Anatomical Image Registration Using Volume Conservation to Assess Cardiac Deformation From 3D Ultrasound Recordings

Abstract

Myocardial deformation imaging can provide valuable insights in myocardial mechanics and help in the diagnosis, prognosis and follow-up of cardiac diseases. However, extracting these indices in 3D is challenging due to the limitations in spatial and temporal resolution of the current volumetric ultrasound systems. For this purpose, we developed an anatomical free-form deformation image registration framework which is locally adapted to the anatomy of the heart. In this work we explored whether incorporating a myocardial volume conservation regularizer would improve strain estimates. We evaluated our technique on in silico echo sequences featuring realistic speckle textures and showed the volume conservation regularizer to be beneficial in reducing strain errors further when used in combination with a smoothness penalty. This combination led to more physiological boundary conditions. It also made distinguishing ischemic from normal segments easier in clinical images.

A Symmetric Prior for the Regularisation of Elastic Deformations: Improved Anatomical Plausibility in Nonlinear Image Registration

“…Most commonly, the Jacobian determinant itself is used as a hard constraint that bounds the allowed range (Haber and Modersitzki, 2004; Sdika, 2008, 2013; Haber et al, 2010), or fixes it to a particular value (Mansi et al, 2011), or a smoothly varying function of the Jacobian determinant is used as a soft constraint (Borz`i et al, 2003; Yanovsky et al, 2007; Leow et al, 2007; Heyde et al, 2016; Mang et al, 2018). It should be noted that when modelling the warps as a velocity field the local divergence can be used as a proxy for the Jacobian determinant.…”

Fast left ventricle tracking using localized anatomical affine optical flow

“…Moreover, it might also be linked to the fact that no constraints were added in this direction (ie, not directly affected by the chosen localized support parameters), with an independent propagation being applied for each myocardial layer. Such result could be improved by the use of explicit constraints, such as the assumption of myocardial volume conservation, or by a postprocessing regularization step …”

“…Moreover, it might also be linked to the fact that no constraints were added in this direction (ie, not directly affected by the chosen localized support parameters), with an independent propagation being applied for each myocardial layer. Such result could be improved by the use of explicit constraints, such as the assumption of myocardial volume conservation, 29 or by a postprocessing regularization step. 33,40 In the clinical experiments (Section 3.2 and 3.3), the proposed tracking strategy when combined to previously presented automatic segmentation strategies 16,45 proved to improve performance in both 3D US ( Table 1 and Table 2) and cine CMR datasets ( Table 3 and Table 4).…”

“…11,12 The latter focuses on following the motion of specific patterns along the cardiac sequence, thus allowing to extract the motion and deformation of the target structure. 6 Tracking methodologies include techniques based on block matching, 28 nonrigid registration, [29][30][31][32][33] optical flow, 16,[34][35][36][37] deformable models, 38 among others. 39,40 These methods differ on the way the motion is estimated over the image sequence, presenting different computational burdens and accuracies according to the modality and specific application.…”

Incompressible Image Registration Using Divergence-Conforming B-Splines

“…Proof. The proof follows from (5) and the fact that the value of a B-spline at any point m is bounded by the values of the B-spline coefficients associated to the knots that are close to m (see supplementary material for more details).…”