Science 3 min read

Tracking Blood Glucose Using AI System and ECG Signals

An AI system that can detect hypoglycemic events using ECG signals paves the way for better non-invasive blood glucose monitoring devices to be developed.

JPC-PROD / Shutterstock.com

JPC-PROD / Shutterstock.com

A team of scientists has developed an AI system that can track blood glucose using non-invasive ECG signals.

It’s essential for anyone with diabetes to test their blood sugar level regularly. Unfortunately, current traditional methods are still invasive.

It entails pricking your finger with a small, sharp needle called a lancet. Next, you have to place a drop of blood onto a test strip, which in turn goes into a meter. The meter then displays the blood sugar level.

It’s safe to say that the method can be uncomfortable. So, a non-invasive glucose monitor may be one of the most sought after medical diagnostic devices.

Sadly, advancements in this area have been slow. While a few wearable devices promised to monitor glucose levels without breaking the skin, these innovations are yet to become a reality.

But that’s about to change.

Now, a team of researchers from the University of Warwick is demonstrating a new non-invasive technique that involves using artificial intelligence to detect hypoglycemic events from simple ECG signals.

In a statement, co-author of the study, Leandro Pecchia explained:

“Our innovation consisted in using artificial intelligence for [automatically] detecting hypoglycemia via few ECG beats. This is relevant because ECG can be detected in any circumstance, including sleeping.”

So, how does the new diagnostic technique work?

Developing an AI System that Can Learn Individual ECG Signals

Previous studies that sought to track blood glucose through ECG signals have been unsuccessful due to the variation in signals found in individual subjects.

Each patient has a unique fluctuation, and this results in a heterogenous ECG data. So, machine learning systems have never been able to correlate individual blood glucose measurements from a large cohort.

Pecchia noted:

“The differences highlighted above could explain why previous studies using ECG to detect hypoglycemic events failed. These inter-subject differences would hinder the performance of AI algorithms trained over cohort ECG-data.”

The critical breakthrough came when the University of Warwick team developed an AI system that did just that – learn the ECG rhythm of an individual patient.

In their article in the journal Scientific Reports, the researchers reported the efficacy of the new non-invasive method. The AI system was able to detect low-blood glucose with an 82 percent accuracy in healthy volunteers.

With that said, this is not the first non-invasive glucose monitoring system. However, the researchers noted that the framework might prove useful in a broader glucose tracking system that uses other non-invasive physiological signals. These include skin conductivity, physical activity level, and nutrition information.

Now the researchers are working to verify and refine the method for a larger patients population.

Read More: Google’s AI Can Spot Lung Cancer Months Before Doctors

First AI Web Content Optimization Platform Just for Writers

Found this article interesting?

Let Sumbo Bello know how much you appreciate this article by clicking the heart icon and by sharing this article on social media.


Profile Image

Sumbo Bello

Sumbo Bello is a creative writer who enjoys creating data-driven content for news sites. In his spare time, he plays basketball and listens to Coldplay.

Comments (0)
Most Recent most recent
You
share Scroll to top

Link Copied Successfully

Sign in

Sign in to access your personalized homepage, follow authors and topics you love, and clap for stories that matter to you.

Sign in with Google Sign in with Facebook

By using our site you agree to our privacy policy.