机器学习在儿童创伤后应激障碍识别及转归预测中的应用

创伤后应激障碍(PTSD)会给儿童发展带来负面效应, 其影响甚至延续至成年期。然而传统诊断方式难以做到快速、客观、准确的识别和诊断儿童PTSD, 机器学习作为一种处理大量变量和数据的新兴方法, 逐渐被应用到儿童PTSD的早期预测、识别及辅助诊断等研究中。机器学习凭借其性能、原理等方面的优势, 可被应用在儿童PTSD的识别与转归领域。相比自我报告式的诊断, 通过机器学习辅助识别和诊断儿童PTSD的过程具有效率高、客观准确、节约资源等独特优势。然而, 机器学习也在硬件成本、算法选择和预测准确度等方面存在局限性。未来研究人员需要进一步提高机器学习诊断识别儿童PTSD的准确率, 并将机器学习算法同传统诊断方法结合进行更多的探索和应用。     

人工智能辅助的自闭症早期患者的筛查与诊断

自闭症谱系障碍(Autistic Spectrum Disorders, ASD)的症状早在婴幼儿期就会显现, 越早发现, 越早干预, 治疗效果越好。传统自闭症早期筛查与诊断在评估方法、流程上存在局限, 无法满足大规模筛查和诊断需求。随着人工智能技术的快速发展, 使用智能化方法进行自闭症早期大规模无感筛查与诊断逐渐成为可能。近10年间, 国内外对自闭症智能化识别方法的探索在经典任务行为、面部表情和情绪、眼动、脑影像、运动控制和运动模式、多模态6个领域积累了丰富的研究成果。未来研究应围绕构建国内自闭症早期智能医学筛查与诊断体系, 开发针对婴幼儿患者的筛查工具, 构建融合多模态数据的自闭症婴幼儿智能化识别模型, 建立结合脑影像技术的自闭症精细化诊断方法等方面来开展。     

机器学习在抑郁症领域的应用

抑郁症患者疾病意识的不足以及早期筛查方法的缺乏导致患者在被诊断时大多已发展至重性抑郁障碍。为改善现状, 近年来机器学习被逐渐应用到抑郁症的早期预测、早期识别、辅助诊断和治疗决策中。在应用中, 机器学习模型准确性的影响因素包括样本集种类及规模、特征工程、算法类型等。建议未来将机器学习进一步融入医疗健康系统及移动应用程序等, 不断优化机器学习模型, 通过充分挖掘患者健康数据来改善抑郁症的预防、识别、诊断和治疗等相关问题。     

关于自闭症的临床、实验心理学的研究

最近,国际上的临床心理学、实验心理学对自闭症的诊断、分类、发生率以及教育与治疗等,从各个角度展示了许多新的研究结果。实验心理学的调查结果表明,自闭症儿童的注意力异常,对视、听、触等感觉反应处理样式,不同于弱智儿童和正常儿童。语言障碍是自闭症儿童发展障碍中的核心症侯,这与自闭症在社会性和人际关系上存在障碍有着密切的关系。进入９０年代以后,对自闭症“心的理论”的各种实验研究,引起各国研究者的注目和争论     

自闭症谱系障碍的早期筛查工具

基于35项实证研究, 遵循诊断准确性研究的质量评估(QUADAS-2)的循证程序对自闭症谱系早期预警特征筛查工具开展质量评估, 包括婴儿阶段(10个筛查工具, 159388人)、幼儿阶段(14项筛查工具, 11712人), 旨在为自闭症谱系风险的早期识别提供全面的理解。结果表明：幼儿阶段的筛查工具的分类准确性要高于婴儿阶段, 在婴儿阶段达到良好水平的筛查工具有M-CHAT-R/F、PDQ-1; 在幼儿阶段达到优秀水平的筛查工具有OERA、TIDOS。其中, “改良的幼儿自闭症检查表-有修订的后续随访” (M-CHAT-R/F)是目前最具潜力的自闭症谱系风险筛查工具之一。最后, 我们探讨了应用QUADAS-2评估研究质量的局限性, 强调需要规范工具的质量评估标准与进一步验证研究的必要性。     

儿童自闭症的历史、现状及其相关研究

Kanner于1943年最先报告了关于自闭症的详细研究.自闭症作为一种症候群,其具体症状因人而异.社会性障碍、交流障碍与想象障碍是自闭症的三个主要障碍,同时又构成诊断自闭症的基础.研究者已经弄清了自闭症的发生率、性别差异、与智商和社会阶层的关系等部分事实.不同的研究者从不同方面解释和说明自闭症,其中“心理理论”说是近年来非常有吸引力的自闭症理论之一.这一理论主张自闭症在“读心理”方面具有障碍.     

