摘要
目的验证智能化改良简易认知筛查测试(Mini-Cog)工具用于认知障碍筛查的可靠性及有效性。方法选择2022年6至12月北京大学第一医院神经内科门诊及住院的老年患者217例,分为痴呆组61例,轻度认知障碍组114例,认知正常组42例。首先采用智能化技术自动测评Mini-Cog,然后专家面对面测评,比较结果一致性,进而分析其区分痴呆与认知正常及轻度认知障碍的预测价值。结果3组年龄、受教育程度、蒙特利尔认知评估量表评分、简易智能状态检查量表评分、Mini-Cog机器评分和专家评分比较,差异有统计学意义(P<0.05)。各组Mini-Cog机器评分与专家评分取得较好的一致性(r=0.85,P<0.01)。区分痴呆与认知正常患者的任务中,智能化改良Mini-Cog工具敏感性为93.4%,特异性为90.5%,曲线下面积为0.97。区分痴呆与非痴呆患者的任务中,智能化改良Mini-Cog工具敏感性为93.4%,特异性为80.1%,曲线下面积为0.93。结论智能化改良Mini-Cog工具机器评分可靠性较高,在痴呆患者筛查任务中具有较高的敏感性和特异性。
Objective To verify the reliability and efficacy of an intelligently modified Mini-Cog tool for screening cognitive impairment.Methods A total of 217 elderly patients admitted to Depart-ment of Neurology of Peking University First Hospital from June to December 2022 were enrolled in this study.They were 61 patients of dementia,114 of mild cognitive impairment and 42 cognitively normal individuals(NCI).The artificial intelligence(AI)Mini-Cog assessment was first performed,followed by face-to-face cognitive assessment by experts of neuropsychology.The consistency between AI and expert rated scores were first evaluated.The predictive values of intelligently modified Mini-Cog tool in dementia screening tasks(differentiating dementia patients from NCI and non-demented participants)were then evaluated.Results Age,education level,and MoCA,MMSE and Mini-Cog scores differed significantly across 3 groups(P<0.05).The accordance between AI scores and expert assessed scores were satisfactory(r=0.85,P<0.01).When differentiating dementia patients from NCI participants,the tool's sensitivity and specificity was 93.4%and 90.5%respectively,with an AUC value of 0.97.When differentiating dementia patients from the participants without dementia,the sensitivity and specificity was 93.4%and 80.1%respectively,with an AUC value of 0.93.Conclusion Our modified digital Mini-Cog tool has satisfactory reliability,and shows palatable sensitivity and specificity in dementia screening tasks.
作者
冷方达
徐国海
龙如蛟
许文深
王婷
姚聪
孙永安
张佶
解恒革
Leng Fangda;Xu Guohai;Long Rujiao;Xu Wenshen;Wang Ting;Yao Cong;Sun Yong'an;Zhang Ji;Xie Hengge(Department of Neurology,Peking University First Hospital,Beijing 100034,China)
出处
《中华老年心脑血管病杂志》
CAS
北大核心
2023年第8期793-796,共4页
Chinese Journal of Geriatric Heart,Brain and Vessel Diseases
基金
国家重点研发计划项目(2018YFC1314203)
科技部阿尔茨海默病及其他痴呆临床队列研究项目(2021ZD0201800)。