Method Core

方法核心

• 批准号: 10707147
• 负责人: Erik Allen Imel
• 金额: $ 23.83万
• 依托单位: INDIANA UNIV-PURDUE UNIV AT INDIANAPOLIS
• 依托单位国家: 美国
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-09-19 至 2027-07-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10707147
• 关键词: Back; Big Data; Biology; Blood specimen; Characteristics; Clinical; Clinical Research; Clinical Trials Design; Code; Communities; Complex; Consultations; Data; Data Mart; Data Science; Data Set; Data Sources; Development; Diagnosis; Disease; Disparate; Electronic Health Record; Environmental Risk Factor; Foundations; Fracture; Functional disorder; Funding; Genomics; Genotype; Goals; Grant; Health; Human; Image; Indiana; Individual; Informatics; Information Systems; Interdisciplinary Study; Laboratories; Link; Longevity; Measurement; Measures; Mediating; Metastatic Neoplasm to the Bone; Methodology; Methods; Molecular; Morphology; Muscle; Musculoskeletal; Musculoskeletal Diseases; Natural Language Processing; Osteoporosis; Outcome; Participant; Patient Care; Patient Outcomes Assessments; Patients; Pharmaceutical Preparations; Phenotype; Physical Function; Physical Performance; Population; Prospective Studies; Proteomics; Public Health Informatics; Research; Research Design; Research Personnel; Resource Informatics; Resources; Retrospective Studies; Risk Factors; Science; Secure; Services; Source; Specimen; Standardization; Statistical Data Interpretation; Systems Biology; Technology; Text; Tissue Sample; Training; Translating; Translational Research; World Health; artificial intelligence method; bench to bedside; biobank; bone; bone imaging; clinical center; clinical phenotype; cohort; collaborative approach; computable phenotypes; data integration; data quality; data standards; design; disability; diverse data; electronic health information; electronic health record system; epidemiology study; exome sequencing; falls; genomic data; image processing; improved; innovation; machine learning method; machine learning model; metabolomics; multidisciplinary; multiple data sources; musculoskeletal injury; new technology; novel; predict clinical outcome; public health intervention; query tools; real world application; sarcopenia; social; social determinants; social factors; social health determinants; structural health determinants; success; tool

Musculoskeletal (MSK) injuries and diseases are the leading cause of disability across the lifespan. The pathophysiology of MSK disease is complex, and clinical research requires a broad range of data types including electronic health record (EHR) data (patient characteristics, treatments, and clinical outcomes), standardized physical performance measures, bone imaging, laboratory measures, omics data (e.g. genomics, proteomics, metabolomics), patient reported outcome tools, and social determinants of health. A lack of integration of these complex ‘big data’ sources impedes access and utilization by investigators. The goal of the Indiana Center for Musculoskeletal Health-Clinical Research Center is to provide access to state-of-the-art informatics resources and technology to define and characterize MSK biology and disease genotypes; computable, molecular, functional, and clinical phenotypes; and the social factors, collectively mediating MSK health, disease and disability across the lifespan, and to develop cures. In our initial P30 funding period, we established a Musculoskeletal Informatics Methodology (MIM) Core, leveraging a large inter-organizational integrated health information exchange of EHR data to detect several computable (assessed from the EHR) MSK phenotypes, that incorporate diagnosis and other codes, medications, laboratory values and/or clinical text notes. We integrated such computable phenotypes with the measurements (physical performance and bone assessments) and associated biospecimens from the ICMH-CRC FIT Core, to support investigators in clinical study design, feasibility assessments, and translating bench discoveries to humans. In the current proposal, the MIM Core will expand our support of investigators in multi-disciplinary clinical and translational research leveraging the use of novel informatics methods and resources, to improve musculoskeletal health. The MIM Core will develop a large facile, secure Musculoskeletal Data Mart that integrates and makes ready for clinical research, data from multiple resources including: 1) EHR systems of the Indiana Network for Patient Care, using computable phenotypes to generate a cohort of more than 880,000 patients with MSK disorders, 2) community data systems information on social determinants of health and other risk factors, and 3) the full data from the FIT Core musculoskeletal phenotyping, biospecimens, and genomic data. This Data Mart will facilitate feasibility analyses, grant support, and retrospective and prospective studies to support translational research from bench to bedside and back. The MIM Core will also provide statistical support for clinical trial design, and informatics and data science approaches such as applying natural language processing to extract EHR text data, health informatics analyses of outcomes, integrative genomic analyses, and applying machine learning models to predict clinical outcomes. These new and innovative initiatives will link the researchers of the Indiana Center for Musculoskeletal Health to state-of-the-art resources, advancing musculoskeletal science to ultimately benefit public and individual health and mobility.

肌肉骨骼损伤和疾病是导致终生残疾的主要原因。这个 MSK病的病理生理学是复杂的，临床研究需要广泛的数据类型，包括 电子健康记录(EHR)数据(患者特征、治疗和临床结果)，标准化 体能测量、骨骼成像、实验室测量、组学数据(例如基因组学、蛋白质组学、 代谢学)、患者报告的结果工具和健康的社会决定因素。缺乏对这些的整合 复杂的大数据来源阻碍了调查人员的访问和利用。印第安纳中心的目标是 肌肉骨骼健康-临床研究中心将提供最先进的信息学 用于定义和表征MSK生物学和疾病基因类型的资源和技术；可计算， 分子、功能和临床表型；以及共同调节MSK健康、疾病的社会因素 和一生中的残疾，并开发治疗方法。在我们最初的P30资助期，我们建立了 肌肉骨骼信息学方法(MIM)核心，利用大型跨组织集成 EHR数据的健康信息交换以检测几个可计算的(从EHR评估的)MSK 表型，包括诊断和其他代码、药物、化验值和/或临床文本注释。 我们将这些可计算的表型与测量结果(身体性能和骨骼)结合起来 评估)和来自ICMH-CRC FIT核心的相关生物检验品，以支持临床研究人员 研究设计，可行性评估，并将长凳发现转化为人类。在目前的提案中， MIM Core将扩大我们在多学科临床和翻译研究方面对研究人员的支持 利用新的信息学方法和资源，改善肌肉骨骼健康。MIM CORE将开发一个大型、便捷、安全的肌肉骨骼数据集市，集成并为临床做好准备 研究，来自多个资源的数据，包括：1)印第安纳州患者护理网络的电子病历系统，使用 可计算的表型产生超过88万名MSK障碍患者的队列，2)社区 关于健康的社会决定因素和其他风险因素的数据系统信息，以及3)来自FIT的完整数据 核心肌肉骨骼表型、生物标本和基因组数据。这个数据集市将促进可行性 分析、资助、回顾和前瞻性研究，以支持BASCH的转化性研究 在床边和床后。MIM Core还将为临床试验设计和信息学提供统计支持 以及数据科学方法，如应用自然语言处理来提取EHR文本数据、健康 结果的信息学分析，综合基因组分析，并应用机器学习模型 预测临床结果。这些新的和创新的举措将把印第安纳中心的研究人员联系起来 将最先进的资源转化为肌肉骨骼健康，推动肌肉骨骼科学最终受益 公共和个人健康和流动性。