Encrypted login | home

Program Information

Improved Primary Modulation Method for Scatter Correction in Cone-Beam CT Using Local Filtration

no image available
L Zhu

L Zhu1*, (1) Georgia Institute of Technology, Atlanta, GA, (2) University of Science and Technology of China, Anhui, China

Presentations

TH-AB-601-6 (Thursday, August 3, 2017) 7:30 AM - 9:30 AM Room: 601


Purpose: Excessive scatter contamination fundamentally limits the image quality of cone-beam CT (CBCT). The author has previously proposed an effective scatter correction method for CBCT using Fourier Transform based primary modulation (FTPM), with a few limitations including the assumption of uniform modulation frequency and magnitude. This work aims to overcome the above drawbacks by developing a new algorithm for the primary modulation method with improved accuracy and reliability.

Methods: A new signal relationship is developed to obtain a first scatter estimate from an x-ray projection with the insertion of a non-uniform primary modulator. The method empirically adjusts the effective modulation magnitude for each projection ray to account for the beam-hardening effects. Estimated scatter signals with high expected errors are discarded in the generation of the final scatter distribution. The author proposes a technique of local filtration to accelerate the signal processing, and the algorithm is referred to as local filtration based primary modulation (LFPM).

Results: The study on the Catphan600 phantom shows that LFPM effectively reduces the CT image error from 222 HU to 15 HU, and the image contrast is enhanced by a factor of 2 on average. On an anthropomorphic head phantom, LFPM reduces the CT image error from 153 HU to 18 HU, and eliminates the streak artifacts observed on the result of FTPM with substantially improved image uniformity. On the Rando phantom, LFPM reduces the CT image error from 278 HU to 4 HU around the object center.

Conclusion: Distinct from the FTPM algorithm, LFPM does not require projection data downsampling or uniform modulation frequency and magnitude. It also discards suspicious scatter estimates and models the beam-hardening effects for improved scatter estimation accuracy. The presented research further exploits the potential of the primary modulation method on scatter correction, and facilitates its clinical adoption in CBCT imaging.

Funding Support, Disclosures, and Conflict of Interest: Supported by National Institute Of Biomedical Imaging And Bioengineering of National Institutes of Health under Award Number R21EB019597, R21EB021545. Partially supported by Ministry of Science and Technology of China Key Research and Development Projects (Grant No. 2016YFC0101400) and by National Natural Science Foundation of China (Grant No. 81671681).


Contact Email: