AI Insight
A multimodal artificial intelligence system called APOLLAI was tested against emergency medical services personnel for detecting stroke-related neurological deficits in 295 patients with suspected stroke. APOLLAI demonstrated significantly higher sensitivity than EMS personnel (78.5% versus 59.4%) while maintaining comparable specificity (87.0% versus 85.9%) when evaluated against expert neurologist consensus. The AI system analyzed video and audio recordings to assess five key neurological indicators including gaze, facial palsy, motor function, aphasia, and dysarthria.
Why it matters
Nearly one-third of strokes go unrecognized before hospital arrival, and this AI system could provide a standardized tool to improve early stroke detection in prehospital settings. By catching more strokes that EMS personnel might miss while maintaining accuracy, the system could enable faster treatment initiation, which is critical for stroke outcomes.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Background and Purpose: Prehospital stroke recognition remains suboptimal, with nearly one-third of strokes unidentified before hospital arrival, and neurological assessment being inherently subjective, with substantial interobserver variability even among certified clinicians. APOLLAI, a multimodal artificial intelligence system, was developed to detect stroke-related neurological deficits from video and audio recordings. We compared the diagnostic performance of APOLLAI and emergency medical services (EMS) personnel against an expert consensus reference standard. Methods: In this prospective study, video recordings of 295 patients with suspected stroke (mean age, 70.3 {+/-}13.3 years) were obtained at Nimes University Hospital (August 2023-July 2025). Five National Institutes of Health Stroke Scale items (best gaze, facial palsy, motor arm, aphasia, dysarthria) were independently annotated by seven certified stroke neurologists, combined into consensus labels using the Dawid-Skene algorithm, and separately assessed by EMS personnel (mean of 15 raters per patient). APOLLAI, a multimodal machine-learning framework that combines computer vision, audio-based deep learning, and a final support vector machine (SVM) fusion classifier, was evaluated against EMS using sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC). Results: Interrater agreement among the seven neurologists was moderate overall (mean {kappa}=0.541, range, 0.367-0.732), indicating an intrinsic difficulty of neurological assessment. APOLLAI showed higher sensitivity than EMS for all five neurological tasks and for the global assessment (78.5% versus 59.4%), with broadly comparable specificity (87.0% versus 85.9% for the global assessment). AUC values ranged from 0.839 (facial palsy) to 0.966 (best gaze), with a global AUC of 0.882. Conclusions: APOLLAI demonstrated higher sensitivity than EMS personnel for detecting stroke-related neurological deficits, with preserved specificity. These findings support its potential as a standardized, scalable adjunct to prehospital stroke recognition, warranting prospective validation under real-world EMS conditions.