Research Article
Application of Artificial Neural Network Models in Segmentation and Classification of Nodules in Breast Ultrasound Digital Images
Table 1
Evaluation metrics applied to phantoms images regarding the five segmentation techniques applied in the tests.
| ā | AOM (%) | AUM (%) | AVM (%) | CM (%) | CP (%) | CR (%) | (%) | (%) | Err. (%) | FPR (%) |
| Active contour | 83.07 | 15.44 | 1.90 | 88.57 | 84.55 | 98.10 | 83.07 | 94.73 | 5.27 | 0.87 | Region growing | 75.02 | 22.50 | 3.23 | 83.09 | 77.49 | 96.76 | 75.02 | 91.51 | 8.48 | 1.06 | Fuzzy -means | 72.38 | 19.75 | 18.51 | 78.03 | 80.24 | 81.48 | 72.38 | 87.81 | 12.18 | 8.02 | -means | 60.73 | 9.72 | 36.78 | 71.41 | 90.27 | 63.21 | 60.73 | 77.63 | 22.36 | 27.69 | SOM | 82.50 | 13.17 | 5.17 | 88.05 | 86.82 | 94.82 | 82.50 | 94.32 | 5.67 | 2.02 |
| Required value | 100 | 0 | 0 | 100 | 100 | 100 | 100 | 100 | 0 | 0 |
|
|