Key result
A deep learning model analyzing clinical data and resting 12-lead ECG achieved moderate discrimination for pre-participation screening in athletes, with an AUC of 0.72 (95% CI 0.66-0.78).
Its licence doesn't allow us to show a copy here.
Why the study?
Pre-participation cardiovascular screening prevents sudden cardiac death in athletes, but ECG interpretation requires expertise and remains resource-intensive.
Does a multimodal deep learning model analyzing ECG and clinical data accurately classify fitness for competitive sports in athletes undergoing pre-participation screening?
Observational (n=526)
No
Does a multimodal deep learning model analyzing ECG and clinical data accurately classify fitness for competitive sports in athletes undergoing pre-participation screening?
Effect estimate: AUC 0.72 (95% CI 0.66-0.78)
An AI-based multimodal analysis of ECG and clinical data showed moderate diagnostic performance for cardiovascular risk stratification in competitive athletes, suggesting potential as an adjunct to physician assessment.
No takes yet. Share an insight, caveat, or question.
Davide et al. (2026) conducted an observational in Competitive athletes undergoing pre-participation cardiovascular screening (n=526). Deep learning model for analysis of clinical data and resting 12-lead ECG vs. Clinical evaluation (PPS clinical classification) was evaluated on Diagnostic performance against PPS clinical classification (AUC 0.72, 95% CI 0.66-0.78). A deep learning model analyzing clinical data and resting 12-lead ECG achieved moderate discrimination for pre-participation screening in athletes, with an AUC of 0.72 (95% CI 0.66-0.78).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: