Exploring the future of diabetes management

By Emily Leighton, MA’13

Lightly prick your finger pad with a lancet to form a small drop of blood. Place the blood on a test strip and insert it into a blood glucose meter, or glucometer. Wait for the glucometer to analyze your blood and give you a blood glucose reading.

If you have diabetes, you likely repeat this process several times a day. Regular glucose monitoring is an essential tool in diabetes management.

But what if we could measure blood sugar levels without the hassle and pain of a finger prick?

New research points to artificial intelligence (AI) as a promising alternative. Researchers at Schulich Medicine & Dentistry and the University of Warwick have developed and tested a new technology that uses non-invasive, wearable electrocardiogram (ECG) sensors to detect low blood sugar levels.

Published in Scientific Reports, the pilot study followed a group of healthy volunteers for 14 consecutive days. The AI model was trained with each subject's personal ECG data in order to automatically detect hypoglycemic events, which occur when a person’s blood sugar falls below the normal level.

The research team found the sensitivity and reliability of the new technology comparable to current, more invasive monitoring methods, including Continuous Glucose Monitoring that requires inserted sensors below the skin’s surface. It also allows for continuous monitoring, even during sleep.

Co-author Dr. Saverio Stranges, Chair of the Department of Epidemiology and Biostatistics, contributed to the research from a clinical and population health perspective. “Hypoglycemia is a major clinical issue among diabetic patients and is often the outcome of inappropriate management,” he said. “We wanted to develop a cost-effective and non-invasive alternative for timely detection, which allows clinicians to adequately respond and change the therapy or dosage of insulin administration.”

The research team is now looking to study the technology using a larger sample, one that includes people with diabetes and pre-diabetes.

Pre-diabetes, the transition period between normal glycemic status and diabetes, is particularly significant because of its prevalence – more than 20 per cent of the adult population (aged 25 to 64) is pre-diabetic. “This window of time is so important,” explained Dr. Stranges. “If we can detect high-risk individuals with pre-diabetes, we can implement lifestyle management to delay the onset of diabetes and, in some cases, even return people to normal blood sugar levels.”

Dr. Stranges continues to study this high-risk ‘population sub-group.’ He led a 2019 study published in BMJ Open that explored the public health burden of pre-diabetes and diabetes in Luxembourg using data from the 2013-2015 European Health Examination Survey

“This is where public health gains can be highest, targeting this much larger group,” said Dr. Stranges. “Diabetes is not something that happens from one day to another, it is a lifelong process.”