BIGDATA: Causal Inference in Large-Scale Time Series

大数据：大规模时间序列中的因果推断

• 批准号: 10415027
• 负责人: SAMANTHA KLEINBERG
• 金额: $ 28.15万
• 依托单位: STEVENS INSTITUTE OF TECHNOLOGY
• 依托单位国家: 美国
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-06-01 至 2025-02-28
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10415027
• 关键词: Ablation; Address; Admission activity; Adult; Age; Algorithms; Altered Level of Consciousness; Behavior assessment; Big Data; Brain; Brain Injuries; Cardiovascular system; Caring; Classification; Clinical; Clinical Trials; Cognitive; Complex; Conscious; Consensus; Consumption; Critical Illness; Data; Disease; Early Intervention; Electroencephalography; Environment; Evaluation; Goals; Health; Hospitals; Human; Impairment; Individual; Intelligence; Intensive Care Units; Knowledge; Lead; Learning; Link; Machine Learning; Measurement; Measures; Medical; Methods; Modeling; Neonatal; Neonatal Intensive Care Units; Neurologic; Neurological status; Outcome; Output; Patient Care; Patients; Pattern; Performance; Physiological; Population; Prevalence; Privacy; Prognosis; Proxy; Quality of life; Research; Research Personnel; Resolution; Risk Factors; Series; Signal Transduction; System; Testing; Time; Unconscious State; Withdrawal; Withdrawing Care; Work; causal model; clinical decision support; clinical practice; effective intervention; functional outcomes; improved; insight; large datasets; learning strategy; machine learning algorithm; neonatal patient; neonate; open source; personalized intervention; premature; pressure; prevent; privacy preservation; respiratory; response; simulation; stroke patient; symptom treatment; tool

Project summary Large datasets generated by hospitals could have a transformative effect on medical knowledge and patient care. Yet currently the volume of data is more likely to overwhelm clinicians and the challenges of the data can overwhelm machine learning algorithms. Intensive care units (ICUs) generate data at a resolution of seconds, for the entirety of a patient's stay. Our long-term goal is to turn these data into actionable knowledge, like risk factors for a disease, early intervention targets, and real-time information to support clinical decisions. This is a broad problem, but particularly important in ICUs, which involve high stakes decisions being made in a complex environment under time pressure. We focus in particular on understanding consciousness in adults, and neurologic status in neonates. While 7% of ICU admissions are due to loss of consciousness, and degree of consciousness is critical to evaluating prognosis, making difficult choices such as when to withdraw care, and providing early interventions to improve quality of life, there are no objective or automated assessments for consciousness (adults) or neurologic status (neonates). We have shown that unresponsive patients with brain activation were twice as likely to regain the ability to follow commands compared to unresponsive patients without such activation, yet these assessments are too time consuming for regular clinical use. However we also showed that physiological data routinely collected in ICUs can be used as a proxy to classify consciousness. It is still not known why it changes and we must be sure that the patterns we find are in fact causal to avoid treating symptoms instead of a disease or launching unsuccessful clinical trials. There have been two key barriers preventing a causal understanding of consciousness. First, variables measured for each ICU patient differ, and can differ within a patient over the course of their admission. This leads to confounding when attempting to infer causal models, and has prevented learning a single model for all patients, which limits generalizability. Second, while the challenges of medical data require new methods, researchers are rarely able to rigorously evaluate and compare them, since real-world data lacks ground truth and often cannot be shared for privacy reasons. To address these challenges, we aim 1) to develop methods that learn generalizable causal models with latent variables (by intelligently sharing and combining information across patients), 2) to develop data driven simulations methods for testing machine learning algorithms while preserving privacy, and 3) to apply these methods to neonatal and neurological ICU data. We aim to create better indicators for consciousness and to uncover causes of both neurological status in ICU and its link to long-term functional outcomes. Our work turns potential weaknesses of medical data (different variables measured across individuals) into a strength, and will enable better use of large-scale observational biomedical data for real-time treatment decisions.

项目概要 医院生成的大型数据集可能会对医学知识和患者产生变革性影响 关心。然而，目前的数据量更有可能让临床医生不知所措，数据带来的挑战可能会 压倒机器学习算法。重症监护病房 (ICU) 以秒的分辨率生成数据， 患者整个住院期间。我们的长期目标是将这些数据转化为可操作的知识，例如风险 疾病的影响因素、早期干预目标以及支持临床决策的实时信息。这是 这是一个广泛的问题，但在 ICU 中尤其重要，因为这涉及到在医院中做出高风险的决策。 时间压力下环境复杂。我们特别关注理解成年人的意识， 和新生儿的神经系统状态。 7% 的 ICU 入院是由于意识丧失和意识丧失程度 意识对于评估预后、做出艰难的选择（例如何时撤回治疗）以及 提供早期干预措施以改善生活质量，但没有客观或自动的评估 意识（成人）或神经状态（新生儿）。我们已经证明，大脑反应迟钝的患者 与无反应的患者相比，激活的患者恢复遵循命令能力的可能性是其两倍 如果没有这样的激活，但这些评估对于常规临床使用来说太耗时。然而我们也 研究表明，在 ICU 中常规收集的生理数据可以用作意识分类的代理。这是 仍然不知道为什么会发生变化，我们必须确保我们发现的模式实际上是避免治疗的因果关系 症状而不是疾病或启动不成功的临床试验。有两个主要障碍 阻碍对意识的因果理解。首先，每个 ICU 患者测量的变量都不同，并且 患者在入院期间可能会有所不同。这会在尝试时导致混淆 推断因果模型，并且阻止了为所有患者学习单一模型，这限制了普遍性。 其次，虽然医学数据的挑战需要新的方法，但研究人员很少能够严格地 评估和比较它们，因为现实世界的数据缺乏基本事实，并且通常无法出于隐私原因共享 原因。为了应对这些挑战，我们的目标是 1）开发学习可推广因果模型的方法 潜在变量（通过智能地共享和组合患者之间的信息），2）开发数据驱动 在保护隐私的同时测试机器学习算法的模拟方法，以及 3）应用这些方法 新生儿和神经科 ICU 数据的方法。我们的目标是创建更好的意识指标并 揭示 ICU 神经状态的原因及其与长期功能结果的联系。我们的工作 将医疗数据的潜在弱点（针对个体测量的不同变量）转化为优势， 并将能够更好地利用大规模观察生物医学数据进行实时治疗决策。