A Clinical Decision Support System for Femoral Peripheral Arterial Disease Treatment

Table 10

Comparison of MLP-23 and RBFNN.

	MLP-23		RBFNN		Statistical significance
	Mean ± SD	95% CI	Mean ± SD	95% CI
AUC	0.839±0.018	0.828–0.850	0.949±0.028	0.931–0.966	+
Cutoff	0.542±0.016	0.532–0.552	0.510±0.011	0.503–0.517	+
Accuracy	0.838±0.017	0.827–0.848	0.950±0.022	0.936–0.964	+
Sensitivity	0.835±0.020	0.823–0.847	0.953±0.015	0.944–0.963	+
Specificity	0.840±0.018	0.829–0.851	0.948±0.030	0.929–0.966	+
PPV	0.824±0.021	0.811–0.836	0.942±0.034	0.920–0.963	+
NPV	0.851±0.019	0.839–0.862	0.958±0.013	0.949–0.966	+
F-score	0.829±0.018	0.818–0.840	0.947±0.024	0.932–0.962	+
Yuden index	0.675±0.034	0.654–0.697	0.901±0.044	0.873–0.928	+
H-L	10.211±3.409	8.098–12.324	7.880±1.557	6.915–8.845	−