Bringing prevost's sign into the third dimension: Artificial intelligence estimation of conjugate gaze adjusted length (CGAL) and correlation with acute ischemic stroke
Bibliographic record
Abstract
Conjugate gaze deviation is associated with acute ischemic stroke (AIS), although previously only measured on a 2D plane. The current study evaluates 3D imaging efficacy to assess conjugate gaze deviation and correlate direction and strength of deviation to neuro-clinical findings.A retrospective analysis of 519 patients who had CT scans for suspected AIS at our institution. Direction and angle of eye deviation were calculated based on 2D axial images. Volumetric reconstruction of CT scans allowed for calculation of 3D conjugate gaze adjusted length (CGAL). Angle, direction, and vector strength of both 2D and 3D scans were calculated by an artificial intelligence algorithm and tested for agreement with hemispheric ischemia location. CGAL measurements were correlated to NIHSS scores. Follow up MRI data was used to evaluate the sensitivity and specificity of CGAL in the identification of AIS.The final analysis included 122 patients. A strong agreement was found between 3D gaze direction and hemispheric ischemia location. CGAL measurements were highly correlated with NIHSS score (r = .72, P = .01). A CGAL >0.25, >0.28, and >0.35 exhibited a sensitivity of 91%, 86%, and 82% and specificity of 66%, 89%, and 89%, respectively, in AIS identification. A CGAL >0.28 has the best sensitivity-specificity balance in the identification of AIS. A CGAL >0.25 has the highest sensitivity.Given CED's correlation with NIHSS score a 1/4 deviation in the ipsilateral direction is a sensitive ancillary radiographic sign to assist radiologists in making a correct diagnosis even when not presented with full clinical data.
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 | 0.007 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".