Multi-objective representation learning methods for interpetable predictions of patient outcomesusing electronic health records

使用电子健康记录对患者结果进行可重复预测的多目标表示学习方法

• 批准号: 10477327
• 负责人: William La Cava
• 金额: $ 23.66万
• 依托单位: BOSTON CHILDREN'S HOSPITAL
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10477327
• 关键词: Address; Affect; Archives; Area; Automobile Driving; Cardiovascular Diseases; Categories; Clinical; Communities; Complex; Couples; Data; Data Reporting; Data Set; Databases; Development; Disease; Electronic Health Record; Engineering; Feedback; Goals; Graph; Hand; Health; Health Sciences; Health system; Heart Diseases; Heart failure; Hospitals; Inpatients; Knowledge; Learning; Machine Learning; Measurement; Measures; Mentors; Methodology; Methods; Mining; Modeling; Nature; Outcome; Pathology; Patient Care; Patient-Focused Outcomes; Patients; Pennsylvania; Performance; Pharmaceutical Preparations; Phase; Population; Process; Property; Protocols documentation; Quality of life; Recommendation; Replacement Arthroplasty; Research; Research Personnel; Risk; Structure; Techniques; Testing; Text; Time; Training; Translating; University Hospitals; Visit; Work; archive data; archived data; base; care costs; cluster computing; data complexity; deep learning; deep neural network; design; disease diagnosis; disorder subtype; gradient boosting; heart disease risk; high dimensionality; hospital readmission; improved; insight; learning strategy; machine learning method; machine learning model; network architecture; neural network; novel; open source tool; operation; patient health information; point of care; predictive modeling; readmission rates; readmission risk; statistical and machine learning

Project Summary/Abstract This project proposes new methods for representing data in electronic health records (EHR) to improve pre- dictive modeling and interpretation of patient outcomes. EHR data offer a promising opportunity for advancing the understanding of how clinical decisions and patient conditions interact over time to influence patient health. However, EHR data are difficult to use for predictive modeling due to the various data types they contain (con- tinuous, categorical, text, etc.), their longitudinal nature, the high amount of non-random missingness for certain measurements, and other concerns. Furthermore, patient outcomes often have heterogenous causes and re- quire information to be synthesized from several clinical lab measures and patient visits. The core challenge at hand is overcoming the mismatch between data representations in the EHR and the assumptions underly- ing commonly used statistical and machine learning (ML) methods. To this end, this project proposes novel wrapper-based methods for learning informative features from EHR data. Both methods propose specialized operators to handle sequential data, time delays, and variable interactions, and have the capacity to discover underlying clinical rules/decisions that affect patient outcomes. Importantly, both methods also produce archives of possible models that represent the best trade-offs between complexity and accuracy, which assists in model interpretation. These method advances are made possible by encoding a rich set of data operations as nodes in a directed acyclic graph, and optimizing the graph structures using multi-objective optimization. The central hypothesis of this research is that multi-objective optimization can learn effective data representations from the EHR to produce accurate, explanatory models of patient outcomes. Preliminary work has shown that these methods can effectively learn low-order data representations that improve the predictive ability of several state- of-the-art ML methods. This technique demonstrates good scaling properties with high-dimensional biomedical data. Aim 1 (K99) is to develop a multi-objective feature engineering method that pairs with existing ML methods to iteratively improve their performance by constructing new features from the raw data and using feedback from the trained model to guide feature construction. In Aim 2 (K99), this method is applied to form predictive models of the risk of heart disease and heart failure using longitudinal EHR data. The resultant models will be inter- preted with the help of mentors in order to translate predictions into clinical recommendations. For Aim 3 (R00), a second method is proposed that uses a similar framework to optimize existing neural network approaches in order to simplify their structure as much as possible while maintaining accuracy. The goal of Aim 4 (R00) is to identify hospital patients who are at risk of readmission and propose point-of-care strategies to mitigate that risk. This goal is facilitated through the application of the proposed methods to patient data collected from the Hospital of the University of Pennsylvania, the Geisinger Health System, and publicly available EHR databases.

项目总结/摘要 该项目提出了新的方法来表示电子健康记录（EHR）中的数据，以改善预 对患者结果的直观建模和解释。EHR数据提供了一个有希望的机会， 了解临床决策和患者状况如何随着时间的推移相互作用，以促进患者健康。 然而，EHR数据很难用于预测建模，因为它们包含各种数据类型（例如， 连续的、分类的、文本等），他们的纵向性质，高量的非随机缺失， 测量和其他问题。此外，患者结局通常具有异质性原因， 从多个临床实验室测量和患者访视中获取综合信息。的核心挑战 目前的问题是克服EHR中的数据表示与基础假设之间的不匹配- 使用常用的统计和机器学习（ML）方法。为此，该项目提出了新的 基于包装器的方法，用于从EHR数据中学习信息特征。这两种方法都提出了专门的 运算符来处理顺序数据、时间延迟和变量交互，并有能力发现 影响患者结局的潜在临床规则/决策。重要的是，这两种方法也产生档案 代表复杂性和准确性之间的最佳权衡的可能模型，这有助于模型 解释。通过将一组丰富的数据操作编码为节点， 在有向无环图中，并使用多目标优化来优化图结构。中央 本研究假设多目标优化可以从 电子健康记录，以产生准确的，解释性的模型，病人的结果。初步研究表明， 方法可以有效地学习低阶数据表示，提高几个状态的预测能力， 最先进的ML方法。该技术在高维生物医学领域具有良好的标度特性。 数据目标1（K99）是开发一种与现有ML方法配对的多目标特征工程方法 通过从原始数据中构建新特征并使用来自 训练好的模型来指导特征构建。在目标2（K99）中，应用该方法来形成预测模型 心脏病和心力衰竭的风险。由此产生的模型将在- 在导师的帮助下进行预测，以便将预测转化为临床建议。对于目标3（R 00）， 提出了第二种方法，该方法使用类似的框架来优化现有的神经网络方法， 以在保持精度的同时尽可能地简化它们的结构。目标4（R 00）的目标是 识别有再次入院风险的住院患者，并提出护理点策略以减轻风险 风险通过将所提出的方法应用于从患者数据库中收集的患者数据， 宾夕法尼亚大学医院，Geisinger健康系统和公开可用的EHR数据库。