A statistical shape model is a suitable tool for high accuracy eye modeling required for treatment planning for external beam radiotherapy of intraocular tumors. In this study, the automatically adapted eye model has been compared to manual segmentation based on CT volume scans of 17 patients. The presented solution outperforms state of the art methods in terms of accuracy, reliability and robustness. Increased modeling possibilities will contribute to an improved treatment planning accuracy.
Jérôme Yerly, Lorenzo Di Sopra, Josefina Adriana Maria Bastiaansen, João Pedro Forjaco Jorge, Silvio Ionta, Benedetta Franceschiello, David Zeugin