Abstract 50: Co-localization of NCCT hypodensity and CTA spot sign to predict intracerebral hematoma expansion and severity: development and validation of the Black,-&-White sign
Bibliographic record
Abstract
Introduction: Hematoma expansion (HE) occurs in one-fourth of patients with acute intracerebral hematoma (ICH) and is associated with worse outcomes. Existing radiological markers of HE show modest predictive accuracy. We aim to investigate a novel radiological marker that co-localizes findings from non-contrast CT (NCCT) and CT angiography (CTA) to predict HE. Methods: We analyzed 200 consecutive acute ICH patients admitted at Foothills Medical Centre in Calgary, Canada (development cohort) and analyzed 304 patients from the multicenter observational study PREDICT (validation cohort). The Black-&-White (B&W) sign was defined as any visually identified spot sign on CTA co-localized with a hypodensity sign on the corresponding NCCT ( Figure 1 ). The primary outcome was hematoma expansion (≥6mL or ≥33%). Secondary outcomes included absolute (<3mL, 3-6mL, 6-12mL, ≥12mL) and relative (0%, <25%, 25-50%, 50-75%, or >75%) hematoma growth scales. Results: In the development cohort, 22% (n=44) showed the spot sign, 34.5% (n=69) the hypodensity sign, and 7% (n=14) the B&W sign. Those with the B&W sign had higher proportions of HE (100% vs. 19.4%, p<0.001), greater absolute hematoma growth (23.37 mL [IQR=15.41-30.27] vs. 0 mL [IQR=0-2.39], p<0.001) and relative hematoma growth (120% [IQR=49-192] vs. 0% [0-15%], p<0.001) in comparison to those without. The B&W sign yielded a specificity of 100%, a PPV of 100%, and an accuracy of 82%. In the validation cohort, 25% (n=76) showed the spot sign, 39.1% (n=119) the hypodensity sign, and 9.5% (n=29) the B&W sign. Those with B&W signs had higher proportions of HE (79.3% vs. 30.2%, p<0.001), greater absolute hematoma growth (19.1 mL [IQR=6.4-40] vs. 0.8 mL [0-5.6], p<0.001), and relative hematoma growth (92% [IQR=16-151%] vs. 9% [0-34%], p<0.001) in comparison to those without. The B&W sign yielded a specificity of 97%, a PPV of 79%, and an accuracy of 71%. We found a consistent progressive improvement in predicting HE when hypodensity and spot signs are absent, individually present, both present but not co-localized, and co-localized ( Figures 2-3 ). The inter-rater agreement was excellent (k=0.84) in the development cohort, with a uniform imaging protocol of NCCT immediately followed by CTA at a single center, and moderate (k=0.54) in the validation cohort at multiple centers. Conclusion: The Black-&-White sign is a robust predictor of hematoma expansion occurrence and hematoma expansion severity.
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 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 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 | 0.000 | 0.001 |
| 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".