26th Annual Student Research Conference
Algebraic Reconstruction Techniques in Computed Tomography
Rachel E. McCarroll
Dr. Eduardo Velasco, Faculty Mentor
X-ray Computed Tomography (CT), developed in the 1970's, acts as a cornerstone in medical imaging. Often used as a first look into injury and disease, accurate reconstruction of images is crucial. The Algebraic Reconstruction Technique (ART) is an algorithmic method used for reconstruction which, while no longer used in the clinical setting, has distinct advantages over clinically preferred methods. In ART, a single ray iterative technique serves as a standard. Additional methods include a dual-ray technique and a constrained method which prepares images with the assumption that the absorption coefficient for all materials is greater than or equal to zero. The ability of each method (single-ray, dual-ray, and non-negative) to produce an accurate reconstruction is addressed. Additionally, in the presence of varying levels of random noise, each method is evaluated on the basis of time needed for reconstruction, image quality, and number of iterations needed for accurate reconstruction.
Keywords: Physics, Imaging, Medicine, Reconstruction, Linear Algebra, Computed Tomography, Mathematics
Topic(s):Physics
Presentation Type: Oral Paper
Session: 206-1
Location: MG 1096
Time: 9:30