Physics-Informed Deep Neural Network for Inter-Crystal Scattering Recovery in ToF-CT
Bibliographic record
Abstract
Inter-crystal scattering (ICS) fundamentally constrains high-resolution Time-of-Flight Computed Tomography (ToF-CT) by enforcing trade-offs between sensitivity and spatial resolution. Lowering energy thresholds to capture more photons exacerbates spatial ambiguity, while raising thresholds to suppress ICS events reduces detection efficiency, an increasingly critical compromise under stringent dose constraints. As an alternative, this paper present a novel physics-informed deep learning approach that fundamentally resolves this dilemma by exploiting spatiotemporal energy deposition patterns. Our approach employs a two-stage architecture: first, a specialized Convolutional Neural Network (CNN) discriminates between primary and scattered energy deposits; subsequently, a Multi-Layer Perceptron (MLP) precisely reconstructs the originating crystal location for identified scatter events, effectively recovering otherwise lost spatial information. Both network designs are informed by empirical characterization of ICS behavior in finely segmented LYSO detectors. Monte Carlo simulations of LYSO-based ToF-CT detector arrays achieved$>94 \%$accuracy in ICS recovery. By transforming previously rejected scatter events into valuable diagnostic information, our method enables operation at lower energy thresholds (20 keV), yielding a 34 % improvement in absolute detection efficiency while maintaining spatial integrity. Quantitative image reconstruction using a modified ACR phantom revealed a 20 % enhancement in spatial resolution alongside substantial improvements in signal-to-noise ratio (SNR), without compromising contrast fidelity.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".