Modern health care systems are conducting continuous, automated surveillance of the electronic medical record (EMR) to identify adverse events with increasing frequency; however, many events such as sepsis do not have elucidated prodromes (i.e., event chains) that can be used to identify and intercept the adverse event early in its course. Currently, there does not exist reliable framework for discovering or describing causal chains that precede adverse hospital events. Clinically relevant and interpretable results require a framework that can (1) infer temporal interactions across multiple patient features found in EMR data (e.g., labs, vital signs, etc.) and (2) can identify patterns that precede and are specific to an impending adverse event (e.g., sepsis). In this work, we propose a linear multivariate Hawkes process model, coupled with ReLU link function, to recover a Granger Causal (GC) graph with both exciting and inhibiting effects. We develop a scalable two-phase gradient-based method to maximize a surrogate-likelihood and estimate the problem parameters, which is shown to be effective via extensive numerical simulation. Our method is subsequently extended to a data set of patients admitted to an academic level 1 trauma center located in Atalanta, GA, where the estimated GC graph identifies several highly interpretable chains that precede sepsis. Here, we demonstrate the effectiveness of our approach in learning a GC graph over Sepsis Associated Derangements (SADs), but it can be generalized to other applications with similar requirements.

翻译：现代医疗系统正持续且自动化地监测电子病历（EMR），以日益频繁地识别不良事件；然而，许多事件（如脓毒症）尚未阐明其前驱症状（即事件链），以便在不良事件发展早期进行识别和干预。目前，尚不存在可靠的框架来发现或描述医院不良事件发生前的因果链。临床相关且可解释的结果需要一个能够：（1）推断EMR数据中多个患者特征（如实验室检查、生命体征等）的时间交互作用；（2）识别即将发生的不良事件（如脓毒症）前特定模式的框架。本文提出一种结合ReLU链接函数的线性多变量霍克斯过程模型，以恢复同时包含兴奋性和抑制性效应的格兰杰因果图。我们开发了一种可扩展的两阶段梯度方法，通过最大化代理似然函数估计问题参数，并通过大规模数值模拟证明其有效性。随后，我们将该方法应用于乔治亚州亚特兰大一所学术一级创伤中心收治的患者数据集，该数据集估计的格兰杰因果图识别出脓毒症前多个高度可解释的因果链。本研究展示了该方法在识别脓毒症相关紊乱因果图上的有效性，但其可推广至具有类似需求的其他应用场景。