NIH R01 (PA 11-260): DASSIM-RT and Compressed Sensing-Based Inverse Planning Project Summary The goal of this project is to establish a novel paradigm of dense angularly sampled and sparse intensity- modulated radiation therapy (DASSIM-RT). In this scheme, the redundant or dispensable modulation of the incident intensity-modulated beams is removed effectively by using a compressed sensing (CS) technique. The delivery time saved in this way is used to increase the angular sampling for improved dose conformality. By balancing the angular sampling and intensity modulation, DASSIM-RT enables us to fully utilize the technical capabilities of modern digital linacs to produce highly conformal dose distributions that can be delivered efficiently. Specifically, we will 1) set up a compressed sensing (CS)-based framework for inverse treatment planning;2) investigate a new type of treatment scheme termed DASSIM-RT;and 3) show the advantage of DASSIM-RT through a series of phantom cases and previously treated patients. DASSIM-RT represents a truly optimal RT scheme with uncompromised angular sampling (including non-coplanar beams), beam intensity modulation, and possible field-specific energy and collimator angle. If successful, the project will allow us to overcome many of the limitations of existing treatment schemes to meet the unmet clinical demand for highly conformal dose distributions in radiation oncology.
This project is directed at establishing a dense angularly sampled and sparse intensity modulated radiation therapy (DASSIM-RT) scheme to advance RT treatment techniques to a new paradigm. The project will develop enabling concepts and technologies including: compressed sensing-based inverse planning, DASSIM-RT, single arc delivery scheme of DASSIM-RT, and prior knowledge guided search of optimal beam configuration. The proposed research promises to overcome many of the limitations of existing treatment schemes and empower the radiation oncology discipline with substantially improved tools for cancer management.
|Ibragimov, Bulat; Toesca, Diego; Chang, Daniel et al. (2018) Development of deep neural network for individualized hepatobiliary toxicity prediction after liver SBRT. Med Phys 45:4763-4774|
|Mo?nik, Domen; Ibragimov, Bulat; Xing, Lei et al. (2018) Segmentation of parotid glands from registered CT and MR images. Phys Med 52:33-41|
|Liu, Hongcheng; Xing, Lei (2018) Isodose feature-preserving voxelization (IFPV) for radiation therapy treatment planning. Med Phys 45:3321-3329|
|Qin, Wenjian; Wu, Jia; Han, Fei et al. (2018) Superpixel-based and boundary-sensitive convolutional neural network for automated liver segmentation. Phys Med Biol 63:095017|
|Ren, Shangjie; Hara, Wendy; Wang, Lei et al. (2017) Robust Estimation of Electron Density From Anatomic Magnetic Resonance Imaging of the Brain Using a Unifying Multi-Atlas Approach. Int J Radiat Oncol Biol Phys 97:849-857|
|Wang, Huan; Dong, Peng; Liu, Hongcheng et al. (2017) Development of an autonomous treatment planning strategy for radiation therapy with effective use of population-based prior data. Med Phys 44:389-396|
|Ibragimov, Bulat; Xing, Lei (2017) Segmentation of organs-at-risks in head and neck CT images using convolutional neural networks. Med Phys 44:547-557|
|Ibragimov, Bulat; Toesca, Diego; Chang, Daniel et al. (2017) Combining deep learning with anatomical analysis for segmentation of the portal vein for liver SBRT planning. Phys Med Biol 62:8943-8958|
|Ar?k, Sercan Ö; Ibragimov, Bulat; Xing, Lei (2017) Fully automated quantitative cephalometry using convolutional neural networks. J Med Imaging (Bellingham) 4:014501|
|Han, Bin; Ding, Aiping; Lu, Minghui et al. (2017) Pixel response-based EPID dosimetry for patient specific QA. J Appl Clin Med Phys 18:9-17|
Showing the most recent 10 out of 28 publications