Eko bags $20M to expand digital stethoscope, grow footprint

Doctors currently use the Duo to perform quick ECGs and to monitor heart disease patients remotely. In the future, it could be equipped with algorithms diagnose specific conditions, such as atrial fibrillation and congestive heart failure. (Eko)

Doctors have been using Eko’s digital stethoscope and electrocardiogram (ECG) device to screen patients for heart disease since 2017, but the company isn’t stopping there. It raised $20 million to bankroll the development of machine-learning algorithms that could one day provide digital diagnoses of heart conditions as accurately as a cardiologist. 

Drawn from ARTIS Ventures, DigiTx Partners, NTT Venture Capital, 3M Ventures, Mayo Clinic, Seraph Group and XTX Ventures, the series B funding will also support the commercial buildout of Eko’s device, called Duo, to more clinics and health systems. Although the diagnostic algorithms aren’t FDA-cleared yet, the expansion will set the stage for their eventual rollout. 

Eko CEO Connor Landgraf

“We're focusing on growing the number of sites that are using our product and will be using our machine-learning algorithms to screen for disease appropriately in very quick office exams,” Eko CEO Connor Landgraf told FierceMedTech. “Our end goal is to be able to help these large health systems, clinics, and individual physicians’ offices identify heart disease earlier than they could otherwise.” 

RELATED: Renovacor bags $11M to push precision medicine for rare heart disease

Doctors currently use the Duo to perform quick ECGs and to monitor heart disease patients remotely. According to Landgraf, it takes about 15 seconds and can be done at the patient’s bedside, while a traditional 12-lead ECG may take an hour or longer. In the future, the Duo could be equipped with algorithms diagnose specific conditions, such as atrial fibrillation, valvular heart disease and congestive heart failure. 

“We’re focused on using machine learning to better understand heart sounds and ECG to be able to provide more accurate and timely detection of cardiovascular disease in the clinic... We’re trying to unlock the information about our heart’s function that is locked up in these invisible signals and sounds. We want to use machine learning to provide that information to clinicians,” Landgraf said. 

Last November, Eko reported that one of its algorithms could detect heart murmurs in children as accurately as a cardiologist. 

RELATED: Eko AI study demonstrates accurate digital diagnoses of pediatric heart murmurs without a specialist 

The company envisions the Duo to be “at the frontlines of care,” where it can be used by anyone from nurses to primary care physicians and cardiologists to quickly screen patients for specific cardiovascular diseases, Landgraf said. 

“The ability to do that in seconds at the office or very quickly at the point of care could really change the efficiency of the clinical system, the experience of the patient and in some situations, change the disease pathway because we’d be able to identify disease earlier,” he said. 

Its algorithms could also “close the feedback loop” in the treatment of heart disease, which tends to be approached with a broad brush. 

RELATED: GV leads $58.5M round for Verve, a startup looking to pit gene editing against heart attacks

“[In cardiovascular disease,] we’re usually using a generic at a generic dosage for these patients,” Landgraf said. “We could have the opportunity to use [the Duo] to close that feedback loop and provide data to clinical teams to let them know that the dosage should be adjusted.” 

That use case is still a long way off for Eko, but it could be a way to personalize treatment for heart disease that the company believes is otherwise impossible. 

In addition to its in-house work, Eko has struck partnerships to propel its work. With Mayo Clinic, it is working on an algorithm to detect low ejection fraction, the diminished ability of the heart to pump blood, which can indicate heart failure. And Eko is carrying out a study with Northwestern Medicine to improve the screening of valvular heart disease using machine learning.