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

Malignant and benign clustered microcalcifications: automated feature analysis and classification

Y Jiang1, R M Nishikawa, D E Wolverton

  • 1Department of Radiology, University of Chicago, Illinois 60637, USA.

Radiology
|March 1, 1996
PubMed
Summary

Computer analysis of mammograms accurately differentiates malignant from benign clustered microcalcifications, outperforming radiologists and potentially reducing unnecessary biopsies.

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

Computerized analysis of multiple-mammographic views: potential usefulness of special view mammograms in computer-aided diagnosis.

IEEE transactions on medical imaging·2002
Same author

Ideal observer approximation using Bayesian classification neural networks.

IEEE transactions on medical imaging·2001
Same author

Automatic segmentation of breast lesions on ultrasound.

Medical physics·2001
Same author

Components-of-variance models for random-effects ROC analysis: the case of unequal variance structures across modalities.

Academic radiology·2001
Same author

Continuous versus categorical data for ROC analysis: some quantitative considerations.

Academic radiology·2001
Same author

Computerized classification of benign and malignant masses on digitized mammograms: a study of robustness.

Academic radiology·2000

Area of Science:

  • Medical imaging analysis
  • Artificial intelligence in radiology
  • Breast cancer diagnostics

Background:

  • Clustered microcalcifications on mammograms often require biopsy to determine malignancy.
  • Distinguishing benign from malignant microcalcifications is crucial for patient management.
  • Current diagnostic methods can lead to false-positive findings and unnecessary invasive procedures.

Purpose of the Study:

  • To develop and evaluate a computer-based method for differentiating malignant from benign clustered microcalcifications.
  • To assess the accuracy of automated image feature extraction and analysis in breast cancer detection.
  • To provide a tool that assists radiologists in classifying suspicious microcalcifications.

Main Methods:

  • Analysis of 100 mammograms from 53 patients with suspicious clustered microcalcifications.

Related Experiment Videos

  • Extraction of eight computer-defined features from clustered microcalcifications.
  • Integration of extracted features using an artificial neural network for classification.

    • Main Results:

    • Computer analysis achieved 100% identification of malignant cases and 82% for benign cases.
    • The computer's accuracy was statistically significantly superior to that of five radiologists (P = .03).
    • The method demonstrated high sensitivity and specificity in classifying microcalcifications.

    Conclusions:

    • Quantitative image features can be effectively extracted and analyzed by computers to differentiate malignant from benign clustered microcalcifications.
    • This automated technique shows promise in improving diagnostic accuracy for breast cancer.
    • The developed method may help reduce the rate of false-positive biopsy findings in clinical practice.