Breast cancer is the most common malignancy of women in the USA. It is estimated that 225,000 new breast cancers are diagnosed every year. External beam partial breast irradiation (EB-PBI) is a non-invasive, time- efficient and cost-effective radiation therapy treatment paradigm for stage I and II breast cancer. However, the treatment quality of EB-PBI suffers from inaccurate target location due to the misrepresentation of target volume during treatment planning and lumpectomy cavity and breast deformation during the treatment course. Onboard treatment breast image is currently available by integer a volumetric breast ultrasound scanner (VBUS) into the existing EB-PBI in our institute. The object of this project is to develop a real-time image registration tool to automatically register the planning computed tomography (CT) image to the planning or on- treatment 3-D ultrasound breast image. We plan to test the hypothesis that 1) the implementation of the real- time deformable image registration (DIR) algorithm will enable us to model the breast and lumpectomy cavity deformation as treatment progresses and 2) this new developed computational tool will permit us to improve the precision of delivery dose and sparing of adjunct normal tissue by optimizing the treatment plan in accordant with up-to-date patient anatomy.
Our specific aims and measurable objectives are: 1) to develop a GPU-based ultrafast image registration tool for ultrasound and CT images registration;2) to validate and assess the developed registration tool by sequentially testing on series of digital phantoms, experimental phantoms, and real patients'data sets;3) to demonstrate that development of the real-time image registration tool will enable precise target location and optimal dose distribution. The success of the proposed project will provide a method that utilizes on-board treatment ultrasound images to locate treatment target volume in an efficient and accurate way. Consequently it will enhance the therapeutic quality of breast cancer patient care by improving the precision of treatment delivery while sparing adjacent healthy tissues. Additionally, the general methodology developed in this work has a broad applicability to the radiation therapy of a variety of cancers other than breast cancer.

Public Health Relevance

This project is to innovatively incorporate the 3-D ultrasound breast imaging technique into the online adaptive breast radiotherapy system through developing a real-time image registration computational tool. This online adaptive breast radiotherapy paradigm will substantially improve the therapeutic quality of breast cancer patient care by precisely locating target volume and optimally compensating for breast anatomy changes.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Postdoctoral Individual National Research Service Award (F32)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1-F09-A (20))
Program Officer
Jakowlew, Sonia B
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of California San Diego
Internal Medicine/Medicine
Schools of Medicine
La Jolla
United States
Zip Code
Gu, Xuejun; Dong, Bin; Wang, Jing et al. (2013) A contour-guided deformable image registration algorithm for adaptive radiotherapy. Phys Med Biol 58:1889-901
Gu, Xuejun; Jia, Xun; Jiang, Steve B (2011) GPU-based fast gamma index calculation. Phys Med Biol 56:1431-41
Gu, Xuejun; Jelen, Urszula; Li, Jinsheng et al. (2011) A GPU-based finite-size pencil beam algorithm with 3D-density correction for radiotherapy dose calculation. Phys Med Biol 56:3337-50