Monitor this situation.
Unsubscribe anytime.
[SITUATION] · [QUIET] · [HEALTH]
2 clusters · 7 sources · 13 days · First seen · Last updated
AI voice analysis for type 2 diabetes screening
Overview
Researchers from Thymia and RMIT University have developed an artificial intelligence model designed to identify type 2 diabetes risk by analyzing subtle vocal changes, such as hoarseness, roughness, and breath control difficulties. The model was trained on more than 63,000 voice samples from over 21,000 individuals in the United Kingdom and the United States.
Initial testing on 7,319 adults in the UK showed the model correctly assigned higher risk scores to approximately 80% of those with reported type 2 diabetes. Validation against glycated hemoglobin (HbA1c) blood tests in a subgroup of 801 participants showed the model identified higher risk in roughly 75% of cases. The technology is intended to serve as an initial screening tool via phone or app to prompt follow-up clinical testing rather than providing a formal diagnosis.
When the research was presented at the European Association for the Study of Diabetes congress, researchers noted a limitation in the model’s performance, stating it was “less effective for Black individuals due to underrepresentation in the training data.”
Entities
Thymia · European Association for the Study of Diabetes · RMIT University
Timeline
-
[HEALTH] 2 sourcesAI model detects type 2 diabetes through voice analysis
Researchers have developed an AI model that can detect type 2 diabetes by analyzing vocal patterns, showing high accuracy in UK testing but facing challenges with demographic representation.
-
[HEALTH] 5 sourcesAI model identifies type 2 diabetes risk through voice analysis
An AI model developed by Thymia and RMIT University can identify type 2 diabetes risk by analyzing 20 seconds of speech for vocal changes like hoarseness and breath control issues.
Sources
24-ore.com · abcnews.al · bienpublic.com · laopinion.com · medicaldialogues.in · tempo24.news · ygeiamou.gr