基于社交媒体数据的心理指标识别建模:机器学习的方法

心理指标识别建模是基于海量数据结合计算机机器学习算法识别心理特征的一种新兴方式。由于传统纸笔测量方式所存在的诸多限制,本文对基于社交媒体数据的心理建模方法及应用于心理测量的可行性进行综述,介绍了特征及提取方法、常用机器学习算法以及应用场景,并对心理指标识别建模的优势和不足进行了总结与展望。该测量方法基于社交媒体数据,相比自我报告法具有时效性高、可回溯测量、生态效度好等独特优势。然而,基于社交媒体的心理指标识别建模方法也在学习成本、硬件成本等方面存在局限性。未来研究人员需要进一步探索社会媒体信息与用户心理变量间的关联机制,并将心理指标识别模型同传统心理学研究方法结合进行更多的探索和应用。心理指标识别建模结合心理测量基本原理和计算机领域机器学习的技术,将为心理学研究打开一扇新的大门。     

汉语自闭症儿童分类能力的实验研究

为探查自闭症儿童的分类能力,采用类别判断任务,比较了刺激的典型性对12名汉语自闭症儿童和相匹配控制组儿童分类正确率和反应时的影响。研究发现,自闭症组在典型刺激上的分类能力与控制组大致相同,但在不典型刺激上的分类能力明显更差,表明自闭症儿童既有一定的分类能力,又存在内在的缺陷。研究结果支持自闭症"弱的中心统合理论"。     

机器学习在发展性阅读障碍儿童早期筛查中的应用

发展性阅读障碍严重影响儿童的学业成就、心理健康和社会适应能力。近年来,机器学习因其强大的数据处理和挖掘能力逐渐被应用到阅读障碍儿童的早期筛查中,在标准化心理教育测试、眼动追踪、游戏测试、脑成像等多个领域积累了较为丰富的成果,获得了更加精准高效、灵活可靠的分类结果。然而,机器学习在对象选取、数据采集、转化潜力和安全隐私等方面仍存在局限性。未来研究需要重点关注学龄前阅读障碍儿童的早期筛查系统的科学性,同时积极构建多模态数据库、在多种算法中寻找最佳算法以获取最优参数,最终实现临床实践中的广泛使用。     

五官完整度对自闭症儿童面孔表情性别识别的影响

摘 要 为探讨自闭症儿童在不同面孔遮蔽部位和不同面孔表情下的性别识别能力,本研究选取自闭症、智力障碍和普通儿童各21名对240张面孔表情图片进行性别识别。结果发现,自闭症儿童的性别识别正确率显著低于普通儿童,且表现出对中性和高兴表情的识别优于生气表情,同时在遮蔽眼睛时正确率最低。研究认为,自闭症儿童的总体性别识别能力偏低,且受面孔遮蔽部位和表情类型的影响显著,在面孔表情加工过程中体现出较为明显的局部加工特征。     

Does DSM-IV Asperger's Disorder Exist?

DSM-IV criteria for autistic and Asperger's disorders were applied to 157 children with clinical diagnoses of autism or Asperger's disorder. All children met the DSM-IV criteria for autistic disorder and none met criteria for Asperger's disorder, including those with normal intelligence and absence of early speech delay. The reason for this was that all children had social impairment and restricted and repetitive behavior and interests (required DSM-IV symptoms for both autistic and Asperger's disorders) and all had a DSM-IV communication impairment (which then qualified them for a diagnosis of autistic disorder and not Asperger's disorder). Communication problems exhibited by all children were impaired conversational speech or repetitive, stereotyped, or idiosyncratic speech (or both), which are DSM-IV criteria for autism. These findings are consistent with those of 5 other studies and indicate that a DSM-IV diagnosis of Asperger's disorder is unlikely or impossible.     

