Table 3. Validation results of four cases compared to radiologist data by mean and median values.
| case number | REGION GROWING | ADAPTIVE THRESHOLD | ACTIVE CONTOURS | ||||||
| SMN | std(SMN) | SMD std(SC) | SMN | std(SMN) | SMD std(SC)SMN | SMN | std(SMN) | SMD std(SC)SMN | |
| V1 | 79.4% | 24.8% | 87.0% | 67.3% | 30.5% | 76.4% | 54.5% | 43.0% | 79.5% |
| V2 | 87.1% | 18.0% | 92.5% | 64.5% | 35.9% | 79.0% | 51.7% | 42.6% | 70.2% |
| V3 | 71.3% | 33.1% | 85.5% | 67.5% | 30.1% | 78.0% | 72.8% | 28.5% | 81.9% |
| V4 | 49.6% | 38.1% | 70.5% | 65.9% | 35.7% | 81.9% | 58.2% | 43.6% | 83.7% |
| 71.9% | 16.2% | 86.3% | 66.3% | 1.4% | 78.5% | 59.3% | 9.4% | 80.7% | |
Presented are validation results: mean and median similarities of four cases compared to manually-segmented data from the radiologist. Similarity (S) was evaluated on a slice-by-slice basis using validation-only parameters as mean (SMN) or median (SMD) of all the slices in a case. Std is the standard deviation of mean similarities of slices within a case.