Image-based High-resolution In-silico Modeling of Total Cardiac Function

Informations projet

InsiliCardio

N° de convention de subvention: 750835

DOI

10.3030/750835

Projet clôturé

Date de signature de la CE 20 Février 2017

Date de début 1 Mars 2017

Date de fin 31 Août 2019

Financé au titre de

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Coût total

€ 218 176,20

Contribution de l’UE

€ 218 176,20

Coordonné par

MEDIZINISCHE UNIVERSITAT GRAZ
Austria

Periodic Reporting for period 2 - InsiliCardio (Image-based High-resolution In-silico Modeling of Total Cardiac Function)

Période du rapport: 2018-09-01 au 2019-08-31

Résumé du contexte et des objectifs généraux du projet

OVERALL OBJECTIVES
Our objective was to develop the most advanced biophysically detailed in-silico model of total electro-mechano-fluidic function of the heart.
This model was parametrized, verified, and used to study cause-effect relationships between flow and pressure and their impact upon pumping performance.
A novel set of features such as combined models of both heart and attached outflow vessels and the computational efficiency provides a unique platform for translational research.

This ambitious endeavor was feasible only by combining the expertise of the applicant in modeling soft tissue mechanics and his supervisors in modeling electrophysiology (Gernot Plank, MUG) and blood flow (Shawn Shadden, UC Berkeley).
Clinical input and datasets for model parametrization and validation were provided by Titus Kühne (DHZ Berlin) and by clinical collaborators of Prof. Shadden at UCSF.
During the return-phase, the applicant used the infrastructure of Prof. Plank’s lab and the large network of academic and industrial collaborations as an incubator for building up his own research group by starting an tenure track position in the field of computational hemodynamics and biomechanics at the Medical University of Graz.

CONCLUSION OF THE ACTION
A unique biophysically detailed and anatomically accurate model of total heart function was developed that allows the computer simulation of bidirectionally coupled
electo-mechano-fluidic models of cardiac function. This was achieved by developing novel methodology which enables these multi-physics
simulations at an unprecedented level of biophysical and anatomical detail.
The unique in-silico framework was validated by comparison to clinical data provided by our clinical partners and allows to accurately predict the electromechanical response of heart and aorta to changes in afterload in terms of tissue and wall shear stresses.

In particular, this action addressed the following objectives:
i) Postprocess cardiac image data using semi-automatic algorithms;
ii) Simulate cardiac and vascular hemodynamics with short simulation cycles based on a massively parallel computing approach;
iii) Model the bidirectional interaction between blood flow and tissue deformation to quantify the impact of flow upon heterogeneity in mechanical loading;
iv) Validate and parameterize the biophysically detailed electro-mechano-fluidic model with patient-specific data.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

"WORK PERFORMED FROM THE BEGINNING OF THE PROJECT

WP 1: Postprocessing of tomographic image data
A large number of clinical datasets (imaging, pressure, and laboratory data) have been made available as an outcome of the EU funded project CardioProof. From these datasets a total of 19 cases have been segmented and postprocessed.
Additionally, due to a cooperation with KCL, a virtual cohort of 24 four-chamber models is available for future studies.
WP 2 – Hemodynamics and fluid structure interaction on high-performance computers
Task 2.1 – Efficient computation of hemodynamics
A massively parallel computational fluid dynamics (CFD) code was implemented together and results showed fast computational times and excellent strong scaling properties. Additionally, a code framework based on FEniCS was implemented together with Miguel Rodriguez from the Shadden Lab, UC Berkeley.
Task 2.2 – FSI and the simulation of total heart function
The CFD code implemented for WP 2.1 was coupled to the electromechanical framework previously published. This code is capable of performing weakly-coupled fluid-structure interaction (FSI) problems and results were published in [Karabelas2018]. Additionally, an Arbitrary Lagrangian-Eulerian (ALE) FSI framework based on FEniCS was implemented together with Fanwei Kong (Shadden Lab, UC Berkeley).
WP 3 – Validation and parameterization of the in-silico EMF models
First validation and parameterization results were published during the outgoing phase. Thorough parameterization, personalization, and validation studies are about to be submitted. Sensitivity analysis and uncertainty quantifications were performed.

DISSEMINATION OF RESULTS
* 6 Peer-reviewed journal publications (see Publications)
* additionally 5 submitted journal publications
* 6 talks at international conferences

* Science to public activities:
1) Austrian Research and Innovation Talk in Austin to present the work to an non-scientific audience from industry and media.
2) The research will be featured in the Bridges Blog of the Office of Science and Technology Austria (https://ostaustria.org/bridges-blog)
3) Science to Schools in prepartion with the BRG Kepler, Graz
4) My path as a Postdoc was featured in the University of California Postdoc Newsletter.
5) Member of the ASCINA network and active participant in several activities of ASCINA.
6) Presentation: ""In silico models of cardiovascular mechanics as a tool for medical device development"" at the ""In Silico Get Together""-Meeting in Graz
7) Presentation: ""Digital twin models of cardiac function in precision cardiology"" at the ""Open Campus - Precision Medicine Day"" in Graz
8) Presentation: ""In silico Modelle der Gesamtherzfunktion"" at the ""Informationsveranstaltung zum Thema Anwendung der 3R (Replace, Reduce, Refine)"" in Graz
9) Creation of webpage for dissemination and presentation of results (https://ccl.medunigraz.at)"

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

Computer-based methods, which are indispensable in engineering and industry, are becoming increasingly important in medicine and the life sciences and computational cardiovascular bio- and fluid mechanics have gained great momentum in recent years. The significance of this growing interdisciplinary research area is documented by lighthouse projects within the VPH initiative in Europe, or by NIH pioneering awards and Physiome modeling projects in the US as well as FDA’s Medical Device Development Tools (MDDT) and Software as a Medical Device (SaMD) programs.
Recent advances in simulating cardiac electrophysiology and mechanics have rendered the heart the most highly integrated example of a virtual organ. Today, biophysically detailed and tomographically reconstructed, anatomically accurate models of the heart are considered to be the state-of-the-art.
While major progress has been achieved in modeling the various fields of physics in isolation, this is, to a much lesser extent, the case for multiphysics models of total cardiac function. Such models are way more demanding due to their inherent integrative nature.

Research into advancing methodology for in silico clinical applications are poised to have a significant impact on future health care throughout Europe and as such also on European society and economy.
It is worth noting that the electrophysiological simulation software, co-developed by the applicant, is already in use by a world market leader in cardiac devices to optimize devices for anti-arrhythmia therapy.
Electro-mechanical applications are currently on the brink of being introduced into the clinic as a tool for optimizing cardiac resynchronization therapies and computational blood flow analysis is clearly an emerging market as well (http://www.heartflow.com/).

Patient-specific in-silico models of total heart function.