Synthesis of pediatric brain tumor images with mass effect
Abstract
In children, brain tumors are the leading cause of cancer-related death. The amount of labeled data in children is much lower than that for adult subjects. We would like to develop a new method to synthesize high-quality 3D pathological pediatric MRI brain images from pathological adult ones. To realistically simulate the appearance of brain tumors, the mass effect (i.e., the deformation induced by the tumor to the surrounding tissue) is considered. This project is an extension of our previous work "Synthesis of pediatric brain tumor images with mass effect". We will test the method on a real pediatric tumoral brain dataset. All the datasets used in this project are all publicly available. Since the available data is relatively scarce, we are developing an unsupervised tumor segmentation method that uses the information from adults to help the segmentation of tumors in children.
