Cardiologs raises $6.5M for AI-powered ECG analysis

By Jonah Comstock
Share

Paris-based Cardiologs has raised $6.5 million to support its AI-powered algorithm for ECG analysis. The round was led by a syndicate of investors including Idinvest, ISAI, Kurma Partners, and Partech Ventures, with additional participation from Bpifrance seed fund, an existing investor. This brings the company's total funding to $10 million.

“Ambulatory ECG analysis and reporting is a labor-intensive process that can result in delays for arrhythmia detection and increased costs,” Yann Fleureau, co-founder and CEO of Cardiologs Technologies, said in a statement. “Our Cardiologs team has trained a neural network using more than 500,000 recordings, and this training dataset keeps growing. The result is that Cardiologs is designed to recognize patterns in a cardiac signal for fast and precise analysis of heart diseases in a similar intuitive manner as expert cardiologists."

Cardiologs has both a CE Mark and FDA clearance for its algorithm, with the latter granted just this past July. The software can scan for 10 arrhythmias including atrial fibrillation. It can use data from clinical ECGs as well as ECGs from newer mobile health products such as smart watches, ECG patches, or connected clothing, as long as there are at least 24 hours of continuous data (which excludes products like AliveCor's Kardia Mobile).

In clinical studies included with the FDA submission, Cardiologs' algorithm had a positive predictive value of 91 percent and a sensitivity of 97 percent for detecting atrial fibrillation. Detecting arrhythmias early, especially atrial fibrillation, helps doctors to prevent stroke and heart disease. Because life-threatening arrhythmias are often asymptomatic, early detection is especially important.

While Cardiologs' pure software play is unusual, a few other digital health companies in the ECG-monitoring space have done research using artificial intelligence to screen for arrhythmias.

In July, a team of researchers from Stanford University, working with cardiac monitoring company iRhythm, created an AI algorithm that in a small proof-of-concept trial outperformed board-certified cardiologists at identifying various types of arrhythmias from ECGs, incuding atrial fibrillation. 

Just last month, a study in Circulation found that using AliveCor's Kardia Mobile twice a week led to 19 atrial fibrillation cases being detected in the intervention group, compared to just five in the control group. Many of the intervention group cases were asymptomatic. AliveCor has several FDA-cleared algorithms, including one for detecting atrial fibrillation.

Additionally, last month Apple announced it would be using new functionality of the Apple Watch to amass clinical data through a research project called the Apple Heart Study. In conjunction with Stanford University researchers, Apple will use the sensors built into Apple Watches to observe and analyze arrhythmias. Apple has also been in conversation with the FDA about the study.

The ability of the Apple Watch to monitor heart rhythms has previously been established in a study by Cardiogram, a startup focused on that very thing. They found in May that the Watch could detect atrial fibrilation with 97 percent accuracy.