Abstract
Background Postoperative atrial fibrillation (POAF) affects 20 to 50% of patients undergoing cardiac surgery and is associated with longer hospital stays and adverse outcomes. Although several risk factors for developing POAF have been identified, accurate prediction remains challenging. Wearable ECG patches and remote patient monitoring enable continuous heart rhythm surveillance. Using AI models, subtle yet distinct patterns may be recognized that precede POAF development.
Objective This study evaluates whether combining continuous ECG patch monitoring with deep learning algorithms can improve both early risk stratification and near real-time prediction of POAF.
Methods We analyzed continuous ECG and wearable-derived physiology from 20 postoperative cardiac surgery patients enrolled in a prospective monitoring trial. Each patient wore a 14-day adhesive patch sensor (VivaLNK VV-330) capturing per-second ECG and activity streams. Two complementary deep learning pipelines were developed: (1) a daily-level multimodal Transformer, which downsampled ECG and contextual “TAB tokens” into day-wise units to predict AF occurrence and burden, and (2) an hour-ahead forecasting model, which condensed the last two hours of minute-level physiology into attention-weighted summaries to generate rolling, causal predictions of AF risk in the next hour.
Results Across 162,217 downsampled data elements, the daily-level model showed conservative behavior with very low false negatives, consistently identifying AF-positive days and correctly stratifying high-burden episodes. The hour-ahead forecasting model was trained on 5,607 windows and achieved excellent discrimination (AUC 0.95), high specificity (0.98), and strong predictive value (NPV 0.98). Recall-focused calibration further reduced missed AF hours while maintaining low false alarm rates. Together, the two frameworks provided reliable daily burden stratification and fine-grained, near real-time risk forecasts.
Conclusion Continuous multimodal monitoring combined with AI enables accurate POAF detection, daily risk stratification, and rolling hour-ahead forecasts. This dual-resolution framework has the potential to support perioperative decision-making by enabling earlier intervention, more precise surveillance, and better allocation of preventive therapies in cardiac surgery patients.
Competing Interest Statement
The authors have declared no competing interest.
Clinical Trial
NCT04880265
Funding Statement
This study did not receive any funding
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
The details of the IRB/oversight body that provided approval or exemption for the research described are given below:
The study was reviewed and approved by the Mass General Brigham Institutional Review Board (IRB), Boston, MA, USA. Ethical approval was granted under protocol number 2021P000356, which covers prospective remote patient monitoring after cardiac surgery, including use of wearable ECG patch devices and secondary analysis for AI modeling.
I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.
Yes
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Yes
I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.
Yes
Footnotes
↵* Negin Maddah is co-first author
Data Availability
All data produced in the present study are available upon reasonable request to the authors
