Can your voice reveal Type 2 diabetes? New AI tool can reportedly detect diabetes with 89% accuracy

A groundbreaking study suggests that detecting diabetes in individuals could be as simple as having them speak a few sentences on their smartphones. Researchers at Klick Labs in the United States utilised six to ten seconds of an individual's voice, coupled with basic health information such as age, gender, height, and weight, to develop an AI model capable of discerning Type 2 diabetes presence.

This model, described in the Mayo Clinic Proceedings: Digital Health journal, exhibited an accuracy of 89 per cent for women and 86 per cent for men. To conduct the study, 267 people, both diagnosed as non-diabetic and with Type 2 diabetes, were asked to record a specific phrase into their smartphones six times daily for a duration of two weeks. Analysing over 18,000 recordings, scientists identified 14 acoustic features that differentiated non-diabetic from Type 2 diabetic individuals.

Jaycee Kaufman, the paper's first author and a research scientist at Klick Labs, emphasised that the research reveals significant vocal variations in individuals with and without Type 2 diabetes, potentially transforming diabetes screening methods. She pointed out that current detection methods often demand considerable time, travel, and expense, whereas voice technology has the potential to eliminate these barriers entirely.

The research team examined various vocal characteristics, including subtle changes in pitch and intensity that are imperceptible to the human ear. Leveraging signal processing techniques, they were able to identify voice alterations associated with Type 2 diabetes. Remarkably, these vocal changes manifested differently in males and females.

Globally, nearly half of the 240 million adults living with diabetes are unaware of their condition, and approximately 90 per cent of these cases are Type 2 diabetes, as reported by the International Diabetes Federation. Common diagnostic tests for prediabetes and Type 2 diabetes, such as the glycated haemoglobin (A1C), fasting blood glucose (FBG) test, and oral glucose tolerance test (OGTT), are frequently utilised.

Yan Fossat, Vice President of Klick Labs and the principal investigator of this study, highlighted the potential of this non-intrusive and accessible approach for screening a large number of people and identifying undiagnosed cases of Type 2 diabetes. “Our research underscores the tremendous potential of voice technology in identifying Type 2 diabetes and other health conditions. Voice technology could revolutionise healthcare practices as an accessible and affordable digital screening tool,” Fossat said.

Also Read Over 85% Indian employers expect AI to create new jobs in 1-5 years, says report