MétaCan
Menu
Back to cohort
Record W4414222002 · doi:10.1002/neo2.70034

Deep Learning in Emergency Radiology: Evaluating a Tool for Automated ASPECT Scoring in Acute Ischemic Stroke

2025· article· en· W4414222002 on OpenAlexaboutno aff
Julia G. O'Connor, Angela Ayobi, Maxime Tassy, Christophe Avare, Sarah Quenet, Adam Davis, Peter Chang, Daniel Chow, Christopher G. Filippi, Yasmina Chaibi

Bibliographic record

VenueClinical neuroimaging. · 2025
Typearticle
Languageen
FieldMedicine
TopicAcute Ischemic Stroke Management
Canadian institutionsnot available
Fundersnot available
KeywordsIntraclass correlationThrombolysisStroke (engine)Receiver operating characteristicDeep learningConsistency (knowledge bases)TriageIschemic strokeReliability (semiconductor)

Abstract

fetched live from OpenAlex

ABSTRACT Background and Purpose The Alberta Stroke Program Early Computed Tomography Score (ASPECTS) quantifies ischemic damage in middle cerebral artery territory strokes, guiding treatment decisions including thrombolysis and mechanical thrombectomy. However, inter‐rater variability in scoring limits its clinical reliability. Automated ASPECTS tools like CINA‐ASPECTS (v1.4.3, Avicenna.AI) have shown potential to enhance consistency and efficiency. This study validated CINA‐ASPECTS using a multinational, multi‐scanner external dataset against a ground truth (GT) established by expert neuroradiologists. Methods Head NCCT scans from 339 patients (20 ASPECTS regions each) were collected from clinical sites in the United States, France, and Japan. Five US board‐certified expert neuroradiologists participated in establishing the GT using NCCT and additional imaging when available, blinded to each other and the device output. CINA‐ASPECTS outputs were compared to this GT on a region‐based and score‐based level. Stratified analyses were conducted by scanner manufacturer, cortical versus deep regions, and individual ASPECTS regions. Results CINA‐ASPECTS demonstrated 72.8% sensitivity, 91.8% specificity, and an area under the receiver operating characteristic curve of 0.823 in region‐based analysis. Score‐based agreement showed an intraclass correlation coefficient (ICC) of 0.83 [95% CI: 0.79–0.86]. Dichotomized score‐based analysis showed good specificity, sensitivity, and ICC. Stratified analysis confirmed consistent performance across scanner manufacturers and cortical/deep ASPECTS regions. Performance was robust regardless of ASPECTS region, potentially aiding in the scoring of difficult regions. Conclusions CINA‐ASPECTS reliably identifies ischemic changes across diverse clinical and imaging conditions, achieving good agreement with expert consensus. These findings support its utility in clinical decision‐making for acute stroke management.

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 imitation

Not 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.

metaresearch head score (Codex)0.002
metaresearch head score (Gemma)0.005
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.475
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0020.005
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.049
GPT teacher head0.412
Teacher spread0.363 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

Study designObservational
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueClinical neuroimaging.Same topicAcute Ischemic Stroke ManagementFrench-language works237,207