中国全科医学 ›› 2023, Vol. 26 ›› Issue (10): 1224-1233.DOI: 10.12114/j.issn.1007-9572.2022.0729

所属专题: 神经退行性病变最新文章合集 阿尔茨海默病最新文章合集 老年问题最新文章合集

• 论著·认知障碍专题研究 • 上一篇    下一篇

老年轻度认知障碍者笔迹特征及其应用价值研究

卫珠琴1, 张若愚2, 张晨3, 苏丽明1, 黄诚1, 张军伟1, 钱敏才4, 祁亨年2,*(), 王丽娜1,*()   

  1. 1.313000 浙江省湖州市,湖州师范学院医学院
    2.313000 浙江省湖州市,湖州师范学院信息工程学院
    3.313001 浙江省湖州市,湖州市仁皇山滨湖街道社区卫生服务中心公卫科
    4.313000 浙江省湖州市,湖州市第三人民医院心身科
  • 收稿日期:2022-10-16 修回日期:2023-01-24 出版日期:2023-04-05 发布日期:2023-02-09
  • 通讯作者: 祁亨年, 王丽娜
  • 卫珠琴,张若愚,张晨,等.老年轻度认知障碍者笔迹特征及其应用价值研究[J].中国全科医学,2023,26(10):1224-1233.[www.chinagp.net]

    作者贡献:卫珠琴负责文献查证、数据收集及整理、结果的分析与解释、论文撰写;张若愚负责文献查证、数据收集与整理;张晨负责社区调研的组织与协调及数据收集;苏丽明、黄诚、张军伟负责数据收集;钱敏才负责文稿的修订;祁亨年负责书写笔迹特征分析技术的指导、文章的质量控制及审校;王丽娜负责研究的构思、文章的质量控制及审校,对文章整体负责。
  • 基金资助:
    国家自然科学基金面上项目(72174061); 国家自然科学基金青年项目(71704053); 浙江省卫生健康科技计划(2022KY370); 浙江省大学生科技创新活动计划暨新苗人才计划(2022R431B022)

Characteristics and Application Value of Handwriting in Elderly Patients with Mild Cognitive Impairment

WEI Zhuqin1, ZHANG Ruoyu2, ZHANG Chen3, SU Liming1, HUANG Cheng1, ZHANG Junwei1, QIAN Mincai4, QI Hengnian2,*(), WANG Lina1,*()   

  1. 1. School of Medicine, Huzhou University, Huzhou 313000, China
    2. School of Information Engineering, Huzhou University, Huzhou 313000, China
    3. Department of Public Health, Binhu Subdistrict Community Health Center of Renhuangshan, Huzhou 313001, China
    4. Psychosomatic Department, Huzhou Third People's Hospital, Huzhou 313000, China
  • Received:2022-10-16 Revised:2023-01-24 Published:2023-04-05 Online:2023-02-09
  • Contact: QI Hengnian, WANG Lina
  • About author:
    WEI Z Q, ZHANG R Y, ZHANG C, et al. Characteristics and application value of handwriting in elderly patients with mild cognitive impairment [J]. Chinese General Practice, 2023, 26 (10): 1224-1233.

