2-D Strain Assessment in the Mouse Through Spatial Compounding of Myocardial Velocity Data: In Vivo Feasibility

Abstract Ultrasound assessment of myocardial strain can provide valuable information on regional cardiac function. However, Doppler-based methods often used in practice for strain estimation suffer from angle dependency. In this study, a partial solution to that fundamental limitation is presented. We have previously reported using simulated data sets that spatial compounding of axial velocities obtained at three steering angles can theoretically outperform 2-D speckle tracking for 2-D strain estimation in the mouse heart. In this study, the feasibility of the method was analyzed in vivo using spatial compounding of Doppler velocities on six mice with myocardial infarction and five controls, and results were compared with those of tagged microscopic magnetic resonance imaging (μMRI). Circumferential estimates quantified by means of both ultrasound and μMRI could detect regional dysfunction. Between echocardiography and μMRI, a good regression coefficient was obtained for circumferential strain estimates ( r  = 0.69), whereas radial strain estimates correlated only moderately ( r  = 0.37). A second echocardiography was performed after μMRI to test the reproducibility of the compounding method. This yielded a higher correlation coefficient for the circumferential component than for the radial component ( r  = 0.74 circumferentially, r  = 0.49 radially).