Predicting Advanced Prostate Cancer from Modeling Early Indications in Biopsy and Prostatectomy Samples via Transductive Semi-Supervised Survival Analysis

Table 2

Results of training on Dataset 2 and validating on Dataset 3.

	SVRc	SVRc with Transduction	Cox Model	Cox Model with Transduction
	PSAR Training Performance
CI	0.78	0.79	0.80	0.81
Sensitivity	0.77	0.68	0.82	0.77
Specificity	0.73	0.83	0.71	0.75
Criterion	1.34	1.35	1.38	1.39
	PSAR Validation Performance
CI	0.80	0.81	0.82	0.82
Sensitivity	0.74	0.69	0.79	0.79
Specificity	0.72	0.83	0.72	0.82
Criterion	1.33	1.38	1.39	1.47
	CF Validation Performance
CI	0.88	0.88	0.88	0.88
Sensitivity	1.00	1.00	1.00	1.00
Specificity	0.68	0.78	0.68	0.75
Criterion	1.56	1.66	1.56	1.63