Optimization of CAD System Using Adaptive Simulated Annealing for Digital Mammography

Abstract

Mammography plays an important role in the detection and diagnosis of breast cancer. Although computer-aided detection (CAD) scheme is essential and acts as second opinion for the detection and diagnosis of breast cancer, its performance for SFM is not suitable for clinical trial due to the lack of full optimization for CAD system. In addition, current CAD system is not evaluated on FFDM images. The purpose of this study is to develop a new kind of fully optimized CAD system for PFDM using a global optimization algorithm to improve its performance on sensitivity and specificity in mass and MCCs detection on mammograms. In the initial grant year, the major accomplishments are as follows: (1) Databases for training and testing of CAD system performance have been constructed and corresponding truth files have been generated for FFDM and SFM respectively. (2) Performance of current CAD system for the detection and diagnosis of breast cancer has been retrospectively evaluated on FFDM and SFM images, respectively. (3) CAD modules have been developed or modified for FFDM.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 2002
Accession Number
ADA408701

Entities

People

  • Wei Qian
  • Xuejun Sun

Organizations

  • University of South Florida

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Annealing
  • Biomedical Research
  • Breast Cancer
  • Clinical Trials
  • Computer Vision
  • Computer-Aided Design
  • Department Of Defense
  • Detection
  • Directional
  • Feature Extraction
  • Mammography
  • Neoplasms
  • Optimization
  • Signal Processing
  • Wavelet Transforms

Readers

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