An administrative Supplement is requested to purchase a high-performance computer system to simulate and generate data for predicting diabetic retinopathy hemodynamics. Pathogenesis of diabetic retinopathy is characterized by the appearance of morphological abnormalities in the retinal capillary vessels. Although such abnormalities are used in the clinical evaluation of the disease severity, the hemodynamic mechanisms underlying their development and progression remain unknown. These morphological abnormalities are highly localized in specific regions of the retinal vascular network, and may correlate with the local variations of the hemodynamic parameters and forces. Diabetic conditions significantly alter the biophysical properties of the blood cells, however the influence of such altered biophysical properties on the retinal hemodynamics and pathogenesis of retinopathy are not known. Existing in vivo imaging techniques have limitations in terms of the hemodynamic measurements in the topologically complex and multi- plexus retinal vasculature. Additionally, tissue hypoxia and the loss of blood flow autoregulation are pathogenic factors in retinopathy. No study exists that correlates diabetes-mediated altered biophysics of the individual blood cell to the loss of retinal tissue oxygenation and flow regulation. Our underlying hypotheses are: (i) altered biophysics of diabetic red blood cells (RBC) alone can mediate vascular abnormalities by altering the hemodynamic parameters and forces; and (ii) such changes are spatially heterogeneous across the retinal vascular network, and correlate with the focal and heterogeneous nature of vascular abnormalities. The broad objective of this project is to understand the relationship between the hemodynamics of diabetic blood cells, retinal vascular network topology, and pathogenesis of retinopathy, using a high-fidelity, predictive computational modeling study. Specific aims are: 1) To develop a multiscale computational model of the diabetic retinopathy hemodynamics taking into consideration the precise microstructural and geometric details of the 3D vascular networks as obtained from in vivo images of the human retina, and 3D deformation of every single blood cell with altered biophysical properties representing diabetic conditions. 2) To predict diabetic RBC-mediated alteration in the retinal hemodynamics, and how such changes are correlated to the formation and heterogeneity of microvascular abnormalities and vascular adaptation at different stages of progressive retinopathy. 3) To evaluate the significance of diverse cellular-scale hemodynamic pathways involved. 4) To predict the role of RBC hemodynamics on retinal hypoxia and loss of nitric oxide bioavailability as pathogenic factors in retinopathy. This study is significant and innovative because it will (i) develop the first high-fidelity, predictive computational model that combines the exact 3D geometry of ultra-large-scale and multi-plexus in silico retinal vasculature, and 3D deformation and rheology of every blood cell, (ii) provide a rheology- topology coupling mechanism as a basis of hemodynamics-mediated initiation and progression of vascular abnormalities, (ii) directly model heterotypic individual cell-cell and cell-endothelium interactions, and (iv) couple individual RBC transient deformation with blood and retinal tissue gas transport. Acquisition of the high-performance computer system will enable us to perform the simulations and generate data to accomplish the Specific Aims.

Diabetic retinopathy is the leading cause of blindness in working-age adults with more than 190 million worldwide projected to be affected by 2030. This project will develop a computer model to predict initiation and progression of diabetic retinopathy based on changes in blood flow in retinal vessels that are caused by changes in the properties of blood cells and vessel structure due to their prolonged exposure to diabetic conditions. The detailed and accurate prediction from this model may provide novel insights on blood flow-mediated pathogenesis of retinopathy, and reliable metrics for early disease detection, thereby directly benefiting clinical diagnosis and treatment. An Administrative Supplement is requested to purchase a high-performance computing system to perform the modeling study.