Multiscale Predictive Modeling of Blood Cell Damage with Experimental Verification This proposal aims to develop a multiscale model to characterize blood cell damage under complex flow conditions. Blood damage is an important concern for various blood wetting medical devices. In literature, blood damage criterion is typically obtained through empirical fitting of experimental hemolysis data in a specific device, yet little is known about cellular scale process of blood cell damage, which hinders the accurate evaluation of blood damage in a general medical device. The goal of this proposal is to study blood cell damage at molecular and cellular level using combined computational modeling and experimental approaches. Specifically, we will develop a multiscale model that links molecular scale pores formation to cell membrane damage and hemoglobin release. The multiscale computational modeling will be applied for the first time to study of cellular flow over various channel geometries and clinically relevant devices with consideration of both hydrodynamics and membrane damage dynamics. Specifically, we plan to: 1) Develop a multiscale red blood cell membrane damage model. A localized coarse-grained molecular dynamics model at the high stress region will be concurrently linked with a network based cellular membrane model. 2) Couple the cell membrane damage model with local fluid flow through Immersed Boundary method to study cell deformation, pore formation and membrane rupture. Such computational model will be applied to predict blood cell damages in a channels with different geometries and flow conditions. A generalized cellular level blood cell damage model will be developed. 3) Verify the developed multiscale blood cell damage model using AFM measurements, microfluidic tests, and Couette-type blood-shearing devices. A few designed tests will be performed to evaluate cell damage based on hemoglobin analysis of individual cells under controlled stress history and compared to the simulation results. Finally, the developed blood damage model will be applied to study hemolysis in a ventricular assist device. The proposed multi-scale model can directly correlate the microscale state of the cell membrane to local stresses as well as predict cell damage in device with complex geometry and flow condition. Such model could serve as a predictive tool for hematologic biomedical device design and optimization.
The proposed research will result in a standalone, user-friendly, robust, cloud-based computational software for hemolysis evaluation in medical device. Such software is able to process a large amount of simulation data concurrently, provide reproducible hemolysis analysis on existing blood-wetting device, and predict hemolysis on new device designs.
Showing the most recent 10 out of 11 publications
