This paper reviews the technical ecosystem of Gate, an open-source Monte Carlo simulator for medical physics. Based on the shoulders of Geant4, the principal modules (geometry, physics, scorers) are described together with brief descriptions of some key concepts (Volume, Actors, Digitizer). The main source code repositories are described together with the automated compilation and tests processes (Continuous Integration). We then described how the OpenGATE collaboration managed the collaborative development of about one hundred developers during almost 20 years. The impact of Gate on medical physics and cancer research is then summarized. Finally, future development perspectives are indicated.

Objective: Three different breast PET insert geometries are proposed for integration into an existing MRI breast coil (Breast Biopsy Coil, NORAS MRI products) to be used inside a whole-body PET/MRI scanner (Biograph mMR, Siemens Healthineers) to enhance the sensitivity and spatial resolution of imaging inside the breast. Approach: Monte Carlo simulations were performed to predict and compare the performance characteristics of the three geometries in terms of the sensitivity, spatial resolution, scatter fraction, and noise equivalent count rate (NECR). In addition, the background single count rate due to organ uptake in a clinical scan scenario was predicted using a realistic anthropomorphic phantom. Main results: In the center of the field of view (cFOV), absolute sensitivities of 3.1 %, 2.7 %, and 2.2 % were found for Geometry A (detectors arranged in two cylinders), Geometry B (detectors arranged in two partial cylinders), and Geometry C (detectors arranged in two half cylinders combined with two plates), respectively. The full width at half maximum (FWHM) spatial resolution was determined to be 1.7 mm (Geometry A), 1.8 mm (Geometry B) and 2.0 mm (Geometry C) at 5 mm from the cFOV. Designs with multiple scintillation-crystal layers capable of determining the depth of interaction (DOI) strongly improved the spatial resolution at larger distances from the transaxial cFOV. The system scatter fractions were 33.1 % (Geometries A and B) and 32.3 % (Geometry C). The peak NECRs occurred at source activities of 300 MBq (Geometry A), 310 MBq (Geometry B) and 340 MBq (Geometry C). The background single-event count rates were 17.1×106 cps (Geometry A), 15.3×106 cps (Geometry B) and 14.8×106 cps (Geometry C). Geometry A in the three-layer DOI variant exhibited the best PET performance characteristics but could be challenging to manufacture. Geometry C had the lowest impact on the spatial resolution and the lowest sensitivity among the investigated geometries. Significance: Geometry B in the two-layer DOI variant represented an effective compromise between the PET performance and manufacturing difficulty and was found to be a promising candidate for the future breast PET insert.

No abstract