Accurate tumor positioning from PET-CT by performing registration based segmentation

Abstract

In this work, an efficient tumor positioning method is proposed by performing registration based segmentation from 18-FDG PET-CT scanners. At the first stage, the tumor is segmented from PET scans by region growing using the manual seeds which employs the SUV monotonous features, and then the tumor contours are transferred to corresponding CT images automatically by registration method which is based on edge preserving scale space for following radiation therapy planning. The experiments results demonstrate the efficiency of proposed method.