Image-based estimation of ventricular fiber orientations for personalized modeling of cardiac electrophysiology

Abstract

Technological limitations pose a major challenge to acquisition of myocardial fiber orientations for patient-specific modeling of cardiac (dys)function and assessment of therapy. The objective of this project was to develop a methodology to estimate cardiac fiber orientations from in vivo images of patient heart geometries. An accurate representation of ventricular geometry and fiber orientations was reconstructed, respectively, from high-resolution ex vivo structural magnetic resonance (MR) and diffusion tensor (DT) MR images of a normal human heart, referred to as the atlas. Ventricular geometry of a patient heart was extracted, via semiautomatic segmentation, from an in vivo computed tomography (CT) image. Using image transformation algorithms, the atlas ventricular geometry was deformed to match that of the patient. Finally, the deformation field was applied to the atlas fiber orientations to obtain an estimate of patient fiber orientations. The accuracy of the fiber estimates was assessed using six normal and three failing canine hearts. The mean absolute difference between inclination angles of acquired and estimated fiber orientations was 15.4°. Computational simulations of ventricular activation maps and pseudo-ECGs in sinus rhythm and ventricular tachycardia indicated that there are no significant differences between estimated and acquired fiber orientations at a clinically observable level.

Fig. 1

Our processing pipeline for estimating ventricular fiber orientations in vivo.

Fig. 2

Geometry and fiber orientations of the atlas ventricles. (A) Epicardial (red) and endocardial (green and magenta) splines, and corresponding landmarks (yellow) overlaid on an example slice of the atlas image. (B) Atlas ventricles in 3-D. (C) Atlas fiber orientations.

Fig. 3

Application of the fiber orientation estimation methodology to an example patient heart image. (A) Epicardial (red) and endocardial (green and magenta) splines, and corresponding landmarks (yellow) overlaid on an image slice. (B) Patient ventricles in 3-D. (C) Superimposition of ventricles of atlas [magenta, see Fig. 2(B)] and patient. (D) Patient ventricles and the affine transformed atlas ventricles. (E) Patient ventricles and LDDMM-transformed atlas ventricles. (F) Estimated patient ventricular fiber orientations.

Fig. 4

Validation of the fiber orientation estimation methodology by comparing estimated fiber orientations with DTMRI-derived orientations. (A) Superimposition of DTMRI-acquired fiber orientations (greenish yellow) and one set of estimated fiber orientations (cyan) of heart 1. (B) Acquired and estimated fiber orientations of heart 7. (C) Enlarged portion of (B) showing alignment between acquired and estimated fiber orientations. Note that the streamlines were generated at random locations within the myocardium for visualization purposes only, and so their exact positions are irrelevant. (D) Distribution of mean estimation error in normal ventricles. (E) Distribution of mean estimation error in failing ventricles. (F) Section of tissue extracted from (D). (G) Section of tissue extracted from (E). Colorbar applies to (D)–(G). (H) Histograms of errors in normal and failing ventricles. Frequency denotes the number of voxels having a given error.

Fig. 5

Results from simulations of one beat of sinus rhythm in normal canine ventricular models. (A) Activation map simulated using the model with acquired fiber orientations (model 1). (B) Absolute difference between simulated activation maps obtained from a ventricular model with acquired fiber orientations and that with estimated fiber orientations, averaged over the five estimates. (C) Simulated pseudo-ECGs with models 1 and 3. (D) Simulated activation maps from ventricles with estimated fiber orientations (models 2–6).

Fig. 6

Results from simulations of one beat of sinus rhythm in failing heart models. In the first column, rows 1–3 show activation maps calculated using models 7–9, respectively. In the second column, rows 1–3 display results of simulations with models 10–12, respectively. Rows 1–3 in the third column portray the absolute difference between the activation maps shown in the first and second columns of the corresponding row. Rows in the fourth column display simulated pseudo-ECGs from models in the first and second columns of the corresponding row.

Fig. 7

Results from simulations of VT induction with the failing heart models. Rows 1–3 in the first column show activation maps during one cycle of reentrant activity in simulations with models 7–9, respectively. Rows 1–3 in the second column show activation maps corresponding to models 10–12, respectively. Rows in the third column illustrate pseudo-ECGs from models in the first and second columns of the corresponding row.