Stereoscopic Digital Mammography: Improving Cancer Diagnosis

Abstract

The subject of study in this project was the potential benefit of using stereoscopic digital mammography as an adjunct to standard film mammography to improve the detection and diagnosis of breast cancer. We sought to determine whether stereo mammography might bring significant gains - either in earlier detection of worrisome lesions, or in more accurate discrimination of lesions as benign or malignant. Several main activities were required to accomplish this evaluation. We further developed and refined a system for capturing and viewing stereo mammograms, using state-of-the-art digital methods. Over the duration of the project, stereo mammograms and film studies were acquired on a large number of path-proven cases. Three experienced mammographers served as consultants, and an extensive effort was devoted to developing with them a comprehensive list of visual features to consider in reading the stereo mammograms. In a final reading study with 137 cases, we measured diagnostic accuracy obtained with standard film mammograms alone compared to that obtained with film combined with the stereo mammogram. We found that the stereo mammogram significantly improved diagnostic accuracy. A second, exciting finding was that new, additional lesions, mostly masses, were detected in the stereo mammogram but were not visible in the film study, in 30% of the cases.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 2000
Accession Number
ADA385848

Entities

People

  • David J. Getty

Organizations

  • BBN Technologies

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Abstracts
  • Accuracy
  • Breast Cancer
  • Cancer
  • Detection
  • Detectors
  • Diagnostic Imaging
  • Digital Images
  • Display Systems
  • Health Services
  • Image Processing
  • Imaging Techniques
  • Medical Personnel
  • Neoplasms
  • Standards
  • Three Dimensional
  • Two Dimensional

Fields of Study

  • Medicine
  • Physics

Readers

  • Oncology and Biomarker-Based Cancer Detection.
  • Systems Analysis and Design
  • Vision Science/Vision Psychology/Cognitive Neuroscience.