Code and Data


We believe in the concept of open and reproducible research. The codes of our projects are collected in our Github repositories:
Github MIDAS.lab
Github MIDAS.lab/TUM


To promote research on machine learning-based automated tumor lesion segmentation on whole-body FDG-PET/CT data we host the autoPET challenge and provide a large, publicly available training data set: autoPET


A collection of codes and documentations of previous projects can also be found here: k-space astronauts