The unavoidable dose delivered to nearby radiosensitive structures limits the maximum radiation dose that can be delivered safely to the tumor target during external beam radiation therapy for the treatment of cancer. To reduce the risk of unacceptable morbidity each individual course of radiation therapy must be carefully preplanned to achieve a high tumor dose while sparing normal tissues as much as possible. This guiding principle of radiation therapy dose planning has been refined to the modern concept of dose conformation--shaping the dose distribution to match the target m that pervades most modern research in radiation therapy. Innovative technologies, methods, and ideas for three-dimensional (3-D) and even four-dimensional (3-D + time) image-guided treatment simulation and delivery have pushed this concept at the research level to the extreme limit of essentially shrink-wrapping the high dose region to a conformal glove-like fit around the target volume while enveloping sensitive tissues in protective lower dose regions, even in the context of spatially complex and time-varying geometrical shapes and relationships of the tumor and surrounding normal anatomy. Research in this area is gaining momentum from clinical trials that are producing medical evidence showing highly conformal radiation therapy, both with and without dose escalation, significantly increases the probability of favorable clinical outcome both in terms of improving tumor control and reducing the occurrence of side effects. Graphically sophisticated and analytically complex software tools are essential resources for conducting image guided computer simulations (virtual simulation) and optimizations of treatment delivery, including highly accurate dose calculations. Our overall aim is to fill a national need for a fully documented and supported, open and extensible treatment planning and virtual simulation research platform for investigating new techniques and methods for image-guided radiotherapy treatment simulation and delivery.
|Liao, Shu; Shen, Dinggang (2012) A feature-based learning framework for accurate prostate localization in CT images. IEEE Trans Image Process 21:3546-59|
|Li, Wei; Liao, Shu; Feng, Qianjin et al. (2012) Learning image context for segmentation of the prostate in CT-guided radiotherapy. Phys Med Biol 57:1283-308|
|Shi, Yonghong; Liao, Shu; Shen, Dinggang (2011) Learning Statistical Correlation of Prostate Deformations for Fast Registration. Mach Learn Med Imaging 7009:1-9|
|Shi, Yonghong; Liao, Shu; Shen, Dinggang (2011) Learning statistical correlation for fast prostate registration in image-guided radiotherapy. Med Phys 38:5980-91|
|Lee, Huai-Ping; Lin, Ming C; Foskey, Mark (2008) Physically-based validation of deformable medical image registration. Med Image Comput Comput Assist Interv 11:830-8|