Computer-Aided Diagnosis of Digital Mammograms

Abstract

The long-term goal of our research is to develop computer-aided diagnosis (CAD) techniques to improve the detection and diagnosis of breast cancer. We have developed a computer technique that can classify breast calcifications in mammograms accurately, and this technique as a diagnostic aid has been shown to be able to improve radiologists' diagnostic accuracy. We have determined that Breast Imaging Report and Data System (BI- RADS) lesion descriptions provided by radiologists can be used as supplemental data to computer-extracted image features to improve the performance of computer classification of malignant and benign breast lesions. We have also found that our computer classification technique developed on screen-film mammograms, can achieve equally high performance on full-filed digital mammograms. This high performance is little affected by variability in the way in which radiologists indicate the general location of calcifications to the computer, which is designed as a means for the radiologist to query the computer aid. These results suggest that the computer technique has the potential to become a clinically useful and viable tool for diagnostic mammography.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2004
Accession Number
ADA431258

Entities

People

  • Yulei Jiang

Organizations

  • University of Chicago

Tags

Communities of Interest

  • Energy and Power Technologies
  • Human Systems

DTIC Thesaurus Topics

  • Accuracy
  • Computer Vision
  • Computer-Aided Diagnosis
  • Data Analysis
  • Data Mining
  • Data Science
  • Databases
  • Electronic Mail
  • Gaussian Distributions
  • Information Processing
  • Information Science
  • Knowledge Management
  • Medical Personnel
  • Normal Distribution
  • North America
  • Pattern Recognition
  • Random Variables

Fields of Study

  • Medicine
  • Physics

Readers

  • Computational Modeling and Simulation
  • Oncology and Biomarker-Based Cancer Detection.