Project

Back to overview

A Flexible High Performance Approach to Cardiac Electromechanics

Applicant Krause Rolf
Number 149828
Funding scheme Project funding (Div. I-III)
Research institution Istituto di scienze computazionali (ICS) Facoltà di scienze economiche
Institution of higher education Università della Svizzera italiana - USI
Main discipline Information Technology
Start/End 01.04.2014 - 31.03.2017
Approved amount 172'426.00
Show all

Keywords (7)

nonlinear elasticity; non-nested meshes; cardiac electromechanics; parallel computing; electrophysiology; cardiac mechanics; partial differential equations

Lay Summary (Italian)

Lead
The understanding, diagnosis, and treatment of heart failure is of major importance for our societies and has a strong impact on economic and social life. The mechanisms of development of heart failure are complex. This is especially true when it comes to the interaction of electrical and mechanical activity, which, despite intense and active research, is still not understood satisfactorily. In this project, we aim at developing a simulation framework for the electro-mechanical behavior of the human heart, which will be build on recent ideas and methodological developments in applied mathematics and computational sciences. We aim in particular at developing simulation tools, which will enable us to exploit the computational power of modern super-computers efficiently, thus turning them into a useful tool in cardiac simulation, with the final goal to improve diagnosis and therapy of heart failure.
Lay summary

L'insufficienza cardiaca è una delle patologie più frequenti nei paesi occidentali. Essa è il risultato di diverse malattie cardiache che compromettono a complessa interazione tra l'attività elettrica e meccanica del cuore. L’insufficienza cardiaca provoca una progressiva riduzione delle capacità motorie e della qualità della vita in generale e è in continuo aumento.
Fortunatamente, le sempre maggiori capacità dei moderni supercomputer e il progresso del calcolo ad elevate prestazioni hanno permesso negli ultimi decenni lo studio approfondito di modelli cardiaci “in silico” sempre più dettagliati. I modelli matematici e le simulazioni numeriche si sono rilevate fondamentali per descrivere e capire i diversi processi che hanno luogo nel cuore. Gli strumenti di simulazione più ampiamente utilizzati si basano su modelli, discretizzazioni e metodi di risoluzione che si focalizzano puramente su un approccio meccanico o elettrofisiologico. Sebbene esistano modelli elettromeccanici, che incorporano entrambi gli aspetti, i metodi numerici e gli strumenti di simulazione di questi sistemi accoppiati non sono ancora efficienti ed elaborati se comparati a quelli sviluppati per il caso dei modelli “mono-fisica”. Questo risulta essere chiaramente uno svantaggio rilevante nella ricerca e nelle reali applicazioni cliniche.
L'obiettivo di questo progetto è dunque quello di sviluppare un nuovo framework di simulazione, basato sugli elementi finiti, che permetta di simulare ed indagare tutte le componenti elettrofisiologiche e meccaniche che avvengono alle diverse scale spazio-temporali. In tal modo si vuole creare un “laboratorio virtuale” che permetta di perfezionare la diagnosi e la terapia (delle malattie causa) dell’insufficienza cardiaca.
Basandoci sull’utilizzo delle più moderne tecniche esistenti nel campo della simulazione numerica, sarà quindi possibile sfruttare pienamente le capacità dei moderni supercomputer, trasformandoli in un utilissimo strumento di simulazione del cuore.
Direct link to Lay Summary Last update: 17.04.2014

Responsible applicant and co-applicants

Employees

Name Institute

Publications

Publication
The relation between local repolarization and T-wave morphology in heart failure patients.
(2017), The relation between local repolarization and T-wave morphology in heart failure patients., in International journal of cardiology.
Impact of mechanical deformation on pseudo-ECG: a simulation study.
(2016), Impact of mechanical deformation on pseudo-ECG: a simulation study., in Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working grou.
An in-silico analysis of the effect of heart position and orientation on the ECG morphology and vectorcardiogram parameters in patients with heart failure and intraventricular conduction defects.
(2015), An in-silico analysis of the effect of heart position and orientation on the ECG morphology and vectorcardiogram parameters in patients with heart failure and intraventricular conduction defects., in Journal of electrocardiology.
Verification of cardiac mechanics software: benchmark problems and solutions for testing active and passive material behaviour
(2015), Verification of cardiac mechanics software: benchmark problems and solutions for testing active and passive material behaviour, in Proceedings of the Royal Society A: Mathematical, Physical and Engineering Science.

Collaboration

Group / person Country
Types of collaboration
Free University Berlin, Prof. Dr. R. Kornhuber, Prof. Dr. Dr. h.c. P. Deuflhard Germany (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. F. Schuricht, University of Dresden Germany (Europe)
- in-depth/constructive exchanges on approaches, methods or results
Maastricht University, Cardiovascular Research Institute Maastricht, Prof. Dr. Frits Prinzen Netherlands (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Research Infrastructure
Politecnico di Milano, Dipartimento di Elettronica, Informazione e Bioingegneria, Dr. Enrico Caiani Italy (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Research Infrastructure
Zuse Institut Berlin, Dr. M. Weiser Germany (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Research Infrastructure
- Exchange of personnel
Prof. Christoph Bourauel, University of Bonn Germany (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
University of Lugano, Institute of Computational Science, Dr. W. Kroon Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication

Scientific events

Active participation

Title Type of contribution Title of article or contribution Date Place Persons involved
ECCM - ECFD Eccomas Conference Talk given at a conference Multigrid augmented Lagrangian preconditioners for incompressible cardiac mechanics. 13.06.2018 Glasgow, Great Britain and Northern Ireland Krause Rolf; Pozzi Sonia;
Swiss Numerics Colloquium Poster Higher Order Exponential Time Integrators in Cardiac Electrophysiology. 28.04.2017 Basel, Switzerland Krause Rolf; Pozzi Sonia;
APCOM & WCCM 2016 Talk given at a conference Efficient Solution of Saddle Point Problem 24.07.2016 Seoul, Korean Republic (South Korea) Krause Rolf; Pozzi Sonia;
LATSIS Talk given at a conference Impact of Mechanical Deformation on Pseudo-ECGs: a Simulation Study 26.06.2016 Zürich, Switzerland Krause Rolf; Pozzi Sonia;
ECCOMAS 2015 Talk given at a conference Eccomas Congress, Crete, Title: Fully Coupled Solution Strategies for Electro-Mechanics in the Heart. 05.06.2016 Crete, Greece Krause Rolf; Pozzi Sonia;
Colloque Numerique Suisse Talk given at a conference Numerical Strategies for Saddle Point Formulation in Cardiology 22.04.2016 Fribourg, Switzerland Pozzi Sonia; Auricchio Angelo; Krause Rolf;
ICCM 15 Talk given at a conference A New Reduced Dual Basis Approach for Parametrized Contact Problem in Elasticity 04.05.2015 Hannover, Germany Pozzi Sonia; Krause Rolf;


Self-organised

Title Date Place
MALT meeting – iMaging And eLectrical Technologies – meeting 30.04.2015 USI, Lugano, Switzerland

Knowledge transfer events

Active participation

Title Type of contribution Date Place Persons involved
The Digital Future Talk 12.05.2016 Berlin, Switzerland Krause Rolf;


Communication with the public

Communication Title Media Place Year
Media relations: print media, online media Digital Future Der Tagesspiegel, Berlin International 2016
Media relations: print media, online media L’UNIVERSO Giornale Studentesco Universitario Indipendente. L’UNIVERSO Italian-speaking Switzerland 2016

Awards

Title Year
Best Poster Award on 12th IEEE EMBS International Summer School on Biomedical Imaging Saint-Jacut de la Mer, Emerald Coast, Brittany, France, 16-24 June, 2016 2016

Use-inspired outputs

Software

Name Year

Start-ups

Name Year
Algo4U 2016

Associated projects

Number Title Start Funding scheme
169599 Multilevel Methods and Uncertainty Quantification in Cardiac Electrophysiology 01.10.2016 Project funding (Div. I-III)

Abstract

Heart failure is one of the most common cardiovascular diseases in western countries. Being the result of different heart diseases, by which the complex interplay of electrical activation and mechanical contraction within the heart is severely affected, heart failure leads to a progressive reduction in exercise capabilities and quality of life. Due to the ongoing demographic changes in our societies the incidence of heart failure will increase further.Clearly, the understanding, diagnosis, and treatment of heart failure is of major importance for our societies and has a strong impact on economic and social life. The mechanisms of development of heart failure are complex and differ substantially between patients. This is especially true when it comes to the interaction of electrical and mechanical activity, which is a central and crucial aspect for heart function. In fact, despite intense and active research, significant parts of the electromechanical interplay in the heart muscle are still not understoodsatisfactorily, posing a major obstacle towards diagnostic and therapeutic progress.Fortunately, during the last decades the ever growing capacities of modern supercomputers and advancement of high performance computing have allowed for the realization of more and more detailed “in-silico models” of the heart: Mathematical modeling and numerical simulation have turned out to be an indispensable tool to understand and describe the different mechanisms within the heart muscle. However, most of the widely used simulation tools are based on models, discretizations, and solution methods that focus primarily on either mechanics or electrophysiology.Electromechanical models, which incorporate both aspects, are available, but the numerical methods and tools employed for the numerical treatment of these coupled systems are not yet as efficient and elaborate as it is the case for the respective “mono-physics models.” This obviously is a clear disadvantage for research and clinical application. The number of potential applications for detailed coupled electromechanical models is vast, as they allow for, e.g., linking cellular electrophysiology and tissue mechanics across spatial scales to pumpfunction of the heart as a whole, relating abnormalities in ion channels and calcium handling to abnormalilites in the electrocardiogram, or linking regional infarction to aberrant wall motion.A first reason for the “progress gap” of the coupled models can be found in the adequate and efficient coupling of the different temporal and spatial scales at which electrophysiology and mechanics take place. Whereas the electrophysiology can be considered as meso- (activation potential) and micro-effect (ion channels), the mechanical deformation is related more to the macro-scale. Nevertheless, most state-of-the art approaches employ a single spatial discretization for both displacements and activation potentials, thereby leading to one of the scales beingeither under- or overresolved. Coupled approaches, in which an independent choice of discretizations for the different variables is possible, are not available.A second reason is that the ever-increasing performance of modern supercomputers comes at the price of massive parallelism, which requires many simulation methods and the corresponding scientific software to be redesigned in order to be able to deal with the upcoming generations of supercomputers. This is especially true when it comes to the parallel solution of coupled systems of partial differential equations (PDEs) that arise from the electromechanical models, as new developments usually are first realized for “mono-physics” models.The aim of this proposal is therefore to close this “progress gap” by 1) developing a novel coupled multiscale simulation framework which allows to fully exploit the difference in required spatiotemporal resolution of the electrophysiological and mechanical components by allowing different (finite-element) discretizations on the different scales; 2) developing implicit and monolithic solution techniques for the arising coupled multiscale-systems based on state-of-the-art discretization and solution approaches, and; 3) implementing the newly developed methods within a user-friendly software library, featuring capabilities for flexible choices of models as well as discretizations and solution methods.
-