Seminar "Data-integrated computational tracer transport simulations in cortical brain tissue"

Tracer experiments are used to assess transport of solutes and water in brain tissue (e.g. in the context of investigating the “glymphatic" hypothesis). Only sparse measurement data is available about the mechanisms and major pathways of water and some solutes entering (infiltration, perfusion) and leaving (clearance) the functional brain tissue. Moreover, there are at least three main issues with current experimental data when it comes to quantification of transport:

(1)  the underlying anatomy and transport pathways are complex (e.g. microvascular networks) but are believed to be crucial in facilitating transport,
(2)    the measured tracer transport is often only a proxy of what we are interested in (e.g. water transport),
(3)    the tracer transport can often not be assessed directly but is to be inferred from a proxy signal (e.g. NMR signal or fluorescence microscope image).

In this talk – in an attempt to bridge scales, fill in missing data, and connect various data sources – we present a mathematical model with which we can perform virtual tracer perfusion simulations in brain tissue. We consider the architecture of microvascular networks and a parametrization of the model based on a combination of experimental data and computational estimation. The mathematical model is a two-compartment model (para-vascular, extra-vascular) formulated as a coupled mixed-dimensional system of partial differential equations. We present the advanced numerical methods and software for these embedded transport network systems that allow for simulating domains with thousands of blood vessels (typical domain size: 1mm3).