It’s a statistic model for epilepsy. We have GitLab running alongside dedicated NFS server for the data, build Docker images with the code, run them with access to the data, and put quality control images and reports on GitLab CI as artifacts. Each patient gets a branch, triggers CI runner generating patient reports. It took a while to figure it out but it’s really good for a dev team lacking in discipline ;)

That is awesome, thanks for sharing.