Deep machine learning probes EHR data
Will this be the year that many healthcare providers start using more advanced data approaches to problems like prevention and disease management?
For data and analytics vendor Jvion’s COO Ritesh Sharma, that is among the overarching themes the industry will tangle with heading into HIMSS15.
“How can we leverage the massive amounts of data created by technology — electronic health record and management systems, social media, and wearables — to help improve individual and population health outcomes?” Sharma wonders.
One approach that Jvion uses and will showcase at HIMSS15 is combining clinical rules and deep machine learning to follow and predict healthcare scenarios for individual patients and populations.
Atop Jvion’s agenda of near-term priorities, in fact, is working with providers to lower incidences of target diseases, including sepsis and acute myocardial infarction within their patient populations.
“For us, it comes down to using the data that a provider already has to enable accurate and scalable predictions,” Sharma explained.
EHR data, specifically.
“How do we support new models of care coordination that stretch across hospitals and health systems, other providers, and the community?” Sharma asks. “How do we manage high-risk patients, as aging baby boomers and complex patients with co-morbidities increase?”
In all of these questions resides a common principle of preventing illness, but the current provider technology infrastructure isn’t designed to predict adverse health events, Sharma added, so “it cannot drive the preventative measures that will improve health outcomes.”