Mammogram Screening by Automated Followup: A Feasibility Study

Abstract

This report describes part of a study aimed at developing a computer-based aid for mammogram screening that makes a detailed comparison between mammograms of the same patient acquired at different screenings and detects changes indicative of cancer. The focus of the work in the past three years has been on putting two mammograms acquired at different time into correspondence. The essence of the approach is identification of control points in two mammograms; these points are used to put regions in two mammograms into correspondence. The emphasis of the work in the past year has been on improving the procedure for determining control points, i.e., points that are the same in two images. For this purpose we have developed a model based approach to identify regions of interest in two mammograms. The model encompasses breast tissue characteristics, modeling of compression effects and formation of X- ray images. The model is utilized to develop appropriate segmentation operators and the report discusses utilization of the model to detect lobules and ducts. The model can also be utilized for generating synthetic mammograms. Presently, we are evaluating the improvements this approach offers, relative to our original approach, in terms of determining more reliably control points, namely branching points of ducts.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 1999
Accession Number
ADA381307

Entities

People

  • Dragana Brzakovic

Organizations

  • Lehigh University

Tags

Communities of Interest

  • Biomedical
  • Ground and Sea Platforms

DTIC Thesaurus Topics

  • Adipose Tissue
  • Compression
  • Computer Vision
  • Connective Tissue
  • Detection
  • Detectors
  • Elastic Properties
  • Films
  • Identification
  • Image Processing
  • Materials
  • Radiation
  • Simulations
  • Three Dimensional
  • Trees (Data Structures)
  • Two Dimensional
  • X Rays

Fields of Study

  • Medicine
  • Physics

Readers

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