Assessment and Diagnosis of Autism Spectrum Disorder in Hanoi,Vietnam

For individuals with autism spectrum disorder (ASD), early and accurate diagnosis enables early intervention, which leads to improved quality of life. The diagnostic label of ASD has only been recognized in Vietnam since 2000 and the number of children who have been diagnosed has increased dramatically. However, there is limited understanding about the condition. This article aims to describe the current assessment and diagnostic practices in Hanoi, Vietnam, a low resource setting. Data for this article came from in-depth interviews with parents of children with ASD and service providers, and observation at health clinics and early intervention centers that provide ASD assessment during field work from July 2011 to May 2012. This study reveals that although there is some improvement in assessment services, parents still face a number of challenges in accessing quality assessment and diagnosis of ASD in Vietnam. These include rushed and perfunctory assessment; a lack of clinical guidelines for assessment; limited communication among health professionals in assessment teams, and between professionals and parents of children with ASD. Public awareness-raising of the signs of autism, capacity building for professionals, development of practical and contextual appropriate assessment protocols are needed to enhance the accessibility and quality of assessment services.     

自闭症儿童依恋研究述评

摘要　　研究者曾一度认为自闭症儿童无法建立起依恋关系。本文系统回顾了自闭症儿童依恋研究的最新进展,将自闭症儿童是否存在依恋行为（尤其是安全型依恋行为）、依恋发展解释理论、研究方法、类型分布以及影响因素等问题进行了详细梳理,并结合研究分析,提出有关自闭症儿童依恋研究的几点启示,以供参考。     

Differential rates of skill acquisition and outcomes of early intensive behavioral intervention for autism

Intensive behavioral intervention for very young children with autism has received increased attention in recent years. Researchers have documented unprecedented success in educating some young children with autism, although not every child makes dramatic developmental gains. It might be useful to identify early in treatment those children who will benefit most from the current methodology and who might require slight variations in instructional format or curricular focus. The present study suggests that initial learning rates are moderately correlated with treatment outcomes after two years. Among 20 children receiving early, intensive behavioral intervention, initial acquisition of skills was correlated with later learning rates, severity of autism symptomatology and adaptive behavior profiles two years into treatment. Implications are discussed, especially in light of the universal need for intensive intervention in this population. Copyright © 1999 John Wiley & Sons, Ltd.     

Cognitive abilities and disabilities in infantile autism: A review

The literature on various aspects of learning and performance in autistic children is reviewed and interpreted as indicating very little that is specific to autism. Inadequate and inconsistent methodology precludes generalizations concerning the nature of the disorder. It is suggested that future research should be particularly concerned with controlling for developmental influences on performance, and with investigation of the higher functioning autistic children who are less governed by retardation factors. The current evidence is considered to support a hypothesis concerning abnormal hemisphere functioning in this group of children.     

Joint attention and children with autism: a review of the literature

Preverbal communication and joint attention have long been of interest to researchers and practitioners. Both attending to social partners and sharing attentional focus between objects or events and others precede the onset of a child's first lexicon. In addition, these prelinguistic acts also appear to have important implications with regard to learning to socialize. The construct of joint attention has been noted as an early developing area prior to the transition to symbolic communication. Thus, the importance of joint attention in typically developing children, and the lack thereof in children with autism, has interested researchers for use in diagnosis and intervention for autism. That is, joint attention has been gaining momentum as an area that not only helps characterize children with autism, but also as a prognostic indicator and a potential intervention goal. In this paper, the status of the literature about initiation of joint attention by young typically developing children and young children with autism was examined. Empirical studies regarding joint attention behaviors, including eye gaze alternation, the use of protodeclaratives and protoimperatives, and studies that investigated joint attention as a predictor of language acquisition were reviewed. Possible areas for future research for children with autism are discussed.     

A Conceptual Toolkit for Intensive Early Behavioral Intervention Teachers

Intensive early behavioral intervention (IEBI) for children with autism requires effective team work involving professional therapists, parents, teachers and others. It is important for all team members to employ a common conceptual and procedural vocabulary so that communication among them is clear, unambiguous, and effective. This article outlines the rationale of a small and practical vocabulary or conceptual toolkit that we have found useful in working with teachers and families in designing, implementing, and analyzing the details of behavioral programs to help autistic children learn to speak and understand language. Special attention is given to a heuristic device called the language matrix. Illustrative data describing the course of generative language acquisition by a child with autism taught according to the language matrix approach are included.