Institutional Research Information Service
UCL Logo
Please report any queries concerning the funding data grouped in the sections named "Externally Awarded" or "Internally Disbursed" (shown on the profile page) to your Research Finance Administrator. Your can find your Research Finance Administrator at http://www.ucl.ac.uk/finance/research/post_award/post_award_contacts.php by entering your department
Please report any queries concerning the student data shown on the profile page to:

Email: portico-services@ucl.ac.uk

Help Desk: http://www.ucl.ac.uk/ras/portico/helpdesk
Publication Detail
WE‐E‐213CD‐03: Inverse‐Consistent Symmetric Registration of Inner Colon Surfaces Derived from Prone and Supine CT Colonography
  • Publication Type:
    Conference presentation
  • Authors:
    Roth H, Mcclelland J, Modat M, Hampshire T, Boone D, hu M, Ourselin S, Halligan S, Hawkes D
  • Date:
  • Status:
Purpose: Robust registration of prone and supine colonie surfaces acquired during CT colonography may lead to faster and more accurate detection of colorectal cancer and polyps. Any directional bias when registering one surface to the other could precipitate incorrect anatomical correspondence and engender reader error. Despite this, non‐rigid registration methods are often implemented asymmetrically, which could negatively influence the registration. We aimed to reduce directional bias and so increase robustness by adapting a cylindrical registration algorithm to be both symmetric and inverse‐consistent. Methods: The registration task can be simplified by mapping both prone and supine colonie surfaces onto regular cylinders. Spatial correspondence can then be established in cylindrical space using the original surfaces' local shape indices. We implemented a symmetric formulation of the popular non‐rigid B‐spline image registration method in cylindrical space. A symmetric similarity measure computes the sum of squared differences between both cylindrical representations of prone‐to‐supine and supine‐to‐prone directions simultaneously. Inverse consistency of the transformation is enforced by adding an appropriately weighted penalty term to the optimisation function. Results: We selected 8 CT colonography patient cases with marked variation in luminal distension and surface morphology. We randomly allocated 4 of these for tuning an optimal set of registration parameters and 4 for validation. The mean inverse‐consistency error was reduced by 32% from 4.8mm to 3.2mm by the new symmetric formulation. The mean registration error improved from 8.2mm to 7.3mm for 330 manually chosen reference points on the 4 validation sets. Conclusions: A symmetric formulation of prone and supine surface registration improves the quality of registration. Information from both prone‐to‐supine and supine‐to‐prone directions helps enforce convergence towards a more accurate solution due to reduced directional bias. A more robust and accurate registration will facilitate interpretation of CT colonography and has the potential to improve existing computer‐aided detection methods. The authors gratefully acknowledge financial support for this work from the NIHR program: “Imaging diagnosis of colorectal cancer: Interventions for efficient and acceptable diagnosis in symptomatic and screening populationsâ€. © 2012, American Association of Physicists in Medicine. All rights reserved.
Publication data is maintained in RPS. Visit https://rps.ucl.ac.uk
 More search options
UCL Researchers Show More
Metabolism & Experi Therapeutics
Dept of Med Phys & Biomedical Eng
Dept of Med Phys & Biomedical Eng
Dept of Med Phys & Biomedical Eng
University College London - Gower Street - London - WC1E 6BT Tel:+44 (0)20 7679 2000

© UCL 1999–2011

Search by