PET SFCM image segmentation for alzheimer’s disease

Abstract

Medical images are known to capture the human body in both anatomical and functional view. These images are interpreted with expert domain for clinical analysis. Here, the selection of image sample plays a fundamental role. However, doctors need to manually obtain this process. But, in order to get similarity between the samples automation is definitely required as it reduces the computation time. So, the automation process should be knowledge based to get better results. This paper highlights the knowledge based automation of medical image sample analysis. It presents a significant assessment of PET – SFCM approach for the segmentation of functional medical images which is considered as the value of neighboring pixels in spatial correlation. Here, the proposed method is used to apply the decision support strategy to identify the effective samples from the huge data collection. The proposed algorithm is implemented in Matlab 7.0. The obtained results were analyzed and compared with other two clustering approaches known as K-Means and Fuzzy C-Means. The resultant images encourage the identification and an evaluation of treatment response in a set of oncological constraints.