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

Principal component analysis for content-based image retrieval.

Usha Sinha1, Hooshang Kangarloo

  • 1Department of Radiological Sciences, University of California, Los Angeles, School of Medicine, 924 Westwood Blvd, Suite 420, Los Angeles, CA 90024, USA. usinha@itmedicine.net

Radiographics : a Review Publication of the Radiological Society of North America, Inc
|September 18, 2002
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

Investigating the correlation between force output, strains, and pressure for active skeletal muscle contractions.

Journal of the mechanical behavior of biomedical materials·2026
Same author

Evaluation of Magnetization Transfer Contrast Sequences: Application to Monitor Age-Related Differences in Muscle Macromolecular Fraction.

Tomography (Ann Arbor, Mich.)·2025
Same author

Strain mapping using compressed sensing accelerated 4D flow MRI-Potential for detecting coactivation in thigh muscles.

Frontiers in physiology·2025
Same author

Magnetic Resonance Imaging Biomarkers of Muscle.

Tomography (Ann Arbor, Mich.)·2024
Same author

Investigating the Correlation between Force Output, Strains, and Pressure for Active Skeletal Muscle Contractions.

ArXiv·2023
Same author

Effect of different ankle joint positions on medial gastrocnemius muscle fiber strains during isometric plantarflexion.

Scientific reports·2023
Same journal

MRI of Lesions Growing Along the Pituitary Stalk.

Radiographics : a review publication of the Radiological Society of North America, Inc·2026
Same journal

Invited Commentary: Early Detection of Pancreatic Cancer: Are We Up for the Challenge?

Radiographics : a review publication of the Radiological Society of North America, Inc·2026
Same journal

Radiology Board Examinations: A Fundamental Shift.

Radiographics : a review publication of the Radiological Society of North America, Inc·2026
Same journal

Early Pancreatic Cancer: Clinical Implications, Workup, and Imaging Findings with Histopathologic Correlation for Personalized Surveillance.

Radiographics : a review publication of the Radiological Society of North America, Inc·2026
Same journal

Comprehensive Approach to Prostate Cancer Metastasis Mimics at Prostate-Specific Membrane Antigen PET/CT.

Radiographics : a review publication of the Radiological Society of North America, Inc·2026
Same journal

Invited Commentary: Postdeployment Monitoring of AI in Radiology: Beyond the Test Set.

Radiographics : a review publication of the Radiological Society of North America, Inc·2026
See all related articles

Principal component analysis (PCA) enables content-based retrieval of medical images by reducing data dimensionality. This method achieved 83% accuracy in finding matching brain MRI sections, improving image search capabilities.

Area of Science:

  • Medical Imaging
  • Computer-Aided Diagnosis
  • Machine Learning in Radiology

Background:

  • Current picture archiving and communication systems primarily use patient demographics and study descriptions for image retrieval.
  • Content-based image retrieval (CBIR) offers a more advanced approach by analyzing image content directly.

Purpose of the Study:

  • To investigate the application of principal component analysis (PCA) for content-based retrieval of medical images.
  • To develop and evaluate an automated algorithm for filtering relevant images within an imaging study.

Main Methods:

  • Principal component analysis (PCA) was employed to reduce the dimensionality of medical images into a basis set of prototype images.
  • Each image was represented by its projection vector onto the basis set.

Related Experiment Videos

  • Retrieval accuracy was assessed by comparing query image projection vectors with those in the database, using a 3 mm tolerance for matches.

    • Main Results:

    • The PCA-based algorithm achieved an 83% retrieval accuracy for axial brain magnetic resonance imaging (MRI) scans.
    • Preprocessing for intensity and geometric uniformity significantly improved retrieval performance.
    • The algorithm effectively identified matching image sections within a specified proximity.

    Conclusions:

    • Principal component analysis (PCA) is a viable method for content-based retrieval of medical images.
    • The developed algorithm can serve as an automated module for selecting relevant images in medical studies.
    • This approach enhances the efficiency and accuracy of medical image search and management.