# Automatic Mapping of Atrial Fiber Orientations for Patient-Specific Modeling of Cardiac Electromechanics using Image-Registration

## TL;DR

This paper introduces an automated image registration method to map detailed atrial fiber orientations onto patient-specific heart models, enabling personalized electromechanical simulations without invasive measurements.

## Contribution

The study presents a novel, automated approach for defining atrial fiber architecture in patient-specific models using image registration based on an atlas with histologically detailed fibers.

## Key findings

- Accurately mapped fiber orientations in ten patient-specific atria.
- Electrophysiological activation patterns in patient models matched atlas characteristics.
- Method enabled efficient, personalized electromechanical simulations.

## Abstract

Knowledge of appropriate local fiber architecture is necessary to simulate patient-specific electromechanics in the human heart. However, it is not yet possible to reliably measure in-vivo fiber directions, especially in human atria. Thus, we present a method which defines the fiber architecture in arbitrarily shaped atria using image registration and reorientation methods based on atlas atria with fibers predefined from detailed histological observations. Thereby, it is possible to generate detailed fiber families in every new patient-specific geometry in an automated, time-efficient process. We demonstrate the good performance of the image registration and fiber definition on ten differently shaped human atria. Additionally, we show that characteristics of the electrophysiological activation pattern which appear in the atlas atria also appear in the patients' atria. We arrive at analogous conclusions for coupled electro-mechano-hemodynamical computations.