Fig. 2
From: A deep learning approach identifies new ECG features in congenital long QT syndrome
Identification of ECG features importance. Box plots showing the score Grad-CAM corresponding to the P wave, QRS complex, and the S segment with the T wave calculated for 100 control ECGs correctly classified by the best performing DL models developed for A LQTS1, B LQTS2, and C LQTS3 ECG classification. *** Adjusted p-values ≤.001