https://www.governmentcomputing.com/health/industry-news/university-of-warwick-low-glucose

Government Computing

GC Staff Writer

Researchers from the University of Warwick claim to have developed a non-invasive wearable sensor powered by artificial intelligence (AI) that can detect low-glucose levels through ECG without the need of a fingerpick test.

According to the university, currently, the NHS has continuous glucose monitors (CGM) for the detection of hypoglycaemia, which use an invasive sensor with a little needle for measuring glucose in interstitial fluid. The CGM sensor sends alarms and data to a display device.

In several cases, this needs calibration two times a day with invasive finger-prick blood glucose level tests, said the university.

To address the issues with the traditional process, Dr Leandro Pecchia’s team at the University of Warwick took up a research, the findings of which have been published under the title – ‘Precision Medicine and Artificial Intelligence: A Pilot Study on Deep Learning for Hypoglycemic Events Detection based on ECG’.

The research claims that using the latest findings of artificial intelligence that is deep learning, hypoglycaemic events can be sensed from raw ECG signals picked up by off-the-shelf non-invasive wearable sensors.

University of Warwick said that two pilot studies featuring healthy volunteers showed that the average sensitivity and specificity to be nearly 82% for hypoglycaemia detection with the AI-powered wearable sensor. This, the university said is on par with the current performance of CGMs with the difference being the non-invasiveness of the wearable sensor.

Dr Leandro Pecchia said: “Fingerpicks are never pleasant and in some circumstances are particularly cumbersome. Taking fingerpick during the night certainly is unpleasant, especially for patients in paediatric age.

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

Dr Pecchia said that his team’s approach allows for customised tuning of detection algorithms and stresses on how hypoglycaemic events impact ECG in individuals. Basing on the information, clinicians can adjust the therapy to each individual, said Dr Pecchia.

However, he advises that more clinical research is clearly needed to confirm the team’s results in wider populations for which the researchers are seeking partners to join their efforts.

%d bloggers like this: