SCH: Leveraging Clinical Time Series to Learn Optimal Treatment of Acute Dyspnea

SCH：利用临床时间序列学习急性呼吸困难的最佳治疗方法

• 批准号: 10015336
• 负责人: Michael William Sjoding
• 金额: $ 23.52万
• 依托单位: UNIVERSITY OF MICHIGAN AT ANN ARBOR
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-09-10 至 2023-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10015336
• 关键词: Acute; Address; Affect; Algorithms; Blood Pressure; Chronic Obstructive Airway Disease; Clinical; Clinical Data; Computer Vision Systems; Congestive Heart Failure; Data; Data Set; Decision Making; Diagnosis; Drops; Dyspnea; Electronic Health Record; Engineering; Ensure; Evaluation; Event; Goals; Health; Healthcare; Hospitalization; Hospitals; Image; Infrastructure; Learning; Linear Models; Machine Learning; Methods; Modeling; Morality; Outcome; Pathologic Processes; Patient Rights; Patient-Focused Outcomes; Patients; Performance; Physicians; Pneumonia; Policies; Positioning Attribute; Protocols documentation; Psychological reinforcement; Quality of Care; Research; Research Personnel; Respiratory Failure; Rewards; Right to Treatments; Sample Size; Sampling; Selection for Treatments; Series; Shortness of Breath; Signal Transduction; Signs and Symptoms; Societies; Stream; Symptoms; System; Techniques; Test Result; Theoretical Studies; Time; Training; Variant; Work; base; clinical care; clinical database; clinical decision support; clinical practice; clinically relevant; convolutional neural network; deep learning; deep neural network; design; health data; high dimensionality; improved; innovation; interdisciplinary collaboration; learning algorithm; learning strategy; mortality; network architecture; novel; optimal treatments; patient response; patient stratification; prospective; success; supervised learning; support tools; theories; tool; treatment strategy

The ability to rapidly match the right patients to the right treatments at the right time is critical to ensuring patients receive high quality care. The vast majority of machine learning applications in healthcare focus on diagnosing or stratifying patients for a particular outcome. In contrast, reinforcement learning (RL) aims to learn how clinical states (i.e., sets of signs, symptoms, and test results) respond to specific sequences of treatments, with the goal of optimizing clinical outcomes. RL does not aim to diagnose, but infers diagnosis based on a patient's response to specific treatments--in many ways mimicking how clinicians operate in practice. This proposal will develop a novel clinician-in-the-loop reinforcement learning (RL) framework that analyzes electronic health record (EHR) clinical time-series data to support physician decision making, iteratively providing physicians the estimated outcome of potential treatment strategies. Our topic of focus for this work is the evaluation and treatment of patients hospitalized with acute dyspnea (shortness of breath) and signs of impending respiratory failure. Acute dyspnea is an ideal condition for an RL approach, since it can be due to three overlapping conditions: congestive heart failure, chronic obstructive pulmonary disease and pneumonia. Determining optimal treatment for these patients is clinically difficult, as a patient's presentation is frequently ambiguous, rapidly changing, and often due to multiple causes. Inappropriate treatment may occur in up to a third of patients leading to increased mortality. While developing this RL framework, we will also develop methods to learn more useful representations of high-dimensional clinical time-series data to improve the efficiency of RL model training. In addition, given the challenges of working with observational health data, we will develop new methods for evaluation of learned policies and develop new theory to better understand the limitations of RL using observational data. The project has four aims: 1) create a shareable, de-identified EHR time-series dataset of 35,000 patients with acute dyspnea, 2) develop techniques for exploiting invariances In tasks involving clinical time-series data to improve the efficiency of RL model training, 3) develop and evaluate an RL-based framework for learning optimal treatment policies for acute dyspnea, and 4) prospectively validate the learned treatment model. This research will result in new techniques for learning representations from time-series data and will study both the theoretical and practical limitations of RL using observational clinical data, leading to key advancements in ML and RL for clinical care. The tools developed for clinical decision support in this proposal have the potential for high impact because of their ability to generalize beyond the problem studied here to other conditions, laying the groundwork for clinical systems that directly impact society by aiding in the timely and appropriate treatment of patients.

能够在正确的时间快速将正确的患者与正确的治疗相匹配对于确保患者获得治疗至关重要 接受高质量的护理。医疗保健领域的绝大多数机器学习应用都专注于诊断或 针对特定结果对患者进行分层。相比之下，强化学习（RL）旨在了解临床状态如何 （即一组体征、症状和测试结果）对特定的治疗顺序做出反应，目的是 优化临床结果。 RL 的目的不是诊断，而是根据患者的反应来推断诊断 具体的治疗方法——在很多方面模仿临床医生在实践中的操作方式。该提案将开发一种新颖的 临床医生循环强化学习 (RL) 框架，用于分析电子健康记录 (EHR) 临床 时间序列数据支持医生决策，迭代地为医生提供估计结果 潜在的治疗策略。我们这项工作的重点主题是住院患者的评估和治疗 伴有急性呼吸困难（呼吸短促）和即将发生呼吸衰竭的迹象。急性呼吸困难是理想的 RL 方法的条件，因为它可能是由于三种重叠的条件造成的：充血性心力衰竭、慢性 阻塞性肺病和肺炎。确定这些患者的最佳治疗在临床上很困难， 因为患者的表现常常含糊不清、变化迅速，而且往往是由多种原因造成的。 多达三分之一的患者可能会出现不适当的治疗，导致死亡率增加。在开发这个 强化学习框架，我们还将开发方法来学习更多有用的高维临床表征 时间序列数据来提高RL模型训练的效率。此外，考虑到与合作伙伴合作的挑战 观察健康数据，我们将开发新方法来评估所学政策并开发新理论 使用观察数据更好地理解强化学习的局限性。该项目有四个目标：1）创建一个可共享的、 35,000 名急性呼吸困难患者的去识别 EHR 时间序列数据集，2) 开发利用技术 不变性 在涉及临床时间序列数据的任务中，提高 RL 模型训练的效率，3）开发和 评估基于强化学习的框架，用于学习急性呼吸困难的最佳治疗策略，以及 4) 前瞻性 验证学习到的治疗模型。这项研究将产生学习表征的新技术 时间序列数据，并将使用观察性临床数据研究强化学习的理论和实践局限性， 导致临床护理的机器学习和强化学习取得重大进展。为临床决策支持而开发的工具 提案具有产生巨大影响的潜力，因为它们能够推广到此处研究的问题之外 其他条件，通过及时和有效地帮助，为直接影响社会的临床系统奠定基础 对患者进行适当的治疗。