Which anatomical directions to quantify local right ventricular strain in 3D echocardiography? - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Which anatomical directions to quantify local right ventricular strain in 3D echocardiography?

Résumé

Technological advances in image quality and post-processing have led to the better clinical adoption of 3D echocardiography to quantify cardiac function. However, the right ventricle (RV) raises specific challenges due to its specific half-moon shape, which led to a lack of consensus regarding the estimation of RV motion and deformation locally. In this paper, we detail three ways to estimate local anatomically-relevant directions at each point of the RV surface, in 3D, and the resulting Green-Lagrange strain projected along these directions. Using a database of RV surface meshes extracted from 3D echocardiographic sequences from 100 control subjects, we quantified differences between these strategies in terms of local anatomical directions and local strain, both at the individual and population levels. For the latter, we used a specific dimensionality reduction technique to align the latent spaces encoding the strain patterns obtained from different computations of the anatomical directions. Differences were subtle but visible at specific regions of the RV and partially interpretable, although their impact on the population latent representation was low, which sets a preliminary quantitative basis to discuss these computation standards.
Fichier principal
Vignette du fichier
DiFolco_FIMH_2023.pdf (931.45 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04096724 , version 1 (13-05-2023)

Licence

Paternité

Identifiants

  • HAL Id : hal-04096724 , version 1

Citer

Maxime Di Folco, Thomas Dargent, Gabriel Bernardino, Patrick Clarysse, Nicolas Duchateau. Which anatomical directions to quantify local right ventricular strain in 3D echocardiography?. 12th International Conference on Functional Imaging and Modeling of the Heart, Jun 2023, Lyon, France. pp.607-615. ⟨hal-04096724⟩
22 Consultations
27 Téléchargements

Partager

Gmail Facebook X LinkedIn More