Heart doctors are hailing a new stethoscope empowered with artificial intelligence (AI) as a major breakthrough in the treatment of aortic stenosis, the most common and most serious type of cardiac valve disease problems.
The AI digital stethoscope uses a single-lead electrocardiogram (ECG) and was able to identify low ejection fraction in a test of 100 patients, according to the Mayo Clinic, based in Rochester, Minnesota. AI device maker, Eko, partnered with the healthcare experts to produce the innovative diagnostic tool, called DUO.
The heart’s ejection fraction (EF) is the amount (expressed as a percentage) of blood pumped – or ejected – from the heart’s left ventricle with each contraction. Physicians use the EF to determine how well a human heart is pumping out blood and to diagnose or monitor heart failure.
Normally, EF falls between 50 and 70. A low ejection fraction – typically 45 or below – can signal signs of heart failure or cardiomyopathy (a disease of the heart muscle).
Participants at the American Heart Association’s Scientific Sessions heard about the AI stethoscope on November 16, 2019, which displayed significant accuracy in detecting an EF of less than 35 percent. This low measurement indicates a very high risk of potentially lethal heart arrhythmias (irregular heartbeat) linked to the sudden onset of cardiac arrest and death.
The Eko DUO is the first POC (point-of-care) device that combines a single-lead ECG with an AI algorithm to achieve such successful results. The development project was initiated in October 2018.
The following month, Eko revealed the first murmur-detection computer algorithm that outperformed most of the cardiologists who participated in a research study titled “Artificial Intelligence Detects Pediatric Heart Murmurs With Cardiologist-Level Accuracy.” The neural network AI algorithm was trained on thousands of heart sound recordings. Then, the algorithm was tested on an independent dataset of pediatric heart sounds and compared to echocardiogram imagery.
Five pediatric cardiologists also listened to the heart sound recordings and independently ruled whether a recording contained a murmur. The AI algorithm outperformed all five human heart specialists.
Attendees of the American Society of Echocardiography Scientific Sessions in March 2019 applauded in a standing ovation after a presentation on the Eko CORE digital stethoscope capable of machine learning (AI). The phonocardiograms produced by the high-tech device were fast and effective.
Steve Pham, MD, Vice President of Clinical and Research Affairs at Eko Devices, explained the significance of this medical achievement:
“A machine-learning algorithm trained on heart sounds can rapidly and accurately detect a murmur in patients with clinically significant aortic stenosis. Front-line clinicians may be able to use Eko stethoscopes (Eko CORE) with this algorithm to refer patients for echocardiography to confirm aortic stenosis.”
Brent E. White, MD, of the Bluhm Cardiovascular Institute at Northwestern Memorial Hospital in Chicago, with colleagues analyzed 639 recordings from 161 patients who were undergoing transthoracic echocardiography, the most common noninvasive type of echocardiogram that uses high-frequency soundwaves (ultrasound) to create a moving picture of the heart through the chest wall.
The 15-second phonocardiogram recordings produced by the digital stethoscope were transmitted wirelessly to a mobile app (Eko Mobile).
Pham shared his excitement over the new invention:
“All clinicians are trained to use a stethoscope from Day One of medical school but most are not trained to use ultrasound or other advanced valve disease diagnostics. It is far more affordable than ultrasound. The ubiquity of stethoscopes allows Eko to be an early detector of a notoriously ‘silent’ disease.”
The Eko executive revealed the next steps for his company:
“Further research that is needed includes developing algorithms that can differentiate between valve diseases and that can stage the diseases, which can determine intervention strategy. We also need to study the health economics and outcomes impact on the use of valve disease algorithms in front-line medicine.”
Renowned cardiologists gave high praise to the PARTNER 3 and EVOLUT Low-Risk trial results. Industry insiders believe that transcatheter aortic valve replacement (TAVR) may soon replace surgical aortic valve replace (SAVR) as the preferred treatment for patients with symptoms of severe aortic stenosis.
TAVR is a minimally-invasive surgical procedure that repairs an old, damaged heart valve without removing it. Instead, a replacement valve is wedged into the aortic valve’s place using a catheter (a thin tube made from medical-grade materials).
In the PARTNER 3 study, a team led by Martin B. Leon, MD, FACC, compared the outcomes of TAVR and SAVR in 1,000 patients with severe aortic stenosis and a low risk of death with surgery. The patients randomly received either TAVR with a third-generation balloon-expandable valve or standard SAVR with a bioprosthetic valve.
Dr. Leon commented on the fact that the TAVR vs. surgery comparison found that the length of hospital stay fell from seven to three days with TAVR:
“This is a landmark study because it involves 80 percent of the people who are currently being treated with surgery for aortic stenosis.”
Further findings included:
- Mortality rates of 1.0 percent vs. 2.5 percent
- Stroke rates of 1.2 percent vs. 3.1 percent
- Rehospitalization rates of 7.3 percent vs. 11.0 percent
In the EVOLUT study, Michael J. Reardon, MD, FACC, and colleagues, randomized 1,468 patients to TAVR with a self-expanding bioprosthesis compared with surgical replacement. After 24 months, death or disabling stroke occurred in 5.3 percent of the TAVR group compared with 6.7 percent of the surgery group, meeting the prespecified criteria for noninferiority (superiority).
Eugene Braunwald, MD, Professor of Medicine at Harvard Medical School and founding chairman of the Thrombolysis In Myocardial Infarction (TIMI) Study Group at Brigham and Women’s Hospital commented on the ground-breaking studies:
“This is a historic moment, and all of us here should recognize it as such.”
Eko is currently seeking FDA approval for its proprietary algorithm and has plans to partner it with its existing cardiac monitoring devices (Core and Duo) after securing regulatory clearance.