An image fusion algorithm identifies in-depth parameters of disease variables and creates output images that retain all the significant and viable information that is gathered from the source images without adding additional artifacts or distortions. A variety of performance measures are used to evaluate images prospectively and to fuse images, including structure similarity indexes, standard deviations, edge detection, correlation coefficients, high pass correlations, average gradients, root-mean-square errors, peak signal-to-noise ratios, and entropy. Multi-scale and multi-directional DWT and CWT can highlight pathological information in images. The principal components method is used to extract tumor features by using weighted average fusion rules. We compare the proposed method with existing techniques to prove that it provides high spatial and spectral resolution, as well as high structural and mutual information content in the merged image, thereby enhancing the tumor region.