Texture Snake for Tumor Grading
Autumn 2016
Master Semester Project
Project: 00315
Tumors are characterized by a spatial heterogeneity. The cells of each corresponding region do not behave the same way. According to their organization they have different speed of growing and spreading. In the recent years, a lot of research has been done to analyze the tissue decomposition of a tumor in order to improve therapeutic efficiency. In image processing, active contour methods (also called snakes) and texture analysis are powerful tools for detection and segmentation. The goal of this project is to develop a snake that is able to accurately delineate each region of a tumor and to extract the corresponding texture. More specifically, it consists in the design of an energy based on texture analysis.
REQUIREMENTS: Java, Image Processing.
- Supervisors
- Anaïs Badoual, anais.badoual@epfl.ch, 31136, BM 4142
- Michael Unser, michael.unser@epfl.ch, 021 693 51 75, BM 4.136
- Adrien Depeursinge