Aorta flow characterization and simulation

Analysis of mechanical behaviour in blood flow is a relevant topic for many diseases. Wall Shear Stress (WSS) in aorta has been related to pathologies such as atherosclerosis formation. The study of this topic using simulation opens many possibilities, including the use of Machine Learning (ML) techniques to associate WSS profiles to different aorta geometries. However, the training of ML models requires large aorta datasets. Moreover, these models often rely on the existence of a common numerical description for every case in the dataset.

The main goal of this research line is to build a parametric model of the aorta geometry based on a reduced number of characteristics that can be used in these kind of models. This model will be used to develop a computational pipeline that helps the simulation of blood flow and the obtention of clinically meaningful information.

Unlike approaches that rely on Principal Component Analysis (PCA) applied directly to the wall geometry, the resulting method of this project is based on geometric information that can be given anatomical meaning.

The proposed methodology is able to characterize an aorta from Medical Image (MI) data and synthesize an equivalent geometry with only small visible differences.

Geometric parametrization

The geometric characterization is based on the assumption that the aorta can be modelled, to some extent, as a tubular surface. Without going into technical details, a tubular surface can be considered as the Cartesian product of a regular curve (the centerline) and a regular closed curve. In particular, Canal Surfaces use circumferences as the closed curve that generates the surface when travelling along the centerline. As a result, canal surfaces can be easily parametrized using their centerline and cross sections, which are circumferences.

Our algorithm will generate a similar parametrization using an evolving frame along the curve and ellipses that fit the surface that was given as input.

Validation of geometry

One of the main goals is to be able to generate aorta geometries in order to perform Computational Fluid Dynamics (CFD) simulations. To validate the geometry characterization and generation procedure, simulations of blood flow in an aorta acquired from CT and the reconstructed geometry using our method have been run and compared.

Expected impact of the results

Among other applications, the resulting geometry characterization method can be used to develop a pipeline for thoracic aorta simulation using CFD or to generate large datasets of aorta geometries corresponding to a given population. This can have a considerable impact on in-silico trial studies.

Related publications

Pau Romero, Samuel Santos, Rafael Sebastian, Francisco Martinez-Gil, Dolors Serra, Pilar Calvillo, Alejandro Rodríguez, Rebeca Maldonado Puig, Luis Martí-Bonmatí, Angel Alberich-Bayarri, Miguel Lozano and Ignacio García-Fernández. Reconstruction of the aorta geometry using canal surfaces. In proc. of the International Conference on Computational and Mathematical Biomedical Engineering. 2019. PDF.

Samuel Santos; Pau Romero; Rafael Sebastian; Francisco Martinez-Gil; Dolors Serra; Pilar Calvillo; Alejandro Rodríguez; Rebeca Maldonado Puig; Luis Martí-Bonmatí; Angel Alberich-Bayarri; Miguel Lozano; Ignacio García-Fernández.
Validation of wall shear stress profiles from reconstructed aorta geometries. Poster in the International Congress on Industrial and Applied Mathematics. 2019.

Pau Romero; Samuel Santos; Rafael Sebastian; Francisco Martinez-Gil; Dolors Serra; Pilar Calvillo; Alejandro Rodríguez; Rebeca Maldonado Puig; Luis Martí-Bonmatí; Angel Alberich-Bayarri; Miguel Lozano; Ignacio García-Fernández.
PARAMVALVE: In-silico prediction of aorta wall shear stress from a geometric characterization. Poster in the International Congress on Industrial and Applied Mathematics. 2019.

Alba Peris; Samuel Santos; Pau Romero; Dolors Serra; Ignacio García-Fernández; Miguel Lozano.
Pressure boundary conditions for SPH fluid simulations. Poster in the International Congress on Industrial and Applied Mathematics. 2019.

Funding

This research has been funded through projects


  • PARAMVALVE. Predicción in silico del efecto de prótesis valvular para planificación y optimización de la terapia.
    Ayudas Para El Desarrollo De Proyectos De Innovación Conjuntos Entre Investigadores De La Universitat De Valencia Y Del Hospital La Fe / Iis La Fe. PROGRAMA VLC-BIOMED 2017.
    01/05/2018 - 31/04/2019. 12-PARAMVALVE-GARCIA-ALBERICH-2017-B. Funding: 15000 €.
    PI : Ignacio Garcia-Fernandez.

  • DESSAVALVE. Patient-Specific Modelling of the Aortic valve replacement: Advance towards a Decision Support System.
    CIBER-BBN Intramural Program. Instituto de Salud Carlos III, Ministerio de Economía y Competitividad.
    01/07/2016- 30/06/2018. CIB16-BM019.
    co-PI : Ignacio Garcia-Fernandez. 
  •  
    PS-VAVALVE. Patient-Specific image-based computational flow dynamic simulations for assessment of optimal biological aortic Valve replacement selection and delivery.
    CIBER-BBN Intramural Program. Instituto de Salud Carlos III, 01/01/2014- 31/12/2015. CIB13-BM13.
    co-PI : Rafael Sebastian. 

Shock wave formation in compliant arteries

We focus on the problem of shock wave formation in a model of blood flow along an elastic artery. We analyze the conditions under which this phenomenon can appear and we provide an estimation of the instant of shock formation. Numerical simulations of the model have been conducted using the Discontinuous Galerkin Finite Element Method. The results are consistent with certain phenomena observed by practitioners in patients with arteriopathies, and they could predict the possible formation of a shock wave in the aorta.

