TUMOR SEGMENTATION FROM PET/CT IMAGES USING LEVEL SETS METHOD

In the framework of detection, diagnostic and treatment planning of tumors, the Positron Emission Tomography (PET) examination coupled with a Computed Tomography (CT) have become recently the most efficient technique. Manual segmentation of regions of interest (ROI) could be time consuming and this is necessary to obtain information about patient pathology. The aim of our study is to propose an automatic solution to this problem. This paper presents an approach of tumor segmentation based on a level set method. One critical step, the algorithm stopping, is automatic. The results obtained by the proposed method dealing with PET-CT images are encouraging.