Improved estimation and visualization of two-dimensional myocardial strain rate using MR velocity mapping.

PURPOSE: To estimate regional myocardial strain rate, with reduced sensitivity to noise and velocities outside the region of interest, and provide a visualization of the spatial variation of the obtained tensor field within the myocardium.
MATERIALS AND METHODS: Myocardial velocities were measured using two-dimensional phase contrast velocity mapping. Velocity gradients were estimated using normalized convolution and the calculated 2D strain rate tensor field was visualized using a glyph representation. Validation utilized a numerical phantom with known strain rate distribution. Strain rate glyph visualizations were created for normal myocardium in both systole and diastole and compared to a patient with an anteroseptal infarction.
RESULTS: In the phantom study the strain rate calculated with normalized convolution showed a very good agreement with the analytic solution, while traditional methods for gradient estimation were shown to be sensitive to both noise and surrounding velocity data. Normal myocardium showed a homogenous strain rate distribution, while a heterogeneous strain rate can be clearly seen in the patient data.
CONCLUSION: The proposed approach for quantification and visualization of the regional myocardial strain rate can provide an objective measure of regional myocardial contraction and relaxation that may be valuable for the assessment of myocardial heart disease. Copyright (c) 2008 Wiley-Liss, Inc.