Table 2.
Performance and confusion matrices of machine learning (ML) algorithms applied for the detection and classification of glioma in the study subjects.
| Sample Type | Algorithm Applied |
Confusion Matrices of ML Algorithms 1 | No. of Features Identified 2 | Group | Precision | Recall | F1-Measure |
|---|---|---|---|---|---|---|---|
| Glioma vs. Control (n = 42) |
Extra Tree Classifier | [16 0] | 104 | Control | 1.00 | 1.00 | 1.00 |
| [0 26] | Tumor | 1.00 | 1.00 | 1.00 | |||
| Logistic Regression | [16 0] | 01 | Control | 0.94 | 1.00 | 0.97 | |
| [1 25] | Tumor | 1.00 | 0.96 | 0.98 | |||
| Random Forest |
[16 0] | 158 | Control | 1.00 | 1.00 | 1.00 | |
| [0 26] | Tumor | 1.00 | 1.00 | 1.00 | |||
| LGG vs. HGG (n = 25) | Extra Tree Classifier | [4 5] | 107 | LGG | 0.80 | 0.44 | 0.57 |
| [1 15] | HGG | 0.75 | 0.94 | 0.83 | |||
| Logistic Regression | [4 5] | 92 | LGG | 1.00 | 0.44 | 0.62 | |
| [0 16] | HGG | 0.76 | 1.00 | 0.86 | |||
| Random Forest |
[2 7] | 88 | LGG | 0.67 | 0.22 | 0.33 | |
| [1 15] | HGG | 0.68 | 0.94 | 0.79 |
1 Key used: [true negative false positive]; [false negative true positive]; 2 details of features/spectral regions identified in each case are provided in the Supplementary Material as Table S1.