MRI/PET IMAGE FUSION USING MULTIWAVELET TRANSFORM

Abstract

Image fusion is the technique of blending multiple images to obtain a fused image with more descriptive and reliable information. The keen motivation behind image fusion is to compound the complementary as well as relevant information of several images captured from a common scene, to generate an image comprising the superior features of source images. With the availability of modern instrumentation (medical imaging tools) and developed technologies, medical image fusion has become a vital tool in medical applications. Medical image fusion is a concept dealing with the idea of improving the image content by fusion of multiple images obtained using various imaging tools such as Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), Computed Tomography (CT), Single photon Emission Computed Tomography (SPECT). The objective of this project work is to develop a novel fusion algorithm to fuse two different modality images of brain of a patient to obtain a resulting image with more clear view and complete description. In this work, a framework for medical image fusion based on Multi-Wavelet transform is proposed. Multi-Wavelet transform is improvement over traditional scalar wavelet. In the first part of project work, PET and MRI images are aligned with each other. In the second part, two images are decomposed using Discrete Multi-wavelet transform. In third part of work, these decomposition coefficients are merged correspondingly using edge detection method. In fourth part of work, using these new coefficients, inverse wavelet transform is applied to obtain a fused image with better human/machine perception. Several data set for different diseases are experimented on.