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-derived nuclear features distinguish malignant from benign breast cytology

W H Wolberg1, W N Street, D M Heisey

  • 1Department of Surgery, University of Wisconsin, Madison, USA.

Human Pathology
|July 1, 1995
PubMed
Summary

Computer analysis of nuclear features accurately distinguishes benign from malignant breast cytology. This automated system achieved high accuracy, aiding in the diagnosis of breast cancer from fine needle aspiration samples.

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

The adrenocortical response of greater sage grouse (Centrocercus urophasianus) to capture, ACTH injection, and confinement, as measured in fecal samples.

Physiological and biochemical zoology : PBZ·2009
Same author

Mycobacteriosis-associated mortality in wild striped bass (Morone saxatilis) from Chesapeake Bay, U.S.A.

Ecological applications : a publication of the Ecological Society of America·2008
Same author

Successive overrelaxation for support vector machines.

IEEE transactions on neural networks·2008
Same author

An adaptive resource-allocating network for automated detection, segmentation, and classification of breast cancer nuclei topic area: image processing and recognition.

IEEE transactions on neural networks·2008
Same author

Nonlinear knowledge in kernel approximation.

IEEE transactions on neural networks·2007
Same author

Analysis of tumor characteristics and survival in liver transplant recipients with incidentally diagnosed hepatocellular carcinoma.

Journal of gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract·2002

Area of Science:

  • Biomedical Engineering
  • Computational Pathology
  • Oncology

Background:

  • Accurate differentiation between benign and malignant breast cytology is crucial for patient management.
  • Traditional methods rely on visual assessment, which can be subjective.
  • Computer-based analysis offers objective quantification of cellular features.

Purpose of the Study:

  • To develop and validate a computer-based system for analyzing nuclear features in breast cytology.
  • To assess the system's ability to differentiate between benign and malignant samples.
  • To evaluate the diagnostic accuracy using machine learning and logistic regression.

Main Methods:

  • Nuclear size, shape, and texture features were extracted from fine needle aspiration (FNA) samples using computer-based analytical techniques.

Related Experiment Videos

  • Nuclei were precisely outlined by a computer algorithm ('snake') after operator input.
  • Classification of samples as benign or malignant was performed using inductive machine learning and logistic regression, with cross-validation for accuracy assessment.

    • Main Results:

    • The computer system calculated 10 distinct nuclear features for each cell.
    • Logistic regression achieved a cross-validated classification accuracy of 96.2%.
    • Inductive machine learning yielded a higher cross-validated accuracy of 97.5% in distinguishing malignant from benign breast cytology.

    Conclusions:

    • Computer-based analysis of nuclear features provides a highly accurate method for classifying breast cytology.
    • The system can generate a probability of malignancy, with a "suspicious" range defined.
    • The automated system demonstrated excellent diagnostic performance in classifying FNA samples, with all recent cases correctly diagnosed.