Here we demostrate how you can constrain a cardiac computational model using clinical measurements such as pressure, volume and regional strain. The problem is formulated as a PDE-constrained optimisation problem where the objective functional represents the misfit between measured and simulated data. There are two phases; passive and active. In the passive phase the material parameters are the control parameters, and in the active phase the contraction parameter is the control parameter. The control parameters can be scalar or spatially resolved. The problem is solved using a gradient based optimization algorithm where the gradient is provided by solving the adjoint system.

The code here is currently undergoing a major refactoring in order for it to be compatible with more recent versions of fenics as well as the standalone cardiac mechanics solver pulse. We expect the master branch to be stable, as development will mainly be done on other branches. The original code, hosted at Bitbucket is no longer maintained.

pulse-adjoint is written in pure python but based on fenics and dolfin-adjoint (using libabjoint). Therefore fenics version 2017 and dolfin-adjoint that uses libadjoint (not pyadjoint) is currently the only supported versions. You also need to install pulse.

As pulse-adjoint is a pure python package you can install it by simply typing

python setup.py install

when you are in the same folder as the setup.py.

Documentation is found at pulse-adjoint.readthedocs.io

In order to get starting with pulse-adjoint it is beneficial to get to know FEniCS as well as the pulse, which has a lot of demos

Next you can check out the demoes in this repository. They should give an idea on how you can incorporate your own data into the pipeline. If you need to generate microstructure you can checkout out the ldrb

You are welcomed to use this code for your own reaseach, but encourage you to cite the following paper:

Finsberg, H., Xi, C., Tan, J.L., Zhong, L., Genet, M., Sundnes, J., Lee, L.C. and Wall, S.T., 2018. Efficient estimation of personalized biventricular mechanical function employing gradient‐based optimization. International journal for numerical methods in biomedical engineering, p.e2982. DOI

GNU LGPL v3

• Henrik Finsberg ([email protected])