Hello everyone!
New research has explored the use of voice analysis to detect non -diagnosed cases with quite precision.
The study, carried out by researchers from the Luxembourg Institute of Health, has used artificial intelligence (AI) to analyze voice recordings and detect differences between people with and without type 2 diabetes.
The most surprising thing is that, using only 25 seconds of voice recording and some basic health data (age, sex, BMI and hypertension state), the model achieved a 66 % accuracy in women and 71 % in men.
The problem of non -diagnosed diabetes is enormous, since half of adults with diabetes do not know that they suffer from it, which can lead to serious complications if it is not treated in time.
This new method would not only be more accessible, but could make the simplest and fastest detection without the need for invasive or expensive procedures.
607 adults participated in the study, who provided voice recordings from their phones or laptops.The AI algorithm analyzed vocal characteristics such as tone, intensity and bell, detecting differences between people with and without type 2 diabetes.
In addition, this method showed a 93 % concordance with the type 2 diabetes risk score of the American Diabetes Association, which reinforces its potential.
Although the results are very promising, researchers clarify that more research is still needed before this technology can be implemented on a large scale.However, it is a great step towards a more accessible and scalable detection strategy.
What do you think of this idea of being able to detect diabetes just by analyzing the voice?
It is impressive how technology continues to progress to improve our health!
Source: [EASD] (https:/5
