Table 4 Performance analysis on spatial features.
From: A novel hybrid approach for multi stage kidney cancer diagnosis using RCC ProbNet
Model | Acc. | Class | Prec. | Rec. | F1 |
|---|---|---|---|---|---|
KNC | 0.58 | Grade 0 | 0.75 | 0.86 | 0.80 |
Grade 1 | 0.45 | 0.56 | 0.50 | ||
Grade 2 | 0.47 | 0.64 | 0.54 | ||
Grade 3 | 0.50 | 0.44 | 0.47 | ||
Grade 4 | 1.00 | 0.62 | 0.76 | ||
Avg. | 0.63 | 0.62 | 0.61 | ||
SGD | 0.71 | Grade 0 | 0.75 | 0.86 | 0.80 |
Grade 1 | 0.00 | 0.00 | 0.00 | ||
Grade 2 | 0.70 | 0.64 | 0.67 | ||
Grade 3 | 0.62 | 0.94 | 0.75 | ||
Grade 4 | 0.86 | 0.92 | 0.89 | ||
Avg. | 0.59 | 0.67 | 0.62 | ||
RF | 0.91 | Grade 0 | 0.88 | 1.00 | 0.93 |
Grade 1 | 1.00 | 0.78 | 0.88 | ||
Grade 2 | 0.91 | 0.91 | 0.91 | ||
Grade 3 | 0.82 | 0.88 | 0.85 | ||
Grade 4 | 1.00 | 1.00 | 1.00 | ||
Avg. | 0.79 | 0.80 | 0.81 | ||
NB | 0.75 | Grade 0 | 0.50 | 0.57 | 0.53 |
Grade 1 | 0.43 | 0.67 | 0.52 | ||
Grade 2 | 0.59 | 0.91 | 0.71 | ||
Grade 3 | 0.83 | 0.31 | 0.45 | ||
Grade 4 | 1.00 | 0.85 | 0.92 | ||
Avg. | 0.67 | 0.66 | 0.63 | ||
LR | 0.71 | Grade 0 | 0.88 | 0.89 | 0.88 |
Grade 1 | 0.75 | 0.83 | 0.84 | ||
Grade 2 | 0.81 | 0.90 | 0.88 | ||
Grade 3 | 0.90 | 0.91 | 0.90 | ||
Grade 4 | 0.89 | 0.90 | 0.90 | ||
Avg. | 0.79 | 0.77 | 0.80 | ||
GNB | 0.20 | Grade 0 | 0.00 | 0.00 | 0.00 |
Grade 1 | 0.00 | 0.00 | 0.00 | ||
Grade 2 | 0.20 | 0.36 | 0.26 | ||
Grade 3 | 0.29 | 0.38 | 0.32 | ||
Grade 4 | 0.38 | 0.38 | 0.38 | ||
Avg. | 0.17 | 0.22 | 0.19 |
