A Deep Learning Approach to Classify Fabry Cardiomyopathy from Hypertrophic Cardiomyopathy Using Cine Imaging on Cardiac Magnetic Resonance

Table 4

The -score, accuracy, sensitivity, specificity, and AUC were compared on 5-fold cross-validation, TVGH testing data, and TCVGH external validation data.

Dataset	-score	Accuracy	Sensitivity	Specificity	AUC
5-fold	0.861	0.924	0.895	0.992	0.967
TVGH	0.846	0.909	0.917	0.906	0.914
TCVGH	0.727	0.806	0.727	0.850	0.918