Type:
General Education
Session ID:
26
Title:
Closing the Loop in Sepsis Prediction With ML and ISLET Visualization
Description:
While 80-85 percent of sepsis cases present within the first 48 hours of admission (ED), they have lower mortality (5-10 percent) as compared to 15-20 percent of cases that present later and have higher mortality (15-30 percent). To better (and earlier) identify sepsis cases not present on admission, at a large safety-net hospital, an end-to-end early sepsis prediction and response workflow was created in the inpatient setting. First, an ML model was built to predict the risk of a patient becoming septic in real-time. Next, the model baked into clinical workflows through FHIR® APIs to make the model actionable at point of care. The model accesses EMR every 15 minutes and alerts the care providers when the risk exceeds a certain threshold, which can be tailored to local populations. Finally, an EHR-integrated decision support app (ISLET) was added to enable clinicians to easily view and understand model output to improve actionability. Prediction, alerting, visualizing the root causes and acting on the case completes the workflow. This full workflow has been running for thousands of patients every 15 minutes in the last year. This session will focus on the challenges, achievements and impact of this workflow on healthcare outcomes.
Level:
Intermediate
Format:
60-Minute Lecture
Learning Objective #1:
Identify scalable ML solutions for triaging patients on high risk of sepsis within a large inpatient population
Learning Objective #2:
Recognize the EMR integrated HL7® -SMART on FHIR® visualization app to demystify the black box machine learning models
Learning Objective #3:
List the lessons learned in implementing a workflow of sepsis prediction and response model in a clinical setting