Our technology has been vetted through fourteen peer-reviewed publications.

Sepsis Randomized Clinical Trial

The NIH funded randomized clinical trial showed statistically significant  reduction in mortality with InSight, our sepsis AlgoDiagnosticTM.

Sepsis Randomized Clinical Trial

Shimabukuro DW, Barton CW, Feldman MD, et al. Effect of a machine learning-based severe
sepsis prediction algorithm on patient survival and hospital length of stay: a randomised
clinical trial. BMJ Open Respiratory Research 2017;4:e000234. READ THE PAPER HERE >

LEARN MORE       POSTER 

Higher accuracy, hours earlier

InSight makes predictions by comparing basic vital signs in
the patient electronic health record with statistical patterns
gathered from a reference library of prior cases. This method
allows for significant improvements in sensitivity and
specificity over commonly used systems.1

Higher accuracy, hours earlier On Set Higher accuracy, hours earlier

1. Calvert J, Desautels T, Chettipally U, Barton C, Hoffman J, Jay M, Mao Q, Mohamadlou H, Das R. High-performance detection and early
prediction of septic shock for alcohol-use disorder patients. Annals of Medicine and Surgery. 2016 Jun 30;8:50-5. Read the paper here >

Delivering real-time alerts

By delivering accurate alerts in real-time, InSight allows clinicians to intervene early,
leading to improved care outcomes. InSight continually re-evaluates a patient’s
condition when new vitals are collected, leading to up to 60% higher accuracy.2

2. Desautels T, Calvert J, Hoffman J, Jay M, Kerem Y, Shieh L, Shimabukuro D, Chettipally U, Feldman MD, Barton C, Wales DJ. Prediction of sepsis in the
intensive care unit with minimal electronic health record data: a machine learning approach. JMIR medical informatics. 2016 Jul;4(3).
Read the paper here >

Under the hood

InSight was derived from NSF and NIH funded fundamental research
in bioinformatics and machine learning, and tracks the evolution
of patient trajectories to identify patients at high risk of sepsis.3

 

 

Under the hood
3. Calvert JS, Price DA, Barton CW, Chettipally UK, Das R. Discharge recommendation based on a novel technique of homeostatic
analysis. Journal of the American Medical Informatics Association. 2016 Mar 28;24(1):24-9. Read the paper here >

More accurate with less data

The use of advanced machine learning techniques makes InSight
robust to data scarcity. Even in units without frequent measurements,
InSight provides accurate alerts with few false alarms.
4

More accurate with less data
4. Desautels T, Calvert J, Hoffman J, Jay M, Kerem Y, Shieh L, Shimabukuro D, Chettipally U, Feldman MD, Barton C, Wales DJ. Prediction of sepsis in the
intensive care unit with minimal electronic health record data: a machine learning approach. JMIR medical informatics. 2016 Jul;4(3). Read the paper here >

Validated on the pediatric population

We have applied InSight to the highly diverse pediatric population, where
it detects and predicts pediatric severe sepsis more accurately than commonly
used detection methods. Using only vital sign data, InSight can detect pediatric
severe sepsis at onset with a specificity of 87% and a sensitivity of 80%.5

5. Desautels T, Hoffman J, Barton C, Mao Q, Jay M, Calvert J, Das R. Pediatric Severe Sepsis
Prediction Using Machine Learning. bioRxiv. 2017 Jan 1:223289.
 Read the paper here >

Interested in learning more?

Contact Us