Skip to Main Content
Echocardiographic assessment of myocardial perfusion is currently achieved by measuring indices of contrast replenishment following destructive high-energy ultrasound impulses @ask-echo), We developed a technique for automated detection of perfusion defects based on quanritative analysis of parametric pelfusion images created from these indices. Parametric images were obtained a[ rest and during dipyridamole stress in 18 pts with suspected coronary artery disease. Each image was divided into 6 segments, and mean parameter value (MPV) was calculated for each segment. Changes in MPV from rest to stress were used to automatically detect stress-induced pe&sion defects. ROC analysis was used to optimize (1) the threshold for MPV stress-to-rest ratio, and (2) the minimal number of abnormal segments, required for the diagnosis of ischemia, using coronary stenosis > 70% as rhe "gold standard". The sensitivity, specificity and accuracy of the automated detection of ischemia were 63, 75 and 69% in the LAD, and 67,100 and 75% in the non-LAD territories. In conclusion, automated quantitative analysis of echocardiographic porumetric perjfusion images is feasible and m y be useful for objective detection of myocardial ischemia.