Research Article
ContourTL-Net: Contour-Based Transfer Learning Algorithm for Early-Stage Brain Tumor Detection
Table 6
Experimental result (%) evaluation of the proposed methodology with the state-of-the-art methods for BR35H dataset.
| | Methods | Sensitivity (recall) | Specificity (TN) | Precision (PPV) | NPV | Accuracy |
| | DenseNet-169-based FC layer model [54] | — | — | — | — | 98.83 | | DCNN with SGD optimization [51] | — | — | — | — | 99.0 | | Proposed model (no augmentation) | 98.44 | 98.57 | 98.75 | 98.22 | 98.50 | | Proposed model | 99.96 | 99.92 | 99.92 | 99.96 | 99.94 |
|
|
Significant performances are denoted in bold.
|
