Boston start-up QPID hopes marry the best of big data and biology, allowing doctors and clinicians to sift through Electronic Health Records (EHR) using natural language and historical analysis. The founders say the technology can save lives and money.
QPID, pronounced “Cupid,” is entering the fast-growing field of medical technology, pushed in part by regulation mandating Electronic Health Records as well as a drive to cut down healthcare costs through smarter caregiving.
What separates QPID from the pack, chief executive Michael Doyle told me, is a focus on medical practitioners over theoretical process (the co-founders, Michael Zalis, MD and Mitch Harris, PhD, both have extensive backgrounds within hospitals), as well as smart nods from the latest trends in big data to reduce overtesting while helping catch potential problems before they cause harm.
“Electronic health records are great at capturing data, but what we’re not very good at is getting information out to clinicians in a form, factor and speed that makes a difference, particularly unstructured data,” Doyle said.
He said that the use of MRIs, for example, has increased 500 percent since 1999, although often the scans are duplicative.
“So QPID can alert the physician that the patient had an MRI six months ago and here are the findings,” he said.
Or, before the patient gets in the scanner, QPID can identify potential trouble spots, such as a pacemaker or bullet wound that could make such a generally routine scan deadly.
In addition, the system learns what kinds of language describe what problems so that, as it scans an EHR database, it knows what a doctor’s notes actually mean in terms of treatments, dangers, and impact.
Up until now, the system has been in testing with Massachusetts General Hospital and Partners HealthCare and is in use by about 5,000 clinicians.
With funding from Matrix Partners, Partners Innovation Fund, and Mass General Physicians Organization (MGPO), QPID is now working to expand its reach to other hospitals.