基于机器学习的抽动障碍共患注意缺陷与多动障碍的风险评估模型研究

刘芳; 王华倩; 牛萍; 姚宝珍; 叶静萍; 王高华

doi:10.14188/j.1671-8852.2024.0237

您当前的位置：

首页 >

文章列表页 >

基于机器学习的抽动障碍共患注意缺陷与多动障碍的风险评估模型研究

亲子关系与儿童心理疾病专题研究 | 更新时间：2026-11-05

- 基于机器学习的抽动障碍共患注意缺陷与多动障碍的风险评估模型研究
- Risk assessment models of tic disorders with comorbid attention deficit hyperactivity disorder based on machine learning
- 武汉大学学报（医学版） 2025年46卷第10期页码：1247-1253
- 作者机构：
  
  1.武汉大学人民医院儿科湖北武汉 430060
  2.武汉大学人民医院精神科湖北武汉 430060
- 作者简介：
  
  刘芳，女，1985—，医学博士，主治医师，主要从事儿童保健、神经心理及发育行为研究，E‑mail：liufangann@163.com
  叶静萍，女，1970—，医学博士，副主任医师，主要从事儿科、儿童保健、发育行为研究，E‑mail：jpy7012@163.com
  王高华，男，1964—，医学博士，教授，主任医师，主要从事精神病与精神卫生研究，E‑mail：WGH6402@163.com
- 基金信息：
  
  武汉大学医学腾飞计划项目(TFLC2018001)
- DOI：10.14188/j.1671-8852.2024.0237
  中图分类号： R749
- 收稿：2024-04-08，
  
  纸质出版：2025-10-25
- 稿件说明：
移动端阅览
刘芳,王华倩,牛萍等.基于机器学习的抽动障碍共患注意缺陷与多动障碍的风险评估模型研究[J].武汉大学学报（医学版）,2025,46(10):1247-1253.

LIU Fang,WANG Huaqian,NIU Ping,et al.Risk assessment models of tic disorders with comorbid attention deficit hyperactivity disorder based on machine learning[J].Medical Journal of Wuhan University,2025,46(10):1247-1253.
刘芳,王华倩,牛萍等.基于机器学习的抽动障碍共患注意缺陷与多动障碍的风险评估模型研究[J].武汉大学学报（医学版）,2025,46(10):1247-1253. DOI： 10.14188/j.1671-8852.2024.0237.

LIU Fang,WANG Huaqian,NIU Ping,et al.Risk assessment models of tic disorders with comorbid attention deficit hyperactivity disorder based on machine learning[J].Medical Journal of Wuhan University,2025,46(10):1247-1253. DOI： 10.14188/j.1671-8852.2024.0237.

摘要

目的

基于机器学习方法构建抽动障碍（TD）共患注意缺陷与多动障碍（ADHD）的风险评估模型及筛选风险因素。

方法

调查收集363例TD患儿的临床及人口学和家庭环境信息作为研究数据。采用准确率、精确率、召回率/灵敏度、特异度、F1值及ROC曲线下面积（

AUC

），比较基于Logistic回归、朴素贝叶斯（NB）、随机森林（RF）、K最近邻（KNN）分类算法和支持向量机（SVM）5种机器学习方法的TD共患ADHD风险评估模型的性能。采用RF算法进行特征重要性排序和采用多因素Logistic回归分析筛选风险因素。

结果

在准确率指标上，NB模型表现最优；在F1值、召回率/灵敏度指标上，Logistic回归模型表现均最优；在精确率、特异度指标上，SVM模型表现均最优；NB、RF、Logistic回归模型的ROC曲线

AUC

较大（分别为0.896、0.891、0.882）。RF算法显示TD病情严重程度、家庭教养方式、合并行为问题为特征重要性排序前三位的因素，多因素Logistic回归分析显示TD病情中重度（

=25.92，95%

：12.18~55.13）、合并行为问题（

=3.72，95%

：1.89~7.33）、放纵型家庭教养方式（

=2.59，95%

：1.05~6.40）是TD共患ADHD的危险因素。

结论

机器学习模型可用于TD共患ADHD的风险评估。TD病情严重程度、合并行为问题、家庭教养方式是TD共患ADHD的显著风险因素。

Abstract

Objective

To develop risk assessment models for tic disorders (TD) with comorbid attention deficit hyperactivity disorder (ADHD) and to screen for risk factors using machine learning.

