Model Based Scatter Calculations for A Dedicated Cardiac SPECT Camera
Bibliographic record
Abstract
123 Objectives In CZT-based dedicated cardiac SPECT cameras, energy-window-based scatter estimation is more difficult due to the detection of a large fraction of unscattered photons with reduced energy (“low-energy tail”). Consequently the unscattered photon signal in the scatter window data leads to an increase in noise in the scatter corrected projections. Model-based methods of scatter estimation have less noise and have been shown to be more accurate for cameras with parallel-hole collimators. Thus, model-based approaches may be advantageous for dedicated cardiac systems, but these methods are more complicated in multi-pinhole cameras due to the small field-of-view and distance-dependent variations in sensitivity and magnification. In this study, accuracy of a model based SC method was assessed for a multi-pinhole cardiac SPECT camera using physical phantom studies in comparison to a dual energy window (DEW) SC method. Methods The analytical photon distribution (APD) method was implemented. This method calculates the distribution of probabilities that photons emitted inside the body will scatter in the surrounding scattering medium and be subsequently detected. Scatter calculations were validated by 15 99mTc-SPECT phantom experiments using an anthropomorphic torso phantom with a cardiac insert, in which activity ratios were selected to resemble a clinical scan. Varying levels of photon scatter inside the myocardial compartment was implemented by increasing the activity concentration in the soft tissue compartment of the phantom. The activity inserted into the myocardial compartment of the phantom was first measured using a dose calibrator. SPECT images (140 +/- 14 keV) were acquired on a Discovery NM530c (GE Healthcare) cardiac camera. CT images were acquired on a Infinia-Hawkeye (GE Healthcare) SPECT/CT and co-registered with emission data for AC. MLEM image reconstruction was performed off-line. APD-scatter projections were generated using the reconstructed images and attenuation maps. For comparison, DEW scatter projections (120 +/- 6 keV) were also extracted from the acquired listmode SPECT data. Either APD or DEW scatter projections were subtracted from corresponding 140-keV measured projections and then reconstructed with AC (APD-SC and DEW-SC, respectively). Activity in the heart was recovered using heart masks generated based on a CT based binary template of the myocardial compartment. of the difference in the total cardiac activity from the dose calibrator measurement was compared between APD-SC and DEW-SC images. The difference between modeled and acquired projections was measured as the root mean squared error (RMSE). APD-modeled projections for a clinical cardiac study were also evaluated. Results APD modeled projections showed good agreement with SPECT measurements. While APD-SC reduced mean error in activity measurement compared to DEW-SC in images, T-tests showed the reduction to be statistically significant only where the scatter fraction (SF) was large (mean SF = 28.5%, p = 0.007). APD-SC reduced measurement uncertainties as well however the difference was not found to be statistically significant (F-test p > 0.5). RMSE comparisons showed that elevated levels of scatter did not significantly contribute to a change in RMSE (p > 0.2). Comparison of modeled and acquired projections from a clinical study showed good agreement. Conclusions An APD model-based scatter estimation method produces projections that agree well with data acquired on a dedicated cardiac SPECT scanner with pinhole collimators for both phantom and clinical studies. APD-SC images have lower noise than DEW-SC images and provided a more accurate measure of cardiac activity in high-scatter scenarios.
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 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".