目的 构建并评价孤独症谱系障碍(ASD)儿童合并智力功能缺陷的GBM预测模型,以期为该群体的早期筛查提供新视角。方法 2017年1月—2021年12月,选取浙江大学医学院附属儿童医院明确诊断为ASD的241名儿童纳入分析。本研究使用社会人口学与行为观察数据训练了GBM的预测模型,并与传统的Logistic回归(LR)对比。超参数调整使用网格搜索与十折交叉验证,特征选择使用交叉验证的LASSO方法,模型性能评价使用区分度与校准度。可解释性分析采用SHAP方法。结果 在241例ASD儿童中,98例(40.66%)合并智力功能缺陷。LASSO特征选择筛选出语言能力、母亲学历、行为观察时的年龄、刻板语言、指物和(或)姿势、主动表达社交意向的品质、不寻常感官兴趣、重复行为或刻板兴趣共计8个预测变量。特征选择前后的LR和GBM模型都能较好区分ASD儿童是否合并智力功能缺陷。特征选择后的GBM模型曲线下面积(AUC)(0.870, 95%CI:0.749~0.989)与传统LR(0.851, 95%CI: 0.704~0.921)接近;校准度方面,除全变量的LR校准度较差,其他模型均观测到了较好的校准度。特征重要性方面,语言能力是预测ASD儿童合并智力功能缺陷的第一重要特征。结论 特征选择后搭建的基于GBM的ASD儿童合并智力功能缺陷预测模型具有良好性能,具有一定的临床应用价值。
Abstract
Objective To construct and evaluate a Gradient Boosting Machines(GBM) prediction model for children with autism spectrum disorder (ASD) and comorbid intellectual impairment, so as to provide a new perspective for early screening of this population. Method From January 2017 to December 2021, 241 children with a clear diagnosis of ASD in the Children's Hospital, Zhejiang University School of Medicine were included in the analysis. The prediction model of GBM was trained using sociodemographic and behavioral observation data and compared with traditional Logistic regression (LR) in this study. Hyperparameter adjustment was performed using grid search with ten-fold cross-validation, feature selection methods were performed using cross-validation LASSO, and the performance of the model was evaluated using discrimination and calibration. Explainability analysis was evaluated using SHapley Additive exPlanation (SHAP). Results The sample totaled 241 children with ASD, of whom 98 (40.66%) had intellectual impairments. Eight predictor variables were screened by the LASSO method, including language ability, mother's education attainment, age at the time of behavioral observation, stereotyped speech, pointing/gestures, social quality, unusual sensory interest and repetitive stereotyped behaviors. Both LR and GBM models before and after feature selection were better at distinguishing whether children with ASD had combined intellectual impairment. The area under curve (AUC) of the GBM model after feature selection (0.870, 95%CI: 0.749 - 0.989) was close to that of the conventional LR (0.851, 95%CI: 0.704 - 0.921). Regarding calibration, good calibration was observed for all models, except for the poor calibration of the full variable LR. In terms of feature importance, language ability contributed the most to the prediction of combined intellectual functioning deficits in children with ASD. Conclusion The prediction model for children with ASD complicated with intellectual impairment constructed by the feature selection method of LASSO and the GBM model has good performance and some clinical value.
关键词
孤独症谱系障碍 /
智力功能缺陷 /
机器学习 /
预测模型
Key words
autism spectrum disorder /
intellectual impairment /
machine learning /
prediction model
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] American Psychiatric Association. The Diagnostic and Statistical Manual of Mental Disorders: DSM-V [M]. 5th ed. Washington D C: American Psychiatric Public Inc, 2013: 53-59.
[2] 赵亚楠,罗雅楠,王翔宇,等.中国2～6岁孤独症儿童家庭直接经济负担研究[J].中华疾病控制杂志,2021,25(9):1085-1090.
Zhao YN, Luo YN, Wang XY. Research on the direct financial burden on families with 2-6 years old children having autism spectrum disorder in China[J]. Chin J Dis Control Prev, 2021,25(9):1085-1090.
[3] Maenner MJ, Shaw KA, Bakian AV, et al. Prevalence and characteristics of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 sites, United States, 2018[J].MMWR Surveill Summ, 2021, 70(11):1-16.
