Segmentation of organs at risk in Head and Neck CT scans

Do you want to contribute to improving the life of patients with head and neck cancer? Help us to develop a better algorithm for the segmentation of organs at risk. The accurate segmentation allows designing radiotherapy treatment plans that expose organs at risk to low radiation dose, leading to improved quality of life after the treatment. The segmentation is performed on 3D computed tomography (CT) images. We apply machine learning techniques to assign labels to patches based on a repository of manually labeled images. Implementation is mainly done in MATLAB. Our main goal is to refine our existing algorithm and to develop new methods to achieve a better performance. Our existing code base makes it easy to get started.

Prerequisites: Programming experience is required. Experience in MATLAB, image processing and machine learning is of benefit, but not necessary.

Contact: Christian Wachinger, wachinge@mit.edu