According to a study published in “Mayo Clinic Proceedings: Digital Health”, researchers from Klick Labs have developed an AI model that can detect Type 2 diabetes by analyzing just six to ten seconds of a person’s voice.
The model works by analyzing a person’s voice for subtle changes in pitch and intensity that are not perceivable by the human ear. It uses acoustic voice features and basic health data such as age, sex, height, and weight to distinguish whether an individual has Type 2 diabetes.
The study involved 267 participants, both diabetic and non-diabetic. They were asked to record a phrase on their smartphones six times a day for two weeks.
The researchers analyzed over 18,000 recordings for more than 14 different acoustic features, which varied between diabetic and non-diabetic individuals and the model reported an accuracy rate of 89% for women and 86% for men, according to a statement detailing the finding.
Surprisingly, these vocal changes manifest differently in males and females, a discovery that could lead to more personalized screening methods.
The latest research demonstrates the ever-growing role AI plays in healthcare, with the convergence of machine learning models, data science helping to improve patient treatment and assisting medical discoveries.
According to the International Diabetes Federation, nearly 240 million adults worldwide are living with diabetes without knowing they have the condition. Almost 90% of these cases are Type 2 diabetes. Current diagnostic tests for prediabetes and Type 2 diabetes include the glycated hemoglobin (A1C), fasting blood glucose (FBG) test, and the OGTT.
All of these tests require a trip to a healthcare provider for patients. However, the AI model developed by Klick Labs could change the way people screen for diabetes, offering better access and lower costs than current screening methods.
Yan Fossat, Vice President of Klick Labs, believes that this non-intrusive and accessible approach offers the potential to screen vast numbers of people and help identify the large percentage of undiagnosed people with Type 2 diabetes.
He stated that “voice technology could revolutionize healthcare practices as an accessible and affordable digital screening tool”.
The next step for Klick Labs is to replicate the study and expand the vocal search to look for prediabetes, hypertension, and other diseases. This research underscores the tremendous potential of voice technology in identifying not only Type 2 diabetes but also other health conditions.
Image Credit: Towfiqu barbhuiya @Unsplash
