This website requires JavaScript.

Granger Causal Chain Discovery for Sepsis-Associated Derangements via Multivariate Hawkes Processes

Song WeiYao XieChristopher S. JosefRishikesan Kamaleswaran
Sep 2022
摘要
Modern health care systems are conducting continuous, automated surveillanceof the electronic medical record (EMR) to identify adverse events withincreasing frequency; however, many events such as sepsis do not have clearlyelucidated prodromes (i.e., event chains) that can be used to identify andintercept the adverse event early in its course. Currently there does not exista reliable framework for discovering or describing causal chains that precedeadverse hospital events. Clinically relevant and interpretable results requirea framework that can (1) infer temporal interactions across multiple patientfeatures found in EMR data (e.g., labs, vital signs, etc.) and (2) can identifypattern(s) which 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 $g(x) = x^+$ link function to allow potential inhibition effects,in order to recover a Granger Causal (GC) graph. We develop a two-phasegradient-based scheme to maximize a surrogate of likelihood to estimate theproblem parameters. This two-phase algorithm is scalable and shown to beeffective via our numerical simulation. It is subsequently extended to a dataset of patients admitted to Grady hospital system in Atalanta, GA, where thefitted Granger Causal graph identifies several highly interpretable chains thatprecede sepsis.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答