Table 3.
Dataset imbalance for binary classification analysis
| Rabbit # | 3D:P(cm3) | 3D:N(cm3) | 3D:P/(P+N) | 2D:P(cm2) | 2D:N(cm2) | 2D:P/(P+N) |
|---|---|---|---|---|---|---|
| 1 | 0.10 | 14.48 | 0.7% | 0.14 | 7.15 | 1.9% |
| 2 | 0.10 | 14.48 | 0.7% | 0.16 | 7.13 | 2.2% |
| 3 | 0.09 | 14.49 | 0.6% | 0.11 | 7.18 | 1.5% |
| 4 | 0.07 | 14.51 | 0.5% | 0.13 | 7.16 | 1.8% |
| 5 | 0.06 | 14.52 | 0.4% | 0.12 | 7.17 | 1.7% |
| 6 | 0.05 | 14.53 | 0.3% | 0.13 | 7.16 | 1.8% |
| 7 | 0.06 | 14.52 | 0.4% | 0.10 | 7.19 | 1.4% |
| 8 | 0.03 | 14.55 | 0.2% | 0.06 | 7.23 | 0.8% |
| 9 | 0.03 | 14.55 | 0.2% | 0.06 | 7.23 | 0.8% |
| Global | 0.59 | 130.63 | 0.5% | 1.01 | 64.60 | 1.6% |
Class distribution for the MRI2D and MRI3D datasets. The positive (P = damaged tissue) and negative (N = undamaged tissue) class sizes (MRI3D: volume, MRI2D: area) are provided both on an animal-wise basis (cohort #1) and for the global datasets.