Using Deep Learning to Analyze ECGs and Assess Heart Health for Competitive Sports Eligibility in Adults Aged 18 to 60

What this Trial Is About

Researchers are investigating the use of a Deep Learning DL model to analyze electrocardiogram ECG traces from athletes to provide a probability-based assessment of cardiovascular disease risk. This observational study aims to develop and validate this model to help determine athletes eligibility for competitive sports. The study follows current guidelines and cardiological standards for evaluating athletes heart health. The study involves two groups of athletes. The first group of 455 athletes has already undergone standard clinical evaluations for sports participation clearance their ECGs and clinical results will be used to train the DL model. The second group of 76 athletes will be evaluated using standard fitness tests and the DL model to validate its accuracy. Participants in both groups are classified as fit or unfit for competitive sports based on medical assessments. Participants will undergo ECG tests and clinical evaluations according to established cardiological guidelines. Researchers will collect and analyze ECG data to train and test the DL model, measuring its accuracy, sensitivity, and specificity in classifying athletes fitness for competition. The study duration for each participant can last up to 12 months from the first medical evaluation to the final decision on sports eligibility.

Deep Learning ECG Evaluation and Clinical Assessment for Competitive Sport Eligibility