Validation of a Synthetic Cohort of Aortic Stenosis Patients

Image-based modelling for diagnosis and treatment planning for aortic stenosis became increasingly relevant in cardiovascular research. In theory, the method allows non-invasive calculation of diagnostic parameters. Furthermore, prediction of hemodynamic outcome after different treatment strategies is feasible. This approach might help to identify optimal treatment strategies for a patient as well as support development of novel implantable devices. A relevant problem for translation into clinical or industrial application is the lack of available data sets due to data privacy regulations. A promising approach to mitigate this problem is the generation of synthetic data. This type of data can be shared freely, supporting reproducibility studies and comparison of different in-silico approaches. However, synthetic data must be validated by demonstrating, that the data matches the cohort it is intended to mimic. In this study, we generated synthetic data of aortic stenosis patients. The data set includes general demographics, functional parameters, and geometries of the aorta and the aortic valve. Peak-systolic hemodynamics of the real patients as well as the synthetic data set was calculated and compared against each other.This work has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 101017578 (SIMCor - In-Silico testing and validation of Cardiovascular IMplantable devices)