Index terms: magnetic resonance imaging; real-time imaging; left ventricle; regional function; wall motion; wall thickening ABNORMALITIES OF REGIONAL left ventricular wall dynamics at rest or during stress are sensitive markers of myocardial ischemia (1,2). Cardiac magnetic resonance imaging (MRI) provides accurate and reproducible measurements of regional left ventricular wall motion (WM) and wall thickening (WT) for both qualitative and quantitative analysis (3-12). MRI is a noninvasive imaging modality that offers excellent spatial resolution and tissue contrast and, unlike echocardiography and x-ray angiography, provides excellent depiction of both endocardial and epicardial contours. This may lead to higher diagnostic accuracy of stress MRI for detection of inducible WM abnormalities, compared with stress echocardiography (13,14). However, until recently MRI has had the disadvantage of relatively slow image acquisition. With conventional techniques, a 10 -20-second acquisition time is required per slice and up to 15 minutes for a complete data set of the left ventricle (LV) (15). Furthermore, signals in conventional MRI acquisition are time averaged, and each slice has to be acquired over several heartbeats in one breath hold. This is unsuitable in imaging of patients with cardiac arrhythmias and in patients who cannot breath hold. Faster image acquisition is therefore crucial in establishing cardiac MRI as a useful and cost-effective clinical tool. This is particularly important in stress MRI studies in order to shorten the exposure of patients to pharmacological stress agents and to allow early detection of WM abnormalities.With high-performance gradient systems and recently developed ultrafast hybrid MRI acquisition sequences, significant reductions in image acquisition times can be achieved in cardiac . With these sequences, multiphase cine images of the heart can be acquired in real time in a single RR interval, i.e., without time averaging. Measurements of ventricular dimensions and global ventricular function from realtime-acquired MRI data sets have shown good correlation with conventional fast gradient echo acquisition (22)(23)(24). Visual assessment of regional function from real-time MRI has been shown to be feasible (25). However, to date no comparison has been made between real-time and conventional MRI acquisition for qualitative and quantitative assessment of WM and WT. It is uncertain how the lower spatial resolution and signalto-noise ratio (SNR) of real-time-acquired images will affect the accuracy and reproducibility of such measurements.