Methods

The clinical

demographic

and family environmental information of 363 children with TD were investigated and collected as study data. The performance of five machine learning models of TD with comorbid ADHD based on logistic regression

Naive Bayes (NB)

Random Forest (RF)

K‑Nearest Neighbor (KNN) classification algorithm

and Support Vector Machine (SVM) was compared using metrics such as accuracy

precision

recall/sensitivity

specificity

F1‑score

and area under the curve (

AUC

) of ROC. Feature importance was ranked using the RF algorithm

and multivariate logistic regression analysis was conducted to identify risk factors.

Results

The NB model demonstrated the highest accuracy; the logistic regression model performed best in terms of F1‑score and recall/

sensitivity; the SVM model achieved the highest precision and specificity.The

AUC

values of the NB

and logistic regression models were relatively large (0.896

0.891

and 0.882

respectively). Multivariate logistic regression analysis showed that moderate/severe tic severity (

=25.92

95%

: 12.18⁃55.13)

behavioral problems (

=3.72

95%

: 1.89⁃7.33)

and indulgent parenting pattern (

=2.59

95%

: 1.05⁃6.40) were risk factors for TD with comorbid ADHD.

Conclusion

Machine learning models can be effectively used for risk assessment of TD with comorbid ADHD. Tic severity

comorbid behavioral problems

and parenting pattern were identified as significant risk factors.

关键词

Keywords

references

卢青 , 孙丹 , 刘智胜 . 中国抽动障碍诊断和治疗专家共识解读 [J ] . 中华实用儿科临床杂志 , 2021 , 36 ( 9 ): 647 ‑ 653 .

LU Q , SUN D , LIU Z S . Interpretation of expert consensus for diagnosis and treatment of tic disorders in China [J ] . Chin J Appl Clin Pediatr , 2021 , 36 ( 9 ): 647 ‑ 653 .

JIANG J L , CHEN M X , HUANG H F , et al . The aetiology of tourette syndrome and chronic tic disorder in children and adolescents: A comprehensive systematic review of case‑control studies [J ] . Brain Sci , 2022 , 12 ( 9 ): 1202 .

LIU Z S , CUI Y H , SUN D , et al . Current status, diagnosis, and treatment recommendation for tic disorders in China [J ] . Front Psychiatry , 2020 , 11 : 774 .

HIRSCHTRITT M E , LEE P C , PAULS D L , et al . Lifetime prevalence, age of risk, and genetic relationships of comorbid psychiatric disorders in Tourette syndrome [J ] . JAMA Psychiatry , 2015 , 72 ( 4 ): 325 ‑ 333 .

GROTH C , MOL DEBES N , RASK C U , et al . Course of tourette syndrome and comorbidities in a large prospective clinical study [J ] . J Am Acad Child Adolesc Psychiatry , 2017 , 56 ( 4 ): 304 ‑ 312 .

HIRSCHTRITT M E , DARROW S M , ILLMANN C , et al . Genetic and phenotypic overlap of specific obsessive‑compulsive and attention‑deficit/hyperactive subtypes with Tourette syndrome [J ] . Psychol Med , 2018 , 48 ( 2 ): 279 ‑ 293 .

ROTHENBERGER A , HEINRICH H . Co‑occurrence of tic disorders and attention‑deficit/hyperactivity disorder‒Does it reflect a common neurobiological background? [J ] . Biomedicines , 2022 , 10 ( 11 ): 2950 .

Cross‑Disorder Group of the Psychiatric Genomics Consortium . Genomic relationships, novel loci, and pleiotropic mechanisms across eight psychiatric disorders [J ] . Cell , 2019 , 179 ( 7 ): 1469 ‑ 1482 . e11.

TERMINE C , LUONI C , FONTOLAN S , et al . Impact of co‑morbid attention‑deficit and hyperactivity disorder on cognitive function in male children with Tourette syndrome: A controlled study [J ] . Psychiatry Res , 2016 , 243 : 263 ‑ 267 .

GREIMEL E , WANDERER S , ROTHENBERGER A , et al . Attentional performance in children and adolescents with tic disorder and co‑occurring attention‑deficit/hyperactivity disorder: new insights from a 2×2 factorial design study [J ] . J Abnorm Child Psychol , 2011 , 39 ( 6 ): 819 ‑ 828 .

LEE D Y , OH A R , PARK J , et al . Machine learning‑based prediction model for postoperative delirium in non‑cardiac surgery [J ] . BMC Psychiatry , 2023 , 23 ( 1 ): 317 .

DU L , DU JJ , YANG M , et al . Development and external validation of a machine learning‑based prediction model for the cancer‑related fatigue diagnostic screening in adult cancer patients: A cross‑sectional study in China [J ] . Support Care Cancer , 2023 , 31 ( 2 ): 106 .

