COMHEART2CLINIC · Translation of Computational Heart Models into the Clinic

Translation of Computational Heart Models into the Clinic

Project context and objectives The long-term goal of COMHEART2CLINIC was to enhance the management of cardiovascular diseases. The main objective fulfilled was the solution of specific translation challenges of the integration of multi-scale computational models of the heart into the clinic. Key to this translation is the development of model personalisation strategies, to which the project has made significant contributions. Project results A new robust and automatic method for the personalisation of computational mechanical meshes of the cardiac anatomy has been developed. The solution works for both healthy and diseased anatomical conditions, and eliminates the tedious manual construction and refinement process that was necessary before. It also has the capability to work with low-resolution medical images, solving an additional unmet need of mesh customisation from sparse datasets. In addition, a new strategy for the personalisation of the material properties of the myocardium has been proposed. Preliminary results show the potential of this tool, not only for personalisation of the models, but also for diagnosing the status of heart failure. Results illustrate how the myocardium is significantly stiffer in heart-failure patients compared to healthy ones, and that during diastole the profile of the decaying active tension is significantly smaller and faster in healthy cases compared to diseased ones. Finally, work has also been done towards the personalisation of the electrophysiological properties of the myocardium. The project has also addressed the need to develop a Clinical User Interface of a Cardiac Resynchronisation Therapy support system. The proposed design has been validated by a group of cardiologists, in coordination with partners of the euHeart consortium (FP7). Preliminary results on the predictive power of models for CRT patient selection have not been as encouraging as expected, as was reported in the update of the project research plan. A validation in-silico study has provided a new important variable in the decision-making process for the selection of the CRT therapy: the role of the auto-regulatory mechanism of the heart. On the training and transfer aspects, it is important to highlight that the researcher has been able to integrate his efforts effectively and efficiently in both his research team in the University of Oxford, (UK) and in the clinic, at the Hospital of St Thomas (UK). The interaction and integration with potential users, cardiologists, and other researchers in related fields (computer science, physiology, medical imaging, etc.) has been particularly fruitful. The researcher has also invested quite a considerable effort in the development of a robust and easy-to-use mesh personalisation tool, which has been made available to the scientific community as a web service. In conclusion, the clinical adoption of computational models of the heart for the better selection and therapy planning of CRT is today much closer to becoming a reality, thanks to the contributions of COMHEART2CLINIC.

Data: CORDIS, © European Union