The development and validation of patient-specific three-level model of the vasculature, integrating artery anatomy, blood flow and particle dynamics that describe the formation and growth of atherosclerotic plaque.
Though, numerous 3D anatomical arterial models coupled with computational fluid dynamics for bloodstream evaluation exist, none of them describes the biological processes involved in the development of plaque formation which are triggered by low or oscillatory shear stresses. Our aim is to evaluate the patient vulnerability and the vessel vulnerability using patient specific risk prediction model where patient risk score, systemic inflammation (monocyte activation, circulating levels of Adhesion molecules, PCR), artery geometrical and plaque features and flow/mass transport analysis will be sequentially included in a logistic regression analysis for overall patient risk prediction. Moreover, we will combine anatomical information with blood flow and particle dynamics of the regions of interest, in order to provide a model for the simulation of the plaque formation and growth according to patient’s characteristics (e.g. bifurcations, LDL concentration, and pharmacological treatment). For this purpose the arterial tree will be reconstructed from MRI/CT and IVUS, and computational flow dynamics will provide the patient-specific blood flow analysis. In sites of interest, which are potentially crucial for plaque formation, based on the computed artery wall shear stress, local particle dynamics will be computed based on patient specific blood analysis data. Coupled with the expressive capability of suitable biology programming languages it will provide a patient-specific model for plaque formation and evolution, which can be used to simulate various patient medical status conditions.
In order to provide the integrated model several in vitro and in vivo studies will take place:
- Laboratory & Bench Study, to study the biological factors and the reaction of endothelium cells to different shear stresses blood molecules concentrations.
- Animal Studies for both carotid and coronary stenosis for the collection and analysis of clinical data for the development of an integrated animal model
- Patient Studies for the parameterization of the generic model to human specific data as well as acquiring the desired population for developing the patient specific risk prediction model.
The development and testing of Treatment Decision Support Tool for cardiologists’ assistance in selecting appropriate patient treatment.
The aim is to support cardiologists in clinical practice by assisting them to select appropriate pharmacological treatment or invasive interventions based on the patient profile, predicted values from patient-specific models, medical knowledge and patient classification. Two types of expert systems will be developed supporting the carotid and the coronary cases. Apart from validated knowledge supported by clinical guidelines, both expert systems will exploit partly-validated medical knowledge as well as un-validated knowledge created within ARTreat. The latter feature of the Treatment Decision Support System indicates the system’s capability of exploiting previously recorded patient information, in order to produce new patient data associations that indicate potential risks. The treatment decision support system will use information from the patient-specific three-level model (objective 1) in order to enhance the diagnostic and treatment capabilities and recognize and characterize patient vulnerability. This tool will allow the cardiologists to explore new knowledge and potentially integrate it within the system since this knowledge will lack validation by being supported by a considerable number of clinical trials. The treatment decision support will be evaluated with respect to functional testing and user acceptance by installing the coronary application in UCAM and PRM and the carotid application in CNR.
The development and testing of Interventional Decision Support Tool for cardiologists’ assistance in stent positioning, in clinical interventions.
The Interventional Decision Support System (IDSS) assists cardiologists in planning the stenting intervention procedure for the treatment of a patient’s coronary arterial disease at the artery branch level. The cardiologist’s decision is supported by an application that simulates the patient’s revascularization based on the positioning and subsequent deployment of an appropriate stent placed in the stenosed artery of the patient. The artery is represented as a 3-D model derived from CT images and includes a representation of the outer wall, the lumen and regions of soft and hard plaque types. The end-user will be able to simulate various interventional scenarios, applying different stent types and positions, obtaining for each one of them an estimation of the resulting stented artery geometry, a blood flow simulation and an animation showing mechanical stresses in both artery and stent during balloon inflation. Through an appropriately designed graphic interface, the end user will compare the simulated scenarios and use this information on top of his clinical judgment in order to determine the intervention parameters sufficiently, according to patient specific characteristics.
The consideration of the real outer wall as well as the spatial distribution of plaque types in the arterial model will lead to more realistic simulations of the stent deployment. The testing of the interventional decision support will be performed by interventional cardiologists in order to test the Interventional Decision Support System for accuracy of results, user friendliness and usability.
The development and testing of a virtual-training environment for stent-positioning for interventional cardiologists.
The training environment will exploit the three-level patient model and produce virtual patient cases describing usual medical cases in different levels of difficulty. It will provide also a virtual-interface close to the real-conditions, to allow for virtual-stent positioning. It will be mainly used for educational and skill development processes.
The testing of the virtual training environment will be performed by interventional cardiologists.
Concluding, the overall aim is the improved prediction of plaques at risk for infarction or stroke through the combination of previously disjoint markers and diagnostics on separate biological levels. The improved model will be used to provide accurate decision support applications for the clinical cardiologist and an advanced training environment for the development of interventional skills.