Visit our website: http://bit.ly/2GtXaiw
Using Al to Improve Heart Murmur Screening 4X
Andreas Schriefl, Founder and CEO, eMurmur
Abstract:
eMurmur received both FDA (class II) and CE (class IIa) regulatory clearances in April 2019 for it’s AI driven clinical product, eMurmur ID.
eMurmur ID is a mobile and cloud solution which operates in conjunction with a 3rd party electronic stethoscope. It uses advanced machine learning to identify and classify pathologic and innocent heart murmurs, the absence of a heart murmur, and S1, S2 heart sounds. The end-to-end solution supports the workflows of healthcare providers performing cardiac auscultation and has multiple applications including primary and specialty care, hospitals, health system, health plans and telehealth. It is the first cleared product globally capable of the determining the absence or presence of heart murmurs across all patient age groups. This technology will put expert-level murmur detection capability (~90% accuracy for cardiologists) into the hands of front-line health care workers (~20% accuracy), most of whom readily acknowledge they need help with this most basic health assessment procedure.
The presentation will tell the origin story of eMurmur, how AI was used to solve the challenge, the critical role of a gold-standard reference database to train and validate the AI algorithms, the unique steps necessary to gain FDA clearance, and the platform created that can now be leveraged to other heart health issues as well as auscultations in general (ie. lung). Along the way the presentation will touch on the critical role of having medical advisers to guide the development approach, how the solution was designed to fit seamlessly into the current standard of care, and the importance of building relationships with key health partners in the market to lend credibility to the value AI can bring to the field of auscultation.
Add comment