Efficient high cone-angle artifact reduction in circular cone-beam CT using deep learning with geometry-aware dimension reduction

Physics in Medicine and Biology, vol. 66, pp. 135015

Abstract

High cone-angle artifacts (HCAAs) appear frequently in circular cone-beam computed tomography (CBCT) images and can heavily affect diagnosis and treatment planning. To reduce HCAAs in CBCT scans, we propose a novel deep learning approach that reduces the three-dimensional (3D) nature of HCAAs to two-dimensional (2D) problems in an efficient way. Specifically, we exploit the relationship between HCAAs and the rotational scanning geometry by training a convolutional neural network (CNN) using image slices that were radially sampled from CBCT scans. We evaluated this novel approach using a dataset of input CBCT scans affected by HCAAs and high-quality artifact-free target CBCT scans. Two different CNN architectures were employed, namely U-Net and a mixed-scale dense CNN (MS-D Net). The artifact reduction performance of the proposed approach was compared to that of a Cartesian slice-based artifact reduction deep learning approach in which a CNN was trained to remove the HCAAs from Cartesian slices. In addition, all processed CBCT scans were segmented to investigate the impact of HCAAs reduction on the quality of CBCT image segmentation. We demonstrate that the proposed deep learning approach with geometry-aware dimension reduction greatly reduces HCAAs in CBCT scans and outperforms the Cartesian slice-based deep learning approach. Moreover, the proposed artifact reduction approach markedly improves the accuracy of the subsequent segmentation task compared to the Cartesian slice-based workflow.

Authors 9

1. Jordi Minnema corresponding

   Academic Center for Dentistry Amsterdam · Amsterdam University Medical Centers · Amsterdam Movement Sciences · Vrije Universiteit Amsterdam

   Affiliation as printed

   Vrije Universiteit Amsterdam

   Amsterdam UMC and Academic Centre for Dentistry Amsterdam (ACTA), Vrije Universiteit Amsterdam, Department of Oral and Maxillofacial Surgery/Pathology, 3D Innovationlab, Amsterdam Movement Sciences, 1081 HV Amsterdam, The Netherlands

2. Maureen A.J.M. van Eijnatten

   Centrum Wiskunde & Informatica · Eindhoven University of Technology

   Affiliation as printed

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam,

   Eindhoven University of Technology

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam, The Netherlands

   Medical Image Analysis Group, Department of Biomedical Engineering, Eindhoven University of Technology, P.O. Box 513, 5600 MB Eindhoven, The Netherlands

3. Henri Der Sarkissian

   Centrum Wiskunde & Informatica

   Affiliation as printed

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam,

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam, The Netherlands

4. Shannon Doyle

   Centrum Wiskunde & Informatica

   Affiliation as printed

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam,

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam, The Netherlands

5. Juha Koivisto

   University of Helsinki

   Affiliation as printed

   University of Helsinki

   Department of Physics, University of Helsinki, Gustaf Hällsströmin katu 2, FI-00560, Helsinki, Finland

6. Jan Wolff

   Aarhus University · University Medical Center Hamburg-Eppendorf · Fraunhofer Research Institution for Additive Manufacturing Technologies IAPT

   Affiliation as printed

   Aarhus University

   Fraunhofer Research Institution for Additive Manufacturing Technologies IAPT, Am Schleusengraben 13, D-21029 Hamburg,

   University Hospital

   Department of Dentistry and Oral Health, Aarhus University, Vennelyst Boulevard 9, DK-8000 Aarhus C, Denmark

   Department of Oral and Maxillofacial Surgery, Division for Regenerative Orofacial Medicine, University Hospital Hamburg-Eppendorf, 20246 Hamburg, Germany

   Fraunhofer Research Institution for Additive Manufacturing Technologies IAPT, Am Schleusengraben 13,21029 Hamburg, Germany

7. Tymour Forouzanfar

   Academic Center for Dentistry Amsterdam · Amsterdam University Medical Centers · Amsterdam Movement Sciences · Vrije Universiteit Amsterdam

   Affiliation as printed

   Vrije Universiteit Amsterdam

   Amsterdam UMC and Academic Centre for Dentistry Amsterdam (ACTA), Vrije Universiteit Amsterdam, Department of Oral and Maxillofacial Surgery/Pathology, 3D Innovationlab, Amsterdam Movement Sciences, 1081 HV Amsterdam, The Netherlands

8. Felix F. Lucka

   Centrum Wiskunde & Informatica · University College London

   Affiliation as printed

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam,

   University College London

   Centre for Medical Image Computing, University College London, WC1E 6BT London, United Kingdom

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam, The Netherlands

9. Kees Joost Batenburg Aachen Leiden Institute of Advanced Computer Science

   Leiden University · Centrum Wiskunde & Informatica

   Affiliation as printed

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam,

   Leiden University

   Centrum Wiskunde & Informatica (CWI), 1090 GB Amsterdam, The Netherlands

   Leiden Institute of Advanced Computer Science (LIACS), Leiden University, 2333 CA Leiden, The Netherlands