AI uses tongue colour to detect illness with 98% accuracy

AI uses tongue colour to detect illness with 98% accuracy

An AI computer algorithm that analyses the colour of the human tongue to diagnose diseases such as diabetes has been developed

Other illnesses that can be detected include anaemia, asthma, COVID-19, liver and gallbladder disease, and a range of vascular and digestive problems.

Senior author Ali Al-Naji said ‘the colour, shape and thickness of the tongue can reveal a litany of health conditions’.

For example, a yellow tongue indicates diabetes while white is linked to anaemia. Cancer patients typically have purple tongues with a greasy coating, and indigo or violent indicates vascular or digestive issues or asthma. Severe COVID-19 can cause the tongue to be deep red, while an unusually shaped red tongue is a sign of stroke.

‘Secure, efficient, user friendly and affordable’

Developed by Iraqi and Australian researchers, the artificial intelligence (AI) programme was trained with 5,260 images of tongues. A further 60 images from patients with various health problems were used to test the model, of which 59 were successfully diagnosed.

A camera was placed 20cm from each patient to capture their tongues and the system predicted their health condition in real time.

Professor Al-Naji said the technology replicates the 2000-year-old practice of examining the tongue from traditional Chinese medicine.

The researchers said a smartphone could be used to diagnose disease in this way in the future.

Co-author Javaan Chahl said: ‘These results confirm that computerised tongue analysis is a secure, efficient, user-friendly and affordable method for disease screening that backs up modern methods with a centuries-old practice.’

Is AI accepted in healthcare?

In a recent survey, three quarters of NHS staff said they were in favour of using AI for patient care. Eight in 10 (81%) also supported its use for administrative tasks.

Though patients were less positive about AI than staff, 54% still supported its use in patient care. Around 18% of the public believed that AI would make patient care worse. Many respondents were concerned about its accuracy and impact on the interpersonal aspect of care.

While 57% of all staff were looking forward to using AI within their role, 65% said they thought it would make them feel more distant from patients.

Tim Horton is assistant director of insight and analysis at the Health Foundation, which conducted the survey. He said: ‘Interest is growing rapidly in the potential of artificial intelligence to improve health care.

‘While it is not a panacea, it could play a significant role in helping to ease NHS pressures and support the workforce.’


Follow Dentistry.co.uk on Instagram to keep up with all the latest dental news and trends.

Favorite
Get the most out of your membership by subscribing to Dentistry CPD
  • Access 600+ hours of verified CPD courses
  • Includes all GDC recommended topics
  • Powerful CPD tracking tools included
Register for webinar
Share
Add to calendar