skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Prior-based artifact correction (PBAC) in computed tomography

Abstract

Purpose: Image quality in computed tomography (CT) often suffers from artifacts which may reduce the diagnostic value of the image. In many cases, these artifacts result from missing or corrupt regions in the projection data, e.g., in the case of metal, truncation, and limited angle artifacts. The authors propose a generalized correction method for different kinds of artifacts resulting from missing or corrupt data by making use of available prior knowledge to perform data completion. Methods: The proposed prior-based artifact correction (PBAC) method requires prior knowledge in form of a planning CT of the same patient or in form of a CT scan of a different patient showing the same body region. In both cases, the prior image is registered to the patient image using a deformable transformation. The registered prior is forward projected and data completion of the patient projections is performed using smooth sinogram inpainting. The obtained projection data are used to reconstruct the corrected image. Results: The authors investigate metal and truncation artifacts in patient data sets acquired with a clinical CT and limited angle artifacts in an anthropomorphic head phantom data set acquired with a gantry-based flat detector CT device. In all cases, the corrected imagesmore » obtained by PBAC are nearly artifact-free. Compared to conventional correction methods, PBAC achieves better artifact suppression while preserving the patient-specific anatomy at the same time. Further, the authors show that prominent anatomical details in the prior image seem to have only minor impact on the correction result. Conclusions: The results show that PBAC has the potential to effectively correct for metal, truncation, and limited angle artifacts if adequate prior data are available. Since the proposed method makes use of a generalized algorithm, PBAC may also be applicable to other artifacts resulting from missing or corrupt data.« less

Authors:
;  [1];  [2]; ;  [3]
  1. Medical Physics in Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg (Germany)
  2. Ziehm Imaging GmbH, Donaustraße 31, 90451 Nürnberg (Germany)
  3. Medical Physics in Radiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany and Institute of Medical Physics, Friedrich–Alexander–University (FAU) of Erlangen–Nürnberg, Henkestraße 91, 91052 Erlangen (Germany)
Publication Date:
OSTI Identifier:
22251110
Resource Type:
Journal Article
Journal Name:
Medical Physics
Additional Journal Information:
Journal Volume: 41; Journal Issue: 2; Other Information: (c) 2014 American Association of Physicists in Medicine; Country of input: International Atomic Energy Agency (IAEA); Journal ID: ISSN 0094-2405
Country of Publication:
United States
Language:
English
Subject:
62 RADIOLOGY AND NUCLEAR MEDICINE; ALGORITHMS; ANATOMY; COMPUTERIZED TOMOGRAPHY; CORRECTIONS; IMAGE PROCESSING; IMAGES; INHIBITION; PATIENTS; PHANTOMS; PLANNING

Citation Formats

Heußer, Thorsten, Brehm, Marcus, Ritschl, Ludwig, Sawall, Stefan, and Kachelrieß, Marc. Prior-based artifact correction (PBAC) in computed tomography. United States: N. p., 2014. Web. doi:10.1118/1.4851536.
Heußer, Thorsten, Brehm, Marcus, Ritschl, Ludwig, Sawall, Stefan, & Kachelrieß, Marc. Prior-based artifact correction (PBAC) in computed tomography. United States. https://doi.org/10.1118/1.4851536
Heußer, Thorsten, Brehm, Marcus, Ritschl, Ludwig, Sawall, Stefan, and Kachelrieß, Marc. 2014. "Prior-based artifact correction (PBAC) in computed tomography". United States. https://doi.org/10.1118/1.4851536.
@article{osti_22251110,
title = {Prior-based artifact correction (PBAC) in computed tomography},
author = {Heußer, Thorsten and Brehm, Marcus and Ritschl, Ludwig and Sawall, Stefan and Kachelrieß, Marc},
abstractNote = {Purpose: Image quality in computed tomography (CT) often suffers from artifacts which may reduce the diagnostic value of the image. In many cases, these artifacts result from missing or corrupt regions in the projection data, e.g., in the case of metal, truncation, and limited angle artifacts. The authors propose a generalized correction method for different kinds of artifacts resulting from missing or corrupt data by making use of available prior knowledge to perform data completion. Methods: The proposed prior-based artifact correction (PBAC) method requires prior knowledge in form of a planning CT of the same patient or in form of a CT scan of a different patient showing the same body region. In both cases, the prior image is registered to the patient image using a deformable transformation. The registered prior is forward projected and data completion of the patient projections is performed using smooth sinogram inpainting. The obtained projection data are used to reconstruct the corrected image. Results: The authors investigate metal and truncation artifacts in patient data sets acquired with a clinical CT and limited angle artifacts in an anthropomorphic head phantom data set acquired with a gantry-based flat detector CT device. In all cases, the corrected images obtained by PBAC are nearly artifact-free. Compared to conventional correction methods, PBAC achieves better artifact suppression while preserving the patient-specific anatomy at the same time. Further, the authors show that prominent anatomical details in the prior image seem to have only minor impact on the correction result. Conclusions: The results show that PBAC has the potential to effectively correct for metal, truncation, and limited angle artifacts if adequate prior data are available. Since the proposed method makes use of a generalized algorithm, PBAC may also be applicable to other artifacts resulting from missing or corrupt data.},
doi = {10.1118/1.4851536},
url = {https://www.osti.gov/biblio/22251110}, journal = {Medical Physics},
issn = {0094-2405},
number = 2,
volume = 41,
place = {United States},
year = {Sat Feb 15 00:00:00 EST 2014},
month = {Sat Feb 15 00:00:00 EST 2014}
}