American Psychiatric Association . Diagnostic and Statistical Manual of Mental Disorders [M ] . 5th ed . Washington DC : American Psychiatric Publishing , 2013 .

中华医学会儿科学分会神经学组 . 儿童抽动障碍诊断与治疗专家共识(2017实用版) [J ] . 中华实用儿科临床杂志 , 2017 , 32 ( 15 ): 1137 ‑ 1140 .

The Neurology Group of Chinese Pediatric Society, Chinese Medical Association . Expert consensus on diagnosis and treatment of tic disorder in children (2017 practical edition) [J ] . Chin J Appl Clin Pediatr , 2017 , 32 ( 15 ): 1137 ‑ 1140 .

LECKMAN J F , RIDDLE M A , HARDIN M T , et al . The Yale Global Tic Severity Scale: Initial testing of a clinician‑rated scale of tic severity [J ] . J Am Acad Child Adolesc Psychiatry , 1989 , 28 ( 4 ): 566 ‑ 573 .

忻仁娥 , 唐慧琴 , 张志雄 , 等 . 全国22个省市26个单位24 013名城市在校少年儿童行为问题调查: 独生子女精神卫生问题的调查, 防治和Achenbach's儿童行为量表中国标准化 [J ] . 上海精神医学 , 1992 , 4 ( 1 ): 47 ‑ 55 .

XIN R E , TANG H Q , ZHANG Z X , et al . Investigation on behavior problems of 24 013 school children in 26 units in 22 provinces and cities: Nationwide⁃investigation and prevention of mental health problems of the only child and standardization of Achenbach's children's behavior scale in China [J ] . Shanghai Arch Psychiatry , 1992 , 4 ( 1 ): 47 ‑ 55 .

MACCOBY E E , MARTIN J A . Socialization in the Context of the Family: Parent⁃Child Interactions [M ] // Mussen PH, Hetherington EM. Handbook of child psychology . 4th ed . NewYork, USA : Wiley , 1983 .

DEO RC . Machine learning in medicine [J ] . Circulation , 2015 , 132 ( 20 ): 1920 ‑ 1930 .

HARRINGTON P . Machine learning in action [M ] . Shelter Island, NY : Manning Publications Co. , 2012 .

GHANIZADEH A , MOSALLAEI S . Psychiatric disorders and behavioral problems in children and adolescents with Tourette syndrome [J ] . Brain Dev , 2009 , 31 ( 1 ): 15 ‑ 19 .

STEINBERG T , HARUSH A , BARNEA M , et al . Tic‑related cognition, sensory phenomena, and anxiety in children and adolescents with Tourette syndrome [J ] . Compr Psychiatry , 2013 , 54 ( 5 ): 462 ‑ 466 .

ACHENBACH T M , EDELBROCK C S . Behavioral problems and competencies reported by parents of normal and disturbed children aged four through sixteen [J ] . Monogr Soc Res Child Dev , 1981 , 46 ( 1 ): 1 ‑ 82 .

WANG Y J , LIU Y X . The development of internalizing and externalizing problems in primary school: Contributions of executive function and social competence [J ] . Child Dev , 2021 , 92 ( 3 ): 889 ‑ 903 .

BLACK K J , KIM S , YANG N Y , et al . Course of tic disorders over the lifespan [J ] . Curr Dev Disord Rep , 2021 , 8 ( 2 ): 121 ‑ 132 .

LOWE T L , CAPRIOTTI M R , MCBURNETT K . Long‑term follow‑up of patients with Tourette's syndrome [J ] . Mov Disord Clin Pract , 2019 , 6 ( 1 ): 40 ‑ 45 .

LI Y M , YAN X F , LI Q S , et al . Prevalence and trends in diagnosed ADHD among US children and adolescents, 2017‑2022 [J ] . JAMA Netw Open , 2023 , 6 ( 10 ): e2336872 .

DARYANANI I , HAMILTON J L , ABRAMSON L Y , et al . Single mother parenting and adolescent psychopathology [J ] . J Abnorm Child Psychol , 2016 , 44 ( 7 ): 1411 ‑ 1423 .

MAK K K , HO S Y , THOMAS G N , et al . Family structure, parent‑child conversation time and substance use among Chinese adolescents [J ] . BMC Public Health , 2010 , 10 : 503 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

多种机器学习模型预测膀胱癌肺转移预后

多重线性回归与机器学习在预测初诊乳腺癌患者癌症复发恐惧的比较

机器学习在心脏淀粉样变辅助诊断中的研究进展

基于生物信息学分析预测和验证实验性自身免疫性心肌炎的关键基因

人参皂苷防脱精华治疗雄激素性秃发的临床有效性和安全性研究