A newly developed computer algorithm has achieved an impressive 98% accuracy in diagnosing various diseases by analyzing the color of the human tongue, according to a study by researchers from Iraq and Australia.
The imaging system, created through a collaboration between Middle Technical University (MTU) in Iraq and the University of South Australia (UniSA), is capable of diagnosing a wide range of health conditions. These include diabetes, stroke, anemia, asthma, liver and gallbladder diseases, COVID-19, and multiple vascular and gastrointestinal disorders.
In a series of experiments, researchers trained machine learning algorithms using 5,260 tongue images to detect disease based on tongue color. The AI model demonstrated remarkable accuracy, identifying diseases by matching tongue color to the specific condition in almost all cases.
Two teaching hospitals in the Middle East contributed 60 tongue images from patients suffering from various health conditions to aid in the development of this groundbreaking AI system.
The researchers detailed their findings in the journal Technologies, where they explained how the system analyzes tongue color for real-time diagnosis. This innovation, they argue, highlights AI’s potential to revolutionize medicine.
Adjunct Associate Professor Ali Al-Naji, a senior author from both MTU and UniSA, pointed out that the AI technology mirrors a traditional practice used in Chinese medicine for over 2,000 years, which involves examining the tongue to identify health issues.
“The color, shape, and thickness of the tongue can reveal a variety of health conditions,” said Prof. Al-Naji. “For instance, people with diabetes often have a yellow tongue, while cancer patients may exhibit a purple tongue with a thick, greasy coating. Stroke patients frequently present with a uniquely shaped red tongue.”
He further elaborated that a white tongue could signal anemia, while individuals with severe cases of COVID-19 may have a deep red tongue. An indigo or violet tongue could suggest vascular or gastrointestinal issues, or even asthma.
During the study, cameras placed 20 centimeters away from the patient captured the color of their tongue, allowing the AI system to predict their health condition instantly.
Co-author Professor Javaan Chahl from UniSA hinted at the future potential of this technology, noting that soon, smartphones could be used to diagnose diseases in a similar manner.
“These results confirm that computerised tongue analysis is a reliable, efficient, user-friendly, and affordable method for disease screening. It supports modern diagnostic techniques with a practice rooted in centuries of medical tradition,” Prof. Chahl stated.
This innovative approach to medical diagnosis may signal a major shift in how diseases are detected, combining ancient practices with cutting-edge technology.
