目的 分析人工智能(AI)在儿童神经发育障碍研究中的应用现状、热点主题及变化趋势。方法 检索Web of Science 核心合集中2010年1月1日—2026年5月31日发表的英文文献,采用Excel、VOSviewer和CiteSpace分析发文趋势、合作网络、关键词共现及突现,并基于题名、摘要和关键词进行多标签内容编码。结果 共纳入文献1 776篇。发文量总体呈上升趋势,2018年后增长较明显,2025年达到峰值(297篇)。美国(480篇)和中国(451篇)发文量居前两位。关键词共现显示,研究主题主要包括孤独症谱系障碍(ASD)早期筛查、社交行为识别与智能干预,以注意缺陷多动障碍(ADHD)为代表的脑功能评估与深度学习,以及学习/语言障碍识别与特征提取。关键词突现显示,深度学习、异质性、自然语言处理、严重程度评估、工作记忆和图神经网络等为近年研究热点。内容编码显示,研究对象以ASD(1 228篇,69.14%)和ADHD(430篇,24.21%)为主;应用方向中疾病识别/辅助诊断最常见(1 374篇,77.36%);数据来源包括脑影像、脑电图、视觉行为、语音语言、眼动、运动轨迹、行为量表及机器人/虚拟现实等交互资料;算法类型以传统机器学习(1 148篇,64.64%)和深度学习/神经网络类方法(670篇,37.73%)为主。结论 AI在儿童神经发育障碍领域的研究持续增加,研究内容以疾病识别与辅助诊断为主,并向早期筛查、多模态评估、智能干预、异质性识别和学习/语言障碍识别等方向拓展。未来应进一步开展多中心研究与验证,为儿童神经发育障碍早期识别、评估和干预提供更充分的证据支持。
Abstract
Objective To analyze the application status, research hotspots, and evolving trends of AI in childhood neurodevelopmental disorders. Methods English articles published from January 1, 2010 to May 31, 2026 were retrieved from the Web of Science Core Collection.Excel, VOSviewer, and CiteSpace were used to analyze publication trends, collaboration networks, keyword co-occurrence, and keyword bursts.Multi-label content coding was performed based on titles, abstracts, and keywords. Results A total of 1 776 articles were included.The annual number of publications increased overall, with marked growth after 2018 and a peak in 2025 (n=297).The United States (n=480) and China (n=451) were the leading contributors.Keyword co-occurrence analysis identified three major clusters: autism spectrum disorder(ASD) early screening, social behavior recognition, and intelligent intervention; attention-deficit/hyperactivity(ADHD)-related brain function assessment and deep learning; and learning/language disorder recognition and feature extraction.Keyword burst analysis identified deep learning, heterogeneity, natural language processing, severity assessment, working memory, and graph neural networks as recent hotspots.Content coding showed that studies mainly focused onASD (1 228, 69.14%) andADHD(430, 24.21%).Disease identification and assisted diagnosis (1 374, 77.36%) was the most common application.Data sources included neuroimaging, electroencephalogram, visual behavior, speech and language, eye tracking, movement trajectories, behavioral scales, and robot/virtual reality interaction.Traditional machine learning (1 148, 64.64%) and deep learning/neural network-based methods (670, 37.73%) were the predominant methods. Conclusions Research on AI in childhood neurodevelopmental disorders has continued to increase.Current studies mainly focus on diseaseidentification and assisted diagnosis, and have gradually extended to early screening, multimodal assessment, intelligent intervention, heterogeneity identification, and recognition of learning/language disorders. Further studies should strengthen multicenter research and validation, so as to provide evidence for early identification, assessment, and intervention of childhoodneurodevelopmental disorders.
关键词
人工智能 /
神经发育障碍 /
文献计量学 /
机器学习 /
深度学习 /
儿童
Key words
artificial intelligence /
neurodevelopmental disorders /
bibliometrics /
machine learning /
deep learning /
children
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] American Psychiatric Association.Diagnostic and statistical manual of mental disorders, fifth edition, text revision[M].Washington, DC: American Psychiatric Association Publishing, 2022.
[2] Shahini A, Kamath AP, Sharma E, et al.A systematic review for artificial intelligence-driven assistive technologies to support children with neurodevelopmental disorders[J].Inform Fusion, 2025, 124:103441.
[3] Jia SJ, Jing JQ, Yang CJ.A review on autism spectrum disorder screening by artificial intelligence methods[J].J Autism Dev Disord, 2025, 55(9): 3011-3027.
[4] Wang X, Jia Q, Liang L, et al.Artificial intelligence in ADHD: A global perspective on research hotspots, trends and clinical applications[J].Front Hum Neurosci, 2025, 19: 1577585.
[5] Donthu N, Kumar S, Mukherjee D, et al.How to conduct a bibliometric analysis: An overview and guidelines[J].J Bus Res, 2021, 133: 285-296.
[6] Thabtah F.Machine learning in autistic spectrum disorder behavioral research: A review and ways forward[J].Inform Health Soc Care, 2019, 44(3):278-297.
