The photo is not directly related to the article. Photo = Image Today
In Canada, artificial intelligence (AI) has been developed to diagnose diabetes using only basic health data such as age, gender, height, and weight, and a 10-second recorded voice.
On the 19th, the research team of Jaycee Kaufman of Klick Labs in Canada published a study in the medical journal ‘Mayo Clinic Proceedings: Digital Health’ that can determine type 2 diabetes using 6 to 10 seconds of voice. They announced that they had developed an AI model.
This research is a system that combines voice technology and artificial intelligence. The research team said that it would bring about a big change as it would be possible to easily check for type 2 diabetes with just a few sentences recorded on a smartphone.
According to the International Diabetes Federation (IDF), about half of adult diabetic patients around the world do not know that they have diabetes, and about 90% of diabetic patients have type 2 diabetes, in which the insulin produced in the body does not function properly.
In this study, the research team examined basic health data such as age, gender, height, and weight of 192 healthy men and women and 75 patients with type 2 diabetes, and recorded phrases presented on a smartphone for 6 to 10 seconds six times a day for two weeks. I was told to do it.
By analyzing the 18,465 recorded samples obtained in this way, we extracted 14 acoustic features that can detect differences between healthy people and type 2 diabetes patients, and used these to create an AI model that distinguishes type 2 diabetes. made.
The research team said that this AI model was able to determine type 2 diabetes with an accuracy of 89% for women and 86% for men.
The research team explained that the accuracy of this AI model is similar to that of the fasting blood sugar test (FBG, 85%), glycated hemoglobin test (A1C, 91%), and oral glucose tolerance test (OGTT, 92%). This test is an existing test method that diagnoses diabetes by collecting blood.
“Through signal processing, we were able to detect voice changes caused by type 2 diabetes, and these changes appear differently in men and women,” said researcher Kaufman. “This technology could change the way diabetes is tested.” reported.
He continued, “This test method has the potential to eliminate the barriers to current testing methods, which take a lot of time and money,” and added, “In the future, research will be conducted to further verify the results of this study and expand negative diagnosis to pre-diabetes, women’s health, hypertension, etc. “We will do it,” he added.
2023-10-19 23:23:11
#Diagnosis #diabetes #listening #10second #recorded #voice #Diabetes #diagnosis #developed #Canada #Seoul #Economic #Daily
