CT and MRI multi-modal medical image fusion using weight-optimized anisotropic diffusion filtering

The purpose of multi-modal medical image fusion is to improve the accuracy of clinical diagnosis; the fused image is produced by retaining the essential characteristics and precise details of the source images. In this paper, we proposed a weight-optimised anisotropic diffusion filtering (WOADF) method to merge computed tomography (CT) image and magnetic resonance (MRI) image. Diffusion of the intensities at the boundaries and detecting significantly meaningful edges in a medical image are two manifestations of an edge-preserving filter. WOADF is an effective method for realising these specifications within a scale space that consists of multiple dimensions. First, the CT and MRI images are decomposed into base layer and detail layer then weight map layers are created through image smoothing with an edge-preserving approach. And these weights are optimised by using the proposed WOADF. Lastly, by using the fusion rule, the optimized weight map layers and the fusion decision map are used to make a fused image. Compared to state-of-the-art standards, the proposed WOADF method outperforms them experimentally in terms of detail retention and visual effect.