[4] Maenner MJ, Shaw KA, Baio J, et al. Prevalence of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 sites, United States, 2016[J].MMWR Surveill Summ, 2020, 69(4):1-12.
[5] Tonnsen BL, Boan AD, Bradley CC, et al. Prevalence of autism spectrum disorders among children with intellectual disability[J].Am J Intellect Dev Disabil, 2016,121(6):487-500.
[6] Miller JS, Bilder D, Farley M, et al.Autism spectrum disorder reclassified: A second look at the 1980s Utah/UCLA autism epidemiologic study[J]. J Autism Dev Disord, 2013, 43(1):200-210.
[7] Taheri A, Meghdari A, Alemi M. et al. Clinical interventions of social humanoid robots in the treatment of a pair of high-and low-functioning autistic Iranian twins [J].Sci Iran, 2018, 25(3):1197-1214.
[8] Eldevik S, Titlestad KB, Aarlie H, et al. Community implementation of early behavioral intervention: Higher intensity gives better outcome [J]. Eur J Behav Anal, 2019,21(1):92-109.
[9] White SW, Oswald D, Ollendick T, et al. Anxiety in children and adolescents with autism spectrum disorders [J].Clin Psychol Rev, 2009, 29(3):216-229.
[10] Waugh C,Peskin J. Improving the social skills of children with HFASD: An intervention study [J]. J Autism Dev Disord, 2015, 45(9):2961-2980.
[11] Song C, Jiang ZQ, Hu LF, et al. A machine learning-based diagnostic model for children with autism spectrum disorders complicated with intellectual disability[J]. Front Psychiatry, 2022,13:993077.
[12] Song C, Jiang ZQ, Liu D, et al. Application and research progress of machine learning in the diagnosis and treatment of neurodevelopmental disorders in children [J]. Front Psychiatry, 2022, 13:960672.
[13] 李星珠,王献娜,张通.机器学习在孤独症谱系障碍疾病中的研究进展[J]. 中国康复,2022,37(4):244-247.
Li XZ, Wang XN, Zhang T. Research progress of machine learning in autism spectrum disorders[J]. Chinese Journal of Rehabilitation, 2022,37(4):244-247.
[14] 潘秀雨,李洪华,王冰,等. 孤独症谱系障碍儿童语言发育状况分析[J]. 教育生物学杂志,2021,9(4):262-265,295.
Pan XY, Li HH, Wang B, et al. Analysis of language development profile in children with autism spectrum disorder[J].Journal of Bio-education, 2021,9(4):262-265,295.
[15] Ellis Weismer S, Kover ST. Preschool language variation, growth, and predictors in children on the autism spectrum [J]. J Child Psychol Psychiatry, 2015, 56(12):1327-1337.
[16] 黎文倩,刘晓,代英,等.孤独症谱系障碍儿童的诊断年龄及其影响因素[J]. 中国当代儿科杂志,2018,20(10):799-803.
Li WQ, Liu X, Dai Y, et al. Age of diagnosis of autism spectrum disorder in children and factors influencing the age of diagnosis[J]. Chin J Contemp Pediatr,2018,20(10):799-803.
[17] Krakovich TM, McGrew JH, Yu Y, et al. Stress in parents of children with autism spectrum disorder:An exploration of demands and resources[J]. J Autism Dev Disord, 2016, 46(6):2042-2053.
[18] Zhou W, Liu D, Xiong X, et al. Emotional problems in mothers of autistic children and their correlation with socioeconomic status and the children's core symptoms [J]. Medicine (Baltimore), 2019, 98(32):e16794.
[19] 杨育林, 代英. 孤独症谱系障碍与智力障碍共患与鉴别研究进展[J]. 中国实用儿科杂志, 2022, 37(4): 302-307.
Yang YL, Dai Y. Research progress in autism spectrum disorder with intellectual impairment and its differential diagnosis [J] Chinese Journal of Practical Pediatrics, 2022, 37(4): 302-307.
基金
浙江省自然科学基金(LGF20H090015)
PDF(1554 KB)
