False-Negative Interpretation in a CAD Environment

Abstract

The purpose of this project is to examine the impact of CAD schemes on the diagnostic performance of radiologists, in particular, the change of false-negative interpretations under a CAD cueing environment. Based on the proposed schedule of this project, we have completed the selection of 120 subtle cases that are used in the observer performance study. All the images have been processed, and the sen- sitivity of the CAD schemes has been adjusted to generate different cueing levels in each image. We have designed and implemented an automatic image display sys- tem. This computer-controlled system has the capability of randomizing case selec- tion for each reading session and to record the diagnostic results. After final- izing the study protocol and performing pre-study training for participating radio- logists, the main reading experiment is now underway. The radiologists began reading cases in May. Once the readings are completed, data analyses will be per- formed using ROC-type methodology.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 1999
Accession Number
ADA383984

Entities

People

  • Bin Zheng

Organizations

  • University of Pittsburgh

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Bayesian Networks
  • Computational Science
  • Computer Programming
  • Computer Programs
  • Computer Vision
  • Computers
  • Feature Extraction
  • Image Segmentation
  • Information Science
  • Machine Learning
  • Medical Personnel
  • Neural Networks
  • Pattern Recognition
  • Physicians

Readers

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