Skip to Main Content
Outcome optimization of hyperthermia tumor treatment in the head and neck requires accurate hyperthermia treatment planning. Hyperthermia treatment planning is based on tissue segmentation for 3D patient model generation. We present here an automatic atlas-based segmentation algorithm for the organs at risk from CT images of the head and neck. To overcome the large anatomical variability, atlas registration and intensity-based classification were combined. A cost function composed of an intensity energy term, a spatial prior energy term based on the atlas registration and a regularization term is globally minimized using graph cut. The method was evaluated by measuring Dice similarity coefficient, mean and Hausdorff surface distances with respect to manual delineation. Overall a high correspondence was found with Dice similarity coefficient higher than 0.86 and a mean distance lower than the voxel resolution.