Computer-Aided Interval Change Analysis of Microcalcifications on Mammograms for Breast Cancer Detection

Abstract

The goal of this project is to develop a computer-aided diagnosis (CAD) system for interval change analysis of lesions on mammograms. An important component of the CAD system is the multistage regional registration technique for identifying corresponding microcalcification clusters on temporal pairs of mammograms. In the first stage, an initial search region was estimated on the prior mammogram based on the cluster location on the current mammogram. In the second stage the search region was refined. In the third stage the cluster was detected within the search region. In the first stage we used the regional registration method (RRM), which outperformed the warping techniques. 175 temporal pairs of mammograms were used for evaluation. The average distance between% the estimated and the true cluster centroids on the previous mammogram after the initial stage was 7.95+-4.73mm. In the second stage, automated detection of microcalcification cluster within the search region is performed. Using our current cluster detection program with standard thresholds, 76.6% (134/175) TP with 0.45 false positives (FP) were detected within the search region. Using a high-sensitivity threshold, 89.1%(156/175) TP with 0.43 FP were detected. In the third stage the correspondence classifier was used and it reduced the FP rate to an average of 0.19 FP cluster with sensitivity of 81% (141/175).

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 2004
Accession Number
ADA433041

Entities

People

  • Lubomir Hadjiiski

Organizations

  • University of Michigan

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence Software
  • Assembly
  • Breast Cancer
  • Computer Vision
  • Computer-Aided Diagnosis
  • Data Sets
  • Databases
  • Feature Extraction
  • Feature Selection
  • Information Science
  • Machine Learning
  • Michigan
  • Neural Networks
  • North America
  • Pattern Recognition
  • Training

Fields of Study

  • Medicine
  • Physics

Readers

  • Computer Vision.
  • Oncology and Biomarker-Based Cancer Detection.
  • Regression Analysis.