Digital Twin of cellular blood

Problem

Cellular blood flow models often struggle to combine physical fidelity with large-domain performance. The goal of HemoCell is to simulate blood on a cell-resolved level without sacrificing scalability needed for practical biomedical studies.

Approach

• Build on a high-performance lattice Boltzmann framework for fluid-cell coupling.
• Implement robust high-shear stability and efficient communication strategies.
• Use dynamic load balancing and optimized data structures for large core counts.
• Provide advanced boundary conditions for realistic physiological setups.

Key finding

HemoCell demonstrates strong scalability (including very large core counts) while preserving detailed blood-cell dynamics. This makes it viable for both fundamental studies and translational simulation tasks.

Why it matters

A performant cellular blood digital twin expands what can be studied in silico, from platelet mechanics to patient-relevant microscale transport phenomena.

Outputs

• Project website: hemocell.eu
• Documentation: HemoCell user guide
• Book chapter manuscript: arXiv:2305.02752
• Publications are listed in the References section below.