GATE simulation of an FDG brain source in a 23 cm water cube on a Siemens Biograph mMR, with normalisation weights

A Monte Carlo simulation of a PET scan on the Siemens Biograph mMR, made with GATE by Nicolas Karakatsanis in October 2024. The activity is the [18F]FDG brain phantom of Belzunce (Zenodo 8045458), placed in a 23 cm cube of water. The scan is 54 runs of 10 s (540 s): 116.9 million coincidences and 14.4 million delayed coincidences. PyTomography 4's PET tutorials read it. mMR_voxBrain_withSimplePhantom_1.root to _54.root (about 242 MB each): GATE's ROOT output, one file per 10 s run, with the trees Coincidences (prompts, with GATE's truth for each photon) and delay (delayed coincidences). normalization_weights.npy: one float32 weight for each of the 411,027,456 crystal pairs (torch.combinations order over PyTomography's GATE crystal numbers), computed with PyTomography from a separate simulated calibration scan (a cylinder of radius 318 mm, not included), randoms left out. umap_mMR_brainSimplePhantom.v with its Interfile headers (.hv, .ahv): the attenuation map at 511 keV, in 1/cm, 344 x 344 x 127 voxels of 1.67475 x 1.67475 x 2.0315 mm. mMR_Geometry.mac: the scanner's GATE macro (28,672 LSO crystals of 4 x 4 x 20 mm in 64 rings). gate_braininwater_rois.seg.nrrd: the grey and white matter, as a 3D Slicer segmentation. README.md gives the coordinates (GATE's x, y, z are the patient's left, posterior, inferior, about the centre of the phantom's MR) and how each file was made. The phantom's MR and images are in Belzunce's record. Licence: CC BY 4.0. Please cite this record and the brain phantom: Belzunce MA. High-Resolution Heterogeneous Digital PET [18F]FDG Brain Phantom based on the BigBrain Atlas. Zenodo (2018). doi:10.5281/zenodo.8045458.