Generating fibers for patient specific geometries

In this demo we will show how to generate fiber orientations from a patient specific geometry. We will use a mesh of an LV that is constructed using gmsh (https://gmsh.info), see finsberg/ldrb

It is important that the mesh contains physical surfaces of the endocardium (lv and rv if present), the base and the epicardium. You can find an example of how to generate such a geometry using the python API for gmsh here: finsberg/pulse

First we import the necessary packages. Note that we also import meshio which is used for converted from .msh (gmsh) to .xdmf (FEnICS).

import ldrb

import cardiac_geometries

Convert from gmsh mesh to fenics

mesh, markers, marker_functions = cardiac_geometries.gmsh2dolfin("mesh.msh")

Update the markers which are stored within the mesh

ldrb_markers = {
    "base": markers["BASE"][0],
    "lv": markers["ENDO"][0],
    "epi": markers["EPI"][0],
}

Select a space for the fibers (here linear lagrange element)

fiber_space = "P_2"

Create a dictionary of fiber angles

angles = dict(
    alpha_endo_lv=60,  # Fiber angle on the endocardium
    alpha_epi_lv=-60,  # Fiber angle on the epicardium
    beta_endo_lv=0,  # Sheet angle on the endocardium
    beta_epi_lv=0,
)

Run the LDRB algorithm

fiber, sheet, sheet_normal = ldrb.dolfin_ldrb(
    mesh=mesh,
    fiber_space=fiber_space,
    ffun=marker_functions.ffun,
    markers=ldrb_markers,
    **angles,
)

Save to xdmf with dolfin.XDMFFile(mesh.mpi_comm(), “patient_fiber.xdmf”) as xdmf: xdmf.write(fiber)

Use this function to save fiber with angles as scalars

ldrb.fiber_to_xdmf(fiber, "patient_fiber")