New Notebook Tutorials for Personalized Simulations using our Body Segmentation AI
-
I would like to highlight our new jupyter notebooks released in Sim4Life 9.0 (18820) demonstrating how to setup simulations for personalized modeling, starting from medical image data (MRI or CT). You can find them under Help -> Examples, tutorials H, J and K.
These new tutorials show how to
- import medical images
- run automatic segmentation of the trunc
- extract surface models from the labelfields
- assign material tags to simplify automatic assignment of material properties in the simulations
- extracting anatomical landmarks on the skin for placing electrodes
- setting up a simulation and running it
We would love to get constructive feedback and suggestions. In our experiments the segmentation seems quite robust, but errors may occur (this is not a clinical tool). We try to fix these errors in the post-processing, but there are most certainly cases where the post-processing may do something unintended (like add skin at the border of the field of view, where there should be none). If you have a need for making this robust for your data, contact us at virtualpopulation@itis.swiss so we can discuss.