Using Noncontrast Computed Tomography to Improve Prediction of Intracerebral Hemorrhage Expansion
Bibliographic record
Abstract
Background: Noncontrast computed tomography hypodensities are a validated predictor of hematoma expansion (HE) in intracerebral hemorrhage and a possible alternative to the computed tomography angiography (CTA) spot sign but their added value to available prediction models remains unclear. We investigated whether the inclusion of hypodensities improves prediction of HE and compared their added value over the spot sign. Methods: Retrospective analysis of patients admitted for primary spontaneous intracerebral hemorrhage at the following 8 university hospitals in Boston, US (1994–2015, prospective), Hamilton, Canada (2010–2016, retrospective), Berlin, Germany (2014–2019, retrospective), Chongqing, China (2011–2015, retrospective), Pavia, Italy (2017–2019, prospective), Ferrara, Italy (2010–2019, retrospective), Brescia, Italy (2020–2021, retrospective), and Bologna, Italy (2015–2019, retrospective). Predictors of HE (hematoma growth >6 mL and/or >33% from baseline to follow-up imaging) were explored with logistic regression. We compared the discrimination of a simple prediction model for HE based on 4 predictors (antitplatelet and anticoagulant treatment, baseline intracerebral hemorrhage volume, and onset-to-imaging time) before and after the inclusion of noncontrast computed tomography hypodensities, using receiver operating characteristic curve and De Long test for area under the curve comparison. Results: A total of 2465 subjects were included, of whom 664 (26.9%) had HE and 1085 (44.0%) had hypodensities. Hypodensities were independently associated with HE after adjustment for confounders in logistic regression (odds ratio, 3.11 [95% CI, 2.55–3.80]; P <0.001). The inclusion of noncontrast computed tomography hypodensities improved the discrimination of the 4 predictors model (area under the curve, 0.67 [95% CI, 0.64–0.69] versus 0.71 [95% CI, 0.69–0.74]; P =0.025). In the subgroup of patients with a CTA available (n=895, 36.3%), the added value of hypodensities remained statistically significant (area under the curve, 0.68 [95% CI, 0.64–0.73] versus 0.74 [95% CI, 0.70–0.78]; P =0.041) whereas the addition of the CTA spot sign did not provide significant discrimination improvement (area under the curve, 0.74 [95% CI, 0.70–0.78]). Conclusions: Noncontrast computed tomography hypodensities provided a significant added value in the prediction of HE and appear a valuable alternative to the CTA spot sign. Our findings might inform future studies and suggest the possibility to stratify the risk of HE with good discrimination without CTA.
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 | 0.002 | 0.007 |
| 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 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 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".