Sponsored Webinar: Early Detection of Sepsis Using AI-Based Predictive Analytics Monitoring

Webinar Description: 

Sponsored by Nihon Kohden

Modern clinical medicine often utilizes artificial intelligence (AI) – computer algorithms trained on big data from many patients – to assist clinicians with spotting patients early in the course of subacute potentially catastrophic illness such as sepsis.

In this sponsored webinar, learners will discover how early detection of sepsis is an ideal application of AI-based predictive analytics monitoring. The presenter will provide evidence of the value of visual indicators of clinical deterioration, demonstrating that this approach saves lives.

This webinar was originally presented on 10/27/22. 

Target Audience

Nurses, advanced practice providers, physicians, emergency responders, pharmacists, medical technologists, respiratory therapists, physical/occupational therapists, infection prevention specialists, data/quality specialists, and more. 

Learning Objectives

At the end of the webinar, the learner should be able to: 

  • Recognize the incidence and impact of sepsis and other clinical deterioration in floor and ICU patients;
  • Describe the role of predictive analytics monitoring in early identification of patients at increasing risk of clinical deterioration;
  • Describe the importance of using continuous cardiorespiratory monitoring data along with vital signs and lab tests for predictive analytics monitoring;
  • Recognize the unique value of presenting clinicians with a visual indicator of clinical deterioration.

Sepsis Alliance gratefully acknowledges the support provided by Nihon Kohden for this webinar. 

Course summary
Course opens: 
Course expires: 

J. Randall Moorman, MD

Founding Director, Center for Advanced Medical Analytics, University of Virginia

J. Randall Moorman, MD, a clinical cardiologist, is a professor of medicine and the Founding Director of the Center for Advanced Medical Analytics at the University of Virginia.

He received the MD degree in 1978 from the University of Mississippi and trained in internal medicine and cardiology at Duke University, where he was Chief Medical Resident at Duke Hospital in 1982-3. He trained in molecular electrophysiology and membrane biophysics at Baylor College of Medicine from 1985 to 1990, when he joined the faculty of the University of Virginia. In 2014, he was named UVa Innovator of the Year. In 2017, he gave the keynote address at the meeting of the Association for the Advancement of Medical Instrumentation. In 2017, he became the inaugural director of the UVa Center for Advanced Medical Analytics Engineering, which intends to lead the field in how data advance healthcare. From 2013 to 2019, he was Editor-in-Chief of Physiological Measurement, an Institute of Physics Publishing journal with international readership and authorship. He has mentored more than 40 trainees in basic and clinical research; nearly half have academic appointments. He is Principal Investigator on two NIH grants and leads a Leadership and Data Coordination Center. 

He has published more than 180 full-length peer-reviewed papers in journals including Nature, Science, Physical Review Letters, Journal of Clinical Investigation, Journal of General Physiology, Biophysical Journal, Circulation, Circulation Research, and Critical Care Medicine. His h-index is 47 in Web of Science and 59 in Google Scholar. With JS Richman, he described sample entropy, a variant of Kolmogorov-Sinai entropy that is a general test for non-linear dynamical systems. The paper has been cited >4500 times in Web of Science and >6500 times in Google Scholar and is in the top 10 most-cited papers with only UVa authors. 

The overarching hypothesis of the past 20 years of his work is that subacute potentially catastrophic illnesses have prodromal signatures in the physiological monitoring data. A robust example is the abnormal heart rate characteristics of reduced variability and transient decelerations prior to neonatal sepsis. He developed mathematical techniques to characterize these, related them to clinical findings, and developed the heart rate characteristics index. This is the fold-increase in risk relative to average of the diagnosis of neonatal sepsis in the next 24 hours. He was the Principal Investigator of the largest individually randomized clinical trial ever of premature infants, showing that display of the monitor and no mandated clinical intervention led to improved survival, especially in the 25% of infants who developed sepsis, where mortality was reduced from 20% to 12%. In 2018, this was named one of the top 12 research discoveries in the past 50 years at the University of Virginia. 

Since 2014, he and his coworkers have applied these concepts and approaches to adult illness in the hospitalized patient. In an observational study, they found that the display of risk estimates for respiratory failure leading to urgent unplanned intubation and hemorrhage leading to large, unplanned transfusion led to a 50% reduction in the number of cases of septic shock in a surgical ICU. These predictive algorithms are now standard of care in medical, surgical, cardiac, and pediatric ICUs. A large RCT of predictive analytics monitoring for acute care wards will be complete in 2022. 

No continuing education credits are offered for the Sponsor Innovation Webinar.

Medical Disclaimer

The information on or available through this site is intended for educational purposes only. Sepsis Alliance does not represent or guarantee that information on or available through this site is applicable to any specific patient’s care or treatment. The educational content on or available through this site does not constitute medical advice from a physician and is not to be used as a substitute for treatment or advice from a practicing physician or other healthcare provider. Sepsis Alliance recommends users consult their physician or healthcare provider regarding any questions about whether the information on or available through this site might apply to their individual treatment or care.

Please login or register to take this course.