Evaluation of SRW-OSEM Using Clinical Data
Bibliographic record
Abstract
We describe evaluations of the SRW-OSEM algorithm using clinical patient data. SRW-OSEM is an iterative reconstruction method which incorporates scatter and randoms corrections within the weighting component of the system matrix analogous to the attenuation weighted reconstruction algorithm. Our previous results obtained from small animal phantom data showed that SRW-OSEM can accelerate the reconstruction task and reduce the storage cost as well as improving the image quality. In this work, further evaluations were conducted using clinical patient data whose scatter fraction is much higher than the phantom data previously used. As a result, the trues fraction was ~30%-50% for the patient data as compared to ~80% for the small animal phantom data. Convergence rate in contrast recovery and image profiles were compared between the SRWOSEM and OP-OSEM. Higher improvement in convergence rate with respect to OP-OSEM was observed from SRW-OSEM for the patient data as compared to the improvement observed previously from the small animal phantom data. As expected, the higher the background contamination (e.g. scatter and randoms fractions) the higher the improvement in convergence rate achieved by SRW-OSEM with respect to OP-OSEM. In particular, 3-4 times faster convergence rate with respect to OPOSEM was achieved by SRW-OSEM in this case; e.g. image reconstructed with 3 iterations of SRW-OSEM contains similar contrast as compared to that reconstructed with 12 iterations of OP-OSEM. Furthermore, lower noise was observed from the SRW-OSEM image as compared to the OP-OSEM image due to the lower number of iterations used in the reconstruction when the noise in the trues fraction estimate was low.
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 | 0.012 | 0.027 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".