LONDON: Artificial intelligence (AI) that analyses speech could be used to screen people for type 2 diabetes, a study suggests.
The technology "opens a new route” for diabetes testing, researchers said, with voice recordings able to be collected over the phone or by using an app.
People are usually diagnosed with diabetes through a blood test that determines average blood sugar levels over the previous two to three months.
The test is available to those having symptoms, or during routine health checks which are offered to people aged between 40 and 74.
Hundreds of millions of people around the world live with diabetes, around 90% of them with type 2, according to the World Health Organization. If identified, it can be treated and its consequences avoided or delayed.
The new technology was developed by researchers at tech company thymia and at RMIT University in Melbourne, Australia.
It was trained to detect changes in speech which have been linked to type 2 diabetes, such as increased hoarseness and an inability to control breath.
The tool was trained using more than 63,000 voice samples from more than 21,000 people in the UK and US.
Researchers then tested the model using 20-second recordings of people reading Aesop’s fables out loud.
The study included 7,319 people in the UK and found that the speech model gave a higher risk score to those who reported having type 2 diabetes 80% of the time.
The technology performed well across different ages and genders, but its performance was lower on black patients, though this was likely because of the low number of black patients taking part.
Performance was also lower on recordings from people with heart disease, high blood pressure or obesity.
The second analysis included a sub-group of 801 people who took diabetes blood tests at home within three months of the speech recording.
The AI tool gave these people higher risk scores 75% of the time.
"This is the largest real-world study of speech-based screening for type 2 diabetes to date, which also checks the model’s predictions against blood test results as well as against what people reported about their own diagnosis. Those flagged as higher risk by the model had blood results to match," said Giedre Cepukaityte, research scientist at thymia, who will be presenting the findings at the European Association for the Study of Diabetes (EASD) in Milan.
"This has the potential to change what screening looks like.
"A speech sample can be taken over the phone or through an app, so we can reach far more of the people who need a blood test than current pathways do, particularly those who never get to a health check," Cepukaityte said.
"Our model opens a new route to screening for diabetes. It is not a replacement for a blood test, and it should never stop anyone who thinks they need one from getting one.
The next step, according to Cepukaityte, is to test the model in clinical settings and to understand how well it works for different groups of people.
Dr Lucy Chambers, head of research impact and communications at Diabetes UK, said AI-based technologies could help identify more people who may benefit from diagnostic blood tests. "But it’s crucial that they are rigorously designed and tested to make sure no-one slips through the net," she said. – PA Media/dpa
