Research Article
High-Performance Method for Brain Tumor Feature Extraction in MRI Using Complex Network
Table 3
Results of evaluation parameters of five classification algorithms on T2 and T2-FLAIR MRI.
| Type MRI | Classifier | Accuracy (%) | Precision (%) | Recall (%) | F1-score (%) |
| T2 | SVM | 99.38 | 99.23 | 100 | 99.61 | Random forest | 99.71 | 99.91 | 99.72 | 99.81 | Logistic regression | 99.84 | 99.93 | 99.98 | 99.96 | Naive Bayes | 95.78 | 97.34 | 97.33 | 97.29 | kNN | 98.84 | 99.56 | 98.98 | 99.26 |
| T2-FLAIR | SVM | 99.69 | 100 | 99.61 | 99.80 | Random forest | 99.66 | 99.89 | 99.68 | 99.78 | Logistic regression | 99.69 | 99.71 | 99.89 | 99.80 | Naive Bayes | 98.55 | 99.60 | 98.58 | 99.07 | kNN | 98.77 | 99.07 | 99.38 | 99.22 |
|
|