Abstract
The use of computer-assisted diagnosis has become widespread in cardiac nuclear medicine. Quantitative programs are commercially available for ventriculography, phase analysis, and thallium 201 perfusion studies. The goal of these programs is to eliminate interobserver variability by objectively analyzing the studies without causing a loss of accuracy. In addition, by using quantitative data not apparent on visual inspection, some programs attempt to increase sensitivity for disease above that possible by the visual reading of images. Programs that analyze perfusion studies to detect coronary artery disease have received the most interest. Results have varied, but sensitivities as high as 95% has been reported. This review discusses the techniques of computer-assisted diagnosis for thallium 201 myocardial-perfusion studies. The circumferential and washout profile methods are discussed in detail.
Original language | English (US) |
---|---|
Pages (from-to) | 209-222 |
Number of pages | 14 |
Journal | Journal of Digital Imaging |
Volume | 5 |
Issue number | 4 |
DOIs | |
State | Published - Nov 1992 |
Keywords
- artificial intelligence
- computer-assisted diagnosis
- computers
- coronary artery disease
- nuclear medicine
ASJC Scopus subject areas
- Radiological and Ultrasound Technology
- Radiology Nuclear Medicine and imaging
- Computer Science Applications