SAN FRANCISCO--(BUSINESS WIRE)--Dascena has received a grant from the National Institute of Health (NIH) to conduct a clinical study that evaluates the efficacy of its HindSight algorithm to accurately and efficiently predict the onset of sepsis in patients. The study will gauge the algorithm’s ability to adapt to site-specific data analysis and workflows to trigger more meaningful alerts for physicians – thereby reducing alert fatigue and unlocking the value that machine learning algorithms present to aid sepsis detection and diagnosis.
Sepsis is a dysregulated host response to an infection, affecting at least 1.7 million adults in the U.S. each year and driving an estimated $62 billion in health costs annually. While studies have demonstrated that early diagnosis and treatment of sepsis can significantly reduce adverse outcomes for patients, disease detection has historically been extremely challenging. That said, there have been significant advances in the development and use of machine learning algorithms (MLAs) to better predict the onset of sepsis in patients in the past decade. These MLAs analyze patient data (such as vital sign information and lab measurements) in real time to identify high-risk sepsis patients and then alert clinicians. Many MLAs have been shown to significantly outperform traditional clinical scoring systems used for diagnosing sepsis patients; Dascena’s flagship sepsis prediction algorithm, InSight, has been clinically proven to reduce patient mortality by 39.5%, readmission to the hospital by 22.7%, and in-hospital length of stay by 32.3%.
Despite these advancements, many hospitals remain hesitant to adopt machine learning-based clinical decision support (CDS) tools due to false or irrelevant alerts that lead to alert fatigue among physicians and nurses. This is where Dascena’s HindSight solution comes in.
Building upon the company’s InSight algorithm, HindSight adapts to the idiosyncrasies of real-world clinical deployments to reduce false sepsis alerts. In contrast to other machine learning-based CDS tools, HindSight undergoes periodic retraining using new site-specific patient data. Additionally, the algorithm is designed to analyze data pertaining to the time of clinical evaluation and treatment in relation to the onset of sepsis. As a result, HindSight can learn and adapt to a site's clinical practices over time and reduce irrelevant alerts.
“Machine learning presents a tremendous opportunity in healthcare, helping us accelerate disease detection, diagnosis, and treatment,” said Andrew Pucher, CEO, Dascena. “However, adoption of machine learning-based CDS tools has stalled in large part due to alert fatigue. To take full advantage of the clinical opportunities machine learning affords, we need to eliminate irrelevant alerts and arm physicians and nurses only with the actionable intelligence they need to identify and treat patients suffering from sepsis and other diseases.”
To learn more about Dascena’s innovative algo-enabled care for better patient outcomes, visit the company’s website.
About Dascena
Dascena develops and implements clinically proven machine learning algorithms to improve patient outcomes. These innovations help predict or identify disease, improve the delivery of care, and help save lives. Please visit us to learn more at dascena.com and on Twitter and LinkedIn.
