Application of Data Mining Methods to Radiation Therapy Planning Automation

Intensity-modulated radiation therapy (IMRT) is an advanced technique for cancer treatment through high-precision radiation delivery. It generally requires a complex planning procedure, where the crucial step is manual delineation of the organs at risk in three-dimensional (3-D) computed tomography (CT) images, performed using simple drawing tools. It is a labour-intensive and error-prone process which, depending on the clinical case, may often require several hours of intensive work. This project is aimed at developing novel technological solutions to automate radiation therapy planning by analyzing large populations of patients who underwent the procedure and automatically transferring the relevant  information from the existing to the new clinical cases. In particular, automation of anatomy contouring in the head and neck area will be addressed. Carried out in partnership with Philips Healthcare, the leading provider of radiation treatment planning software systems, the project will facilitate the development of new generations of advanced radiation planning solutions and increase the standards of patient care.

Alex Levinshtein
Faculty Supervisor: 
Dr. Sven Dickinson