Login / Signup

Design and performance simulation studies of a breast PET insert integrable into a clinical whole-body PET/MRI scanner.

Christian M PommranzFabian P SchmidtJulia G MannheimSebastian J DieboldChris TenzerA SantangeloBernd J Pichler
Published in: Physics in medicine and biology (2023)
Objective . Three different breast positron emission tomography (PET) insert geometries are proposed for integration into an existing magnetic resonance imaging (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 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 × 10 6 cps (Geometry A), 15.3 × 10 6 cps (Geometry B) and 14.8 × 10 6 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.
Keyphrases
  • positron emission tomography
  • computed tomography
  • magnetic resonance imaging
  • pet ct
  • contrast enhanced
  • pet imaging
  • monte carlo
  • high resolution
  • mass spectrometry
  • atomic force microscopy