Table 4 Average diagnostic performance of inexperienced pathologists with/without pathomics assisted.
From: Identifying invasiveness to aid lung adenocarcinoma diagnosis using deep learning and pathomics
Model name | ACC | AUC | 95% CI | SE | SP | PPV | NPV | Cohort |
|---|---|---|---|---|---|---|---|---|
Pathomics | 0.817 | 0.897 | 0.8572–0.9373 | 0.795 | 0.865 | 0.925 | 0.667 | Training |
Pathomics | 0.814 | 0.807 | 0.6892–0.9239 | 0.900 | 0.632 | 0.837 | 0.750 | Test |
Intermediate | 0.716 | 0.675 | 0.6168–0.7338 | 0.788 | 0.562 | 0.799 | 0.606 | Training |
Intermediate | 0.695 | 0.656 | 0.5324–0.7799 | 0.765 | 0.547 | 0.785 | 0.615 | Test |
Intermediate+AI | 0.799 | 0.760 | 0.7054–0.8138 | 0.871 | 0.649 | 0.850 | 0.738 | Training |
Intermediate+AI | 0.807 | 0.769 | 0.6618–0.7690 | 0.875 | 0.663 | 0.855 | 0.757 | Test |
Junior | 0.550 | 0.539 | 0.4705–0.6083 | 0.568 | 0.511 | 0.709 | 0.361 | Training |
Junior | 0.566 | 0.547 | 0.4103–0.6845 | 0.600 | 0.495 | 0.714 | 0.373 | Test |
Junior+AI | 0.773 | 0.772 | 0.7144–0.8302 | 0.774 | 0.770 | 0.876 | 0.622 | Training |
Junior+AI | 0.797 | 0.759 | 0.6447–0.8729 | 0.865 | 0.653 | 0.847 | 0.700 | Test |