The overall goal of this research project is to improve the accuracy and reliability of cone-beam computed tomography (CBCT) guidance of radiation treatment in lung and liver. The treatment of inoperable tumors in lung and liver remains a therapeutic challenge. Increasing the radiation dose improves local control but requires high-precision techniques. There is widespread interest in linear accelerator-mounted CBCT systems for guiding radiation treatment, but respiratory motion adversely affects CBCT image quality and limits its accuracy. We propose to develop and evaluate methods to reduce respiratory-induced motion artifacts in CBCT. Our hypotheses are: 1) motion-reduction techniques will improve tumor and organ-at-risk (OAR) visibility and localization accuracy in lung and liver, relative to current CBCT methods;2) motion-reduced CBCT will enable the use of contrast enhancement for localizing liver metastases at treatment, as an alternative to implanted fiducials;and 3) the improved localization accuracy will permit smaller treatment volumes, thereby enabling safe delivery of higher radiation doses.
Our specific aims are to 1) develop and validate a model of respiration-induced motion adapted to each patient from respiration-correlated CT and CBCT images;2) develop a method to reduce motion artifacts in CBCT using the patient-specific motion model;and 3) evaluate motion- reduced CBCT in patients treated with radiation for nonsmall cell lung cancer or liver metastases. The patient-specific motion model uses nonrigid image registration between CT or CBCT images at different respiratory motion states to generate 3D trajectories of all voxels in the volume images. An important component of the model is a principal component analysis that reduces the high-dimensional complex data set to a lower dimension by removing noise and redundancy. The motion reduction procedure sorts the CBCT projections into motion states according to the respiratory signal, reconstructs and morphs the volume image at each motion state to a reference image by means of the patient's motion model, then combines the morphed images to produce a high quality CBCT image. Evaluation of the accuracy of the proposed methods will use physical and analytical deformable phantoms, and manual delineations of clinical images. The first phase of patient studies will compare the ability to visualize and localize tumor and OAR before and after model-based motion reduction improvements to CBCT image quality, using gated CBCT as a reference. The imaging study in liver will test the feasibility of detecting metastatic lesions using motion-reduced CBCT in combination with intravenous contrast injection. The second phase of studies will use motion-reduced CBCT to correct patient position prior to hypofractionated radiotherapy in lung, and determine the fraction of patients for whom higher prescribed dose is achieved through margin reduction made possible by the improved localization accuracy.

Public Health Relevance

There is widespread interest in medical accelerator-mounted cone-beam computed tomography (CBCT) systems for improving the accuracy of radiation treatment;however, breathing motion can blur the images and limit our ability to locate and target tumors in lung and liver. We propose to develop methods to reduce blur and distortion in CBCT images caused by breathing motion. The effectiveness of these methods will be evaluated in patients receiving radiation treatment for lung cancer or liver metastases.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Research Project (R01)
Project #
Application #
Study Section
Radiation Therapeutics and Biology Study Section (RTB)
Program Officer
Deye, James
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Sloan-Kettering Institute for Cancer Research
New York
United States
Zip Code
Dzyubak, Oleksandr; Kincaid, Russell; Hertanto, Agung et al. (2014) Evaluation of tumor localization in respiration motion-corrected cone-beam CT: prospective study in lung. Med Phys 41:101918
Zhao, Qingyu; Chou, Chen-Rui; Mageras, Gig et al. (2014) Local metric learning in 2D/3D deformable registration with application in the abdomen. IEEE Trans Med Imaging 33:1592-600
Regmi, Rajesh; Lovelock, D Michael; Hunt, Margie et al. (2014) Automatic tracking of arbitrarily shaped implanted markers in kilovoltage projection images: a feasibility study. Med Phys 41:071906
McNamara, Joseph E; Regmi, Rajesh; Michael Lovelock, D et al. (2013) Toward correcting drift in target position during radiotherapy via computer-controlled couch adjustments on a programmable Linac. Med Phys 40:051719
Chou, Chen-Rui; Frederick, Brandon; Mageras, Gig et al. (2013) 2D/3D Image Registration using Regression Learning. Comput Vis Image Underst 117:1095-1106
Kincaid Jr, Russell E; Yorke, Ellen D; Goodman, Karyn A et al. (2013) Investigation of gated cone-beam CT to reduce respiratory motion blurring. Med Phys 40:041717
Liu, F; Hu, Y; Zhang, Q et al. (2012) Evaluation of deformable image registration and a motion model in CT images with limited features. Phys Med Biol 57:2539-54
Santoro, Joseph; Kriminski, Sergey; Lovelock, D Michael et al. (2010) Evaluation of respiration-correlated digital tomosynthesis in lung. Med Phys 37:1237-45
Zhang, Qinghui; Hu, Yu-Chi; Liu, Fenghong et al. (2010) Correction of motion artifacts in cone-beam CT using a patient-specific respiratory motion model. Med Phys 37:2901-9
Liu, Xiaoxiao; Saboo, Rohit R; Pizer, Stephen M et al. (2009) A SHAPE-NAVIGATED IMAGE DEFORMATION MODEL FOR 4D LUNG RESPIRATORY MOTION ESTIMATION. Proc IEEE Int Symp Biomed Imaging 2009:875-878