Case-Specific Reliability Assessment for Improved False Positive Reduction with an Information-Theoretic CAD System

  • Authors:

  • Piotr A. Habas;Jacek M. Zurada;Georgia D. Tourassi

  • Affiliations:

  • Computational Intelligence Lab, Department of Electrical and Computer Engineering, University of Louisville, Louisville, USA KY 40292;Computational Intelligence Lab, Department of Electrical and Computer Engineering, University of Louisville, Louisville, USA KY 40292;Duke Advanced Imaging Laboratories, Department of Radiology, Duke University Medical Center, Durham, USA NC 27710

  • Venue:
  • IWDM '08 Proceedings of the 9th international workshop on Digital Mammography
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

The study presents an application of the case-specific reliability analysis framework for quality evaluation of mammographic CAD detections. First, the original framework is modified so that it can be applied to a featureless case-based CAD system that relies on principles of information theory (IT). Then, the IT-CAD system is applied for false positive reduction of visual cues provided by a pre-screening algorithm. The study is based on mammograms from the Digital Database of Screening Mammography (DDSM) and is focused on mass detection. Experimental results show that assessment of the expected accuracy for each individual CAD cue facilitates 11%-18% further reduction of false positives while operating at high mass detection rates.