Functional Analysis
Daniel Ennis1 and Rob J van der Geest2

1UCLA, 2Dept. Radiology, Leiden University Medical Center, Leiden, Netherlands


Assessment of LV dimensions and global systolic function is a key element of every CMR examination. Accurate and reproducible quantification is highly important and requires an appropriate MR acquisition protocol as well as careful use of image analysis methods.

Ventricular size and global function

Assessment of LV size and global systolic function is a key element of every CMR examination. Assessment of global ventricular function requires quantification of the LV cavity volume at different points in the cardiac cycle. For accurate quantification of the sometimes complex shape of the LV, 3D imaging is required in order to obtain accurate quantification results. A typical imaging protocol includes cine imaging in multiple long-axis views and a parallel stack of short-axis slices covering the LV from base to apex. For RV size assessment the use of trans-axial slices has shown to be more accurate and reproducible. The imaging slices are acquired during repeated patient breath holds using steady state free precession (SSFP) at a temporal resolution of approximate 40-50 ms. Automated or manual contour detection is required to segment the LV blood pool in the end-diastolic and end-systolic phase to derived global function parameters such as stroke volume an ejection fraction. Various factors will influence the accuracy and precision of the measurements. Obviously, the accuracy and precision the global function parameters are influenced by the image segmentation. Of especial importance is the precise and consistent definition of the blood pool in the most basal slice. Since the cross-sectional area at the basal level is relatively large, incorrect inclusion or exclusion of the most basal slice will have large influence on the quantification. Cross-referencing the short-axis images with the additionally available long-axis views may help to make the proper judgement. However, inconsistencies in the patient breath-hold position during acquisition of the individual slices may be misleading. The use of novel 3D cine imaging techniques with patient motion correction techniques may be a solution to overcome this limitation. While 3D cine imaging may prolong the total scanning time, it has the advantage that it requires less expertise from the scanner operator in planning the required cardiac views. Retrospectively, every desired view can be generated from the acquired 3D volume and image analysis will not be hampered by mis-registration of the acquired views.

Regional function

Several approaches can be employed to study the regional contractile performance of the LV. Visual assessment of wall motion abnormalities in short-axis and long-axis views is clinically the most common approach. With endocardial and epicardial contours defined, the Centerline method can be applied to quantify regional wall motion or wall thickening. The Centerline method allows to either quantify the regional displacement of the endocardial wall, or to quantify the local thickening of the LV myocardium by quantifying the local change in distance between the endocardial and epicardial contours. For quantitative assessment MR tagging is considered the gold standard approach. MR tagging allows quantitative assessment of intra-myocardial deformation during systole, which can be used to accurately quantify regional LV systolic strain in longitudinal and circumferential direction. Other MR techniques, such as DENSE (Displacement Encoding with Stimulated Echos), SENC (Strain-ENCoded MRI, or phase-contract tissue velocity mapping are alternative imaging method used for assessment of regional contractile function. In recent years, the use of feature tracking (FT) has gained a lot of attention as an alternative method to quantify LV strain from standard cine MR imaging. FT is a technique that has been known from cardiac ultrasound as speckle tracking that is used to track the motion of landmarks (speckles) within the LV myocardium throughout the cardiac cycle, which can subsequently be used to quantify LV strain. This technique, with appropriate modifications, has been shown to also to be applicable to standard cine MRI, which is attractive as it obviates the need for a separate tagging acquisition. Current literature on this technique suggests FT can be used for reproducible assessment of global circumferential and longitudinal strain. Regional strain assessment using FT however, suffers from relatively poor reproducibility.

Diastolic function

Diastolic LV function can be assessed by quantifying the mitral inflow using MR flow velocity mapping. Due to the motion of the mitral valve annulus over the cardiac cycle, a single fixed imaging with through-plane velocity encoding cannot be used to obtain reliable measurements of transvalvular flow. Instead, the use of 4D Flow, with retrospective tracking of the mitral valve position has shown to allow accurate quantification of mitral flow. Using this approach, a 3D volume needs to be acquired covering the region of the mitral valve over the complete cardiac cycle. At each acquired phase the appropriate measurement plane for transvalvular flow can be defined to obtain accurate measurements of mitral inflow and mitral regurgitation.


No acknowledgement found.


No reference found.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)