AI Assistants Show Promise in Emergent Neurologic Care

Background: Artificial Intelligence (AI) is increasingly used in medical systems, yet our confidence in their performance is still an issue of debate. Objective: To assess whether AI assistants improve neurologic care and to identify cases that may benefit most from AI support. Methods: The authors examined the function of 3 AI assistants in evaluating the clinical information of 10 patients. Cases were selected on the basis of confirmed diagnoses and representation of common clinical presentations. Final diagnoses included common acute neurologic conditions such as ischemic stroke, aneurysmal subarachnoid hemorrhage, delirium, bacterial meningitis, benign paroxysmal positional vertigo, status epilepticus, acute inflammatory demyelinating polyneuropathy, idiopathic facial palsy, status migrainosus, and trigeminal neuralgia. The 3 AI assistants included 2 ChatGPT versions and an AI assistant by Entelai focussed on neuroimaging. Results: Use of the AI assistant required a median of 5.5 more...