The health technology company Prenosis announced on Wednesday that its artificial intelligence-powered diagnostic tool for sepsis is the first to receive approval from the U.S. Food and Drug Administration.
“Sepsis occurs when the body has an extreme response to an infection, and it is often deadly,” stated the Centers for Disease Control and Prevention, citing that more than 350,000 adults with sepsis die during hospitalization or are discharged to hospice each year.
Sepsis poses a significant challenge for diagnosis, but early detection is critical due to the rapid deterioration of patients.
The prenosis tool, known as Sepsis ImmunoScore, utilizes 22 different parameters such as temperature, heart rate, and cell counts to assist clinicians in evaluating a patient’s sepsis risk, as reported by CNBC.
Traditionally, doctors and nurses have had to monitor each parameter individually, whereas Prenosis’ solution employs AI to assess all markers simultaneously.
The tool generates a general risk score and four categories that correspond to a patient’s risk of deterioration.
Sepsis ImmunoScore was developed using Prenosis’ Immunix platform, which relies on a dataset comprising over 100,000 blood samples from 25,000 patients.
According to the release, the tool seamlessly integrates into electronic health records, facilitating the creation and management of patients’ medical records.
By embedding Sepsis ImmunoScore into existing workflows, Prenosis ensures ease of use and accessibility for clinicians. Additionally, clinicians will have a display showing precisely which parameters were utilized to calculate the prevailing risk score.
Prenosis’ solution received FDA approval through the De Novo pathway, marking a significant achievement for the 10-year-old Chicago-based startup.
Although Prenosis is the pioneer in obtaining FDA approval for its AI diagnostic tool for sepsis, various organizations have developed similar solutions.
Notably, Johns Hopkins University has engineered an AI system aimed at swiftly identifying sepsis symptoms.
A study published in Nature Medicine in 2022 revealed that the university’s AI model detected sepsis, particularly in severe cases, an average of six hours earlier than traditional methods.
Epic Systems, a prominent healthcare software vendor, also introduced an AI-powered sepsis prediction tool. However, the company’s model has faced significant criticism in recent years.
Despite being utilized in numerous hospitals nationwide, a study published in JAMA Internal Medicine in 2021 found that Epic’s sepsis model “poorly predicts sepsis.” This led to concerns about sepsis management on a national scale.
Epic reportedly revamped its sepsis model in the subsequent year to enhance its performance.
Prenosis disclosed to CNBC that although Sepsis ImmunoScore was ready approximately three years ago, the company opted not to market the product without FDA approval.
It prioritized collaborating with regulators to address safety concerns and ensure the technology’s harmlessness.
Likewise, in September 2022, the FDA updated its guidance for companies and emphasized specific device software functions warranting regulatory oversight.
Among these are software functionalities that analyze patient-specific medical information to detect life-threatening conditions like stroke or sepsis. Essentially, the FDA encourages companies developing sepsis detection tools to seek official authorization.
Prenosis dedicated around 18 months to demonstrate the safety and efficacy of Sepsis ImmunoScore to the FDA.
With the agency’s approval secured, the company plans to conduct further studies to validate the tool’s accuracy and its impact on clinical decision-making.
Prenosis intends to market the tool to hospitals in the U.S. initially, with plans to expand its availability to hospitals worldwide.
