Automated detection of large vessel occlusion using deep learning: a pivotal multicenter clinical trial and reader assessment study
Bibliographic record
Abstract
Abstract Background This multicenter clinical trial evaluated the stand-alone efficacy and the improvements in diagnostic accuracy of early-career physicians using a deep learning-based software to detect large vessel occlusion (LVO) in CT angiography (CTA). Methods This multicenter pivotal clinical trial included 595 ischemic stroke patients from January 2018 to September 2023. Standard reference and LVO locations (intracranial internal carotid artery [ICA], M1, or M2) were determined by consensus among three expert vascular neurologists after reviewing CTA, MR imaging, and symptom data. The performance of the JLK-LVO software was evaluated against a standard reference, and its impact on the diagnostic accuracy of four residents involved in stroke care was assessed. Performance metrics included the area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Results Among the 595 patients (mean age 68.5 ± 13.4 years, 56% male), 275 (46.2%) had LVO. The median time interval from the last known well moment to the CTA was 46 hours (IQR 11.8 to 64.4). For LVO detection, the software demonstrated a sensitivity of 86% and a specificity of 97%. For isolated M2 occlusions, it achieved a sensitivity of 69% and a specificity of 96%. The reader assessment study showed that reading with software assistance improved the sensitivity by 4.0% and AUROC by 2.4% (all p < 0.001) compared to readings without AI assistance. Conclusion The software demonstrated a high detection rate for proximal LVO and moderate sensitivity for isolated MCA-M2 occlusion. In addition, the software improved diagnostic accuracy of early-career physicians in detecting LVO.
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 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".