PERFEX
Description: 

PERFEX (Perfusion Expert) is an expert system for the interpretation of 3D tomograms of myocardial perfusion distribution.

The overall goal is to assist in the diagnosis of coronary artery disease. This system infers extent and severity of coronary artery disease (CAD) from perfusion distribution, and provides as output a patient report summarizing the conditions of the three main arteries and other pertinent information.

PERFEX has to

  1. characterize the image information symbolically in terms of location, size, and shape of stress perfusion defects,
  2. relate this symbolic information to evidence that confirms (or disconfirms) disease in the three major vessels: the left anterior descending (LAD), left circumflex (LCX) and right coronaryarteries (RCA),
  3. recognize shapes that, although suggestive of multiple-vessel disease because the defects “spill” over several territories, they are actually associated with only one vessel,
  4. recognize the presence of artifacts,
  5. detect and localize regions of reversibility,
  6. relate this temporally derived information regarding reversibility to specific arterial vessels (LAD, LCX and RCA),
  7. recognize spill over shapes associated with perfusion redistribution,
  8. recognize the presence of reversibility artifacts and
  9. help track of how multiple defects effect each of the three major coronary arteries of interest. Then the system suggests an overall diagnostic evaluation of the patient and provides an automatically generated report.
Evaluation: The system is undergoing extensive evaluation- especially multicenter testing, and the implementation in different nuclear imaging systems is planned. Preliminary test results indicate that the diagnostic interpretation made by PERFEX are highly consistent with those made by human experts.

Availability: 

The system is designed to be integrated in an actual clinical setting.

A demo of the system is available at: http://www.gvu.gatech.edu/biovis/perfex

Contact: 

 Norberto F.Ezquerra, norberto.ezquerra@cc.gatech.edu

Graphics, Visualization & Usability (GVU) Center at the Georgia Institute of Technology 801 Atlantic Drive , Atlanta, GA, USA http://www.gvu.gatech.edu
Back System Home