Researchers at NuraLogix have created new AI models to predict a person’s risk of HbA1C levels and fasting blood glucose levels that are above a clinical threshold for pre-diabetes.
Canada-based NuraLogix said that the model will enable people to screen themselves using any gadget with a camera setting such as a smartphone or a tablet in the future.
The machine learning-based models were trained using the facial blood flow patterns of participants who had undergone a blood test for HbA1C (glycated haemoglobin) and fasting blood glucose.
According to the firm, the models can predict whether a subject’s HbA1C is higher than 5.7% or their fasting blood glucose is higher than 5.5mmol/L.
Both cases showed that the models’ AUC predictions were greater than 0.80 when compared with the traditional Framingham model, which had an AUC in the region of 0.70.
NuraLogix director of research Dr Naresh Vempala said: “Our models are ground-breaking because we can now predict the likelihood of a person being above an at-risk cut-off for pre-diabetes with significant accuracy.
“It is exciting that our machine learning-based classifiers are touchless and have learned relationships between facial blood flow patterns and diabetes risk.
“This was unimaginable and non-existent in AI until now. Potentially, our AI models can empower people to monitor their own health on a daily basis, thus allowing early detection and prevention.”
In January this year, the firm launched Anura Web, the browser-based version of its AI-powered contactless health and wellness personal measurement solution.
