Registration and Segmentation of Magnetic Resonance Images of Patients with Brain Tumor

• Grant number: 23/16229-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: February 01, 2024
• End date: November 30, 2024
• Field of knowledge: Health Sciences - Medicine - Medical Radiology
• Principal Investigator: Renata Ferranti Leoni
• Grantee: Sara Nunes Mader Garcia
• Host Institution: Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto (FFCLRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil

Abstract

Due to the high overall survival rate of patients with treated brain tumors, the relationship of the incidence and severity of radiation-induced brain injuries to the dose received by healthy brain tissue is becoming a frequently asked question. Radiation-induced morphological changes in white and gray matter can be seen on routine magnetic resonance imaging (MRI). As radiotherapy (RT) techniques are becoming more sophisticated, preserving critical regions can be accomplished without compromising tumor control. Therefore, it is essential to identify brain regions susceptible to radiotherapy-induced injuries, such as atrophy and microstructural changes. To this end, it is necessary to segment the structural MRIs into regions of interest and register them to the dose maps originating from the RT. Several studies reported in the literature focused on tumor segmentation; however, few focused on the segmentation of cortical and subcortical regions. This segmentation is based on templates and atlases of brains without lesions, which may not have good results in the segmentation of images with the presence of a tumor. Additionally, registration between images can add artifacts and distortions. Therefore, the present project aims to test different approaches for MRI registration and segmentation of patients with brain tumors and, thus, obtain the methodology with the best performance for these cases.