By speaking you will know whether you have diabetes or not, AI will reveal the truth in 10 seconds

Medical experts are increasingly adopting Artificial Technology (AI) in the healthcare sector. A lot of research is being done on AI related projects to provide better treatment to people. Recently, a research claimed that through AI it will be known in just 10 seconds whether you have diabetes or not. In future, this technology can be used to detect type 2 diabetes. If everything goes well then you will not have to spend money on lab or glucometer for blood sugar level test.

AI can test diabetes by listening to the patient’s voice. Under the new research, voice sample will be required to detect blood sugar level. According to a study published in the medical journal ‘Mayo Clinic Proceedings: Digital Health’, a person with type 2 Diabetes To test whether it is there or not, a short sound recording is enough. The surprising thing is that the results of the study are quite accurate.

AI will detect diabetes

Worldwide, approximately 24 crore adults are suffering from diabetes and are not aware of it. According to the International Diabetes Federation, about 90% of cases are type 2 diabetes. People with type 2 diabetes have an increased risk of cardiovascular disease, such as heart attack, stroke, and poor blood circulation in the legs and toes. Diabetes screening test using voice analysis will significantly improve the detection of the disease.

diabetes test

Currently, there are medical tests like Fasting Blood Glucose Test (FBG), Oral Glucose Tolerance Test (OGTT), or Glycated Hemoglobin Test (A1C) to detect diabetes. Now let’s see how voice analysis works. Human voice keeps changing, which is identified by AI. The software just needs a phone recording to test.

This system checks things like speech melody, rhythm, pauses and pitch. AI algorithms can filter all this and identify specific voice patterns that match certain characteristics.

This is how AI test is done

The AI ​​screens voice recordings lasting between 6 and 10 seconds, looking for differences in vocal pitch and intensity. With basic health data like age, gender, height and weight, it can predict whether the speaker has type 2 diabetes. Its results are quite accurate, but differ slightly depending on gender.

The tests were 89% accurate when screening women due to differences in voice between male and female speakers. In the case of men, the tests were 86% accurate. To train the AI, Jesse Kaufman and his team at Ontario Tech University in Canada recorded the voices of 267 people who either did not have diabetes or had already been cured of type 2 diabetes.

Over the course of two weeks, participants recorded a short sentence on their smartphone six times a day. This generated more than 18,000 voice samples.

Leave a Reply

Your email address will not be published. Required fields are marked *