Mednition launches early sepsis detection initiative
Mednition announced an early sepsis detection initiative to address one of the leading causes of death in the US and a condition that drives the highest cost of hospitalization each year. Sepsis consumes more than $27 billion in hospital costs annually.
The company’s machine learning powered solution, KATE, is designed specifically to improve sepsis identification while decreasing sepsis related mortality. Preliminary results of this new prototype model have a sensitivity (True Positive Rate) of 90% and specificity (True Negative Rate) of 87% with an Area Under the Curve (AUC) of 0.949.
As part of its sepsis initiative, Mednition also announced that it has appointed Stephen Liu, MD, FACEP, a leading emergency medicine physician, educator and researcher, to the company’s clinical advisory board and to help lead the sepsis program. Dr. Liu is an emergency physician with VEP Healthcare, Inc., and Medical Director and an attending staff physician with emergency services at Adventist Health White Memorial (AHWM) in Los Angeles.
Each year, approximately 1.7 million people in the US develop sepsis resulting in more than 275,000 deaths. Along with being the number one cost of hospitalization in the US, it is the leading cause of hospital readmissions. Additionally, one in three in-hospital deaths is caused by sepsis.
“Sepsis doesn’t generate public attention like cancer or heart disease does,” said Steven Reilly, CEO and co-founder of Mednition. “But when people understand how prevalent it is, how many lives it claims each year and how much it costs us as a society, they see why we need to get it under control.”
Dr. Liu was instrumental in the implementation of KATE. “I’ve been working hands on with the team at Mednition and the emergency clinicians at AHWM for the past year, and I’ve seen the promise of machine learning delivered to ED nurses on a daily basis right where it is needed most, in real-time at the point of care. Mednition’s early sepsis detection prototypes show high accuracy with a low false positive rate. We’re going to use machine learning to turn the tables on sepsis!”
