The current preoperative way of assessing the degree of malignancy in brain glioma is based on magnetic resonance imaging (MRI) findings and clinical data. We studied 280 cases, of which 111 were high-grade malignancies and 169 low-grade, to acquire regular and interpretable patterns of the relations between glioma MRI features and the degree of malignancy. However, as uncertainties in the data and missing values existed, a fuzzy rule extraction algorithm based on Fuzzy Min-Max Neural Network (FMMNN) was proposed. The performance of Multi-Layer Perceptron network (MLP) trained with error Back-Propagation algorithm (BP), the well-known decision tree algorithm ID3, Nearest Neighbor, and the original Fuzzy Min-Max Neural Network were also evaluated. The results showed that two fuzzy decision rules on only 6 features achieved an accuracy of 84.6% (89.9% for low-grade cases and 76.6% for high-grade ones). Investigations with the proposed algorithm revealed that age, mass effect, edema, post–contrast enhancement, blood supply, calcification, hemorrhage, and signal intensity of the T1-weighted Image were important diagnostic factors.