Monitoring fluctuations in blood glucose levels is crucial to control diabetic patients. Strict glucose control reduces the risk of hypoglycemia, which can be life-threatening if not treated promptly.
Previously, glucose monitoring is performed through a needle puncture test or a capillary blood glucose test. Now, thanks to artificial intelligence (AI), doctors can control blood sugar levels using a few heartbeats from unprocessed ECG signals recorded through portable sensors.
A team of researchers from the University of Warwick developed a new technology to detect low glucose levels through an electrocardiogram (ECG) using a non-invasive portable device. The device is linked to the latest artificial intelligence that can detect hypoglycemic pairs from raw ECG signals.
Currently, patients who must undergo continuous monitoring of blood sugar levels receive continuous glucose monitors (MCG) for the detection of hypoglycemia. However, CGM uses an invasive needle that sends alarms when blood sugar levels become low. The device needs calibration twice a day with invasive blood tests with digital puncture.
Credit: Warwick University
AI works just as well
The new artificial intelligence system works just as well as MCGs, providing data on glucose levels without invasive needles. The researchers tested the device in two pilot studies with healthy participants and found that the average sensitivity and specificity was 82 percent in the detection of hypoglycemia, which works as well as the current method of MCG.
“The finger clips are never nice and in some circumstances they are particularly cumbersome. Taking a finger during the night is certainly unpleasant, especially for pediatric patients. Our innovation consisted in the use of artificial intelligence for the automatic detection of hypoglycemia through a few beats of ECG. This is relevant because the ECG can be detected in any circumstance, including sleep, “said Dr. Leandro Pecchia of the Faculty of Engineering at the University of Warwick.
Trained with the participant’s own data.
One of the innovations in the artificial intelligence device is that it was trained using the subject’s own data. The researchers said the new method provides the ability to customize detection algorithms. They reiterated how nocturnal hypoglycemia or hypoglycemic agents affect ECG in people. With the new method, doctors can customize the treatment for their patients.
The researchers emphasized that intersubjective differences are also important and that training the device using cohort data would not provide the same results. Therefore, personalized system-based therapy may be more effective than the therapies used today.
The new approach was able to customize the adjustment of detection measurements, shedding light on how hypoglycemic events affect ECG readings. However, the researchers said that more research is still needed to confirm the results using more participants and a wider population.
The study was published in the Scientific reports
What is hypoglycemia and how severe is it?
Hypoglycemia occurs when blood glucose drops to dangerously low levels. This usually occurs when blood sugar drops less than 70 mg / dL.
Signs and symptoms of hypoglycemia may occur abruptly depending on the person. These include tremors, anxiety, sweating, cold and moist skin, chills, irritability, rapid heartbeat or rapid heartbeat, confusion, dizziness, nausea, hunger, feeling sleepy, weakness, blurred vision, headache, numbness or tingling of the lips , tongue or cheeks. and coordination problems.
In the worst case, the patient may experience loss of consciousness, seizures or even death. It can also contribute to injuries, falls and vehicle accidents.
Porumb, M., Stranges, S., Pescape, A. and Pecchia, L. (2019). Precision medicine and artificial intelligence: a pilot study on deep learning for the detection of hypoglycemia events based on ECG. Scientific reports https://www.nature.com/articles/s41598-019-56927-5