Translational biomedicine studies depend on the integration of multiple datasets that, together, represent the complex plethora of features from patients transiting between health and disease states. Cardiovascular research projects create a vast amount of data that could be used to better understand the development of cardiovascular diseases (CVD), such as: heart congenital conditions, heart valves diseases, heart attack and arrhythmias.
Data contained in electronic health records (EHR) from subjects either suffering of CVD or participating on research cohorts (as health donors) could be integrated to their OMICS information (genome variants and/or gene expression, if known) and also to their imaging derived biomarkers. Two major imaging techniques are currently applied to gather information about patient’s heart conditions: cardiac computer tomography (CT) and cardiac magnetic resonance imaging (MRI). These techniques deliver high-quality cardiovascular data in the format of images such as: heart/artery angiograms and detailed heart scans, respectively.
Angiograms provide the morphology of the coronary anatomy in relationship to the interventricular and atrioventricular valve planes, depicting the different coronary branches lengths, whereas MRI scans provide multi-plan DICOM images of the heart that can be used for the segmentation and estimation of the heart chambers volumes, as well as the thickness of the heart wall. Such features display alterations when patients transit from healthy to diseased states.
We aim to use the CORBEL services to identify, reuse, store and analyze available EU CVD cohorts in order to develop tools based on machine learning algorithms to stratify subjects based on their EHR, OMICS and CT/MRI data.