MRI-Guided Soft Tissue Alignment for Head-and-Neck Radiotherapy and Margin Adaption Assessed by a GPU-Based Dose Deformable Regsitration Framework

X. Sharon Qi, Jack Neylon, Yingli Yang, Lui Yang, Anand Santhanam, Allen Chen, and Daniel A. Low

18th International Conference on the use of Computers in Radiation Therapy

London, UK | 01 Jun 2016

Introduction

Bony landmark based registration using daily cone beam CT image is a widely adopted but compromised alignment procedure used for head-and-neck (H&N) radiotherapy (RT). MR image provides a superior soft tissue contrast, which allows excellent IGRT alignment on soft tissues such as target or critical structures nearby. The purposes of this work are 1) to analyze the inter-fraction patient setup using soft-tissue target and design an appropriate clinical target volume (CTV)-to-planned target volume (PTV) margin for H&N MRI-guided IGRT; and 2) to assess the resulting dosimetric consequences when margin adaption is performed based on weekly MR images via an in-house GPU-based dose deformable registration framework.