Informatics researchers combine algorithm with EHR data to predict secondary stroke risks

Stroke patients that experience an irregular heartbeat are far more susceptible to a second stroke. Identifying that risk factor often requires significant resources, including 24/7 monitoring by physicians. 

But a team of cardiologists and informatics researchers in Northern California has developed a way to predict which patients will experience an irregular heartbeat, known as atrial fibrillation, following a stroke, allowing hospitals to focus energy and resources on those high-risk individuals.

Using EHR data, a team of researchers led by a biomedical data scientists at Stanford University School of Medicine and a cardiologist at Santa Clara Valley Medical Center developed a scoring system made up of seven risk factors that assign patients to three risk groups. The results of their research was published in Cardiology.  

Patients are supposed to be monitored for atrial fibrillation for at least a month after their first stroke. Although stroke victims are frequently monitored in the hospital, “clinicians aren’t usually too vigilant about monitoring them for atrial fibrillation,” once they go home. Calvin Kwong, M.D., an internist at Santa Clara Valley Medical Center told Stanford Medicine. The scoring system allows physicians to identify which patients require home monitoring.

RELATED: Bring expectations for artificial intelligence back down to earth, Stanford researchers say

Stanford has made it clear it wants to be a leader when it comes to healthcare informatics, algorithms and precision health. The university recently created the Center for Digital Health with the goal of doing “cool stuff to improve health care with technology,” according to the center’s director of research and innovation. One of its first projects involves using Apple Watches to improve outcomes for stroke patients and those with behavioral health disorders.

The medical school has also made a point to focus on advancing algorithms that it believes could shape precision health. A recent report issued by Stanford Medicine highlighted the need to improve data literacy among physicians in order to reap the benefits of big data in healthcare.