Detecting Low-Glucose Levels via ECG Using AI
Sun, April 18, 2021

Detecting Low-Glucose Levels via ECG Using AI

A new AI approach can detect when blood sugar levels fall below normal by recording the heart's electrical activity through wearable sensors / Credits: dolgachov via 123RF

 

Testing glucose levels in the blood can be a painful procedure for patients, especially for children. Usually, the process requires the use of needles as well as repeated finger-pricks throughout a single day. “Fingerpicks are never pleasant and in some circumstances are particularly cumbersome. Taking fingerpick during the night certainly is unpleasant, especially for patients in pediatric age,” Dr. Leandro Pecchia from the University of Warwick said.

Currently, medical professionals use Continuous Glucose Monitors (CGM) to measure glucose in an interstitial fluid using an invasive sensor with a little needle. This would send an alarm and data to a display device. But, artificial intelligence can change this. 

A recent study titled ‘Precision Medicine and Artificial Intelligence: A Pilot Study on Deep Learning for Hypoglycemic Events Detection based on ECG’ was published in the Nature Springer journal Scientific Reports proved that AI can detect hypoglycemic events from raw ECG signals. The new approach can detect when blood sugar levels fall below normal by recording the heart's electrical activity through wearable sensors. This could potentially transform how diabetics monitor their glucose levels. 

According to Science Daily, an American website that aggregates press releases and publishes lightly edited press releases about science, two pilot studies, which tested healthy volunteers for low glucose levels, were conducted about using AI for automatic detecting hypoglycemia via few ECG beats. The researchers discovered that the average sensitivity and specificity of the system is approximately 82%, which is much better compared to the current CGM performance. The model also offers personalized therapy that could be more effective than current approaches.

Dr. Pecchia stated that the study emphasized how hypoglycaemic events affect ECG in individuals. “Basing on this information, clinicians can adapt the therapy to each individual. Clearly more clinical research is required to confirm these results in wider populations. This is why we are looking for partners,” he said.