Objective To explore and compare the efficiency of early identification of acute large vessel occlusion （LVO） ischemic stroke in the emergency department using the Vision， Aphasia， Neglect （VAN） assessment and the National Institutes of Health Stroke Scale （NIHSS） score. Methods A retrospective analysis was conducted on the clinical data of 189 patients with acute ischemic stroke who received treatment in the emergency department of the hospital from January 2023 to March 2024. The VAN and NIHSS scores were determined by trained emergency physicians upon admission. Two trained emergency physicians （with a title of attending physician or above） independently and retrospectively scored all patients based on emergency medical records without being aware of the vascular imaging results. In case of inconsistent evaluations， a consensus was reached through consultation. Using head-neck CTA or MRA as the gold standard for LVO diagnosis， the independent predictors of LVO were analyzed via multivariate logistic regression analysis， and the diagnostic efficiency of the two scoring methods for LVO was assessed by receiver operating characteristic （ROC） curves. Results Among the 189 patients with acute ischemic stroke， the gold-standard method （CTA/MRA） indicated that 91 cases （48.15%） had LVO， including 59 cases （64.84%） with anterior circulation LVO and 32 cases （35.16%） with posterior circulation LVO. Multivariate logistic regression analysis revealed that after adjusting for confounding factors such as age and atrial fibrillation， a positive VAN （OR = 12.541， 95%CI： 4.800 ~ 32.766， P < 0.05） and the NIHSS score （OR = 1.319， 95%CI： 1.115 ~ 1.561， P < 0.05） were both independent predictors of LVO. ROC curve analysis demonstrated that the area under the curve （AUC） of the NIHSS score for the diagnosis of LVO was 0.863 （95%CI： 0.805 ~ 0.921）. Using the optimal cut-off value （10 points） as the criterion， its sensitivity， specificity， positive predictive value， and negative predictive value were 82.42%， 87.76%， 86.21%， and 84.31%， respectively. The AUC， sensitivity， specificity， positive predictive value， and negative predictive value of VAN for the diagnosis of LVO were 0.857 （95%CI： 0.799 ~ 0.915）， 84.62%， 86.73%， 85.56%， and 85.86%， respectively. The DeLong test showed that there was no significant difference in the AUC between the two scoring methods （Z = 0.166， P = 0.869）. Regarding anterior circulation LVO， the diagnostic efficiency of VAN was higher than that of the NIHSS score ［（AUC = 0.851， 95%CI： 0.795 ~ 0.908） vs. （AUC = 0.760， 95%CI： 0.682 ~ 0.839）， Z = 2.013， P = 0.044］. For posterior circulation LVO， the diagnostic efficiency of the NIHSS score was significantly higher than that of VAN ［（AUC = 0.839， 95%CI： 0.766 ~ 0.912） vs. （AUC = 0.618， 95%CI： 0.512 ~ 0.724）， Z = 3.814， P < 0.001］. Conclusions Both the VAN and NIHSS score are independent predictors of LVO， which can effectively assess the occurrence of acute LVO ischemic stroke. Furthermore， the VAN demonstrates superior identification efficiency for anterior circulation LVO， whereas the NIHSS score exhibits better identification efficiency for posterior circulation LVO.