摘要: 背景 书写笔迹特征分析技术已在痴呆、帕金森病等相关认知障碍检测领域得到了广泛研究,针对老年轻度认知障碍(MCI)人群的笔迹特征研究仍有待于拓展。 目的 揭示老年MCI者笔迹特征与认知功能正常老年人的差异性,探讨笔迹特征在MCI筛查中的应用价值。 方法 于2022年1—4月,采用便利抽样法,选取湖州市社区老年MCI者33例作为观察组,同期另选取在年龄、性别和受教育程度上与33例社区老年MCI者相匹配的社区认知功能正常老年人43例作为对照组,采用一般资料调查表、简易精神状态量表(MMSE)、蒙特利尔认知评估基础量表(MoCA-B)、日常生活能力量表(ADL)和老年抑郁量表(GDS-15)对其进行调查,并邀请受试者使用点阵数码笔完成6项书写任务(包含4项汉字书写任务、2项图形书写任务),采集受试者笔迹特征的运动学参数,通过判别分析、受试者工作特征(ROC)曲线分析笔迹特征及另设区分方案在MCI识别中的应用价值。 结果 与对照组相比,在图形书写任务中,观察组受试者书写平均压力更高(Z=-2.122,P=0.034),书写思考时间(Z=-4.302,P<0.001)、落笔时间(Z=-3.663,P<0.001)及任务完成总时间(t'=-5.565,P<0.001)更长,书写平均速度(Z=-2.458,P=0.014)、x方向平均速度(Z=-2.950,P=0.003)、y方向平均速度(Z=-2.094,P=0.040)更慢,书写x方向速度最大值(Z=-2.206,P=0.027)、x方向平均加速度(Z=-2.667,P=0.008)更小,书写正确性总分更低(Z=-3.593,P<0.001);在汉字书写任务中,观察组受试者书写思考时间(Z=-3.464,P=0.001)、任务完成总时间(Z=-2.940,P=0.003)更长。与汉字书写任务完成总时间相比,图形书写任务完成总时间在鉴别老年MCI者和认知功能正常老年人中的特异度更高(93.0%比55.8%),ROC曲线下面积(AUC)为0.828;采用图形书写任务笔迹特征集对受试者进行MCI筛查时,灵敏度为87.9%,特异度为79.1%,准确率达80.3%(61/76),且图形书写任务笔迹特征集对老年MCI者的识别效能高于MMSE(Z=1.993,P=0.046)及汉字书写任务笔迹特征集(Z=2.408,P=0.016)。 结论 图形书写笔迹特征可能在筛查可疑MCI人群方面具有潜在应用价值。社区卫生服务机构在借助MCI成套神经心理学测试开展MCI筛查服务之前(的同时),可采用图形书写笔迹特征对社区居民进行MCI筛查。

关键词: 老年人, 轻度认知障碍, 笔迹特征, 运动学参数, 筛查工具

Abstract:

Background

Handwriting synthesis techniques have been extensively studied in the detection of cognitive impairment in dementia and Parkinson's disease. But handwriting characteristics in older adults with mild cognitive impairment (MCI) still need to be studied further.

Objective

To explore the differences between the handwriting characteristics of elderly patients with MCI and normal elderly people, and to assess the value of handwriting features in MCI screening.

Methods

By use of convenience sampling, 33 older adults with MCI were recruited from Huzhou communities from January to April 2022 (observation group), and were compared to age-, sex- and education level-matched 43 community-living older adults with normal cognitive function (control group). The General Information Questionnaire, the Mini-Mental State Examination (MMSE), the Montreal Cognitive Assessment-Basic (MoCA-B), the Activity of Daily Living Scale (ADL), and the 15-item Geriatric Depression Scale (GDS-15) were used to survey subjects. Subjects were invited to complete six handwriting tasks (four are Chinese characters tasks and the other two are graphical drawing tasks) using a dot matrix digital pen to collect their kinematic parameters of handwriting. The classification accuracy, sensitivity and specificity of handwriting characteristics for the diagnosis of MCI were analyzed by discriminant analysis and receiver operating characteristic (ROC) curve, and predictive values of different schemes for MCI were analyzed.

Results

Compared with the control group, the observation group had higher average pressure in writing (Z=-2.122, P=0.034), longer time in air (Z=-4.302, P<0.001), writing time (Z=-3.663, P<0.001) and total time (t'=-5.565, P<0.001), lower average writing velocity (Z=-2.458, P=0.014), horizontal (Z=-2.950, P=0.003) and vertical (Z=-2.094, P=0.040) average writing velocity and maximum horizontal writing velocity (Z=-2.206, P=0.027), lower average acceleration of writing in horizontal direction (Z=-2.667, P=0.008) and overall score for writing correctness (Z=-3.593, P<0.001) in completing graphical drawing tasks. The observation group had relatively longer time in air (Z=-3.464, P=0.001) and total time (Z=-2.940, P=0.003) in completing Chinese characters tasks. Compared with the total time for completing Chinese characters tasks, the total time for completing graphical drawing tasks had higher specificity (93.0% vs 55.8%) in differentiating between MCI and control groups, with an area under the curve (AUC) of 0.828. The summary of handwriting characteristics for graphical drawing tasks correctly classified 80.3% (61/76) of older adults with MCI, with 87.9% sensitivity and 79.1% specificity, and had higher diagnostic efficacy for those with MCI than the MMSE scale (Z=1.993, P=0.046) and the summary of handwriting characteristics for Chinese characters tasks (Z=2.408, P=0.016) .

Conclusion

Handwriting characteristics of graphical drawing tasks may have potential application in screening of older adults at risk for MCI, which can be used simultaneously or prior to sets of neuropsychological tests conducted for the diagnosis of MCI in community health care facilities.

Key words: Aged, Mild cognitive impairment, Handwriting characteristics, Kinematic parameters, Screening tool