Cristóbal Rodero, J. Alberto Conejero and Ignacio García-Fernández.
Shock wave formation in compliant arteries. Evolution Equations & Control Theory. Volume 8, issue 1. 2019. Full text.

Automatic Estimation of Purkinje-Myocardial Junction hot-spots from Noisy Endocardial Samples: A simulation study

The reconstruction of the ventricular cardiac conduction system (CCS) from patient‐specific data is a challenging problem. High‐resolution imaging techniques have allowed only the segmentation of proximal sections of the CCS from images acquired ex vivo. In this paper, we present an algorithm to estimate the location of a set of Purkinje‐myocardial junctions (PMJs) from electro‐anatomical maps, as those acquired during radio‐frequency ablation procedures. The method requires a mesh representing the myocardium with local activation time measurements on a subset of nodes. We calculate the backwards propagation of the electrical signal from the measurement points to all the points in the mesh to define a set of candidate PMJs that is iteratively refined. The algorithm has been tested on several Purkinje network configurations, with simulated activation maps, subject to different error amplitudes. The results show that the method is able to build a set of PMJs that explain the observed activation map for different synthetic CCS configurations. In the tests, the average error in the predicted activation time is below the amplitude of the error applied to the data.

F. Barber, M. Lozano, I. García‐Fernández, R. Sebastián.
Automatic Estimation of Purkinje-Myocardial Junction hot-spots from Noisy Endocardial Samples: A simulation study. International Journal for Numerical Methods in Biomedical Engineering. Volume 34, issue 7. Article e2988. 2018. PDF.

Atrial fibrosis hampers non-invasive localization of atrial ectopic foci from multi-electrode signals: A 3D simulation study

Introduction: Focal atrial tachycardia is commonly treated by radio frequency ablation with an acceptable long-term success. Although the location of ectopic foci tends to appear in specific hot-spots, they can be located virtually in any atrial region. Multi-electrode surface ECG systems allow acquiring dense body surface potential maps (BSPM) for non-invasive therapy planning of cardiac arrhythmia. However, the activation of the atria could be affected by fibrosis and therefore biomarkers based on BSPM need to take these effects into account. We aim to analyze the effect of fibrosis on a BSPM derived index, and its potential application to predict the location of ectopic foci in the atria.

Methodology: We have developed a 3D atrial model that includes 5 distributions of patchy fibrosis in the left atrium at 5 different stages. Each stage corresponds to a different amount of fibrosis that ranges from 2 to 40%. The 25 resulting 3D models were used for simulation of Focal Atrial Tachycardia (FAT), triggered from 19 different locations described in clinical studies. BSPM were obtained for all simulations, and the body surface potential integral maps (BSPiM) were calculated to describe atrial activations. A machine learning (ML) pipeline using a supervised learning model and support vector machine was developed to learn the BSPM patterns of each of the 475 activation sequences and relate them to the origin of the FAT source.

Location of ectopic foci

Results: Activation maps for stages with more than 15% of fibrosis were greatly affected, producing conduction blocks and delays in propagation. BSPiMs did not always cluster into non-overlapped groups since BSPiMs were highly altered by the conduction blocks. From stage 3 (15% fibrosis) the BSPiMs showed differences for ectopic beats placed around the area of the pulmonary veins. Classification results were mostly above 84% for all the configurations studied when a large enough number of electrodes were used to map the torso. However, the presence of fibrosis increases the area of the ectopic focus location and therefore decreases the utility for the electrophysiologist.

Conclusions: The results indicate that the proposed ML pipeline is a promising methodology for non-invasive ectopic foci localization from BSPM signal even when fibrosis is present.

Eduardo Jorge Godoy, Miguel Lozano, Ignacio Garcia-Fernandez, Ana Ferrer-Albero, Javier Saiz, Rafael Sebastian.
Atrial fibrosis hampers non-invasive localization of atrial ectopic foci from multi-electrode signals: A 3D simulation study. Frontiers in Physiology. Volume 9. Article 404. 2018. PDF.

Conformation Constraints for Efficient Viscoelastic Fluid Simulation

Cake Teaser
Conformation Constraints for Efficient Viscoelastic Fluid Simulation

The simulation of high viscoelasticity poses important computational chal-lenges. One is the difficulty to robustly measure strain and its derivativesin a medium without permanent structure. Another is the high stiffness ofthe governing differential equations. Solutions that tackle these challengesexist, but they are computationally slow. We propose a constraint-basedmodel of viscoelasticity that enables efficient simulation of highly viscousand viscoelastic phenomena. Our model reformulates, in a constraint-basedfashion, a constitutive model of viscoelasticity for polymeric fluids, whichdefines simple governing equations for a conformation tensor. The modelcan represent a diverse palette of materials, spanning elastoplastic, highlyviscous, and inviscid liquid behaviors. In addition, we have designed a con-strained dynamics solver that extends the position-based dynamics methodto handle efficiently both position-based and velocity-based constraints. Weshow results that range from interactive simulation of viscoelastic effects tolarge-scale simulation of high viscosity with competitive performance.

Publication:

Héctor Barreiro, Ignacio García-Fernández, Iván Alduán, Miguel A. Otaduy.
Conformation Constraints for Efficient Viscoelastic Fluid Simulation. ACM Trans. on Graphics (Proc. of ACM SIGGRAPH Asia), Volume 36, Number 6 – 2017. PDF preprint, suplemental material, video.