Simulation Studies of the SAVANT High Resolution Dedicated Brain PET Scanner Using Individually Coupled APD Detectors and DOI Encoding
Bibliographic record
Abstract
531 Objectives: The SAVANT (Scanner Approaching in Vivo Autoradiographic Neuro Tomography) is a dedicated human brain PET scanner designed to achieve ultra-high spatial resolution using fully pixelated APD-based detectors with depth-of-interaction (DOI) encoding. The objectives of this work are (a) to predict the performance of the scanner by simulations following the NEMA NU4-2008 and NEMA NU2-2001 standards, and (b) to investigate its capability for imaging the human brain using a mini hot-spot phantom and a 3-D voxelized brain phantom. Methods: The SAVANT brain scanner is based on 4032 front-end detector arrays forming a 39-cm diameter by 23.5-cm axial length cylinder with 144 rings of 896 pixel detectors, defining a 26-cm diameter FOV. The basic detector element consists of a 4 x 8 dual-layer phoswich array of 1.12 x 1.12 x 12 mm3 pixels made of Lu1.8Gd0.2SiO5:Ce (LGSO) and Lu1.9Y0.1SiO5:Ce (LYSO) scintillators. Each 4 x 8 crystal array is read out by a 4 x 8 pixelated monolithic APD array at a 1.2 mm pitch, ensuring one-to-one coupling between individual scintillator and photodetector pixels. The highly integrated electronic front-end, based on a multiple-threshold time-over-threshold method, enables the signal from individual pixel detectors to be processed and recorded independently with fully parallel signal readout and processing, including DOI encoding. Simulation data were generated for various crystal lengths and DOI encoding accuracy to investigate the effect on scanner performance and image quality. The NEMA procedures were used to simulate the spatial resolution, the sensitivity, the scatter fraction and the count rate performance of the scanner, while image quality was evaluated with phantoms. Simulations were performed using Geant4 Application for Tomographic Emission (GATE) and images were reconstructed using the Customizable and Advanced Software for Tomographic Reconstruction (CASToR). Results: For a phoswich crystal length of 4.6 + 7.4 mm ensuring uniform detection efficiency and assuming perfect DOI encoding of the two detector layers, a reconstructed spatial resolution of less than 1.3 (2.1) mm FWHM is obtained at 1 (10) cm from the center of the FOV. A spatial resolution of less than 2 mm FWHM is predicted over 75% of the FOV, enabling both cortical and subcortical structures of the brain to be imaged with unprecedented accuracy. With an energy window of 250-650 keV, the absolute sensitivity is estimated at 3.5% and maximum NECR reaches 13 kcps at 12 kBq/cc. The reconstructed image of an ultra-high resolution hot spot phantom illustrates the expected imaging capabilities of the scanner for small structures where 1.0 (1.2) mm objects can be resolved with a high contrast at ~1 (~10) cm from the center. The reconstructed image of a 3-D voxelized human brain phantom shows that the SAVANT scanner will be particularly useful to investigate the small deep structures of the brain, enabling details of the medial temporal lobe known to be involved in the onset of Alzheimer’s disease to be potentially differentiated. Conclusion: A new high resolution PET scanner design featuring small truly pixelated detectors with coarse DOI encoding is proposed to reach spatial resolution in the millimeter range for imaging the human brain. The simulation results provide evidence of the promising capabilities of the scanner for high performance brain imaging applications such as β-amyloid deposition, tau protein accumulation and neuroreceptor distribution. Acknowledgments: Funding from NIH U01EB027003 and MEDTEQ 32128.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".