Brain tumor detection through image fusion using whale optimization and edge preserving filter

In diagnostic imaging, image fusion remains a significant difficulty, particularly in medical applications like guided image operations and radiation therapy. By maintaining the pertinent details and characteristics of the original images, medical image fusion aims to increase the precision of disease diagnosis. This study suggests a novel methodology for MRI and CT image fusion that uses the proposed tri-scale decomposition with Gaussian and guided filters to decompose the source images into base and detail layers. Each source image is guided through guided filtering using Gaussian curvature as guidance. While the detail layers are fused based on maximum energy assessed using the Krisch compass operator, the base layers are fused based on the whale optimization method, for which the objective function is maximization of entropy, edge strength and pixel intensity. Thirty different kinds of slices of five medical datasets from diverse sources were used to assess the effectiveness of the proposed algorithm both visually and statistically compared to existing approaches. Based on both objective evaluation and qualitative image analysis, the experimental results demonstrated that the suggested strategy performed better than other widely used techniques. In comparison to the existing methods under consideration, the quantitative results show that the proposed algorithm improves the standard deviation by 16%, mutual information by 41%, spatial frequency by 12%, image entropy by 6.5%, edge strength of the fused image by 37%, and structural similarity index by 31%.