Efficient Statistical Learning Methods for Personalized Medicine Using Large Scale Biomedical Data

使用大规模生物医学数据进行个性化医疗的高效统计学习方法

• 批准号: 9891071
• 负责人: Yuanjia Wang
• 金额: $ 32.89万
• 依托单位: UNIV OF NORTH CAROLINA CHAPEL HILL
• 依托单位国家: 美国
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-04-01 至 2022-03-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9891071
• 关键词: Academic Medical Centers; Address; Adverse event; Algorithms; Benefits and Risks; Categories; Center for Translational Science Activities; Classification; Clinical; Collaborations; Collection; Complement; Complex; Computer software; Data; Data Collection; Databases; Decision Making; Dependence; Dimensions; Documentation; Electronic Health Record; Ensure; Equilibrium; Exclusion Criteria; Formulation; Gaussian model; Goals; Health; Health Status; Healthcare; Healthcare Systems; Heterogeneity; Indiana; Informatics; International; Knowledge; Laboratories; Learning; Length; Measurement; Measures; Medical; Methodology; Methods; Modeling; Nature; Non-Insulin-Dependent Diabetes Mellitus; Outcome; Patient Care; Patient-Focused Outcomes; Patients; Pattern; Pharmaceutical Preparations; Process; Quality Control; Randomized Controlled Trials; Records; Research; Risk; Safety; Sampling; Structure; Testing; Time; Treatment Protocols; adaptive learning; adverse event risk; algorithmic methodologies; analytical tool; base; big biomedical data; clinical data warehouse; clinical decision-making; clinical encounter; clinical practice; comorbidity; complex data; data modeling; data space; design; evidence base; feature extraction; heterogenous data; individual patient; individualized medicine; knowledge base; learning strategy; machine learning method; novel; optimal treatments; outreach; patient population; personalized decision; personalized medicine; population based; scale up; statistical learning; temporal measurement; theories; treatment effect; treatment guidelines; treatment response; treatment strategy

Project Summary/Abstract Current medical treatment guidelines largely rely on data from randomized controlled trials that study average effects, which may be inadequate for making individualized decisions for real-world patients. Large-scale electronic health records (EHRs) data provide unprecedented opportunities to optimize personalized treatment strategies and generate evidence relevant to real-world patients. However, there are inherent challenges in the use of EHRs, including non-experimental nature of data collection processes, heterogeneous data types with complex dependencies, irregular measurement patterns, multiple dynamic treatment sequences, and the need to balance risk and benefit of treatments. Using two high-quality EHR databases, Columbia University Medical Center's clinical data warehouse and the Indiana Network for Patient Care database, and focusing on type 2 diabetes (T2D), this proposal will develop novel and scalable statistical learning approaches that overcome these challenges to discover optimal personalized treatment strategies for T2D from real-world patients. Specifically, under Aim 1, we will develop a unified framework to learn latent temporal processes for feature extraction and dynamic patient records representation. Our approach will accommodate large-scale variables of mixed types (continuous, binary, counts) measured at irregular intervals. They extract lower-dimensional components to reflect patients' dynamic health status, account for informative healthcare documentation processes, and characterize similarities between patients. Under Aim 2, we will develop fast and efficient multi-category machine learning methods, in order to evaluate treatment propensities and adaptively learn optimal dynamic treatment regimens (DTRs) among the extensive number of treatment options observed in the EHRs. The methods will provide sequential decisions that determine the best treatment sequence for a T2D patient given his/her EHRs. Under Aim 3, we will develop statistical learning methods to assist multi-faceted treatment decision-making, which balances risks versus benefits when evaluating a DTR. Our approach will ensure maximizing benefit to the greatest extent while controlling all risk outcomes under the safety margins. For all aims, we will develop efficient stochastic resampling algorithms to scale up the optimization for massive data sizes. We will identify optimal DTRs for T2D using the extracted information from patients' comorbidity conditions, medications, and laboratory tests, as well as records-collection processes. Our methodologies will be applied and cross-validated between the two EHR databases. The treatment strategies learned from the representative EHR databases with a diverse patient population will be beneficial for individual patient care, assisting clinicians to adaptively choose the optimal treatment for a patient. Finally, we will disseminate our methods and results through freely available software and outreach to the informatics and clinical experts at our Centers for Translational Science and elsewhere.

项目摘要/摘要 目前的医疗指南在很大程度上依赖于随机对照试验的数据，这些试验研究 平均效果，这可能不足以为现实世界的患者做出个性化的决定。大规模 电子健康记录(EHR)数据为优化个性化治疗提供了前所未有的机会 策略和生成与真实世界患者相关的证据。然而，存在着内在的挑战 使用EHR，包括数据收集过程的非试验性、异类数据类型和 复杂的依赖关系、不规则的测量模式、多个动态治疗序列以及需求 平衡治疗的风险和收益。使用两个高质量的电子病历数据库，哥伦比亚大学医学院 中心的临床数据仓库和印第安纳患者护理网络数据库，并专注于类型2 糖尿病(T2D)，这项提议将开发新的和可扩展的统计学习方法，以克服这些 从现实世界的患者中寻找针对T2D的最佳个性化治疗策略的挑战。具体来说， 在目标1下，我们将开发一个统一的框架来学习潜在的时间过程以进行特征提取和 动态患者记录表示。我们的方法将适应混合类型的大规模变量 (连续的，二进制的，计数)以不规则的间隔测量。它们提取较低维度的分量以反射 患者的动态健康状态，说明信息丰富的医疗记录流程，并确定 病人之间的相似之处。在目标2下，我们将开发快速高效的多类别机器学习 方法评价治疗倾向，自适应学习最优动态治疗方案 在电子病历中观察到的大量治疗选择中有(DTRs)。这些方法将提供 为给定EHR的T2D患者确定最佳治疗顺序的顺序决策。在AIM下 3、我们将开发统计学习方法，以辅助多方面的治疗决策，平衡 评估DTR时的风险与收益。我们的方法将确保最大限度地实现效益最大化。 同时将所有风险结果控制在安全边际之内。对于所有目标，我们将开发有效的随机性 重采样算法，以扩大海量数据大小的优化。我们将确定T2D的最佳DTR 使用从患者的合并症情况、药物和实验室测试中提取的信息 作为记录收集过程。我们的方法将在两个电子健康记录之间应用和交叉验证 数据库。从具有代表性的不同患者的电子病历数据库中学习的治疗策略 人口将有利于个别患者的护理，帮助临床医生自适应地选择最佳治疗方案 对一个病人来说。最后，我们将通过免费提供的软件和外联活动来传播我们的方法和结果 给我们转化科学中心和其他地方的信息学和临床专家。