Image Segmentation Method with Positron Emission Tomography Time Sequence Images

Abstract

Positron emission tomography(PET)images are often used to detect physiology function.However,PET images have more blurs than anatomic images,such as magnetic resonance imaging(MRI)and computed tomography(CT).With the graylevel of PET images,Doctors need to manually obtain the region of interest(ROI).THis paper presents a PET image segmentation technology based on the wavelet theory.Through sampling,recombination,and fetching the thresholds of images,theimage segmentation with 1960 dynamic PET images of abdominal organs were conducted. Compared with a sigle PET image, the image segmentation technology proposed can accurately and rapidly identify the boundaries of organs,such as the small intestines,the liver,and the kidney,as well as efficiently provide the ROI information for doctors.