Table 2:

Section-level and patient-level test results for ViT, ResNet50, and DenseNet121 models

ViTResNet50DenseNet121
Section-level classification accuracy82%72%78%
Patient-level accuracy93%62%62%
Patient-level sensitivity0.901.00.8
Patient-level specificity0.950.420.52
Patient-level PPV0.900.470.47
Patient-level NPV0.951.00.83