[7] Jiao Y, Chen R, Ke XY, et al.Predictive models of autism spectrum disorder based on brain regional cortical thickness[J].NeuroImage, 2010, 50(2):589-599.
[8] Heinsfeld AS, Franco AR, Craddock RC, et al.Identification of autism spectrum disorder using deep learning and the ABIDE dataset[J].NeuroImage Clin, 2018, 17: 16-23.
[9] Tang CG, Zheng WM, Zong Y, et al.Automatic identification of high-risk autism spectrum disorder: A feasibility study using video and audio data under the still-face paradigm[J].IEEE Trans Neural Syst Rehabil Eng, 2020, 28(11):2401-2410.
[10] Chen H, Song Y, Li XL.A deep learning framework for identifying children with ADHD using an EEG-based brain network[J].Neurocomputing, 2019, 356:83-96.
[11] Feng AC, Zhi DM, Feng Y, et al.Functional imaging derived ADHD biotypes based on deep clustering:A study on personalized medication therapy guidance[J].eClinicalMedicine, 2024, 77:102876.
[12] Nilsson Benfatto M, Öqvist Seimyr G, Ygge J, et al.Screening for dyslexia using eye tracking during reading[J].PLoS One, 2016, 11(12): e0165508.
[13] Beccaluva EA, Catania F, Arosio F, et al.Predicting developmental language disorders using artificial intelligence and a speech data analysis tool[J].Human-Computer Interaction, 2024,39(1-2): 8-42.
[14] Wen HW, Liu Y, Rekik I, et al.Combining disrupted and discriminative topological properties of functional connectivity networks as neuroimaging biomarkers for accurate diagnosis of early Tourette syndrome children[J].Mol Neurobiol, 2018, 55(4): 3251-3269.
[15] Letts E, King-Dowling S, Kwan MYW, et al.Machine learning derived physical activity in preschool children with developmental coordination disorder[J].Dev Med Child Neurol, 2025, 67(7): 910-917.
[16] Ahmed M, Hussain S, Ali F, et al.Summarizing recent developments on autism spectrum disorder detection and classification through machine learning and deep learning techniques[J].Appl Sci, 2025, 15(14):8056.
[17] 陈靓影.迈向精准化与普惠化: 人工智能赋能孤独症儿童康复教育的发展图景[J].中国听力语言康复科学杂志, 2026, 24(2):116-120.
Chen JY.Toward precision and inclusion: The development landscape of artificial intelligence-empowered rehabilitation education for children with autism spectrum disorder[J].Chin Sci J Hear Speech Rehabil,2026, 24(2):116-120.(in Chinese)
[18] Mottron L, Bzdok D.Autism spectrum heterogeneity: Fact or artifact?[J].Mol Psychiatry, 2020, 25(12): 3178-3185.
[19] Eslami T, Mirjalili V, Fong A, et al.ASD-DiagNet: A hybrid learning approach for detection of autism spectrum disorder using fMRI data[J].Front Neuroinform, 2019, 13: 70.
[20] Zou L, Zheng J, Miao C, et al.3D CNN based automatic diagnosis of attention deficit hyperactivity disorder using functional and structural MRI[J].IEEE Access, 2017, 5:23626-23636.
[21] Hu WY, Jiang GQ, Han JX, et al.Regional-asymmetric adaptive graph convolutional neural network for diagnosis of autism in children with resting-state EEG[J].IEEE Trans Neural Syst Rehabil Eng, 2024, 32:200-211.
[22] Moon HJ, Cho SB.Exploring implicit biological heterogeneity in ASD diagnosis using a multi-head attention graph neural network[J].J Integr Neurosci, 2024, 23(7): 135.
[23] Tariq Q, Daniels J, Schwartz JN, et al.Mobile detection of autism through machine learning on home video: A develo-pment and prospective validation study[J].PLoS Med, 2018, 15(11): e1002705.
[24] Kim HH, An JI, Park YR.A prediction model for detecting developmental disabilities in preschool-age children through digital biomarker-driven deep learning in serious games: Development study[J].JMIR Serious Games, 2021, 9(2):e23130.
[25] Rudovic O, Lee J, Dai M, et al.Personalized machine learning for robot perception of affect and engagement in autism therapy[J].Sci Robot, 2018, 3(19):eaao6760.
[26] Medina R, Bouhaben J, de Ramón I, et al.Electrophysiological brain changes associated with cognitive improvement in a pediatric attention deficit hyperactivity disorder digital artificial intelligence-driven intervention: Randomized controlled trial[J].J Med Internet Res, 2021, 23(11): e25466.
[27] Zhang TT, Zhou XY, Li X, et al.Noninvasive brain-computer interfaces for children with neurodevelopmental disorders: Attention deficit hyperactivity disorder and autism spectrum disorder[J].Displays, 2025, 86:102886.
基金
西安交通大学第二附属医院院基金管理项目[YJ(QN)202410]
PDF(2968 KB)
