Study on the Incremental Predictive Value of Early Resting-State EEG Phenotypes for Functional Prognosis and Risk Stratification in Acute Ischemic Stroke

This is a single-center, prospective, observational cohort study that plans to consecutively enroll 400 patients with acute ischemic stroke from July 2026 to July 2029. The study will not interfere with acute-phase treatment decisions; information on reperfusion and non-reperfusion treatment will be recorded according to the actual clinical care pathway. After completion of necessary acute treatment, once vital signs are stable and clinical conditions permit, all participants will undergo one resting-state EEG recording as early as possible, within 7 days after symptom onset or the last-known-well time. For patients receiving reperfusion therapy, EEG will be acquired after completion of the reperfusion procedure, and intervals such as onset-to-reperfusion initiation and completion of reperfusion-to-EEG acquisition will be recorded. For patients not receiving reperfusion therapy, EEG will be acquired after the stroke diagnosis is established, routine treatment has been initiated, and the clinical condition is stable; onset-to-admission and onset-to-EEG acquisition intervals will be recorded. EEG features will be extracted and combined with clinical and imaging variables to construct prognostic models. The primary outcome is functional outcome at 3 months after onset, dichotomized as mRS 0-2 versus 3-6. The primary analysis will evaluate the incremental predictive value of EEG phenotypes for poor 3-month functional outcome beyond a conventional clinical-imaging model. Secondary analyses will include validation of 6-month outcomes, functional and cognitive scale outcomes, and differences in EEG phenotypes across treatment pathways.

Inclusion Criteria: * Age \>=18 years, regardless of sex; * Meets diagnostic criteria for acute ischemic stroke, supported by cranial CT/MRI and/or vascular imaging; * Symptom onset time or last-known-well time is clear; * Clinical assessment indicates that resting-state EEG can be completed without delaying acute treatment; * Pre-stroke mRS \2, severe prior dementia, or severe psychiatric disorder that would make follow-up functional or cognitive outcomes difficult to interpret; * Persistent deep sedation, coma, mechanical ventilation, severe metabolic disturbance, severe infection, or other conditions at the time of EEG acquisition that the investigator judges would markedly affect resting-state EEG background activity; * Status epilepticus or recent frequent clinical seizures around the time of EEG acquisition that the investigator judges would substantially affect interpretation of resting-state EEG; * Any other condition that the investigator judges unsuitable for participation in this study.