AI in a Heartbeat
Although most heart murmurs aren’t serious, abnormal heart murmurs can cause shortness of breath, chronic coughing, chest pains, an enlarged liver, and sudden weight gain. They can be signs of a congenital defect like a hole in the heart, a cardiac shunt, or a heart valve problem that has been present since birth. They can be caused by fever, exercise, pregnancy, low red blood cell count, an overactive thyroid gland, or even during times of rapid growth in children, and they shouldn’t be taken lightly.
Eko, a cardiopulmonary digital health company, has recognized the seriousness of the disease and recently announced its heart murmur-detecting AI algorithm is performing comparably to that of expert cardiologists in detecting heart murmurs. The FDA-cleared AI algorithm recorded a sensitivity of 90.0% and specificity of 91.4% for detecting murmurs, when difficult to hear grade 1 murmurs were excluded.
“We designed the study to see whether the algorithm could not only identify murmurs in an experimental setting, but whether it could help clinicians actually make difficult diagnostic decisions at the bedside. And that’s what the results show,” said Dr. John Chorba, the study’s lead author.
The competition for AI algorithm solutions is heating up, with products like eMurmur ID, a smartphone app that utilizes a third-party digital stethoscope and a machine learning algorithm to automate the detection of heart murmurs also recently receiving FDA approval. The platform is designed to fit into a healthcare provider’s current workflow and the heart murmur recordings can be securely shared with other experts for consultation.
By detecting heart disease quickly with AI, patients can get immediate treatment, which lowers costs, improves outcomes, and reduces unnecessary testing. AI has the potential to disrupt the status-quo of heart murmur screening, as well as substantially improve patient care, while significantly reducing costs to healthcare systems around the world.