Encrypted login | home

Program Information

A Novel Iterative Reconstruction Algorithm for Improving CBCT Image Quality

no image available
W Mao

W Mao*, C Liu , K Snyder , A Kumarasiri , B Zhao , S Gardner , J Kim , N Wen , I Chetty , F Siddiqui , Henry Ford Health System, Detroit, MI


TU-L-GePD-JT-2 (Tuesday, August 1, 2017) 1:15 PM - 1:45 PM Room: Joint Imaging-Therapy ePoster Theater

Purpose: We have assessed the image quality of a novel, iterative reconstruction algorithm to determine the potential to improve image quality, ultimately to enhance the accuracy of CBCT-based localization over the standard reconstruction algorithm.

Methods: The current TrueBeam CBCT reconstruction removes scatter using a kernel-based correction followed by filtered back-projection-based reconstruction (FDK). In the prototype CBCT reconstruction pipeline these steps have been replaced by a finite element solver (AcurosCTS)-based scatter correction and a statistical (iterative) reconstruction. Image quality improvements due to the prototype reconstruction pipeline have been quantitatively analyzed on scans of a standard phantom. Standard full-fan Head, half-fan full-rotation Head, and standard Pelvis CBCT protocols have been quantitatively investigated, including evaluation of noise level, uniformity, constancy, spatial resolution, and modulation transfer function (MTF), using a commercially available software package.

Results: Image quality analysis results show that noise level is reduced to from 28.9 HU, 17.3 HU, and 7.2 HU to 19.1 HU, 7.9 HU, and 2.7 HU, for full-fan Head, half-fan Head, and Pelvis scans, respectively, while MTF measurements indicate that spatial resolution is maintained. HU uniformity improved from 8.6 ± 2.0 to 3.7 ± 2.3 for full-fan Head CBCT and from 7.4 ± 4.0 to 4.3 ± 2.4 for Pelvis CBCT. Contrast to noise ratio (CNR) was analyzed based on a 1% contrast insertion with a diameter of 15 mm. CNR was improved from 0.6 to 0.9, from 1.5 to 5.0, and from 1.7 to 3.8, for the full-fan, half-fan Head, and Pelvis CBCT, respectively. This is mainly due to the significant reduction in image noise.

Conclusion: Noise and other image quality characteristics are significantly improved using the iterative reconstruction algorithm, over the current (FDK-based) method. This suggests the potential for the iterative algorithm to improve image quality, ultimately to enhance the accuracy of image-guided applications using CBCT.

Funding Support, Disclosures, and Conflict of Interest: This project has been supported by a research grant of Varian Medical Systems.

Contact Email: