Methods Core

方法核心

基本信息

批准号：
10475475
负责人：
PATRICK J HEAGERTY
金额：
$ 29.9万
依托单位：
UNIVERSITY OF WASHINGTON
依托单位国家：
美国
项目类别：
财政年份：
2017
资助国家：
美国
起止时间：
2017-09-21 至 2027-07-31
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10475475
关键词：
Adopted Algorithms Area Artificial Intelligence Automobile Driving Biomedical Research Biometry Characteristics Clinical Clinical Data Clinical Informatics Clinical Research Clinical Services Clinical Trials Code Cohort Studies Collaborations Collection Communities Data Data Collection Data Coordinating Center Data Science Development Diagnosis Diagnostic Disease Electronic Health Record Fostering Foundations Funding Goals Health Services Research Health Technology Heterogeneity Informatics Information Retrieval International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10)Lead Leadership Learning Machine Learning Measurement Medical Imaging Methodology Methods Mission Modernization Monitor Musculoskeletal Musculoskeletal Diseases National Institute of Arthritis and Musculoskeletal and Skin Diseases Natural Language Processing Observational Study Outcome Patient Outcomes Assessments Patients Phase Phenotype Positioning Attribute Prevention Procedures Process Radiology Specialty Randomized Randomized Clinical Trials Reporting Reproducibility Research Research Design Research Methodology Role Secure System Telephone Text United States National Institutes of Health Universities Use Effectiveness Validation Washington algorithm development clinical center clinical development cluster trial comparative effectiveness convolutional neural network data quality data science infrastructure design electronic structure flexibility health care delivery high dimensionality image processing innovation learning algorithm mHealth machine learning method novel pragmatic trial quality assurance randomized trial research data dissemination statistical and machine learning statistics treatment effect trial design unstructured data

项目摘要

The goal of the Methodologic Core is to support high-impact clinical research focusing on the diagnosis, treatment, and prevention of musculoskeletal diseases. Specifically, the Methodologic Core will provide the contemporary data science infrastructure necessary for studies that leverage electronic health records (EHR), and for trials that collect patient reported outcomes. We will organize around two major themes encompassing multiple activities: confirmatory learning which includes design and analysis of observational studies and randomized trials; and predictive learning which leverages contemporary machine learning and artificial intelligence for processing electronic health records data and other high-dimensional information. The data sciences are driving innovation in clinical research with the linkage of multiple EMR systems opening new opportunities for large-scale clinical learning. Ultimately rigorous research demands new methods for the extraction of reliable information and requires the design of trials that can tailor to the unique characteristics of patients and the context of health care delivery systems. The University of Washington Center for Biomedical Statistics (CBS) has provided informatics and statistical leadership to multiple observational and randomized studies focusing on musculoskeletal disorders and is uniquely positioned to provide the foundation for expanded impact within the broad scope of the NIAMS mission.

方法核心的目标是支持高影响力的临床研究，重点关注肌肉骨骼疾病的诊断、治疗和预防。具体来说， Methodologic Core 将提供必要的当代数据科学基础设施利用电子健康记录 (EHR) 的研究以及收集患者报告的试验结果。我们将围绕两大主题组织多项活动：验证性学习，包括观察性研究的设计和分析随机试验；以及利用当代机器学习和预测学习用于处理电子健康记录数据和其他高维数据的人工智能信息。数据科学正在推动临床研究的创新多个 EMR 系统为大规模临床学习带来了新的机会。最终严格的研究需要新的方法来提取可靠的信息，并且需要可以根据患者的独特特征和背景进行的试验设计医疗保健提供系统。华盛顿大学生物医学统计中心（哥伦比亚广播公司）为多个观察和统计提供了信息学和统计领导专注于肌肉骨骼疾病的随机研究，具有独特的定位，可以提供为在 NIAMS 使命的广泛范围内扩大影响奠定基础。