Cognitive Impact

Avoiding Sepsis in the Hospital Using the Cognitive Clinical Success Machine

Centers for Disease Control and Prevention (CDC) defines sepsis as the body’s extreme response to infection. It occurs when an infection that a patient already has triggers a life-threatening chain reaction. More than 1.5 million people get sepsis and at least 250,000 Americans die from sepsis each year. While anyone can get an infection that can lead to sepsis, there are groups of people who are at higher risk including adults over 65, people with chronic conditions, people with a compromised immune system, and children younger than one.
Historically, it has been very hard to identify patients at risk of sepsis before onset of the infection. Existing methods have not met performance thresholds and tend to lead to extensive and expensive laboratory testing. However, with the application of the Cognitive Clinical Success Machine’s Eigen-based engine, we are identifying:
  • Individuals at risk of sepsis before they enter the hospital
  • Patients who are at risk of sepsis when they are in the acute care setting
  • Patients who had sepsis on the index admission who are at risk of readmission
This capability is driven by the transformative Eigen approach that underpins Jvion’s machine. By combining the established Eigen Sphere infrastructure—which enable the analysis of more than a quadrillion socioeconomic, behavioral, and clinical factors—with extensive clinical intelligence, the machine is able to more precisely and effectively identify patients on a trajectory toward sepsis. The resulting outputs enable faster and more effective clinical action that ultimately leads to improved outcomes for patients and the hospital.


  • [1] Centers for Disease Control and Infection, "Sepsis - Basic Information," U.S. Department of Health & Human Services, 16 September 2016. [Online]. Available: [Accessed 6 September 2917 ].
  • [2] Centers for Disease Control and Prevention, "Protect Your Patients from Sepsis Infographic," Department of Health and Human Services, 1 January 2016. [Online]. Available: [Accessed 6 September 2017].
  • [3] P. Thomas Desautels, B. Jacob Calvert, P. c. a. Jana Hoffman, B. Melissa Jay, M. Yaniv Kerem, M. P. Lisa Shieh, M. David Shimabukuro, M. M. Uli Chettipally, M. M. Mitchell D Feldman, M. Chris Barton, S. David J Wales and M. Ritankar Das, "Prediction of Sepsis in the Intensive Care Unit With Minimal Electronic Health Record Data: A Machine Learning Approach," JMIR Med Inform, vol. 4, no. 3, p. 28, July - Sept 2016.
Jvion's Thinking Machine

We created a machine that thinks like a clinician. Here is how we did it.

Immerse Yourself in Cognitive Science

Take just 2.5 minutes to dive into the world of cognitive science.

Cognitive Machine in Action

See firsthand how a cognitive machine can help your patients and your organization.