Me and two of my colleagues recently made public on github a tool to make the analysis of the NEMA NU4 Image Quality phantom easier, see [1].
It consists of an Amide template and a python script. The Amide template includes a CT image of the phantom that can be used to make the placement of the volumes of interest (VOI) relative to the PET image(s) easier. For a sequence of NU4 images, e.g., a different number of iterations for a given acquisition, the placement only needs to be done once. Once the extraction of the VOIs with Amide is done, the provided python script will compute the metrics described in the NEMA NU4 standard.
If you have any questions or suggestions, do not hesitate to contact me or to make an issue on github!
I hope that it help out at least one person out there!
Bests,
Maxime Toussaint