Detecting “Invisible Gorillas” in Healthcare Data
Applying machine learning to EHRs could help physicians identify diseases faster

Did you see the gorilla the first time you watched this video? According to Ori Geva, co-founder and CEO of a machine learning company, the healthcare world can learn a lot from the now-famous 1999 study conducted by Daniel Simons and Christopher Chabris, which found that 50% of viewers were so focused on watching a basketball drill that they missed a gorilla walking through the middle of the room. “With healthcare data,” Geva writes, “gorillas may come in a variety of forms, and providers and physicians use guidelines and thresholds to detect the obvious ‘symptomatic’ gorillas.”

To assist physicians with identifying less obvious “gorillas,” machine learning can help pinpoint hidden health risks and catch unexpected diagnoses. By combining EHR data, physician expertise, and machine learning, Geva believes we can tap into the “tremendous potential to positively transform the way healthcare is delivered—ensuring the gorillas will be caught, to enable earlier intervention and help reduce healthcare costs.”

To read more about integrating machine learning with health records, click here.  To read about Epic’s work in integrating machine learning into our own software, click here.