Comparison of different strain-based parameters to identify human left ventricular myocardial infarct during diastole: A 3D finite-element study

Abstract

Assessment of myocardial damage is important to obtain an accurate prognosis after myocardial infarction. Myocardial strains have been shown to be good indicators of the myocardial viability. Strain analysis has been used to identify dysfunctional regions, and several strain-based parameters have been proposed to detect regions of infarct. In this study, nine different parameters were used to detect synthetically generated infarct lesions on left ventricles and their performances were compared. The parameters were investigated on ten virtual 3D cases based on healthy human left ventricles extracted from MRI examinations. Realistic infarcts were generated for each virtual case with different locations, shapes, sizes, and stiffer materials. Diastolic virtual strain data were obtained via finite-element simulations using widely implemented constitutive law and rule-based myofiber orientation. The results showed that stretch-dependent invariance in fiber direction is able to better delineate the infarcts in comparison to the other parameters.