Final Report Summary - S-SARCOMERE (Sarcomeric-Simulation AppRoach in a COmputational Multiscale EnviRonmEnt) Cardiovascular diseases (mainly coronary heart disease and stroke) are the main cause of illness and premature death in the European Union and they account for approximately 40% of deaths. They affect the heart and surrounding blood vessels and can take many forms, such as high blood pressure, coronary artery disease, valvular heart disease, stroke and rheumatic heart disease, etc. Complex interactions between the heart, vasculature and the systemic response to the changing physiological environment make it inappropriate to consider components of the cardiovascular system in isolation. Modeling and simulation enabling the cardiovascular system and its interactions to be investigated in silico bring new insight to the study of these issues. In the past, in-vivo experimentation with animal models was a gold standard to investigate these interactions but that poses a particular problem nowadays as there are ethical issues involved. Additionally, there is great variability in in-vivo experimentation due to the use of different species, techniques, and lab conditions, thus extracting and extrapolating clinical data from these experiments for the human case under conditions of interest, is very complicated. The use of simulation techniques may properly address these issues by focusing on the human case, accelerating advances in biomedical science, particularly at the interfaces between biology, chemistry, physics and engineering. One of the 'success stories' in bioengineering is the study of the fluid dynamics of blood (haemodynamics) within the cardiovascular system and the relationship between haemodynamics and the development of cardiovascular disease. Cardiovascular models require a multidisciplinary vision, presenting a particular challenge in that they require both a multi-physics and multi-scale (both, in length and time) approach. The complexity of the system, the extent of the domain, different time and physical scales and the computational resources needed, amongst others, make the use of finite elements for the simulation of the system prohibitive thus, a compromise is needed. The most sophisticated fluid-solid interaction structures provide exquisite detail in the fluid domain, but are limited in that the boundary conditions are prescribed. As the facility to model the interaction of solid and fluid mechanics in the cardiovascular system has developed, the need for improved and interactive boundary conditions has arisen. To date these have focused exclusively on the coupling of local three dimensional models (computationally expensive to simulate) providing significant detail on local haemodynamics with electrical analogue models of the systemic circulation (peripheral conditions). An alternative solution is to couple lumped parameter models of the boundary conditions with a finite element model of the part where detail and accuracy are needed. Significant improvement can be made in terms of the understanding of the underlying physics if the lumped parameter approach includes more physiologically representative mechanisms rather than traditional 'black-box' models. Usually, boundary conditions are expressed in terms of pressure and flow. These are the macroscopic expression of physiological or pathological conditions. These macroscopic variables can be related to the microscopic level of the physiology/pathology through molecular/cellular aspects of the process. The task of developing and integrating these models is challenging. Special computational tools must be developed in order to provide compatibility between models and to validate the results. Where clinical input data is not available, boundary conditions must be defined. One vision is that, these could be provided in terms of 'physiologically oriented' models of lesser complexity (in terms of the computational resources needed) but they must contain sufficient complexity to be able to represent the interaction of the system and its environment. A very attractive alternative was presented by this research. A multi-scale, hierarchical lumped-parameter model of the left ventricle has been created using the has been coupled to a 3D detailed model of a mitral valve (mechanical). A more complex model is required to translate biochemical information into physiological changes, i.e to link pathology with changes in preload, afterload, contractility, etc. Such a model could be applied to study the effects of pharmacological intervention or compensatory changes in the cellular environment of the myocardium developed in response to conditions such as arrhythmias, atrial fibrillation or ventricular hypertrophy. This may have important clinical implications, as it is important to understand the impact of ventricular behaviour on the performance of cardiovascular devices. For mechanical heart valves, cavitation might be a problem, as could lead to catasthropic failure or blood damage. The closure event lasts for only about thirty five milliseconds, and there are no adequate models that can predict or quantify the effect of local/cellular events of chemical, biochemical or mechano-electrical nature at the ventricular level on the closure forces. Patient specific events related to ventricular activity might also be important to consider. For potentially high-risk patients is possible that pharmacological intervention may reduce the cavitation potential of the valve and risk of fracture. In this research we have: - Developed a coherent multi-scale model of the system taking into account the cellular mechanisms of cardiac contraction, with feedback between electrochemical events in the vessel wall and detail haemodynamic characteristics of the flow. -Coupled this model with a 3D model of a cardiac valve -Produced a primer setup to study of the coupling of these models with left ventricular assist devices (LVADs) in order to assess the effect of arrhythmias on the behavior of these devices.