Evaluation of gamma dose comparison algorithms


Prof. Gabor Fichtinger, John Schreiner, Csaba Pinter

info: www.cs.queensu.ca/~gabor, email: gabor@cs.queensu.ca


Dose distribution image is a basic data object in radiation therapy datasets. It quantifies the administered dosage within a certain spatial volume, and is a useful indicator of the quality of a radiation therapy plan. It is many times necessary to qualitatively compare these dose distributions to evaluate the difference between two plans, or as an intermediate step in adaptive radiotherapy.

The Kingston General Hospital radiation oncology team has developed several 3D dose comparison (gamma) algorithms that need to be evaluated and compared. The SlicerRT radiation therapy toolkit (developed in PerkLab) also contains a gamma algorithm that should be involved in this analysis.


Isolate the KGH Matlab-based gamma algorithms, make them easy to run on the same datasets. Create Matlab-bridge modules for the different gamma methods in 3D Slicer. Find good sample datasets and run the competing algorithms on them. Analyze the output and the performance of each algorithm and create a report.


Context: Radiation therapy related algorithm evaluation.

Analytical: 3D image processing algorithm, processing, statistics.

Experimental: Algorithm evaluation, using the 3D Slicer platform, working with a partner team.


Practical aptitude and mindset, good Matlab skills, basic statistics.

Preferred prerequisites: Any combination of Computer Integrated Surgery, Computer Graphics, Medical Informatics or Medical Imaging.

Project type: 

Undergraduate project