Search
  • elenareimalott

The Process

Updated: Oct 26, 2021

The first step was to take MRI scans and put them through a software called FreeSurfer, which uses MRI data to make a composite image of the brain. I used the MRI scan provided, a brain from someone named "Bert".


Surface mesh exported from FreeSurfer.


 

FreeSurfer is a brain imagining software package that was developed by the Laboratory for Computational Neuroimaging at the Athinoula A. Martinos Center for Biomedical Imaging which is affiliated with Harvard Medical School. To learn more about this software, visit the link:

https://surfer.nmr.mgh.harvard.edu/fswiki


I followed this tutorial on YouTube from Perry Radau to learn how to download the application.

https://www.youtube.com/watch?v=P2_sbNFvCek&t=1101s

 

I then took that image and put it through Mesh lab, an app that smoothes down the image and gets rid of any of the imperfections.


First I selected Mesh Layer and choose to "Flatten Visible Layers".


I first choose "Quadratic Edge" and put 150 for the number.


Then I applied some filters to smooth the brain image.


These filters include Laplacian Smooth which can be found under "Smoothing, Fairing, and Deformation."





While doing research to see if anyone has 3D printed a brain before, I found multiple websites that did so- one even allows you to submit your own MRI scans to print your own brain. One website that outlined the process well was on sculpteo.com:

https://www.sculpteo.com/blog/2017/11/08/how-to-3d-print-your-own-brain-using-mri-or-ct-scans-free-software/


This is where I learned about FreeSurfer and Meshlab. All of these filters and its parameters I used, I also got from this website.

 

I then put this digital image through Ultimaker Cura: a slicing software for 3D printing.


Ultimaker Cura converts a 3D model into G-code that desktop 3D printers use to print. The issue is that G-code isn't understood by the robotic 3D printer. I used a software called ropedancer.io developed by AI SpaceFactory which translates that desktop 3D printer code to Rapid code, which the the robotic 3D printer can understand.


The reason the brain was split in half was because there was an issue of support since the brain is not flat at the bottom. I explored many options such as using sand as a support system while printing however, ultimately decided to print the brain in two parts: the top half and the bottom half and printed them separately.


33 views0 comments

Recent Posts

See All