Research

The CCMC performs interdisciplinary research that includes aspects of clinical and basic cardiology, image analysis, numerical mathematics, computational science, informatics, and applied mathematics. Our main areas of interest are ventricular conduction disturbances in heart-failure patients, cardiac resynchronization therapy, and the prediction of atrial fibrillation.

Clinical perspective

Major diagnostic and therapeutic progress resulted in a significant reduction of the burden of cardiac diseases and subsequent improved outcomes. As a consequence, the leading cause of cardiovascular death has shifted from acute event to heart failure. Heart failure often starts with small disturbances to which the heart muscle tries to compensate by dilating or becoming hypertrophic. While in the short term this adaptive process may not result in any functional problem, it is usually associated in the long run with causing functional loss and increased risk of arrhythmias, due to a maladaptive remodeling at the cellular and sub-cellular levels.

Despite pharmacological therapy may slow down the progression of the pathology, heart failure usually progresses to a stage at which additional and more aggressive therapy is needed. This ideal therapy to heart failure should first be able to stop the maladaptive remodeling or even trigger reverse remodeling, but also avoid the risk of sudden cardiac death secondary to malignant ventricular arrhythmia. While effective reverse remodeling can be triggered by biological therapies such intra-cardiac injection of stem cells or through the cardiac resynchronization therapy, it is well known, albeit not completely understood, that not all patients benefit from these treatments. Therefore, a deep and multifaceted mechanistic understanding of the patient's pathology is mandatory to improve the predictability of the therapy results and to tailor the best therapeutic option. This would require a better characterization of the cardiac tissue underlying heart failure, including both electrical and mechanical properties, as well as their interaction.

A similar mechanistic approach is needed for better treatment of atrial fibrillation, the most common arrhythmias and one of the most common cardiac disease, with an estimated incidence of approximately 10% in the population over 65 years. Despite representing a milder condition than heart failure, atrial fibrillation is one of the major cause of stroke, and then it indirectly contributes to a large number of comorbidities, associated with invalidity, high rehabilitation costs, or even death.

Research at the CCMC

From the clinical point of view, the proposed research of the CCMC can be found at the intersection between the causes and treatments of heart failure and atrial fibrillation. Both heart failure and atrial fibrillation are caused by a wide spectrum of pathologic conditions that, beside their direct effect, induce multiple biochemical, electrophysiological and molecular alterations in the heart tissue as well as detrimental changes in cardiac structure. Diagnosis and treatment of these diseases may therefore greatly benefit from increased understanding of the interactions between the various abnormalities and their representation in the measured data. This understanding can be provided by computational modeling.

From the computational point of view, reserach at CCMC focuses on the development and realization of efficient discretization and solution methods for coupled and time dependent systems of non-linear partial differential equations. This includes electrophysiology, cardiac mechanics, and blood flow as well as the combination of those. Finite Element methods on unstructered and adaptive meshes, massively parallel solution methods as domain decomposition and multigrid, and modern approaches for time discretization such as time parallel methods are employed in this context. We moreover work on the development of new methods for parameter estimation (i.e. in the area of inverse problems) as well as sensitivity analysis.

The field of cardiac computer modeling has undergone an immense development since the first ECG computations in the 1960s and, thanks to the arrival of large-scale multiprocessor computers with dozens of gigabytes of memory in the early 21st century, simulation of cardiac mechanics is nowadays possible with highly realistic anatomic models.Knowledge of the myocardial substrate in heart diseases like heart failure and atrial fibrillation can be significantly enhanced through combination of the various image modalities nowadays available in the clinic. The data obtained here can furthermore be used for a more detailed investigation through computational science. A patient-specific model can furthermore be used as a "virtual patient" to test possible treatment strategies.

During the last few years, within the ICS-CCT collaboration significant advancements have been made in the development of image analysis tools and new numerical methods for the simulation of the partial differential equations describing cardiac electrophysiology and mechanics. Within the CCMC we will focus on the further development of these mathematical models, numerical methods, and simulation software  for better understanding of heart failure and atrial fibrillation.
For the first three years the research activity at CCMC is subdivided into 3 main research tracks including a total of 8 cross-linked, independent research projects as depicted in the next figure.

An overview of the interactions between the research tracks and the projects is schematically shown in the following diagram.

 

Projects I-IV are mainly oriented towards computational science and focus mostly on the development of computational tools, yet in continuous interaction with the clinic. Projects V and VI are mainly directed towards improved direct analysis of clinical data, whereas projects VII and VIII are application-oriented and aim at the translation/use of the computational tools to address clinical questions/challenges. The following provides a brief overview for each project:

• Project I covers the pre-processing step from image to computational geometries focusing on semi-automated image segmentation and 3D anatomical reconstruction.
• Project II focuses on the creation of a numerical framework to perform state-of-the-art large-scale coupled cardiac electromechanical and hemodynamic simulations.
• Project III focuses on the development of (semi-)automated parameter estimation/fitting algorithms and quantification of uncertainty to identify the key model parameters and model personalization steps.
• Project IV covers the development of time-series analysis and pattern recognition methods to aid in the patient characterization, enabling automatic and fast pattern-based screening of large amounts of patient data by linking patient data to disease.
• Project V deals with the development of novel techniques for the quantification of electrical activation and electro-mechanical coupling, using recent developments in electro-mechanical contact mapping.
• Project VI covers ongoing efforts in better risk stratification for atrial fibrillation, relating aspects of P-wave morphology extracted from high-resolution body surface ECGs to atrial anatomical remodeling and occurrence of atrial fibrillation.
• Project VII currently forms the data acquisition backbone of the research at ICS-CCT in the context of the clinical studies on cardiac resynchronization and biological therapies.
• Project VIII is focused at unraveling the cause of the previously reported poor correlations between QRS-based indices and the extent of remodeling.