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

Support vector machines committee classification method for computer-aided polyp detection in CT colonography.

Anna K Jerebko1, James D Malley, Marek Franaszek

  • 1Diagnostic Radiology Department, National Institutes of Health, Bethesda, MD 20892, USA. anna.jerebko@siemens.com

Academic Radiology
|April 16, 2005
PubMed
Summary
This summary is machine-generated.

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

Utility of Pancreatic Perivascular Adipose Tissue as a CT Imaging Biomarker for Diagnosing Type 2 Diabetes.

Diabetes·2026
Same author

CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification.

ArXiv·2026
Same author

AI-driven Abdominal Aortic Calcification Extracted From Contrast-enhanced CT Is Predictive of All-cause Mortality and Cardiovascular Events in a Large Adult Population.

Journal of computer assisted tomography·2026
Same author

Automated Delineation of Couinaud Segments at CT for Future Liver Remnant Volumetry.

Radiology. Artificial intelligence·2026
Same author

CT-based Opportunistic Screening for Adding Clinical Value: How I Do It.

Radiology·2026
Same author

Longitudinal evaluation of intra-patient changes in computed tomography-based body composition measures.

Abdominal radiology (New York)·2026

A new computer-aided detection scheme using support vector machines (SVMs) improves colonic polyp detection accuracy. This method enhances sensitivity and significantly reduces false positives in computed tomographic colonography, aiding early diagnosis.

Area of Science:

  • Medical Imaging
  • Machine Learning
  • Gastroenterology

Background:

  • Computed tomographic colonography (CTC) is a key tool for colorectal cancer screening.
  • Accurate detection of colonic polyps in CTC images is crucial for effective screening.
  • Computer-aided detection (CAD) systems aim to improve polyp detection rates and reduce radiologist workload.

Purpose of the Study:

  • To propose and evaluate a novel classification scheme for CAD of colonic polyps in CTC.
  • To enhance the performance of CAD systems by improving sensitivity and reducing false-positive detections.
  • To assess the generalizability and effectiveness of the proposed SVM ensemble method.

Main Methods:

  • Development of a classification scheme using an ensemble of support vector machines (SVMs).

Related Experiment Videos

  • Implementation of bootstrap aggregation (bagging) for training and model selection.
  • Utilized smoothed leave-one-out (SLOO) cross-validation for robust error estimation.
  • Employed independent datasets for model selection and generalization testing.

    • Main Results:

    • The SVM ensemble achieved 75% sensitivity with 1.5 false positives per study on an independent test set.
    • Retraining the ensemble on a higher-resolution test set yielded 81% sensitivity with 2.6 false positives per study.
    • The SVM ensemble demonstrated a 7%-10% increase in sensitivity compared to single SVMs.
    • A 1.5-fold reduction in false-positive detections per study was observed with the ensemble method.

    Conclusions:

    • The proposed SVM ensemble classification method exhibits good generalizability for CAD of colonic polyps.
    • The approach effectively achieves high sensitivity and a low false-positive rate in CTC polyp detection.
    • The model selection and error estimation techniques are valuable for advancing CAD systems in this field.