Fingerpicks can often be painful, deterring patient compliance.
Researchers have developed a new Artificial Intelligence (AI)-based technique that can detect low-sugar levels from raw ECG signals via wearable sensors without any fingerprint test. As it stands continuous glucose monitors or CGMs are available via the NHS for detecting sugar levels in the blood.
Currently Continuous Glucose Monitors (CGM) for hypoglycemia detection measure glucose in interstitial fluid using an invasive sensor with a little needle, which sends alarms and data to a display device.
"Our innovation consisted in using artificial intelligence for automatic detecting hypoglycaemia via few ECG beats". Taking fingerpick during the night certainly is unpleasant, especially for patients in paediatric age. "This is relevant because ECG can be detected in any circumstance, including sleeping".
A member from Warwick's School of Engineering, Dr. Leandro Pecchia, stated: "Fingerpicks are never pleasant and in some circumstances are particularly cumbersome". A comparison of two subjects reveals that they have different ECG waveform changes during the hypoglycemic events. The figure below is an exemplar. The red and green shadows represent the standard deviation of the heartbeats around the mean. Specifically, one of the subjects presents an evidently longer QT interval during a hypoglycemic event, while the other subject does not.
The vertical bars represent the relative importance of each ECG wave in determining if a heartbeat is classified as hypo or normal.
From these bars, a trained clinician sees that for Subject 1, the T-wave displacement influences classification, reflecting that when the subject is in hypo, the repolarisation of the ventricles is slower. This could have an impact on subsequent clinical interventions.
This result is feasible since the Warwick AI model has been trained using the own data of each subject.
Two pilot studies of healthy volunteers showed the system's average sensitivity and specificity was about 82% which is comparable with the current system used to detect hypoglycemia. Because of the heterogeneous nature of ECG data, no machine learning system has been able to successfully take a large cohort of ECG recordings and find universal patterns to correlate with blood glucose measurements in individuals.
What may have made the Warwick scientists' method so effective is that the AI algorithms are trained with the subject's own data.
"Our approach enables personalised tuning of detection algorithms and emphasize how hypoglycaemic events affect ECG in individuals. Basing on this information, clinicians can adapt the therapy to each individual", the authors wrote. However, the researchers said that further research is still needed to confirm the results using more participants and a wider population. "This is why we are looking for partners".