An Intelligent Multimodal Medical Image Fusion Model Based on Improved Fast Discrete Curvelet Transform and Type-2 Fuzzy Entropy

Multi-modal medical image fusion has emerged as a famous and efficient tool in medical applications. The major goal of this fusion is to fuse diverse multi-modal medical images attained from varied imaging modalities into a single fused image, which is broadly utilized by surgeons for the precise diagnosis and treatment of diseases. Multi-modal medical images generally consist of captured images with the specific organ of a patient. These images will indicate a modality, which will provide the observed organ in a different way that leads to dissimilar examinations of a specific incident like stroke. The detection with more suitable clinical decisions is taken by the accurate analysis of each modality. Multi-modal medical imaging is an efficient research area that incorporates the improvement of robust techniques, which can facilitate the information of image fusion attained with diverse sets of modalities. The major goal of this paper is to develop the multi-modal medical image fusion model using the new hybrid meta-heuristic approach. At first, the high-frequency sub-bands and low frequency sub-bands of the images that to be fused split by the weighted fast discrete curvelet transform (W-FDCuT). Once the sub-bands are split, the high-frequency sub-bands of the two images are fused by the optimized Type-2 fuzzy entropy. On the other hand, Averaging approach is used for performing the fusion of low frequency sub-bands. Finally, the inverse W-FDCuT is done for generating the final fused image. To improvise the performance of W-FDCuT and Type-2 fuzzy entropy, the hybrid meta-heuristic algorithm named hybrid jaya with sun flower optimization (HJ-SFO) is adopted, thus enhances the significant parameters by maximizing the structural similarity index measure (SSIM). The comparison over the conventional image fusion models proves the efficiency of the proposed model in terms of diverse analysis.