Table 3:

ROC analysis of individual and combined MRI modalities

	AUC	Sens	Spec	PPV	NPV	P Value^{^a}
C vs PD/MSA-P/PSP						.013
DTI+R2*	0.88	0.80	0.83	0.82	0.81
DTI^{^b}	0.80	0.81	0.71	0.60	0.87
R2*	0.75	0.69	0.69	0.55	0.81
C vs PD						.001
DTI+R2*	0.91	0.86	0.80	0.82	0.89
DTI	0.82	0.74	0.76	0.75	0.76
R2*	0.78	0.71	0.75	0.71	0.74
PD vs MSA-P/PSP						.038
DTI+R2*	0.94	0.86	0.87	0.88	0.84
DTI	0.89	0.83	0.80	0.82	0.81
R2*	0.87	0.87	0.77	0.87	0.77
PD vs MSA-P						.006
DTI+R2*	0.99	0.97	1.00	1.00	0.93
DTI	0.89	0.83	0.86	0.79	0.86
R2*	0.91	0.86	0.86	0.94	0.70
PD vs PSP						.156
DTI+R2*	0.99	0.97	1.00	1.00	0.94
DTI	0.97	0.94	0.94	0.97	0.89
R2*	0.87	0.80	0.83	0.82	0.81
MSA-P vs PSP						.435
DTI+R2*	0.98	0.94	1.00	1.00	0.93
DTI	0.96	0.94	0.92	0.94	0.92
R2*	0.89	0.86	0.80	0.82	0.81

Note:—Sens indicates sensitivity; Spec, specificity; PPV, positive predictive value; NPV, negative predictive value; AUC, area under the curve; C, controls.
↵a ROC curves were compared between the models, including all MRI measurements and that with DTI measurements only.
↵b Models for DTI measurements were generated by including both FA and MD features.