Medical Image Fusion using Interval Type 2 Fuzzy Logic

摘要

In order to improve clinical accuracy and to take better decisions, medical image fusion is used. It involves integration of the essential features present in different medical images into a single image. Different imaging modalities like, Positron Emission Tomography (PET), Computed Tomography (CT), Magnetic Resonance Imaging (MRI) and others capture different details. Dense tissue structures are visible in CT scan whereas soft tissues are visible in MRI scan. If these two scan images are fused into a single image, doctors will be able to diagnose and plan treatment for patients. In this paper, an interval type 2 fuzzy logic based image fusion technique for fusing CT and MRI images is presented. Otsu's segmentation method is used to segment dense tissue from CT image and soft tissue from MRI images. Later fusion is performed using interval type 2 fuzzy logic. Discrete wavelet transform is used to perform multiresolution fusion. Results are compared with type 1 fuzzy logic system and are found to outperform for most of the performance metrics. Sugeno type 2 fuzzy inference system produces better results compared to Mamdani Inference system.