Computer-Aided Diagnosis of Breast Cancer: A Multi-Center Demonstrator.

Abstract

The long range goal of this project is to improve the accuracy and consistency of breast cancer diagnosis by developing a computer aided diagnosis (CAD) system for early prediction of breast cancer using the BI-RADS(trademark)' findings reporting criteria provided by mammographers distributed over a wide geographical area. In the first year of this project, we have hired a Data Technician to set up and manage the mammographic findings database. So far, 700 hundred cases from Duke have been entered, as well as 1000 from the University of Pennsylvania. A further 500 cases from Sloan-Kettering Cancer Center are being processed and entered, as well as an additional 100 from Duke. The data collection process has been delayed by the decreased budget, which in turn delayed CAD system testing. An artificial neural network (ANN) to predict biopsy outcome has been developed. A genetic algorithm has been developed for selecting subsets from the dataset in order to decrease cross-validation variance and increase the network's performance in the ROC area.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 1997
Accession Number
ADA353889

Entities

People

  • Carey E. Floyd Jr.

Organizations

  • Duke University Hospital

Tags

DTIC Thesaurus Topics

  • Accuracy
  • Acquisition
  • Algorithms
  • Biomedical Research
  • Breast Cancer
  • Computer-Aided Diagnosis
  • Computers
  • Consistency
  • Data Acquisition
  • Data Sets
  • Databases
  • Genetic Algorithms
  • Materials
  • Neoplasms
  • Neural Networks
  • Universities
  • Validation

Readers

  • Clinical Trial Research.
  • Computer Vision.
  • Oncology and Biomarker-Based Cancer Detection.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks
  • Biotechnology