Research Article

Radiomics Analysis of Lymph Nodes with Esophageal Squamous Cell Carcinoma Based on Deep Learning

Table 4

The predictive performance of the best model and the two radiologists in the test cohort.

 Training cohortTest cohort
SENSPEACCAUCSENSPEACCAUC

Radiologist 1 (15 years of experience)0.690.660.670.670.650.670.660.66
Radiologist 2 (5 years of experience)0.480.730.610.610.430.740.590.59
Model III-RF0.740.780.760.830.760.760.760.80

SEN, sensitivity; SPE, specificity; ACC, accuracy; AUC, area under the receiver operating characteristic curve.