Enas Al-Banna wrote Thursday, October 19, 2023 12:04 PM
A recent study, the results of which were published in the American journal Mayo Clinic Proceedings: Digital Health, revealed the success of artificial intelligence in diagnosing people with diabetes through the person’s “voice sentences” via a smartphone, which is a major step in discovering the disease, according to what was published in the newspaper “ British Daily Mail.
Canadian researchers trained artificial intelligence to predict accurately Type 2 diabetes Only 6 to 10 seconds of the patient’s voice, after using 14 vocal features to determine the differences between non-diabetic individuals and those with type 2 diabetes.
The AI focused on a range of audio features, including subtle changes in pitch and sound intensity that doctors’ ears couldn’t hear, and linked that data to basic health information, including the patient’s age, gender, height and weight.
The scientists used recordings of between 6 and 10 seconds of voices, along with their basic health data, including age, gender, height and weight, to create an artificial intelligence model that can distinguish whether someone has type 2 diabetes or not, and the model has an accuracy. It is 89% for women and 86% for men.
Scientists analyzed more than 18,000 recordings and 14 audio features for differences between those with diabetes and those without, and then looked at a number of audio features, such as changes in pitch and intensity that cannot be perceived by the ear.
Researchers were able to detect voice changes associated with type 2 diabetes, and found that voice changes appear in different ways between men and women.
Artificial intelligence and diabetes
2023-10-19 09:04:00
#Artificial #intelligence #succeeds #diagnosing #diabetics #seconds #sound
