Brain Tumor Detection through Image Fusion Using Enhanced Dual Channel Pulse Coupled Neural Networks and Grey Wolf Optimizer

Image fusion is particularly crucial for diagnostic imaging in medical applications such as radiation therapy and image-guided surgeries. Medical image fusion seeks to improve diagnostic accuracy by preserving important characteristics and features from the individual pictures in the combined image. This study introduces a novel fusion methodology for MRI and CT medical imaging by decomposing the source images as base and detail layers using a novel three-scale decomposition strategy that employs Gaussian and Guided filters. Gaussian curvature directs the guided filtering procedure for each source image. The base layers are fused using the Proposed Grey Wolf Optimization algorithm (PGWO), which contains an objective function designed to maximize entropy, edge strength, and standard deviation. In order to integrate the detail layers, the activity level information is simultaneously determined using the Enhanced Dual Channel PCNN. To evaluate the effectiveness of the proposed method, thirty slices of seven different types of medical images from various sources were analyzed and compared both visually and statistically with existing approaches. According to experimental data, the suggested approach performs better than traditional approaches in terms of both objective metrics and qualitative image quality. Quantitative findings show notable advancements over current techniques: Standard deviation rises from 15.5 to 32.7%, spatial frequency from 38.2 to 70.5%, mutual information from 42.8 to 62.9%, edge strength from 37.4 to 61.9%, structural similarity index from 37.8 to 43.8%, and image entropy from 12 to 18%.