Single Site performance of AI software for stroke detection and Triage
Bibliographic record
Abstract
Abstract Background Recently developed software utilizing artificial intelligence for fast detection and triage of stroke cases has the potential to accelerate stroke care and improve patient outcomes. We performed this analysis to evaluate the performance and time-to-notification of one such software - RAPID LVO. Methods We created a database of 151 consecutive acute stroke patients for whom CT scans were processed by the RAPID LVO software over a period of eight months. The LVO notification and time to notification of the software were collected, alongside patient information and the CTA findings. Results RAPID LVO achieved a sensitivity of 63.6% and specificity of 85.8% for large vessel occlusion, with an average time to notification of 32.53 minutes. Conclusions RAPID LVO has low sensitivity, moderate specificity and high time-to-notification performance. Our study data demonstrated in particular low overall sensitivity (63%) for distal occlusions (M2-3). The disparity between the observed performance and the performance reported in RAPID LVO’s FDA clearance demonstrates the importance of independent, multi-center evaluation. The gap between the performance in this study compared to published records of RAPID AI may be due to differences in imaging hardware, software implementation, connectivity or clinical definitions.
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 | 0.005 | 0.025 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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".