M.Sc Thesis


M.Sc StudentChernyak Yonatan
SubjectOptimal Anisotropic Irradiation for X-Ray
Computed Tomography
DepartmentDepartment of Electrical and Computer Engineering
Supervisor PROF. Yoav Schechner
Full Thesis textFull thesis text - English Version


Abstract

X-ray Computed Tomography (CT) has revolutionized the medical field. It allows physicians to explore the inside of the patient in a non-invasive manner. Since it was introduced in 1972, the number of scans has reached over 80 million in the united states alone during 2018.

As a result, CT is now the primary source of man-made hazardous ionizing radiation.

Thus, one of the main challenges in CT is balancing the radiation dose absorbed by the patient with the reconstruction quality.

We propose both an estimator for reconstruction quality estimation and an adaptive anisotropic irradiation acquisition method (AIR). AIR is a source optimization method that utilizes the estimator to produce higher quality reconstruction while maintaining the same dose. In addition, we prove that the weighted least squares (WLS) reconstruction is optimal for MSE criterion, with respect to measurement stochasticity. Moreover, we propose an alternative dose criterion that represents better the risk posed on the patient.

AIR method is flexible and requires minor changes to the CT scanners currently in use. The estimator is based on the Cramér-Rao Lower Bound and the WLS. The algorithm is comprised of 5 steps:

(a) Acquisition of an exploratory scan using a low dose traditional isotropic irradiance scan.

(b) A region of interest is defined by the operator.

(c) Optimization of the angular distribution

of the anisotropic irradiation, according to the ROI and the preliminary scan.

(d) Acquisition of the main scan using the optimized irradiance. (e) Reconstruction via WLS.