Although blood-contacting medical devices (BCMDs) have become lifesaving alternatives to organ transplantation for many patients, thrombotic and bleeding events remain the most common postoperative complications that tend to be devastating and often life-threatening. Device-induced hemostatic dysfunction is commonly believed to be the culprit of these complications and is directly related to platelet activation and receptor shedding associated dysfunction caused by the non-physiological shear stress (NPSS) in these devices. Over the years, computational fluid dynamics (CFD)-aided simulation and analysis have been widely adopted to achieve significant research efforts on device-induced platelet dysfunction. With the help of the CFD technique, regions of abnormal NPSS and stagnant flow in blood flow paths can be precisely identified to assess further the potential risk of thrombosis and bleeding in devices. However, the existing CFD models for predicting shear-induced platelet activation and receptor shedding and related hemostatic complications (thrombotic and bleeding), mainly based on the empirical models, have limited success from the device design perspective. This proposal aims to develop a novel platelet activation model based on the art-of-state interpretation of the platelet’s fundamental morphological change upon activation. It will be incorporated in the development of CFD models capable of assessing in-vitro and in-vivo device-induced platelet dysfunction and associated adhesion capacities to substrates. Numerical algorithms and implementation schemes will be developed to link shear- induced platelet damage models to CFD variables to predict device-induced platelet dysfunction. Quantitative adhesion capacities of device-damaged platelets to collagen, vWF, and fibrinogen could be used to represent device-associated potentials for thrombosis and bleeding. Experiments will then be performed to validate the CFD-based predictive modeling. Finally, a sheep study will be performed to test the CFD models for in-vivo predictive modeling. CFD models will incorporate sheep-specific blood properties, device geometry, device operating conditions, platelet consumption, and generation models to predict the in-vivo device-induced platelet dysfunction and altered adhesion capacities of sheep on VADs and ECMO support. The successful completion of this project will result in CFD-based predictive tools. These tools can be used to provide quantitative evaluation of the performance and in-vivo biocompatibilities of BCMDs and aid the development and optimization of new BCMDs with improved functional characteristics and biocompatibility.