Table 4 Validation results for each dataset
From: Development and performance evaluation of a deep learning lung nodule detection system
Dataset | No.of Nodules | Sensitivity | FPs per scan (FPs in total) | CPM score | |||
|---|---|---|---|---|---|---|---|
Total | Solid | Part-solid | GGN | ||||
Internal | |||||||
LIDC-IDRI (127 cases) | 152 | 0.980 (149/152) | 0.985 (134/136) | 1.000 (7/7) | 0.889 (8/9) | 3.10 (394) | 0.960 |
External | |||||||
SPIE-AAAM (70 cases) | 83 | 0.988 (82/83) | 0.984 (63/64) | 1.000 (15/15) | 1.000 (4/4) | 5.88 (352) | 0.893 |
LNDb (177 cases) | 221 | 0.959 (212/221) | 0.970 (193/199) | 1.000 (7/7) | 0.800 (12/15) | 7.25 (1284) | 0.783 |