A new continuous glucose monitor (CGM) that is not only non-invasive but that predicts blood sugar levels from ECG data promises to shake up diabetes management. The device is being showcased at the 2024 Taiwan Innotech Expo.
Continuous glucose monitors (CGMs) have revolutionized diabetes management by providing around-the-clock blood sugar levels and trends, as well as access to comprehensive data that enables medical professionals to make personalized adjustments to insulin therapy. In essence, CGMs allow for better blood sugar control, thereby reducing the risk of diabetics developing disease-related complications.
However, current CGMs have some limitations. They’re invasive, requiring the insertion of a sensor, albeit a very, very thin one, under the skin. Additionally, they can be expensive, can only be used for a limited number of days, and produce waste. However, at Taiwan Innotech Expo 2024, the company Singular Wings Medical has shown off its non-invasive CGM that monitors blood sugar using – wait for it – an electrocardiogram, or ECG, signal. Yes, the heartbeat; from the heart.
“We developed this sensor to collect physiological data, information including ECG, heart rate, respiration, and body temperature,” said Jenny Tseng, Singular Wing’s marketing specialist, when New Atlas spoke with her at the expo. “And at this exhibition, we display the glucose monitoring. So, we use ECG to estimate the glucose level for high, medium, and low glucose level.”
It's understandable that you might be skeptical. How can heart rhythm and blood sugar possibly be connected? How does a device provide continuous blood glucose monitoring when it’s not measuring fluids inside the body? How accurate is it? These are all valid questions.
Let’s start with the basics: what’s an ECG? It’s a non-invasive way of reading the electrical activity of the heart that, ordinarily, is used to diagnose issues with the heart rate or rhythm. In a clinical setting, it might involve placing a number of little gel pads around the chest and on the arms and legs and laying still for about a minute. Of course, these days, many wearables constantly monitor a user’s heart rate and rhythm.
Despite primarily being used for diagnosis in the field of cardiology, over the last few years, with the advancement of the tech found in wearables like smartwatches and smart rings, researchers have looked into whether ECG signals can be used to estimate blood sugar levels in both healthy and diabetic subjects.
Studies have shown that low blood sugar (hypoglycemia) and high blood sugar (hyperglycemia) affect the electrical characteristics of the heart, producing particular changes that can be seen on an ECG. Researchers have used machine learning to analyze ECGs and develop algorithms based on them, using those algorithms to accurately predict blood sugar.
Singular Wings has done the same with its non-invasive CGM, although the product is still being developed.
“Because we use machine learning method and … the accuracy, the average accuracy, I can tell you, is about 80%,” said Dick Hsieh, PhD, account manager for Singular Wings. “But it’s about model training. We still need [a] validation stage. But the result of [the] validation stage, it is [at] the moment unknown.”
A user’s ECG is recorded using Singular Wings’ BEATINFO sensor, which integrates with the BEATINFO Health app available on the Apple Store and Google Play. The sensor, which weighs only 15 grams (0.5 oz), can be worn with a chest strap or removable patch; both are biocompatible and anti-allergenic.
The CGM feeds its data to the Health app in real time and provides a notification to the user if their blood sugar falls outside the normal range. Singular Wings expects its non-invasive CGM to be available in about two years.
Source: Singular Wings
saying it's "high" when it's "high""?
saying it's "normal" when it's "normal"
how are the ranges defined?