Using Machine Learning with Real-World Data to Identify Autism Risk in Children

使用机器学习和真实世界数据来识别儿童自闭症风险

• 批准号: 10591514
• 负责人: Amber M. Angell
• 金额: $ 20.16万
• 依托单位: UNIVERSITY OF SOUTHERN CALIFORNIA
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-03-14 至 2025-02-28
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10591514
• 关键词: Address; Age; Algorithms; American; Base Ratios; Centers for Disease Control and Prevention (U.S.); Child; Clinic; Clinical; Clinical Informatics; Code; Data; Data Set; Databases; Detection; Development; Diagnosis; Diagnostic; Electronic Health Record; Evaluation; Feeling suicidal; Florida; Foundations; Future; Goals; Health Services; Health system; Healthcare; International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10); Intervention; Latino; Location; Logistic Regressions; Los Angeles; Machine Learning; Measures; Methods; Modeling; National Institute of Mental Health; Natural Language Processing; Natural Language Processing pipeline; Notification; Outcome; Parents; Patients; Pediatric Hospitals; Performance; Physicians; Population; Public Health; ROC Curve; Randomized; Records; Reporting; Research; Research Personnel; Research Proposals; Review Literature; Risk; Risk Factors; Sampling; Site; Structure; Testing; Text; Training; Trees; Trust; Underserved Population; Validation; Work; aged; autism spectrum disorder; autistic children; clinical decision support; clinical diagnosis; cohort; computable phenotypes; cost effective; daily functioning; deep learning model; disorder risk; disparity reduction; economic impact; electronic structure; ethnic disparity; girls; gradient boosting; high reward; improved; indexing; machine learning model; phenotyping algorithm; phrases; predictive modeling; psychological distress; random forest; risk prediction model; sex; sex disparity; sociodemographics; structured data; support tools; support vector machine; surveillance network; tool; unstructured data

PROJECT SUMMARY/ABSTRACT Early and accurate identification of autism spectrum disorder (ASD) is important because ASD interventions can support positive long-term developmental outcomes, but there is a delay of >2 years between the age children can reliably be diagnosed and the average age of diagnosis; and 1 in 4 U.S. children aged 8 with ASD have not been diagnosed. Girls and Latino children are disproportionately impacted by the problem of delayed diagnosis and under-identification of ASD, in part because clinicians are less likely to recognize ASD risk factors in them and refer them for an ASD evaluation. Therefore, predicting ASD risk at a population level is needed to enhance early and accurate detection, particularly in these underserved populations. Researchers are beginning to harness clinical informatics methods to identify ASD from real-world data in electronic health records (EHRs), using both structured (e.g., diagnosis codes) and unstructured data (e.g., physician notes). However, existing algorithms suffer from multiple major flaws, including non-representativeness of training samples, outdated diagnosis codes and natural language processing (NLP) methods, and a lack of ‘verified’ ASD diagnosis in their gold standard datasets. This proposed research addresses these gaps by developing a contemporary ASD risk model that uses state-of-the-art machine learning and NLP methods. Using EHR data from Children’s Hospital Los Angeles (including a gold standard dataset with ‘verified’ ASD diagnoses from the Boone Fetter Clinic) and the OneFlorida Data Trust (a Florida state-wide EHR database), we will (1) develop a computable phenotype for ASD using both structured and unstructured EHR data (including parent-reported ASD discriminators and features associated with ASD that are often found in free text in children’s records), and (2) develop a machine-learning risk prediction model for ASD. This will lay the foundation for a clinical decision support tool, to be integrated into EHRs to notify a clinician when a child warrants ASD evaluation. This has potential to improve ASD identification in all children, but it may particularly benefit girls and Latino children, reducing sex and ethnic disparities. Further, it will be easily expandable into a ‘next steps’ study to the overall PCORnet, which provides healthcare to over 24 million children. By using EHRs, this proposal holds promise for future cost-effective health systems interventions that can help to correct a sociodemographic ‘imbalance’ in ASD research by reaching girls and Latino children at risk for ASD.

项目概要/摘要 早期准确识别自闭症谱系障碍 (ASD) 非常重要，因为 ASD 干预措施 可以支持积极的长期发展成果，但年龄之间存在 >2 年的延迟 儿童能够得到可靠的诊断以及诊断的平均年龄；四分之一的美国 8 岁儿童患有自闭症谱系障碍 (ASD) 女孩和拉丁裔儿童受到延误问题的影响尤为严重。 ASD 的诊断和识别不足，部分原因是新来者不太可能认识到 ASD 风险 因此，在人群水平上预测 ASD 风险是很重要的。 需要加强早期和准确的检测，特别是在这些服务不足的人群中。 开始利用临床信息学方法从电子健康的真实数据中识别 ASD 记录（EHR），使用结构化（例如诊断代码）和非结构化数据（例如医生笔记）。 然而，现有算法存在多个重大缺陷，包括训练的非代表性 样本、过时的诊断代码和自然语言处理（NLP）方法以及缺乏“经过验证的” 这项提出的研究通过开发一个黄金标准数据集中的自闭症谱系障碍诊断来解决这些差距。 当代 ASD 风险模型，使用最先进的机器学习和 NLP 方法，使用 EHR 数据。 来自洛杉矶儿童医院（包括黄金标准数据集，其中包含来自洛杉矶儿童医院的“经过验证的”自闭症谱系障碍诊断） Boone Fetter Clinic）和 OneFlorida Data Trust（佛罗里达州范围内的 EHR 数据库），我们将 (1) 开发一个 使用结构化和非结构化 EHR 数据（包括家长报告的 自闭症谱系障碍 (ASD) 鉴别器和与自闭症谱系障碍 (ASD) 相关的特征通常可以在儿童记录的自由文本中找到）， (2) 开发 ASD 的机器学习风险预测模型，这将为临床奠定基础。 决策支持工具，将集成到 EHR 中，以便在儿童需要进行 ASD 评估时通知临床医生。 这有可能改善所有儿童的自闭症谱系障碍识别，但它可能特别有利于女孩和拉丁裔 此外，它还可以轻松扩展到“下一步”研究。 总体而言，PCORnet 通过使用 EHR 为超过 2400 万儿童提供医疗保健，该提案成立。 承诺未来具有成本效益的卫生系统干预措施，有助于纠正社会人口统计问题 通过接触有患自闭症谱系障碍风险的女孩和拉丁裔儿童，自闭症谱系障碍研究中出现“不平衡”。