An algorithm strategy for precise patient monitoring in a connected healthcare enterprise

Abstract

This perspective paper describes the building elements for realizing a precise patient monitoring algorithm to fundamentally address the alarm fatigue problem. Alarm fatigue is well recognized but no solution has been widely successful. Physiologic patient monitors are responsible for the lion's share of alarms at the bedside, most of which are either false or non-actionable. Algorithms on patient monitors lack precision because they fail to leverage multivariate relationship among variables monitored, to integrate rich patient clinical information from electronic health record system, and to utilize temporal patterns in data streams. Therefore, a solution to patient monitor alarm fatigue is to open the black-box of patient monitors to integrate physiologic data with clinical data from EHR under a four-element algorithm strategy to be described in this paper. This strategy will be presented in this paper in the context of its current status as described in our prior publications.

Keywords: Data mining; Translational research.

Fig. 1

A photo of a bedside patient monitor in use at a medical surgical intensive care unit of the UCSF Medical Center. The salient features of the display include: (1) 6 s of multiple channel physiologic signals; (2) vital signs and their upper and lower threshold for alarming; (3) text of last four alarms. In addition, a panel of multiple buttons to access all monitor features and configurations is attached. The look-and-feel of this display has largely remained unchanged for decades as well as core ECG signal processing and arrhythmia detection algorithms

Fig. 2

A schematic representation of integrating data from patient monitors and electronic health record (EHR) system toward a precise patient monitoring solution. The integration takes three key steps. The first step involves representing raw data as time-stamped tokens, the second step then uses training data to identify predictive patterns of co-occurring tokens, and the last step monitors and detects the emerging patterns from the data streams in real-time and further processes the sequence of these pattern triggers to characterize the patient status