US-based Know Labs has reported positive data from a proof-of-concept trial evaluating its device that non-invasively identifies the level of blood sugar in patients with diabetes.
The results, published in Diabetes Technology & Therapeutics Journal, demonstrate the system correctly identified an individual’s glycaemic status as hyperglycaemic, normoglycaemic, or hypoglycaemic with a 93.37% accuracy.
Know Lab’s device uses radiofrequency combined with machine learning algorithms to classify patients into different categories depending on glycaemic status. The sensors measure levels using dielectric spectroscopy, with voltage values at each frequency analysed by the algorithms.
The company says the device has the potential to facilitate early identification and intervention of diabetes. Currently, the regulated non-invasive glucose monitoring and measuring landscape is bare.
Know Labs tested the technology in 31 participants with prediabetes or type 2 diabetes. The device continuously scanned the forearms of patients for two sessions that lasted three hours long each during an oral glucose tolerance test. Water was used in a third session for control.
After training the machine learning algorithms on reference values, the model was able to correctly classify glycaemic status 93.7% of the time. This was compared to a portion of venous blood values from draws taken every five minutes.
Know Labs CEO Ron Erickson said: “This proof-of-concept for the use of our novel radiofrequency sensor as a glycaemic status screening tool indicates the device’s potential to help funnel previously undiagnosed patients more effectively into the healthcare system.”
However, the company stated that “further research is needed to enrich the dataset for categorical screening and improve the accuracy and sensitivity of each glycaemic status”.
KnowLabs already has the KnowU system, which launched earlier this year. The device houses the same technology used in its screening system. KnowU, a non-invasive wearable continuous glucose monitor, is planned to undergo large-scale clinical trials this year.
The diabetes care device market was worth $25bn in 2023 and is estimated to rise to $44bn by 2033, according to analysis by GlobalData.
Elsewhere in the non-invasive monitoring space, Samsung and Apple are both reportedly developing non-invasive blood sugar monitoring technology for their respective smartwatches.
Coupled with advancements in glycaemic status technology, there has also been a rise in unregulated smartwatches claiming to non-invasively measure blood glucose, something the US Food and Drug Administration (FDA) has had to address. To tackle the problem, the agency issued a safety communication warning to the public to avoid smartwatches and smart rings being sold online that claim to directly measure blood glucose.
