The SmartOR: An automated workflow monitoring system for improving OR efficiency and optimizing patient care

Albert Huang, MD1, Guillaume Joerger, MS1, Marc Garbey, PhD2, Brian J Dunkin, MD1. 1Houston Methodist, 2University of Houston

Objective of the Technology or Device

Operating rooms have become increasingly complex and expensive environments, costing as much as $100/minute. Studies show that utilization of this valuable resource averages 68%. Improving efficiency is hampered in most institutions by inaccurate or unavailable data on room utilization. Commercially available workflow monitoring systems have been developed to address this, but are cumbersome to use because of a requirement for manual data entry or complex equipment installations, and they lack an ability to integrate data from multiple sources (e.g. nurse and anesthesia records). Our team has developed the SmartOR – an OR workflow monitoring system that uses a network of sensors to automatically acquire information about room state and progress of the operation in real time.

Description of Technology and Method of its Use or Application

The SmartOR is designed to provide real-time information about OR utilization without the burden of manual data entry. It utilizes a distributed network of low cost sensors installed unobtrusively into the OR to track room state and procedure progress. These small sensors can be installed in less than an hour and are positioned to detect unique steps of the operative procedure. The cornerstone of this method is the generation of a time series with discrete values that correspond to the state value for each action of interest (Figure). Each time a sensor is activated, the value on the user interface changes from 0 to 1. By analyzing the pattern of signals from all sensors, a time portrait of events in that OR can be constructed.

Preliminary Results

The SmartOR has been installed in working operating rooms and used to track over 60 cases across multiple specialties (open and laparoscopic). It accurately identifies the steps of the surgical procedure in a repeatable and reliable way. By analyzing the signals from the network of sensors, the following time and data points are automatically captured: 1) Patient in-room time, 2) Begin ventilation time, 3) Room lights dimmed (surrogate for procedure start time in laparoscopy), 4) Ventilation end time (surrogate for extubation), and 5) Patient out of room time. In addition, room utilization for type of case (open or laparoscopic) and unusual events (e.g. re-intubation indicated by re-start of the ventilator) can be determined.


Conclusions / Future Directions

The Smart OR provides automatic recognition of room state and generates real-time data of OR macro-events. Work is ongoing to further define room state (1. Clean but not set-up, 2. Instruments open but no patient, 3. Surgery ongoing, 4. Room dirty) and to automatically track preoperative patient identification time by the surgeon with respect to first case start time. In addition, work is being done to make the system energy efficient so little intervention is required for maintenance, to build in redundancy in the data acquisition, and to provide real-time data reporting for multiple operating rooms simultaneously. With broader deployment of the system, we will build a statistical model of OR management and provide robust indicators of irregularities and correlate these with patient outcomes.

« Return to SAGES 2015 abstract archive