Li-Yueh Hsu1, Mikhail Gorbachev1,2,
Lin-Ching Chang2, Sujethra Vasu1, Christine Mancini1,
W. Patricia Bandettini1, Andrew E. Arai1
1National Heart Lung &
Blood Institute, National Institutes of Health, Bethesda, MD, United States; 2Department
of Electrical Engineering & Computer Science, the Catholic University of
America, Washington, D.C., United States
Myocardial blood flow can be estimated from the time-signal intensity curves of the left-ventricular (LV) blood cavity and the myocardium of the contrast-enhanced first-pass CMR perfusion images. The aim of this study is to develop a fully automated computer method to extract the arterial input function (AIF) from the LV cavity of the CMR perfusion images. The method was applied to 176 perfusion image series and compared with a reference standard AIF as defined by manual tracing. The results show a close match between computer and manual extracted AIF.