Medical physics research projects
Department of Radiation Oncology, University of Washington

Medical physics research at the UW includes:

• Algorithms for automatic generation of clinical target volumes in head and neck cancer
  (Ira Kalet, Mary Austin-Seymour)
  The success of radiation therapy depends critically on accurately delineating the target volume, which is the region of known or suspected disease in a patient. Methods that can compute a contour set defining a target volume on a set of patient images will contribute greatly to the success of radiation therapy and dramatically reduce the workload of radiation oncologists, who currently draw the target by hand on the images using simple computer drawing tools. The most challenging part of this process is to estimate where there is microscopic spread of disease. We are developing methods for automatically selecting and adapting standardized regions of tumor spread based on the location of lymph nodes in a standard or reference case, together with image registration techniques. The best available image registration techniques (deformable transformations computed using ``mutual information'' optimization) appear promising but will need to be supplemented by anatomic knowledge-based methods to achieve a clinically acceptable match. This project also involves collaboration from Mark Whipple, Otolaryngology/Head and Neck Surgery, Linda Shapiro, Computer Science and Engineering/Electrical Engineering, and Chia-Chi Teng, Electrical Engineering graduate student.


• Intraoperative Dose Optimization For Prostate Brachytherapy
  (Paul Cho)
  
  While brachythearpy has proven to be an effective treatment modality for early-stage prostate cancer, local failure and recurrence do occur. Based on the post-implant analysis correlating the PSA level and the principal dosimetric parameters, it is evident that the probability of cure increases with improved dose distribution. The primary objective of the proposed research is to develop an intraoperative method to measure and modify dose distribution for optimal outcome. Specific aims include: (1) automated detection and localization of seeds from multiple fluoroscopy projections, (2) semi-automated segmentation of prostate volume from ultrasound, (3) automated registration of seeds and prostate volume, (4) development of dose modification supervisor, and (5) clinical evaluation of the target system. The project is funded by NIH/NCI and DoD.


• Advanced Inverse Planning Algorithm For IMRT
  (Paul Cho)
  It has been shown that the inverse problem in IMRT is severly ill-conditioned. The mathematical limitation inherent in inverse planning algorithms has not yet been quantified and properly regulated. The present research exploits the power of singular value decomposition to characterize and regulate the dose matrices for optimal convergence to a feasible solution. Tikhonov method combined with convex projection is being investigated.


• Image Guided Therapy
  (Mark Phillips, Paul Cho, Juergen Meyer)
  Advances in imaging physiological processes, e.g. hypoxia, are an important development in targetting tissues for radiation therapy as well as assessing response to treatment. A collaboration with the Nuclear Medicine/PET group at UWMC is working to develop and apply deformable image registration for two separate clinical studies. The first is to use PET-FDG to reduce the size of target volumes in head and neck cancer, and thus reduce morbidity. The other is to use PET-FMISO to image hypoxia in head and neck tumors and to use the information to design IMRT treatments and to assess the response of the hypoxic regions to radiation therapy. This work is being performed in conjunction with the Nuclear Medicine Department (Paul Kinahan, Joseph Rajendran) and VA-Puget Sound (Eric Ford, David Schwartz).


• Improvements in Radiation Therapy Plan Optimization
  (Mark Phillips, Juergen Meyer, Ira Kalet)
  Treatment planning optimization has recently received much attention due to the advent of inverse planning techniques for IMRT. As useful as these algorithms are, they all have difficulty in handling the predominant situations in radiation therapy. These problems include decisions based on models formulated with incomplete data, incomplete and qualitative presriptions, and mutually contradictory constraints/objectives. Our project is aimed at using belief nets (also known as Bayes' nets) to provide better methods at guiding the optimization process and choosing the most clinically appropriate solution.


• Improved Seeds for Permanent Seed Implants for Prostate Cancer
  (Mark Phillips)
  Classic radiation biology has always categorized tumor response as having a high alpha/beta ratio, similar to that of acutely responding tumors. This has resulted in treatment strategies that make use of prolonged fractionation schedules in order to achieve the most separation between the tumor response and dose-limiting late responding tumors. Recent clinical results have indicated that for prostate tumors the alpha/beta ratio is probably less than 3, similar to late responding tissues. In addition, recently published data indicate that repair is much slower than previous thought. In a project done in collaboration with IsoRay, Inc., a company designing and developing novel isotope-seed combinations, I am investigating the potential advantages that would result from a shorter half-life isotope for permanent seed implants in light of the profound changes in the radiobiological modelling of prostate cancer.