Early Identification of Acute Kidney Injury Using Deep Recurrent Neural Nets, Presented with Probable Etiology

使用深层循环神经网络早期识别急性肾损伤，并提出可能的病因

• 批准号: 9621546
• 负责人: Ritankar Das
• 金额: $ 34.93万
• 依托单位: DASCENA, INC.
• 依托单位国家: 美国
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-09-15 至 2020-02-29
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9621546
• 关键词: Acute Renal Failure with Renal Papillary Necrosis; Address; Admission activity; Affect; Area; Bayesian Modeling; Clinical; Clinical Decision Support Systems; Clinical Trials; Computer software; Data; Data Set; Detection; Development; Diagnosis; Diagnostic; Diagnostic Procedure; Early identification; Electronic Health Record; Etiology; Event; Fatigue; Future; Goals; Hand; Head Start Program; Hospital Costs; Hospital Mortality; Hospitals; Hour; Human; Injury to Kidney; Inpatients; Kidney; Knowledge; Label; Laboratories; Lead; Machine Learning; Measurement; Methods; Monitor; Patient risk; Patient-Focused Outcomes; Patients; Performance; Pharmaceutical Preparations; Phase; Physical Examination; Plant Roots; Process; Prospective Studies; Radiology Specialty; Randomized Controlled Trials; Receiver Operating Characteristics; Recurrence; Renal function; Reporting; Research; Risk; Risk Assessment; Sepsis; Series; Small Business Innovation Research Grant; Supervision; Syndrome; System; Testing; Text; Therapeutic; Time; Training; United States; Work; base; clinical decision support; computer based statistical methods; effective therapy; experience; experimental study; imaging study; improved; improved outcome; insight; learning strategy; mortality; novel; overtreatment; predictive tools; prevent; prototype; rapid diagnosis; recurrent neural network; relating to nervous system; success; tool; trend

Abstract Significance: In this SBIR project we propose to develop Previse, a novel, software-based clinical decision support (CDS) system for predicting acute kidney injury (AKI), and attributing AKI to one of several causal mechanisms (etiologies). Previse will use machine learning methods and information drawn from the electronic health record (EHR) to identify the early signs of acute kidney injury. By doing so before the clinical syndrome of AKI is fully developed, Previse will give clinicians the time to intervene with the goals of preventing further kidney damage, and decreasing the sequelae of AKI. Combining this prediction module with a second module that suggests the underlying causes responsible for an incipient or full AKI, Previse will enable clinicians to make earlier and better-informed treatment decisions for AKI patients. Research Question: Can a machine- learning-based CDS predict the development and progression of AKI in hospitalized patients 72 hours in advance of KDIGO stage 2 or 3, with performance providing an area under the receiver operating characteristic curve (AUROC) of at least 0.85? Is it possible to use a Bayesian model to infer the cause of AKI with high accuracy (AUROC ≥ 0.75)? Prior work: We have developed a prototype version of the Previse system which predicts AKI up to 72 hours in advance of KDIGO stage 2 or 3 criteria, with an AUROC near 0.70. We have previously developed machine-learning-based predictive tools for sepsis, in-hospital mortality, and other adverse patient events with performance significantly improved over commonly used rules-based scoring systems. Specific Aims: To predict the onset of chart-abstracted KDIGO stage 2 or 3 AKI in retrospective data, 72 hours in advance (Aim 1); to use data drawn from the EHR to identify the cause of AKI at time of onset with high accuracy, and to present this causal inference, its likelihood, and relevant evidence supporting it in a human-interpretable fashion (Aim 2). Methods: We will predict the onset of AKI using a deep, recurrent neural network (RNN). This expressive, nonlinear classifier will incorporate time-series information in the qualitative portions of the EHR and will also incorporate features derived from text components, such as radiology reports. Labeling AUROC of 0.85 or higher at 72 hours pre-KDIGO AKI will constitute success in Aim 1. In Aim 2, we will train a dynamic Bayesian network to identify the cause of AKI. We will train this system using semi-supervised methods, where the causes of a set of AKI examples will be hand-annotated by clinician experts; these examples will be split into two groups, with some used for training and the remainder for testing. Aim 2 will be successful if this training results in etiology identification accuracy of at least 0.75 in the test set. Future Directions: Following the proposed work, the combined Previse system will be deployed for prospective studies at partner hospitals.

抽象的 意义：在这个 SBIR 项目中，我们建议开发 Previse，一种新颖的、基于软件的临床决策 用于预测急性肾损伤 (AKI) 的支持 (CDS) 系统，并将 AKI 归因于几个原因之一 机制（病因）。 Previse 将使用机器学习方法和从电子设备中提取的信息 健康记录（EHR）以识别急性肾损伤的早期迹象。在出现临床综合征之前这样做 AKI 完全发展后，Previse 将为临床医生提供干预时间，以预防进一步发生 肾脏损伤，减少 AKI 后遗症。将此预测模块与第二个模块相结合 表明导致早期或完全 AKI 的根本原因，Previse 将使临床医生能够 为 AKI 患者做出更早、更明智的治疗决策。研究问题：机器能否—— 基于学习的 CDS 在 72 小时内预测住院患者 AKI 的发生和进展 KDIGO 第 2 或第 3 阶段的进步，其性能提供了接收器操作下的区域 特性曲线 (AUROC) 至少为 0.85？是否可以使用贝叶斯模型来推断 AKI 的原因 高精度（AUROC ≥ 0.75）？之前的工作：我们开发了 Previse 的原型版本 系统可提前 72 小时预测 AKI，达到 KDIGO 2 或 3 阶段标准，AUROC 接近 0.70。我们之前开发了基于机器学习的脓毒症、院内死亡率、 和其他不良患者事件，其性能比常用的基于规则的方法显着改善 评分系统。具体目标：预测图表抽象 KDIGO 2 期或 3 期 AKI 的发作 提前 72 小时提供回顾性数据（目标 1）；使用从 EHR 中提取的数据来确定 AKI 的原因 在发病时以高精度提供该因果推论、其可能性和相关证据 以人类可解释的方式支持它（目标 2）。方法：我们将使用以下方法预测 AKI 的发作： 深度循环神经网络 (RNN)。这种富有表现力的非线性分类器将结合时间序列 EHR 定性部分中的信息，还将包含源自文本的特征 组件，例如放射学报告。在 KDIGO AKI 前 72 小时将 AUROC 标记为 0.85 或更高 构成目标 1 的成功。在目标 2 中，我们将训练动态贝叶斯网络来识别 AKI 的原因。 我们将使用半监督方法训练该系统，其中一组 AKI 示例的原因将是 由临床医生专家手工注释；这些示例将分为两组，其中一些用于训练 其余部分用于测试。如果该培训能够提高病因识别的准确性，则目标 2 将会成功 测试集中至少为 0.75。未来方向：根据拟议的工作，组合 Previse 系统 将部署在合作医院进行前瞻性研究。