Studies Correlating Glioma Survival with In Vivo, Ex Vivo, or Patient Features by ML Approaches
| Study | Cohort | Feature source | Analysis method | Validation method | Result |
| Zhang et al. (49) | GBM (n = 28) | MRI, ex vivo, patient | Logistic regression, SVM, decision tree, neural network | Leave-one-out cross-validation | 96% ACC |
| Nie et al. (50) | GBM (n = 69) | MRI | Deep learning | Cross-validation | 89% ACC |
| Macyszyn et al. (51) | GBM (n = 105) | MRI | Feature preselection, SVM | Cross-validation | 76% ACC |
| Emblem et al. (52) | Glioma (n = 235) | MRI, tumor volume, patient | SVM | Multicenter cross-validation | 94% SENS, 38% SPEC |
GBM = glioblastoma; SVM = support vector machine; ACC = accuracy; SENS = sensitivity; SPEC = specificity.