TABLE 5.

Effectiveness of nomogram and three features for differentiating the positive pathological response by ROC curve analysis in the training cohort.

Parameters	AUC	95%CI	DeLong test
			z statistic	p‐value
Nomogram	0.908	0.839–0.954	–	–
PPF radiomics signature	0.835	0.754–0.898	2.379 a	0.017
Intratumoral radiomics signature	0.822	0.739–0.888	2.762 b	0.006
PSA level	0.630	0.534–0.719	5.114 c	<0.001

Abbreviations: AUC, area under the receiver operating characteristic curve; CI, confidence interval; PPF, periprostatic fat; ROC, receiver operating characteristic curve.

^a Nomogram versus PPF radiomics signature in distinguishing the positive pathological response.

^b Nomogram versus Intratumoral radiomics signature in distinguishing the positive pathological response.

^c Nomogram versus PSA level in distinguishing the positive pathological response.