Gibras is designed to interpret patient electronic health record data to assist in the early prediction of gastrointestinal bleeding
The US FDA’s Center for Devices and Radiological Health. (Credit: The U.S. Food and Drug Administration)
Machine learning diagnostic algorithm firm Dascena has secured breakthrough device designation from the US Food and Drug Administration (FDA) for its GI Bleed Risk Alert System (Gibras).
Gibras is a cloud-based software application, which uses a machine learning algorithm to interpret patient electronic health record (EHR) data to assist in the early prediction of gastrointestinal bleeding (GIB) in adult patients in a hospital setting.
In a preclinical study, Gibras demonstrated higher sensitivity and specificity compared to the standard of care Glasgow-Blatchford Bleeding Score (GBS) .
Gibras surpassed GBS in predicting GIB in the study, using the first two hours of patient data after admission into the hospital, said the company.
Dascena CEO Ritankar Das said: “GIB is a common cause for hospitalization and unfortunately, is associated with significant morbidity and mortality.
“Our algorithm is designed to provide earlier prediction of GIB, allowing for timely interventions that could result in more favorable outcomes for patients.
“These predictions also provide clinicians with the opportunity to take preventative measures, such as weighing the benefits of discontinuing medications, like NSAIDS and anticoagulants, that are associated with GIB risk but may be prescribed for other clinical indications.”
In May last year, Dascena closed a Series B financing, led by Frazier Healthcare Partners, of $50m to advance the development of its diagnostic algorithm.
The financing round saw the participation of Longitude Capital, Euclidean Capital and an undisclosed investor.
Dascena is engaged in the development of machine learning algorithms to facilitate early disease intervention and enhance care outcomes for patients.