A Multicriteria Method for Cervical Tumor Segmentation in Positron Emission Tomography

Abstract

Accurate delineation of tumors is a fundamental requirement for proper planning and subsequent cancer treatment. In this paper, we propose to model the process of tumor segmentation as a multicriteria decision making problem, considering the information embedded in both Positron Emission Tomography (PET) and Computed tomography (CT) images. A set of images of cervical tumors were semi-automated segmented and the results compared with a manual delineation. The results show that using a multiple criteria approach in the segmentation process can improve sensitivity, and the utilization of both PET and CT images may be a factor for improving precision.