Using voice technology and within 10 seconds... artificial intelligence succeeds in diagnosing diabetes

2023-12-21 2023-12-21T08:37:25Z

Weather of Arabia - Canadian medical research teams were able to train artificial intelligence to accurately recognize type 2 diabetes in a short period of only six to ten seconds , using “voice technology.” These efforts were carried out in cooperation with faculty members at Ontario Tech University in Canada, where scientists from Klick Labs trained AI using audio recordings from 267 individuals selected from India.

Participants were asked to record an audio phrase on their mobile phones several times a day for two weeks. By analyzing 18,000 individual recordings , the scientists focused on 14 audio features to detect recurring and consistent differences between groups with type 2 diabetes and groups without it.

Four of these audio features were shown to be the most useful in accurately predicting who has diabetes and who does not. This innovation is an important step towards rapid and effective diagnosis of diabetes using voice technology and artificial intelligence.

Artificial intelligence uses voice technology to diagnose diabetes very effectively

In an innovative development, artificial intelligence used voice technology to identify type 2 diabetes, focusing on a range of vocal features, including subtle changes in voice pitch and intensity, and linking this data to basic health information, such as the patient's age, gender, height and weight.

The study showed that gender plays a crucial role, as artificial intelligence can diagnose the disease with an accuracy of up to 89% in women, and 86% in men. Jaycee Kaufman, the first author of the study and a research scientist at Klick Labs, said:

“Our research highlights significant vocal differences between individuals with and without type 2 diabetes.”

Through this technology, the need for expensive diagnostic tests such as a blood test can be bypassed, as in the past it was necessary to perform tests such as the glycated hemoglobin (A1C) test and the fasting blood glucose (FBG) test, opening the door to easier and lower cost diagnosis. For type 2 diabetes, Kaufman emphasized that:

“Voice technology has the potential to completely remove these barriers.”