Abstract WP55: Through Thick and Thin: Improved Aspects Grading and Dense Vessel Detection Using Simple Ncct Post-processing
Bibliographic record
Abstract
Introduction: Assessing Alberta Stroke Program Early CT Score (ASPECTS) and identifying hyperdense arteries on non-contrast CT (NCCT) are important components of decision-making in acute stroke. Conventional practice uses 5mm averaged slice thickness NCCT for interpretation of these features. We have systematically evaluated several post processing techniques on NCCT to determine if there is improved reliability in identification of ASPECTS and hyperdense artery. Methodology: We assessed four post-processing techniques on NCCT namely (1) 5mm averaged thickness (2) Minimum Intensity Projection (mIP) - 5mm thickness (3) thin slices (0.625mm) and (4) Maximum Intensity Projection (MIP) - 5mm thickness (Figure 1). Three raters (student, fellow and expert) independently assessed 100 NCCT scans from the PRoveIT database. All scans were read at four different times 10-14 days apart. At each time-point the post processing modality was changed and the patient order randomized. Information on side of suspected infarction was provided. Raters were asked to score ASPECTS and identify presence of hyperdense artery at each reading. Inter-rater reliability was assessed using Intra-cluster correlation (ICC) for ASPECTS and weighted kappa (wKap) for hyperdense artery. Results: The highest inter-rater reliability was found with the MIP technique (ICC 0.42; p<0.001), followed by 5 mm average, mIP and thin slice respectively (ICC 0.33, 0.32, 0.20; all p<0.01). Highest agreement for hyperdense vessel detection was noted with thin slice (wKap 0.30; p<0.001) followed by Average, MIPs and mIPs respectively (wKap 0.25, 0.18, 0.13; all p <0.05). Conclusion: The use of MIP images for ASPECTS grading and thin images for hyperdense vessel detection improves reliability on NCCT. These simple processing steps are easily available on any modern scanner and may help improve patient care.
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 | 0.006 | 0.017 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.008 | 0.002 |
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".