Optimizing Kilovoltage Cone-Beam Computed Tomography for Image-Guided Radiation Therapy

dc.contributor.advisorMedin, Paulen
dc.contributor.committeeMemberWang, Jingen
dc.contributor.committeeMemberJiang, Steve B.en
dc.contributor.committeeMemberPompos, Arnolden
dc.contributor.committeeMemberSun, Xiankaien
dc.creatorOuyang, Luoen
dc.date.accessioned2016-06-27T20:12:42Z
dc.date.available2016-06-27T20:12:42Z
dc.date.created2014-05
dc.date.issued2014-04-14
dc.date.submittedMay 2014
dc.date.updated2016-06-27T19:48:47Z
dc.description.abstractKilovoltage cone-beam computed tomography (kV CBCT) has important applications in image guided radiation therapy (IGRT). In its current state, CBCT is suboptimal for these applications due to its considerable radiation dose, poor soft-tissue contrast and inaccurate Hounsfield Units as a result of in-patient scatter. The specific aims of this dissertation research are to: 1) reduce noise associated with the low-dose CBCT scans acquired with low tube current/time (mAs) protocols by the application of innovative software; and 2) develop a moving blocker system for kV scatter correction in CBCT scans acquired both with and without concurrent megavoltage radiation therapy. The noise reduction of low-dose CBCT was achieved through statistical image reconstruction or image restoration in the CBCT projections. Performance of three statistical iterative image reconstruction algorithms was compared through the noise-resolution-tradeoff analysis. Results of the comparison study suggested that different algorithms were favored in different imaging tasks and conditions. A CBCT projection smoothing algorithm that incorporates prior CBCT imaging data was proposed in this dissertation as an alternate way to improve image quality of the low-dose CBCT. For this approach, the Karhunen-Loève (KL) transform was used to account for the correlation among consecutively acquired CBCT projections. Results from phantom-based studies showed that the KL domain penalized weighted least-square (PWLS) restoration were superior to PWLS smoothing in the projection domain in overall image quality improvement of low-dose CBCT. A moving blocker system was implemented on a LINAC on-board imaging system. Experimental evaluation of the system demonstrated that the moving-blocker based scatter correction strategy in kV CBCT is reliable and practical. Finally, the moving-blocker scatter correction strategy was extended for simultaneous MV-kV scatter correction of kV CBCT acquired during rotational radiation therapy. In both kV-only scatter correction and MV-kV scatter correction, high quality CBCT images were reconstructed following the moving-blocker scatter correction. CBCT image sets acquired with these methods were found to contain accurate CT numbers and minimal shading artifacts and could therefore be used for dose calculation purposes in adaptive radiation therapy.en
dc.format.mimetypeapplication/pdfen
dc.identifier.oclc952355716
dc.identifier.urihttps://hdl.handle.net/2152.5/3321
dc.language.isoen_USen
dc.subjectCone-Beam Computed Tomographyen
dc.subjectRadiotherapy, Image-Guideden
dc.subjectScattering, Radiationen
dc.titleOptimizing Kilovoltage Cone-Beam Computed Tomography for Image-Guided Radiation Therapyen
dc.typeThesisen
dc.type.materialtexten
thesis.degree.departmentGraduate School of Biomedical Sciencesen
thesis.degree.disciplineRadiological Sciencesen
thesis.degree.grantorUT Southwestern Medical Centeren
thesis.degree.levelDoctoralen
thesis.degree.nameDoctor of Philosophyen

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
OUYANG-DISSERTATION-2014.pdf
Size:
23.02 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
LICENSE.txt
Size:
1.84 KB
Format:
Plain Text
Description: