Multimodal Medical Image Fusion Using Hybrid Fusion Techniques for Neoplastic and Alzheimer's Disease Analysis

摘要

A neoplasm is an abnormal growth of cells in the brain, also known as a tumor which causes growth of tumor triggered by DNA mutations within your cells. The Neoplastic disease causes two types of tumor growth. The benign tumors usually grow slowly and cannot spread to other tissues are called as noncancerous growth. The Malignant brain tumors grow quickly and spread to multiple tissues are known as cancerous growth. Alzheimer's disease is a neurological disorder in which the death of brain cells causes memory loss and cognitive decline. During diagnosis of such symptomatic patients, these tumors and abnormal characteristics (structural issues) of brain can be visualized using a feature based fusion of computed tomography, magnetic resonance imaging, positron emission tomography and single photon emission tomography multimodality medical images. In spite of huge progresses, still there is no single modality which can represent all aspects of the human body. In this paper a novel method has been proposed for multimodal medical image fusion using combination of Non-subsampled contourlet transform (NSCT) with Non-subsampled shearlet transform (NSST) hybrid fusion algorithm. Multimodal medical image fusion combines and merges all relevant and complementary information from multiple input images into single composite image which facilitates more precise diagnosis and better treatment. The proposed method is tested on the pilot study datasets of patients infected with Neoplastic and Alzheimer's diseases. The fused hybrid multimodal medical image should convey a better description of the information than the individual input images. Experimental results show the superior solution quality of our approach in comparison to a number of well-known counterpart algorithms.