Study On Cardiac Magnetic Resonance Real-time Cine Imaging

Magnetic resonance imaging has become an important biomedical imaging tool due to its characteristics of non-radiation, excellent soft tissue contrast and imaging at any slice orientations. Cardiac cine MR imaging has been regarded as the gold standard for clinical assessment of cardiac function. Usually adopted segmented acquisition cardiac cine imaging needs ECG triggering and breath-hold to reduce the influence of cardiac and respiratory motion, which however does not apply to pediatric and cardiac patients with severe heart and respiratory diseases. With the development of magnetic resonance imaging technology and hardware systems, real-time cine cardiac magnetic resonance imaging emerges as an important way for cardiac function assessment in clinical utilities. Data acquisition of real-time cine cardiac imaging techniques can be completed in a very short time (<60ms) by appropriate sacrifice of spatial resolution and incorporation of fast imaging techniques, which eliminates the need of breath-hold and ECG gating.For cardiac magnetic resonance real-time cine imaging, the spatial position of multiple slices can be mismatched arising from respiratory motion, leading to error for subsequent measurement of cardiac function. At present, the conventional solution is to visually determine cardiac end-diastolic (ED) and end-systolic (ES) images at end expiratory time period frame by frame, or eliminate the effects of respiratory motion by image registration methods. However, these methods are very laborious and time consuming, or running the risk of high specific absorption rate risk for the high-field MRI systems, which are not desirable in clinical practices.In this study, we proposed an efficient imaging data processing method for free-breathing real-time cine imaging. Without going through all the cine images, ED and ES images at end-expiratory phase could be identified with monitoring the displacement of the left ventricle centroid. The proposed the method was validated on 8 healthy volunteers. A relatively high successful rate was obtained with the results of visual inspection as reference. The results confirmed the feasibility and efficacy of the method, and suggested it to be an efficient and promising way for analysis of real-time cines. The proposed method will benefits the wide adoption of real-time cine imaging in clinical studies with great research significance and application value.