Insights into Imaging

Table 3 Diagnostic performance of radiomics model, combined model, MTD on MRI, and radiologists in validation cohort

From: MRI-based radiomics analysis improves preoperative diagnostic performance for the depth of stromal invasion in patients with early stage cervical cancer

	AUC	SEN	SPE	PPV	NPV
Radiomics Model	0.879 (0.775–0.983)	0.879 (0.727–0.952)	0.846 (0.578–0.973)	0.935 (0.772–0.989)	0.733 (0.448–0.911)
Combined Model	0.886 (0.784–0.988)	0.879 (0.727–0.952)	0.846 (0.578–0.973)	0.935 (0.772–0.989)	0.733 (0.448–0.911)
MTD on MRI	0.844 (0.719–0.969)	0.697 (0.527–0.826)	0.769 (0.497–0.918)	0.885 (0.687–0.970)	0.500 (0.279–0.721)
Senior radiologist	–	75.7%	69.2%	86.2%	52.9%
Junior radiologist	–	63.6%	53.8%	77.8%	36.8%

AUC, Area Under the Curve; SEN, sensitivity; SPE, specificity; PPV, positive predictive value; NPV, negative predictive value; MTD, maximal tumor diameter

Back to article page