Table 3 Direct prediction metrics for tumor cells under different models.
From: Improving lung cancer pathological hyperspectral diagnosis through cell-level annotation refinement
Method | Accuracy(%) | Specificity(%) | Sensitivity(%) | Dice(%) |
|---|---|---|---|---|
RF | 90.51(±5.75) | 99.84(±0.20) | 2.46(±5.27) | 4.26(±8.21) |
SVM | 90.45(±5.96) | 99.93(±0.09) | 2.16(±2.74) | 3.98(±4.82) |
2D CNN | 91.39(±4.66) | 97.81(±1.56) | 34.49(±16.56) | 39.84(±13.88) |
3D CNN | 91.81(±5.19) | 98.81(±0.95) | 31.01(±16.10) | 39.07(±15.09) |
Ours | 92.52(±4.36) | 97.26(±1.66) | 54.34(±17.21) | 55.28(±13.14) |
