人工智能ASPECTS评分的智能卒中一站式体系对急性缺血性卒中的临床价值分析

Clinical Value Analysis of the One-stop Intelligent Stroke System Based on Artificial Intelligence ASPECTS for Acute Ischemic Stroke

目的探究人工智能Alberta卒中项目早期CT评分(ASPECTS)的智能卒中一站式体系在急性缺血性卒中(AIS)中的临床应用价值。方法选取2022年9月至2023年9月本院收治的100例AIS患者作为观察对象,根据患者预后情况为预后不良组和预后良好组。收集两组一般资料,比较两组人工智能ASPECTS评分;受试者工作特征(ROC)曲线分析年龄、人工智能ASPECTS评分和NIHSS评分AIS患者预后不良的预测价值。结果预后良好组和预后不良组在年龄、意识障碍、梗死面积、人工智能ASPECTS评分和NIHSS评分比较有差异(P<0.05)。大面积梗死患者人工智能ASPECTS评分显著低于非大面积梗死患者(P<0.05),有意识障碍患者人工智能ASPECTS评分显著低于无意识障碍患者(P<0.05)。Logistic回归分析得知年龄(OR=3.518,95%CI:1.989~5.012)和NIHSS评分(OR=4.258,95%CI:1.739~10.428)为AIS患者预后不良的危险因素(P<0.05),人工智能ASPECTS评分(OR=0.542,95%CI:0.354~0.831)为AIS患者预后不良的保护因素(P<0.05)。ROC曲线分析显示,年龄、人工智能ASPECTS评分、NIHSS评分预测AIS患者预后不良的曲线下面积(AUC)分别为0.874、0.881、0.909,三者联合预测AIS患者预后不良的AUC为0.983,三者联合优于各自单独预测(Z=2.627、2.734、2.720,P均<0.05)。结论人工智能ASPECTS评分的智能卒中一站式体系对AIS患者的预后有较好的评估价值,可在临床推广。

Objective To explore the clinical application value of the one-stop intelligent stroke system based on artificial intelligence Alberta Stroke Program Early CT Score(ASPECTS)in acute ischemic stroke(AIS).Methods A total of 100 AIS patients admitted to our hospital from September 2022 to September 2023 were collected as observation subjects,and were separated into a poor prognosis group and a good prognosis group based on their prognosis.The general data of two groups were collected,and the artificial intelligence ASPECTS was compared between the two groups;receiver operating characteristic curve(ROC)curve was applied to analyze the predictive value of age,artificial intelligence ASPECTS,and NIHSS score for poor prognosis in AIS patients.Results There were differences in age,conscious disorders,infarct size,artificial intelligence ASPECTS,and NIHSS score between the group with good prognosis and the group with poor prognosis(P<0.05).The artificial intelligence ASPECTS of patients with large area infarction was obviously lower than that of non large area infarction patients(P<0.05),and the artificial intelligence ASPECTS of patients with conscious disorders was obviously lower than that of patients with unconscious disorders(P<0.05).Logistic regression analysis showed that age(OR=3.518,95%CI:1.989-5.012)and NIHSS score(OR=4.258,95%CI:1.739-10.428)were risk factors for poor prognosis in AIS patients(P<0.05),while artificial intelligence ASPECTS(OR=0.542,95%CI:0.354-0.831)was a protective factor for poor prognosis in AIS patients(P<0.05).ROC curve analysis showed that the area under the curve(AUC)of age,artificial intelligence ASPECTS,and NIHSS score in predicting poor prognosis in AIS patients was 0.874,0.881,and 0.909,respectively,the AUC of the three combination in predicting poor prognosis in AIS patients was 0.983,and the combination of the three was better than their individual predictions(Z=2.627,2.734,and 2.720,P<0.05).Conclusion The one-stop intelligent stroke system based on artificial intelligence ASPECTS has good prognostic value for AIS patients and can be promoted in clinical practice.