Table 5 Direct prediction metrics for cell.
From: Improving lung cancer pathological hyperspectral diagnosis through cell-level annotation refinement
Method | Accuracy (%) | Specificity (%) | Sensitivity (%) | Dice (%) |
---|---|---|---|---|
RF | 88.74(±4.97) | 90.88(±4.65) | 80.56(±14.83) | 66.27(±12.90) |
SVM | 87.50(±6.06) | 90.36(±5.12) | 75.93(±17.86) | 62.53(±15.62) |
2D CNN | 89.73(±4.47) | 90.99(±3.98) | 86.61(±11.89) | 70.05(±9.99) |
3D CNN | 90.85(±4.19) | 92.73(±3.29) | 84.70(±12.71) | 72.04(±9.23) |
Ours | 91.33(±3.71) | 92.53(±3.08) | 87.71(±9.69) | 73.84(±8.55) |