Medical News Observer

Key Points

• The AI model analyzes ECG recordings, integrating ECG morphology, demographic, and heart rhythm data to predict atrial fibrillation risks.
• With its deep learning approach, the model accurately identifies near-term atrial fibrillation, outperforming demographic-only models.
• This advancement could lead to more effective atrial fibrillation screening, shorter ECG monitoring periods, and better identification of patients needing further tests or interventions.

A new AI model developed by researchers at Scripps Research offers an advancement in atrial fibrillation (AFib) screening. This model can detect subtle variations in a person’s normal heartbeat, indicative of atrial fibrillation risk. This breakthrough has the potential to improve the detection of individuals at risk for atrial fibrillation, thereby preventing the condition’s severe complications like stroke and heart failure.

Atrial Fibrillation AI Model

The model analyzed atrial fibrillation-free electrocardiogram (ECG) intervals from a single day within two-week recordings from 459,889 single-lead ECG patches. Utilizing deep learning, model integrated ECG morphology, demographic, and heart rhythm data for predicting atrial fibrillation. Impressively, this model achieved high accuracy in predicting near-term atrial fibrillation, surpassing the performance of models using only patient demographic information. 

This model is a significant step towards creating a more effective screening test for atrial fibrillation, allowing for shorter ECG monitoring periods and improving the identification of patients who might benefit from further testing or intervention.

Reference

Ref: Gadaleta, Matteo, Patrick Harrington, Eric Barnhill, Evangelos Hytopoulos, Mintu P. Turakhia, Steven R. Steinhubl, and Giorgio Quer. 2023. “Prediction of Atrial Fibrillation from At-Home Single-Lead ECG Signals without Arrhythmias.” Npj Digital Medicine 6 (1): 1–9. https://doi.org/10.1038/s41746-023-00966-w