Early Brain Tumor Detection Using Fuzzy Logic Decision-Making Models

Abstract

This study presents a fuzzy logic-based decision-making model for the early detection of brain tumors, integrating three key clinical parameters: tumor size (cm), edema volume (ml), and symptom severity (scale 0–10). Triangular membership functions are employed to represent the linguistic variables, while a Mamdani-type inference system evaluates the nonlinear interactions among these inputs to produce a predicted risk percentage. A comprehensive rule base derived from expert knowledge guides the inference process, enabling the system to handle uncertainty and imprecision inherent in medical data. The model was validated using various input combinations, and the results demonstrated strong alignment with clinical reasoning, showing that larger tumors and higher symptom severity significantly increase risk, especially when accompanied by substantial edema. The findings suggest that the proposed fuzzy logic model provides a transparent, flexible, and clinically relevant approach for early brain tumor risk assessment and prioritization in diagnostic workflows.