Robust Detection of Masses in Digitized Mammograms

Abstract

This project is to develop a robust computer aided diagnosis (CAD) system for mass detection with high sensitivity and specificity in digitized mammograms. The research scope in past year is to evaluate the detection performance and robustness of CAD system. Several major progresses have been made including (1). In addition to the training database, two independent testing databases were generated for evaluation. (2). Two testings and comparisons were made between the algorithms before and after the modifications using the methods developed in this project research in past two years: one on performance testing, another on robustness testing. A set of FROC curves was generated. It is demonstrated that the CAD system developed in this project consistently outperformed the CAD method we developed before both in detection performance and generalizability.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2003
Accession Number
ADA417801

Entities

People

  • Lihua Li

Organizations

  • University of South Florida

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Algorithms
  • Biomedical Research
  • Breast Cancer
  • Cancer
  • Classification
  • Computer-Aided Diagnosis
  • Databases
  • Department Of Defense
  • Detection
  • Diagnostic Imaging
  • Electronic Mail
  • Gray Scale
  • Neoplasms
  • Performance Tests
  • Supervised Machine Learning
  • Test And Evaluation
  • Training

Readers

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