In the age of artificial intelligence (AI), many tasks have become easier. A recent study shows that AI can detect Type-2 diabetes just by analyzing a person’s voice. It is now possible to determine whether someone has diabetes using AI-powered voice analysis.
The study was published in Digital Health, a journal by the Mayo Clinic. For women, AI was able to accurately detect diabetes with an 86% success rate, the same as for men. The research involved 267 participants, both diabetic and non-diabetic.
Over a two-week period, each participant's voice was recorded six times a day. With just a 10-second voice sample, the AI system could identify individuals with diabetes. Often, elevated blood sugar levels show no visible symptoms, making early detection challenging. By the time diabetes is diagnosed, the blood sugar level may have already crossed the danger threshold.
However, the simplicity of using AI to diagnose diabetes has sparked discussions.
Ian Fosat, one of the researchers involved in the study, remarked, “This voice-based diabetes detection is like a breakthrough. We are also exploring whether this voice technology can be used to diagnose other diseases.”
