Northwestern Medicine piloting machine learning for heart disease
Bluhm Cardiovascular Institute at Northwestern Medicine announced this week that it has been doing new artificial intelligence work in an effort to improve the efficacy and accuracy of its cardiac screening.
Clinicians there are using a cardiac monitoring platform from Eko, studying how its AI-enabled digital stethoscopes can interpret heart sounds to help screen for heart murmurs and valvular damage.
WHY IT MATTERS
Traditional stethoscopes aren't optimal for detecting cardiac disease. They depend on a "highly trained musical ear that can separate subtle abnormalities from normal sounds with cardiologist-level precision," according to Northwestern researchers.
The idea with the Eko stethoscopes is that AI and machine learning can combine the data from tens of thousands of heart sound patterns to help clinicians better assess what sounds are normal and what's not.
"One of the biggest problems in healthcare is that general practitioners so often miss heart murmurs that if found earlier would allow patients to get treatment before problems arise," said Connor Landgraf, CEO of Eko, in a statement.
"Northwestern is known for their work in valvular heart diseases and together we are working on AI technology that will detect the two most prevalent valve diseases with the sensitivity of a cardiologist," he said. "The results of this study will work to bridge the gap for general practitioners and early detection of these life threatening conditions."
Eko, which is funding studies at two separate sites, plans to enroll 800 patients at Bluhm Cardiovascular Institute's new Center for Artificial Intelligence.
THE LARGER TREND
This just the latest innovative application of AI that could enable clinicians to make more accurate diagnoses. Google and Verily are exploring the ways it can help screen for diabetic retinopathy. Microsoft and Philips have developed augmented reality technology to help with image-guided minimally invasive surgery. Another recent AI application can help predict some cancer symptoms. Frost & Sullivan predicts that this is all only the beginning of the impact AI will have on the future of digital pathology.
ON THE RECORD
"If proven effective, Eko's platform could be a much simpler, lower cost way to identify patients with heart disease," said Dr. James Thomas, director of the Northwestern's Center for Heart Valve Disease and principal investigator for this study, in a statement. "We are looking to support and advance work that broadens access to the best diagnostic tools in healthcare, regardless of whether a patient lives in the city or a more rural area. Deep learning provides that expert knowledge, regardless of a patient's location."
"Artificial intelligence is transitioning into clinical studies with potentially revolutionary implications for the practice of cardiovascular care," added Dr. Patrick M. McCarthy, chief of cardiac surgery at Northwestern Memorial Hospital and executive director of the Bluhm Cardiovascular Institute. "Northwestern Medicine is the perfect incubator for partnering with companies using machine learning in a variety of clinical settings, and it's through advancements like this that we will become even better physicians."
Healthcare IT News is a HIMSS Media publication.