Integration of clinical data with scanned ECGs using deep learning methods for stroke risk prediction in Indian patients with atrial fibrillation: evidence from the KERALA-AF study.

Compared to ensemble and linear methods, our model achieved the highest numerical AUC, though the differences did not reach statistical significance (Random Forest: p = 0.12; Decision Tree: p = 0.31; KNN: p = 0.23; Logistic Regression: p = 0.38).

also in 111,027 other papers