Table 3.

Classification accuracies for the coupled pulmonary hypertension metrics models

LOOCV	AUC	Misclassification error	Sensitivity	Specificity
0D + 1D	0.89	0.21	0.88	0.47
0D + 1D + PA	0.9	0.13	0.93	0.67
0D + 1D + PA + CMR (all)	0.89	0.14	0.97	0.47
0D + 1D + PA + CMR	0.91	0.08	0.97	0.73

^NoteLOOCV: leave-one-out cross validation; AUC: area under the curve. For all models, threshold is not applicable. Zero-dimensional (0D) metrics derived from the 0D Windkessel model. One-dimensional (1D) metrics derived from the 1D wave model. Pulmonary artery (PA) metrics derived from the two-dimensional images of the main PA. Cardiac magnetic resonance (CMR) metrics derived solely from measurements on the cardiac images, with AUC > 0.8. CMR (all) includes all measured metrics derived from the cardiac images.