Abstract: Objective: The aim in this study was to determine the accuracy of multimodal computed tomography (CT) in predicting acute (early) cognitive impairment in patients with acute cerebral infarction. Method: Clinical data were retrospectively collected from 150 patients with acute cerebral infarction who were admitted to Suqian First People’s Hospital between January 2022 and February 2025. All patients underwent multimodal CT examinations within 24 h after admission, including plain CT and CT perfusion imaging. The patients were divided into cognitive impairment (n = 59) and cognitively normal (n = 91) groups based on whether acute cognitive impairment occurred in the early poststroke stage. The characteristics, parameters, and clinical multimodal CT examination data were compared between the patient groups. Factors influencing acute cognitive dysfunction in patients with acute cerebral infarction were analyzed using multivariate logistic regression. Receiver operating characteristic (ROC) and decision curves were drawn and analyzed to verify the predictive value of each feature and parameter of the multimodal CT examination for acute cognitive dysfunction in patients with acute cerebral infarction. Results: Patients in the cognitive impairment group were older than those in the normal cognitive group, the time from onset to admission was longer in people with than without cognitive impairment, proportions of patients with diabetes and NIHSS scores at admission were higher in the cognitive impairment group than those in the normal cognitive group. The multimodal CT examination data were compared between the groups. The proportion of patients with high-density arterial signs was higher in the cognitive impairment group. The cerebral blood volume (CBV) and perfusion parameter cerebral blood flow (CBF) were lower in the cognitive impairment group than in the normal cognitive group. The time to peak (TTP) was higher in the cognitive impairment than in the normal cognitive group, while the lesion volume was larger in the cognitive impairment group than in the normal cognitive group. Multivariate logistic regression analysis revealed that age, NIHSS score at admission, arterial high-density sign, CBV, CBF, TTP, and lesion volume influenced acute cognitive dysfunction in patients with acute cerebral infarction after adjusting for confounding factors. The ROC curve showed that only the area under the curve (AUC) of the arterial high-density sign for predicting cognitive dysfunction was < 0.7, whereas the AUCs of the CBV, CBF, TTP, and lesion volume in predicting cognitive dysfunction were > 0.7. The AUC of the combined prediction model was the highest (0.949). The decision curve showed that when the risk threshold was within the range of 0.01–0.99, the net return rate of the decision curve drawn in combination with multimodal CT was all > 0, and the maximum net benefit was 0.391. Conclusion: Multimodal CT can be used to effectively predict the risk of acute cognitive dysfunction in patients with acute cerebral infarction, providing evidence for the early identification of high-risk populations and guidance for selecting personalized interventions in clinical practice.