RR interval-guided synchronization and analytical reconstruction of augmented limb leads from sequential smartwatch recordings achieve clinically meaningful diagnostic and quantitative agreement with conventional 12-lead ECG in patients with cardiac conditions?
Observational (n=20)
No
Does RR interval-guided synchronization and analytical reconstruction of augmented limb leads from sequential smartwatch recordings achieve clinically meaningful diagnostic and quantitative agreement with conventional 12-lead ECG in patients with cardiac conditions?
This proof-of-concept study demonstrates that a full 12-lead ECG including augmented limb leads can be analytically reconstructed from sequential smartwatch recordings using RR-interval resampling, achieving high clinical and quantitative agreement with standard 12-lead ECGs except in paced rhythms.
Answers are grounded in this paper and cite their sources.
Supports smartwatch 12-lead reconstruction in small cohorts; leaves open larger validation before clinical use.
Background: Smartwatches have significantly expanded personal health monitoring capabilities; however, their application in comprehensive cardiac assessment remains limited by single-lead electrocardiogram (ECG) acquisition. The inability to directly obtain augmented limb leads represents a major limitation for the evaluation of ischemic and arrhythmogenic conditions. Objective: This study aimed to present and preliminarily evaluate a methodology for generating augmented limb leads (augmented voltage right [aVR], augmented voltage left [aVL], and augmented voltage foot [aVF]) from sequential smartwatch recordings through signal synchronization and RR interval alignment, and to combine this approach with precordial smartwatch recordings to reconstruct a full 12-lead ECG in a proof-of-concept setting. Methods: A signal-processing framework was developed to preserve ECG morphology across sequential smartwatch lead recordings. The method is based on RR interval-guided temporal resampling, enabling interlead synchronization and analytical reconstruction of augmented limb leads. A prospective proof-of-concept clinical validation was conducted in 20 cases with diverse cardiac pathologies, including 1 healthy control. Smartwatch-derived ECGs were compared with standard 12-lead ECGs. Evaluation included expert-based clinical interpretation by 85 health care professionals and objective quantitative analysis of key electrocardiographic parameters, including heart rate, PR interval, QRS duration, QT interval, QTc, RR interval, P-wave duration, and PP interval. Mean values, SDs, and intermethod differences were calculated to assess agreement and variability. Results: Reconstruction of a 12-lead ECG from smartwatch-acquired signals was successfully achieved in all cases. Visual correlation scores between smartwatch-derived and conventional ECGs ranged from 78.1% to 91.3% across specialty-level means (mean 81.1, SD 22.7), indicating high perceived concordance; individual case-level scores ranged more widely, from 47.0% to 96.3%. Evaluator-perceived axis agreement reached 95.4% across all cases, while the investigators' direct axis measurements showed complete concordance in all cases. Quantitative analysis demonstrated small mean differences between modalities, with particularly high agreement observed in structurally stable parameters such as heart rate (mean 77.2, SD 22.3 vs mean 75.9, SD 23.1 bpm; Δ=1.3 bpm) and PR interval (mean 176.2, SD 29.1 vs mean 180.8, SD 19.8 ms; Δ=-4.6 ms). Greater variability was observed in atrial activation parameters such as P-wave duration (Δ=-13.3 ms) and time-dependent parameters such as QT (Δ=-19.2 ms) and QTc (Δ=-12.7 ms). Conclusions: This proof-of-concept study demonstrates the feasibility of reconstructing a full 12-lead ECG from smartwatch-derived signals using a synchronization-based approach. The results demonstrated generally small mean differences between conventional and smartwatch-derived ECG measurements, although larger interval differences were observed in some individual cases. Importantly, due to the sequential nature of signal acquisition, the proposed methodology does not aim to achieve direct waveform equivalence but rather clinically meaningful agreement in electrocardiographic parameters. These findings support the potential of smartwatches as accessible tools for advanced cardiac monitoring and broaden their applicability beyond single-lead screening.
No takes yet. Share an insight, caveat, or question.
Suárez et al. (2026) conducted an observational in Diverse cardiac pathologies including ischemic heart disease, bundle branch blocks, atrial arrhythmias, and pacemaker rhythms (n=20). Smartwatch-derived 12-lead ECG (Apple Watch Series 9 with RR interval-guided temporal resampling and analytical reconstruction of augmented limb leads) vs. Conventional 12-lead electrocardiograph (Philips TC30) was evaluated on Visual correlation score, diagnostic concordance, and quantitative interval agreement between smartwatch-derived and conventional 12-lead ECGs. Smartwatch-reconstructed 12-lead ECGs achieved a mean visual correlation score of 81.1%, 79.4% diagnostic concordance, and 95.4% perceived axis agreement compared to conventional 12-lead ECGs.
Synapse has enriched 10 related papers. Consider them for comparative context: