People


Timur Nezlobinsky

PhD student
timur.nezlobinskii@ugent.be

office: 120 008

Tel: +32 - (0)9 264 47 97

 

Research interests:

Scientific computing
Numerical methods
Life science

 

×
Timur Nezlobinsky

Timur Nezlobinsky

PhD student

Hans Dierckx

doctor-assistant
Hans.Dierckx@ugent.be

office: 120 008

Tel: +32 - (0)9 264 47 97

 

Research:

Curved-space approach to cardiac anisotropy

The propagation of excitation waves (action potentials) in the heart is highly anisotropic: they travel about three times faster along the local muscle fiber direction than perpendicular to it. The complex organisation of myofibers renders a general description difficult. Therefore, for a long time the usual way to study the effects of cardiac anisotropy on wave patterns was by blunt numerical simulations.

One particular set of simulations of a cardiac rotor revealed that the equilibrium configuration of the filament curve around which the rotor revolves can be understood in a remarkable manner: it appears to be a geodesic of the space with metric tensor based on the local conduction velocities of the tissue (Wellner et al., PNAS, 2002). Later, we derived the equations of motion for generic rotor filaments in anisotropic cardiac tissue (Verschelde et al., Phys. Rev. Lett. 2007), from which the geodesic principle follows as a special case. In general, it was shown that anisotropic cardiac tissue can be modelled by a Riemannian electrophysiological metric (Young and Panfilov, PNAS, 2010).

Later on, I derived higher order corrections to the geodesic principle in my PhD thesis, and currently we are investigating how to use an effective 2D metric to model 3D cardiac tissue.

Laws of motion for depolarization waves and rotors in cardiac tissue.

The wave fronts and rotors in cardiac tissue exhibit emerging dynamics that is complex, but should nevertheless obey certain physical principles. In the past decade, I have studied how different effects that are present in the heart affect wave and rotor dynamics, including wave front curvature, filament curvature, 2D anisotropy, surface curvature of the atrial walls, smooth and sudden thickness variations, parameter gradients, bidomain models, and mechano-electrical coupling.

Recently, I was able to generalize the above theory to meandering spiral waves (i.e. spirals with a non-circular tip trajectory), which are typically observed in cardiac simulations and experiments.

Computational modelling of electrocardiographic signals.

I am currently performing computational modelling of electrocardiograms with many electrodes (`electrocardiographic imaging' or ECGi). This recent technique enables to reconstruct to some degree the excitation patterns on the outer cardiac surface. Therefore it is a novel tool to unravel underlying mechanisms of cardiac arrhythmias, which is moreover directly applicable in the clinic.

Cardiac electromechanics

Together with PhD-student Sander Arens, I have looked into the effect of mechano-electrical coupling on rotor dynamics. In a simple 2D geometry, we found that the resonant part of the mechanical strain enables spiral wave drift, whose magnitude and direction is mostly determined by boundary conditions, and less by the temporal profile of the developed mechanical tension.

We now have a finite-element code in the group to study cardiac electromechanics and bidomain models in realistic geometries.

Theory for explaining experiments

I find it interesting to work with other colleagues or groups which perform simulations or experiments on excitable systems. In these cases, I try to provide theoretical or geometrical insight to explain the experimentally observed behaviour. Previous examples include the buckling of rotor filaments, phase-locking of spiral waves, restabilisation of scroll waves using a rotating external field, and ectopic wave emission in cardiac monolayers.

 

Teaching:

Mathematical methods in physics

 

Personal Link

Personal Link ×
Hans Dierckx

Hans Dierckx

doctor-assistant

Henri Verschelde

Professor
Henri.Verschelde@ugent.be

office: 120 007

Tel: +32 - (0)9 - 264 47 89

Research interests:
Quantum Field Theory
Elementary Particle Physics
Quantum Computing
Heart waves

Teaching:
Quantum mechanics II
Quantum field theory
Quantum black holes and holography

×
Henri Verschelde

Henri Verschelde

Professor

Nele Vandersickel

Post-doc
Nele.Vandersickel@UGent.be

office: 120 034

Tel: +32 - (0)9 - 264 48 00

 

Research:

Cardiac arrhythmias are one of the largest causes of death in the Western world and it is therefore of great importance to understand the nature of these arrhythmias. From general scientific point of view arrhythmias are associated with vortices formed by non-linear waves of excitation and thus can be studied not only using biological and medical approaches but also using the methods developed in physics. One of such methods which is used in this project is multi-scale biophysical modelling of the heart. We describe the heart by a complex system of partial differential equations and our model correctly represents the anatomy of the heart most of electrophysiological processes resulting in cardiac arrhythmias.

Early Afterdepolarizations and Torsade de Pointes.

In particular I focus on arrhythmias which occur as a result of abnormal excitations of cardiac cells (triggered activity) and study how these abnormal excitations can produce a lethal arrhythmia, namely Torsade de Pointes (TdP). This arrhythmia is very important as many forms of cardiac genetic decease as well as adverse drug actions increase the risk of TdP. My goal is therefore to continue to unravel the underlying mechanism of TdP. I have a collaboration with a world leading experimental group which studies TdP in a dog model (prof. M. Vos). Combining computational studies with experimental data gives us new insights in the mechanisms underlying TdP and possible methods of its management. I published several papers on this topic ( PLoS One 9 (2014), Heart Rhythm 12 (2015), J. Physiol. 594 (2016), Frontiers in physiology 8(2017)). Recently, we found the mechanism of TdP in the CAVB dog model with the help of a new mathematical technique (JACC Clinical Electrophysiology (2017)).

Together with Enid Van Nieuwenhuyse, we also study EADs in the human ventricles in 3D, whereby we characterize the different patterns in 3D. In addition, we are able to compute realistic ECGs so we can compare our simulated ECGs with real ECG recordings of TdP (collaboration with Prof. G. Seemann).

DineHeart

We are currently applying mathematical methods to predict the ablation points the rotational activity in atrial tachycardia. We have a collaboration with the Ghent University hospital with Dr. M. El Haddad and Prof. M. Duytschaever.

Other studies

In the mean time, I broadened my scope by studying fibrosis in patient-specific models (collaboration with Prof. K. Zeppenfeld, Leiden), studying inverse modeling (ECGi, collaboration with Dr. M. Clerx, Dr. M. Cluitmans, Prof. P Volders, Maastricht), and the role of hemichannels on cardiac tissue (collaboration with Prof. L. Leybeart, Ghent).

 

Website

×
Nele Vandersickel

Nele Vandersickel

Post-doc

Sander Arens

PhD Student
Sander.Arens@UGent.be

office: 120 008

Tel:  +32 - (0)9 - 264 47 97

Research interests:
Strongly coupled cardiac electromechanics
Multiphysics modelling
Finite elements
Nonlinear & linear solvers
Parallel computing

×
Sander Arens

Sander Arens

PhD Student

Sasha Panfilov

Professor
Alexander.Panfilov@UGent.be

office: 120 007A

Tel: +32 - (0)9 - 264 49 64

Fax: +32 - (0)9 - 264 49 89

 

Research interests:

My research interests are theoretical studies of various aspects of wave propagation in an excitable media. Such waves are a wide spread phenomenon in biology: action potentials in nerve and in muscular tissue are crucial for normal functioning of most of biological spices. Abnormal regimes of wave propagation (spiral waves) can cause serious problems and diseases, such, for example, as ventricular fibrillation, which is the dominant immediate cause of death in the industrialised world.

×
Sasha Panfilov

Sasha Panfilov

Professor

Tim De Coster

PhD student
timj.decoster@ugent.be

office:  120 034

Tel: +32 - (0)9 - 264 48 00

 

Research:

Atrial modelling

Arrhythmogenicity of fibro-fatty infiltrations

The onset of cardiac arrhythmias depends on electrophysiological and structural properties of cardiac tissue. One of the most important changes leading to arrhythmias is characterised by the presence of a large number of non-excitable cells in the heart, of which the most well-known example is fibrosis. Recently, adipose tissue was put forward as another similar factor contributing to cardiac arrhythmias. Adipocytes infiltrate into cardiac tissue and produce in-excitable obstacles that interfere with myocardial conduction. However, adipose infiltrates have a different spatial texture than fibrosis. Over the course of time, adipose tissue also remodels into fibrotic tissue.

I investigate the arrhythmogenic mechanisms resulting from the presence of adipose tissue in the heart using computer modelling. I study how the size and percentage of adipose infiltrates affects basic properties of wave propagation and the onset of arrhythmias under high frequency pacing in a 2D model for cardiac tissue. Although presence of adipose infiltrates can result in the onset of cardiac arrhythmias, its impact is less than that of fibrosis.

afbeelding-1-tim-de-coster

A-A comparison between simulated geometries and clinically observed ones. For clarity, a similar color scheme was used. Adipose tissue is the lightest color, myocytes the middle shade and fibrotic tissue has been given the darkest colour. In the simulation data (2nd and 3d panel), there was 40% adipose tissue and 9% fibrotic tissue for an adipose radius of 400 μm. The histological cut is from sheep data (Haemers 2015) and has a scale bar which denotes 500 μm. One infiltrate consisting of individual adipocytes can be seen.

B To induce arrhythmias we used a pacing protocol consisting of 10 pulses at a period of 240 ms. If 1.68 s after the burst-pacing, activity was  present in the preparation, the simulation was classified as causing arrhythmia onset, which is the case in the 3d panel. Higher voltages are denoted by a white colour, while lower voltages are shown as black.

Anatomically correct human atria

Abnormal excitation of the heart results in cardiac arrhythmias, which is a major problem in cardiac electrophysiology. Sudden cardiac death  due to ventricular fibrillation remains the largest cause of death in the industrialized world. Another cardiac arrhythmia, atrial fibrillation (AF) affects approximately 1.5% of the population. AF is a progressive disease becoming more prevalent with increasing age and results in increased mortality, morbidity and impaired quality of life. Therefore understanding the mechanisms and processes leading to the onset of cardiac arrhythmias is of great interest.

I have a particular interest in realistic AF-modelling. To achieve this goal, a code was developed to simulate anatomically correct 3D human atria with anisotropy and heterogeneity. An example of a spiral wave, generated with an S1S2-protocol, can be seen in the figure below.

afbeelding-2-tim-de-coster

Realistic human atrial model containing a spiral wave in the right atrium. The spiral was made visible in two different views of the atrium. The left one shows the atrium as it would be seen if you were standing on the right side of a person, while the right one shows the atrium as it would be seen from in front of a person. Both images show a different moment in time of the spiral, where the pictures were chosen to make the dynamics and geometry as clear as possible for each case.

×
Tim De Coster

Tim De Coster

PhD student

Nina Kudryashova

PhD student
nina.kudryashova@ugent.be

office: 120 034

Tel: +32 - (0)9 - 264 48 00

 

Research:

Virtual cardiac monolayers for electrical wave propagation

The complex structure of cardiac tissue is considered to be one of the main determinants of an arrhythmogenic substrate. This project is aimed at developing the first mathematical model to describe the formation of cardiactissue, using a joint in silico-- in vitro approach. First, we performed experiments under various conditions to carefully characterise the morphology of cardiac tissue in a culture of neonatal rat ventricular cells. We considered two cell types, namely, cardiomyocytes and fibroblasts. Next, we proposed a mathematical model, based on the Glazier-Graner-Hogeweg model, which is widely used in tissue growth studies. The resultant tissue morphology was coupled to the detailed electrophysiological Korhonen-Majumder model for neonatal rat ventricular cardiomyocytes, in order to study wave propagation. The simulated waves had the same anisotropy ratio and wavefront complexity as those in the experiment. Thus, we conclude that our approach allows us to reproduce the morphological and physiological properties of cardiac tissue.

afbeelding-profiel-nina

Experimental cases considered in our study. The first column shows isolated cells seeded at a low density to avoid cell--cell interaction. The second column represents cells in the monolayer. The substrate was isotropic in the upper row, whereas in the second row, nanofibres were added. Cardiomyocytes are shown in yellow tints, whereas fibroblasts are shown in blue tints. This image was based on immunohistochemical data but was refined for illustrative purpose.

×
Nina Kudryashova

Nina Kudryashova

PhD student

Enid Van Nieuwenhuyze

assistant
enid.vannieuwenhuyse@ugent.be

office:  120 008

Tel: +32 - (0)9 - 264 47 97

 

Research:

Mathematical implementation and visualization of DineHeart.

Study of EADs in the human ventricles in 3D

The study of EADs, which are defined as the early depolarization of the cell before completion of the repolarization phase. This single cell feature is then studied on a 3D model of the human ventricles. In this model, we found that EADs result in complex pattern formation. Mainly three different patterns were found and were classified as the B, A and O excitation type. We studied these patterns in a broad range of parameter values for which EADs are more susceptible in the single cell. Each type has its own characteristics, and patterns are devised based on this features.

Illustration of the EAD induced patterns called the B, A and O excitation patterns.

afbeelding-voor-profiel-enid

Mathematical model of the human Purkinje fibers

Main goal of this study is to create, test and use a mathematical description of the complex dynamics in the Purkinje fibers of the human heart. Current most advanced Purkinje mathematical model, the Stewart model, is an adaptation of the Ten Tusscher - Panfilov 2006 mathematical model for human ventricular cells. Starting from this world wide used model, we will create a new Purkinje model which is compatible with exclusive human experimental data on non-failing Purkinje fibers.

The mechanisms behind Torsade de Pointes

Torsade de Pointes is a well discussed topic in our research field. It is still unclear whether TdP is a result of a meandering spiral or if it is caused by a chaotic interplay of ectopic foci in the heart. We therefore use multiple initial conditions and simulate in both 2D and 3D. We connect the seen deviations of the patterns and complex shapes to the realistic 9-leads ECG, a intensive used tool to detect cardiac arrhythmias in patients. Understanding the dynamics of TdP can lead to an improvement of current used treatments.

Determine the critical size to sustain fibrillation

We collaborate with an experimental group in the US on this topic. We try to understand the relation between fibrillation and its critical size, thus the minimal size it needs in order to sustain its patterns. By decrease of tissue size, it gets increasingly harder to induce an excitation pattern. Depending on the concentration of Pinacidil (modeled by a FATP gating variable in the TP06 model) patterns differ and tissue-sizes can be changed. The dependency of Pinacidil, corresponding changes in size and induced patterns will be linked during this study.

 

 

×
Enid Van Nieuwenhuyze

Enid Van Nieuwenhuyze

assistant