Artificial Intelligence Can Predict Type 2 Diabetes in Seconds, Canadian Researchers Find
- Advertisement -
Artificial Intelligence Can Accurately Predict Type 2 Diabetes Using Voice
Canadian medical researchers have developed artificial intelligence (AI) that can accurately predict type 2 diabetes within just 6 to 10 seconds of analyzing a patient’s spoken voice.
Training the AI
Researchers from Ontario Technological University in Canada collaborated with scientists from Klick Labs to train the AI. They used recordings from 267 people recruited from India who were asked to record a specific phrase on their mobile phones six times a day for two weeks.
Analyzing Vocal Features
From the 18,000 individual recordings collected, the researchers focused on 14 vocal features. They looked for consistent and repeatable differences between groups with and without type 2 diabetes. Ultimately, they identified four audio features that were most useful in accurately predicting the presence of the disease.
- Advertisement -
Linking Vocal Characteristics to Health Information
The AI analyzed a range of vocal characteristics, including subtle changes in pitch and intensity. It then linked this data to basic health information such as the patient’s age, gender, height, and weight. Interestingly, the AI showed a higher accuracy rate of 89% in diagnosing type 2 diabetes in women compared to 86% in men.
Benefits of Acoustic Technology
Traditionally, screening for prediabetes and type 2 diabetes required expensive personal diagnostic tests like blood tests. However, this new AI technology offers a more efficient and cost-effective alternative. Jacy Kaufman, the first author of the research paper and a postdoctoral fellow at Klick Labs, stated that “acoustic technology could eliminate these barriers entirely.”
Conclusion
This breakthrough in using AI to predict type 2 diabetes based on voice analysis has the potential to revolutionize diagnostic methods. It provides a faster and more accessible approach, making diabetes screening easier for patients and healthcare professionals alike.
Source: Daily Mail
