Performance of an artificial intelligence tool for multi-step acute stroke imaging: A multicenter diagnostic study
Bibliographic record
Abstract
Objective: Several artificial intelligence (AI) tools have been developed to assist in the stroke imaging workflow, which remains a major disease of the 21st century. This study evaluated the combined performance of an FDA-cleared and CE-marked AI-based device with three modules designed to detect intracerebral hemorrhage (ICH), identify large vessel occlusion (LVO), and calculate Alberta Stroke Program Early CT Scores (ASPECTS). Materials & methods: Non-contrast CT (NCCT) and/or computed tomography angiography (CTA) for suspicion of stroke acquired at La Timone and Nord University hospitals (Marseille, France) between March 2019 and March 2020 were retrospectively collected. The AI tool, CINA-HEAD (Avicenna.AI), processed the data to flag ICH, LVO, and calculate ASPECTS. The results were compared to ground truth evaluations by four expert neuroradiologists to compute diagnostic performances. Results: A total of 373 NCCT and 331 CTA from 405 patients (mean age 64.9 ± 18.9 SD, 52.6 % female) were included. The AI tool achieved an accuracy of 94.6 % [95 % CI: 91.8 %-96.7 %] for ICH detection on NCCT and of 86.4 % [95 % CI: 82.2 %-89.9 %] for LVO identification on CTA. The region-based ASPECTS analysis yielded an accuracy of 88.6 % [95 % CI: 87.8 %-89.3 %] and the dichotomized ASPECTS classification (ASPECTS ≥ 6) achieved 80.4 % accuracy. Conclusion: This study demonstrates the reliable, stepwise performance of an AI-based stroke imaging tool across the diagnostic cascade of ICH and LVO detection and ASPECTS scoring. Such robust multi-stage evaluation supports its potential for streamlining acute stroke triage and decision-making.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".