Quantification of Local Mesostructural Deformation of the Myocardium using DENSE MRI and a Novel Set of Strain Attributes

Abstract

Measuring myocardial deformation is central to understanding the mechanical behavior of both the healthy and the diseased heart. While myocardial deformation is quantified via medical imaging in the form of strain, numerous definitions have been used in the engineering and clinical literature, resulting in inconsistencies in interpreting mechanistic changes in myocardial function. Additionally, current strain definitions lack the robustness required to relate quantitative changes in strain to changes in the myocardial mesostructure. Hence, the goal of this work was to propose a novel set of logarithmic, mesostructure-based strain attributes, by leveraging the upper triangular (QR) decomposition theorem. These strain attributes were then compared with the polar Green strain, the most widely used strain definition for the myocardium. Ten healthy subjects were scanned via displacement encoding with stimulated echoes (DENSE). From these images, a 3-dimensional (3D) displacement field was obtained, from which both 3D Green strains and strain attributes were calculated and compared. The results of this work revealed that the two strain definitions yield significantly different deformation profiles of the myocardium (p < 0.05), with strain attributes predicting relatively more spatial heterogeneity compared to the Green strains -- a finding consistent with similar studies that quantified mesostructural myocardial deformation. Additionally, the strain attributes revealed the myocardium to be compressible, challenging the common assumption of incompressibility. By capturing the deformation along the mesostructure, the strain attributes enable intuitive characterizations of cardiac disease, providing clinicians with deeper mechanistic insights into left ventricular function and serving as predictive markers of adverse remodeling.