Submit a comment
Canadian medical researchers have trained artificial intelligence (AI) to accurately diagnose type 2 diabetes in just 6-10 seconds, through the patient's voice.
According to the Daily Mail, the research team at Klick Labs (USA) achieved the above achievement after their AI machine learning model identified 14 distinct sound characteristics between people without the disease and people with type 2 diabetes.
The AI focused on a set of vocal characteristics, including small changes in pitch and intensity that are imperceptible to the human ear. It then paired that data with basic health information, including the study participants' age, gender, height, and weight.
The researchers found that gender played a decisive role: AI was able to diagnose the disease 89% of the time for women, but was slightly less accurate at 86% for men.
This AI model promises to significantly reduce the cost of medical visits. The team said that Klick Labs' model will become more accurate when it is also inputted with data on the patient's age and body mass index (BMI).
Yan Fossat, deputy director of Klick Labs and lead researcher on the model, is confident that their voice technology product has great potential in identifying type 2 diabetes and other health conditions.
Professor Fossat also teaches at Ontario Tech University, with expertise in mathematical modeling and computational science for digital health.
He hopes that Klick’s non-invasive and accessible AI diagnostic method could facilitate diagnosis through a simple phone app, which would help identify and support the millions of people with type 2 diabetes who are not yet screened.
He also expressed hope to extend this new research to other medical areas, such as prediabetes, women's health and hypertension.
According to Tin Tuc newspaper