A Maze-Based Computational Framework for Tumor Localization in the Brain

This paper presents a computational framework that models the brain as a maze-like network for tumor localization and surgical planning. Starting from an abstract structure without prior information about healthy or malignant regions, we generate a voxel-based maze. Once MRI data is available, voxel weights are assigned according to tumor probability. Multiple maze variants are generated and analyzed using A* and Minimum Spanning Tree (Prim) algorithms. This allows the identification of critical voxels—points where malignant tissue is likely concentrated—providing precise targets for neurosurgery.