Examining the colour of a patient's tongue is highly effective at diagnosing diseases such as diabetes, liver disease and cancer, a study has found.
Using machine learning to analyse images of tongues, computer scientists from the University of South Australia and Middle Technical University in Baghdad correctly diagnosed diseases in 94 per cent of cases compared with laboratory results.
The study compared colours from a database of 9000 tongue images with photos from 50 patients with diabetes, renal failure and anaemia.
"Thousands of years ago, Chinese medicine pioneered the practice of examining the tongue to detect illness," MTU and UniSA Adjunct Associate Professor Ali Al-Naji said.
"Conventional medicine has long endorsed this method, demonstrating that the colour, shape and thickness of the tongue can reveal signs of diabetes, liver issues, circulatory and digestive problems, as well as blood and heart diseases."
Diabetes patients typically have a yellow tongue, while cancer patients a purple tongue with a thick greasy coating and acute stroke patients present with a red tongue that is often crooked.
Incorporating AI technology opens up the possibility for diagnosing diseases remotely and can even be done using just a smartphone.
"Computerised tongue analysis is highly accurate and could help diagnose diseases remotely in a safe, effective, easy, painless, and cost-effective way," Prof Al-Naji said.
"This is especially relevant in the wake of a global pandemic like Covid-19, where access to health centres can be compromised."
Previous studies have found tongue analysis can also be used to diagnose appendicitis and thyroid disease.
