March 4, 2026
Amanda Randles delivered an invited talk at the SIAM Conference on Parallel Processing for Scientific Computing, held March 3–6 in Berlin. The presentation, titled Scaling Vascular Digital Twins: From Millions of Heartbeats to Petabytes of Data, explored how advances in high performance computing are enabling new approaches to modeling cardiovascular physiology.
Digital twin models of the vasculature are increasingly being used to simulate blood flow and assess disease risk. Traditionally, these analyses have focused on short time windows representing only a limited number of heartbeats. Recent advances in scalable simulation and data integration are making it possible to extend these models across far longer time horizons, capturing changes in physiology across thousands to millions of cardiac cycles.
The talk discussed how large scale simulation, clinical imaging, and data from wearable devices can be combined to construct longitudinal vascular digital twins. These models generate large volumes of multimodal data and require new approaches to scalable computing, data management, and analysis.
By connecting high fidelity simulation with continuous physiological data, this work aims to support earlier detection of cardiovascular disease and more proactive approaches to patient care.
The Society for Industrial and Applied Mathematics conference on parallel processing for scientific computing brings together researchers working at the intersection of applied mathematics, algorithms, and high performance computing.