Transforms with Type 2 Fuzzy Logic Based Hybrid Medical Image Fusion Technique

Abstract

The main objective of Hybrid Multimodal Medical Image Fusion (HMMIF) technique is to preserve salient image features and detailed information of multiple source images to produce a visual enhanced single fused image. The medical image fusion is the combination of visual information contained in any number of input medical images into a single fused image without introducing distortion or information loss. It improves the image quality by preserving specific features for increasing the clinical applicability of medical images for diagnosis and assessment of medical problems. The fused image has various challenges such as appearance of fusion artifacts, edge strength, contrast of input medical image and computational cost. The hybrid fusion techniques have been developed for pathological studies such as neurocysticercosis, degenerative and neoplastic disease affected medical images. In this work, Transform with fuzzy logic based hybrid fusion techniques have been developed for various MRI-SPECT, MRI-PET and MRI-CT medical image fusion applications. In the proposed method, initially NSCT is applied to decompose the input images which give low and high frequency components. The phase congruency is applied on low frequency components of NSCT. The high frequency components of NSCT are fused by maximum fusion rule.  It is further treated by Type 2 fuzzy logic scheme. The fused image is obtained by taking inverse transforms with the coefficients obtained from all frequency bands. The proposed techniques are compared with the state of the art conventional approaches. It is proved with the experimentation that the proposed techniques are superior in terms of qualitative as well as quantitative evaluation. The fused images using proposed algorithms provide the best of both sources modality content useful in visualization and interpretation of the diseases.