A CT-Based Radiomics Nomogram Model for Differentiating Primary Malignant Melanoma of the Esophagus from Esophageal Squamous Cell Carcinoma

Table 3

The performance of radiomics nomogram model for discriminating PMME from ESCC.

	Primary cohort	Validation cohort
AUC	0.975 (0.948-1.000)	0.906 (0.770-1.000)
Accuracy	91.4% [74/81] (85.2-97.5%)	92.7% [38/41] (84.7-100.0%)
Sensitivity	94.7% [18/19] (74.0-99.9%)	88.9% [8/9] (51.8-99.7%)
Specificity	90.3% [56/62] (80.1-96.4%)	93.8% [30/32] (79.2-99.2%)
PPV	75.0% [18/24] (53.3-90.2%)	80.0% [8/10] (44.4-97.5%)
NPV	98.2% [56/57] (90.6-100.0%)	96.8% [30/31] (83.3-99.9%)

AUC: area under curve; ESCC: esophageal squamous cell carcinoma; NPV: negative predictive value; PMME: primary malignant melanoma of the esophagus; PPV: positive predictive value. Data in parentheses are 95% confidence intervals.