Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Computer-assisted reader software versus expert reviewers for polyp detection on CT colonography.

Stuart A Taylor1, Steve Halligan, David Burling

  • 1Department of Intestinal Imaging, St. Mark's and Northwick Park Hospitals, Watford Rd., Harrow HA1 3UJ, United Kingdom.

AJR. American Journal of Roentgenology
|February 25, 2006
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

UK National Screening Committee position statement on evidence required for multicancer detection tests.

BMJ (Clinical research ed.)·2026
Same author

UK National Screening Committee position statement on surrogate outcomes in cancer screening trials.

BMJ (Clinical research ed.)·2026
Same author

How I Do It: MRI Approach to Perianal Fistulas.

Radiology·2026
Same author

Why Are We Still Implanting Permanent Synthetic Mesh for Hernia Repair Into Young Patients?

Journal of abdominal wall surgery : JAWS·2026
Same author

Impact of using artificial intelligence as a second reader in breast screening including arbitration.

Nature cancer·2026
Same author

Diagnostic accuracy, fairness and clinical implementation of AI for breast cancer screening: results of multicenter retrospective and prospective technical feasibility studies.

Nature cancer·2026

Computer-assisted reader (CAR) software demonstrated high sensitivity in detecting colorectal polyps, outperforming expert reviewers. This advanced software offers a clinically acceptable false-positive rate, potentially exceeding expert performance alone.

Area of Science:

  • Medical Imaging
  • Gastroenterology
  • Artificial Intelligence in Medicine

Background:

  • Colorectal cancer screening relies on accurate polyp detection during CT colonography.
  • Expert reviewers face challenges in consistently identifying all polyps, impacting screening efficacy.

Purpose of the Study:

  • To evaluate the sensitivity of computer-assisted reader (CAR) software for polyp detection.
  • To compare CAR software performance against expert radiologists in CT colonography.

Main Methods:

  • A dataset of colonoscopically validated CT colonography cases was used for training and testing.
  • ColonCAR version 1.2 software was optimized and applied to test cases.
  • Expert reviewers interpreted test cases independently, unaware of the ground truth.

Related Experiment Videos

Main Results:

  • ColonCAR version 1.2 detected 81% of polyps, exceeding the average expert sensitivity of 70%.
  • The software achieved 92% sensitivity for polyps 10 mm or larger.
  • The median false-positive rate was 13 per case, with 91% easily dismissed.

Conclusions:

  • ColonCAR version 1.2 shows high sensitivity for polyp detection in CT colonography.
  • The software demonstrates a clinically acceptable false-positive rate.
  • CAR software can augment expert performance and potentially offer superior standalone diagnostic capabilities.