Image reconstruction and processing for stationary digital tomosynthesis systems Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 21, 2019
  • Wu, Gongting
    • Affiliation: College of Arts and Sciences, Department of Physics and Astronomy
  • Digital tomosynthesis (DTS) is an emerging x-ray imaging technique for disease and cancer screening. DTS takes a small number of x-ray projections to generate pseudo-3D images, it has a lower radiation and a lower cost compared to the Computed Tomography (CT) and an improved diagnostic accuracy compared to the 2D radiography. Our research group has developed a carbon nanotube (CNT) based x-ray source. This technology enables packing multiple x-ray sources into one single x-ray source array. Based on this technology, our group built several stationary digital tomosynthesis (s-DTS) systems, which have a faster scanning time and no source motion blur. One critical step in both tomosynthesis and CT is image reconstruction, which generates a 3D image from the 2D measurement. For tomosynthesis, the conventional reconstruction method runs fast but fails in image quality. A better iterative method exists, however, it is too time-consuming to be used in clinics. The goal of this work is to develop fast iterative image reconstruction algorithm and other image processing techniques for the stationary digital tomosynthesis system, improving the image quality affected by the hardware limitation. Fast iterative reconstruction algorithm, named adapted fan volume reconstruction (AFVR), was developed for the s-DTS. AFVR is shown to be an order of magnitude faster than the current iterative reconstruction algorithms and produces better images over the classical filtered back projection (FBP) method. AFVR was implemented for the stationary digital breast tomosynthesis system (s-DBT), the stationary digital chest tomosynthesis system (s-DCT) and the stationary intraoral dental tomosynthesis system (s-IOT). Next, scatter correction technique for stationary digital tomosynthesis was investigated. A new algorithm for estimating scatter profile was developed, which has been shown to improve the image quality substantially. Finally, the quantitative imaging was investigated, where the s-DCT system was used to assess the coronary artery calcium score.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Lu, Jianping
  • Lalush, David
  • Zhou, Otto
  • Lee, Yueh
  • Evans, Charles
  • Doctor of Philosophy
Degree granting institution
  • University of North Carolina at Chapel Hill
Graduation year
  • 2017

This work has no parents.