A new study conducted by Klick Labs has discovered a groundbreaking method for an AI model to analyze short voice recordings and diagnose patients with Type 2 diabetes. This breakthrough has the potential to significantly reduce the time and cost associated with diagnosing the disease.

The study, published in Mayo Clinic Proceedings: Digital Health, outlines how researchers utilized voice recordings ranging from 6 to 10 seconds in length, along with basic patient information such as age, height, weight, and sex, to develop an AI model capable of determining whether an individual has Type 2 diabetes. Participants in the study, including both individuals with and without diabetes, were instructed to record a specific phrase using their smartphones six times per day for a period of two weeks. Through the analysis of the resulting 18,000 voice recordings, researchers identified 14 acoustic differences between individuals with and without Type 2 diabetes.

The AI model was 89% accurate in diagnosing Type 2 diabetes in women and 86% accurate in men, according to the press release. Klick Labs research scientist and first author of the study Jaycee Kaufman said this breakthrough has the potential to “transform how the medical community screens for diabetes,” removing barriers for diagnosis like time, travel, and cost.

Klick Labs Vice President Yan Fossat said Klick’s new diagnostic process demonstrates “the tremendous potential of voice technology in identifying Type 2 diabetes and other health conditions. Voice technology could revolutionize healthcare practices as an accessible and affordable digital screening tool.”

According to the American Diabetes Association, diabetes was the seventh leading cause of death in the United States in 2019. Each year, approximately 1.4 million Americans are diagnosed with the disease. With the development of this voice-analyzing AI model, the medical community may be able to improve early detection and intervention for individuals at risk of developing Type 2 diabetes.