Adaptive Radiation Therapy (ART) can be used as a feedback control strategy to include patient specific variations in planning and delivery of radiation with the ultimate goal of improving the therapeutic ratio. Deformable image registration (DIR) has the potential to include organ deformation in the adaptive dose delivery. Our goal in this dissertation is to develop methods and tools based on scientific insight gained to account organ deformation which can be thought of as a bio-feedback between physics, anatomy & physiology of organs.In this dissertation we first demonstrate the need for ART in the treatment of localized prostate cancer and propose plan adaptation without deformable image registration tools. This dissertation aims to develop tools to independently perform DIR using open source platform (3D Slicer) and propose a frame work to validate the accuracy of DIR with an emphasis on radiation oncology application. A novel method of verifying DIR accuracy using virtually simulated deformation to mimic clinically observed organ deformation was proposed and validated. Independent validation tools for verifying accuracy of image registration are implemented using open source modules.The second step in implementing ART involves calculating the doses to a deforming anatomy. Dose warping defined as applying the deformation vector field (DVF) arising from DIR on the original dose distribution has the potential to account for organ deformation and accumulate doses. However its application and validity in a clinical environment remains controversial. This dissertation proposes to examine the fundamental science of deformation linked to its causative physical force and use that relationship to verify the validity of dose warping from DIR algorithms. A novel deformable bladder phantom made of viscoelastic polymer with mechanical and tensile properties comparable to human bladder was designed for this purpose. A threshold limit for dose warping was ascertained and established from various commercially available DIR algorithms .The applicability of dose warping to dose painting scenarios was investigated. This dissertation also highlights the need to validate the volume of dose from dose warping as traditionally used 3D gamma pass rates used by most research studies may not accurately describe dose warping for dose painting.
University of Minnesota Ph.D. dissertation. December 2013. Major: Biophysical Sciences and Medical Physics. Advisor: Susanta Hui PhD. 1 computer file (PDF); xxiii, 215 pages, appendix page 194-215.
Towards adaptive radiation therapy--a framework for inclusion of organ deformation in adaptive dose delivery.
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.