Research Article
An Improved Fuzzy Connectedness Method for Automatic Three-Dimensional Liver Vessel Segmentation in CT Images
Table 1
Segmentation performance of the improved fuzzy connectedness method on the synthetic dataset.
| | | Group 1 () | Group 2 () | Group 3 () | Group 4 () | Group 5 () | Group 6 () | | ACC (%) | SEN (%) | SPE (%) | DICE (%) | ACC (%) | SEN (%) | SPE (%) | DICE (%) | ACC (%) | SEN (%) | SPE (%) | DICE (%) | ACC (%) | SEN (%) | SPE (%) | DICE (%) | ACC (%) | SEN (%) | SPE (%) | DICE (%) | ACC (%) | SEN (%) | SPE (%) | DICE (%) |
| | σ2 = 0 | 99.7 ± 0.1 | 96.6 ± 5.1 | 99.7 ± 0.2 | 84.3 ± 3.6 | 99.8 ± 0.1 | 91.2 ± 4.9 | 99.8 ± 0.1 | 86.6 ± 2.6 | 99.8 ± 0.1 | 91.5 ± 3.5 | 99.8 ± 0.1 | 85.4 ± 3.1 | 99.8 ± 0.1 | 91.5 ± 3.5 | 99.8 ± 0.1 | 86.0 ± 3.6 | 99.8 ± 0.1 | 91.5 ± 4.0 | 99.9 ± 0.1 | 85.4 ± 2.4 | 99.7 ± 0.3 | 91.0 ± 3.9 | 99.9 ± 0.1 | 85.6 ± 3.0 | | σ2 = 30 | 99.7 ± 0.2 | 86.7 ± 4.9 | 99.7 ± 0.2 | 83.8 ± 4.1 | 99.7 ± 0.1 | 87.5 ± 7.3 | 99.8 ± 0.1 | 82.2 ± 3.3 | 99.7 ± 0.1 | 89.4 ± 4.7 | 99.8 ± 0.1 | 80.7 ± 3.1 | 99.7 ± 0.1 | 89.9 ± 5.2 | 99.7 ± 0.1 | 80.7 ± 3.6 | 99.7 ± 0.1 | 90.4 ± 5.2 | 99.8 ± 0.1 | 81.0 ± 2.5 | 99.7 ± 0.1 | 88.6 ± 6.6 | 99.8 ± 0.1 | 80.2 ± 2.5 | | σ2 = 45 | 99.5 ± 6.3 | 91.9 ± 4.6 | 99.7 ± 0.1 | 81.3 ± 3.0 | 99.7 ± 0.1 | 88.3 ± 6.1 | 99.8 ± 0.1 | 81.8 ± 4.1 | 99.7 ± 0.1 | 90.4 ± 4.5 | 99.8 ± 0.1 | 80.7 ± 3.1 | 99.7 ± 0.1 | 89.7 ± 5.3 | 99.8 ± 0.1 | 80.8 ± 3.5 | 99.7 ± 0.1 | 89.1 ± 5.3 | 99.8 ± 0.1 | 80.9 ± 2.8 | 99.7 ± 0.1 | 87.8 ± 6.6 | 99.8 ± 0.1 | 80.6 ± 2.7 | | σ2 = 60 | 99.6 ± 0.1 | 91.4 ± 4.3 | 99.7 ± 0.1 | 80.6 ± 2.4 | 99.7 ± 0.1 | 88.2 ± 5.5 | 99.8 ± 0.1 | 82.2 ± 4.1 | 99.7 ± 0.1 | 90.3 ± 4.5 | 99.8 ± 0.1 | 81.0 ± 3.1 | 99.7 ± 0.1 | 90.2 ± 5.1 | 99.8 ± 0.1 | 81.1 ± 3.3 | 99.7 ± 0.1 | 88.6 ± 5.1 | 99.8 ± 0.1 | 81.4 ± 2.6 | 99.7 ± 0.1 | 87.2 ± 6.4 | 99.8 ± 0.1 | 81.1 ± 2.7 |
|
|
ACC = accuracy; SEN = sensitivity; SPE = specificity; DICE = Dice coefficient.
|
