Researchers have used more than 5,000 pictures of tongues to train an AI algorithm in a traditional technique of disease diagnosis.

“One of the most exciting insights from our study is that tongue color can accurately reflect internal diseases,” Ali Al-Naji, a University of South Australia (UniSA) adjunct associate professor and the study’s senior author, told The Epoch Times in an email.

“By analyzing the tongue’s color, we identified various health conditions, demonstrating the potential of this method for noninvasive diagnosis.”

Researchers claim the algorithm can detect illnesses with 98 percent accuracy. The researchers identified diseases such as diabetes, stroke, appendicitis, gastrointestinal problems, and cancer based on the color of the tongue.

The new technology, described in the June issue of Technologies, is based on traditional Chinese medicine (TCM). TCM has used the tongue as an essential diagnostic tool for millennia. The findings suggest that a simple tongue scan could become a low-cost, noninvasive tool to diagnose diseases and track their development.

Study Details

The study states that the human tongue has distinct features linked to the body’s internal organs, including color, shape, and thickness, allowing it to successfully detect diseases and track their progression.

“The tongue can offer clues about overall health and various diseases because it reflects changes in the body’s internal state,” Al-Naji told The Epoch Times.

He described different tongue colors and their associated conditions, which are listed below.

• People with diabetes have a yellow tongue.
• Cancer patients exhibit a purple tongue with a thick greasy coating.
• Acute stroke patients present with an unusually shaped red tongue.
• A white tongue can indicate anemia.
• People with severe cases of COVID-19 are likely to have a deep-red tongue.
• An indigo or violet-colored tongue suggests vascular and gastrointestinal issues or asthma.

Engineering researchers from UniSA and Middle Technical University in Iraq used 5,260 tongue images to train AI to detect and analyze tongue color under various lighting conditions.

Researchers tested six machine learning algorithms using seven classes (red, yellow, green, blue, gray, white, and pink) to determine which was most precise, with diagnostic accuracy ranging from approximately 91 percent to 99 percent. Based on the results, the algorithm with the highest accuracy (98.71 percent), called XGBoost, was chosen for the proposed imaging system and connected to a visual user interface to determine tongue color and its associated diseases.

The imaging system was then tested in real-time using 60 tongue images from patients with various health conditions provided by two teaching hospitals in the Middle East. Diseases included diabetes mellitus, mycotic infection, asthma, anemia, fungiform papillae, and COVID-19.

Even the research team was amazed by the algorithm’s precision.

“We were surprised by the system’s accuracy, especially in real-time conditions and with varying lighting. The high accuracy of over 98% in detecting tongue colors exceeded our expectations, showing the effectiveness of AI in tongue diagnosis,” Al-Naji told The Epoch Times.

“Using our algorithm, we could predict several diseases related to tongue colors, including diabetes, mycotic infections, asthma, COVID-19, and anemia, by analyzing changes in tongue color. Each condition showed distinct color patterns that the system could detect accurately.”

The study used cameras placed 20 centimeters away from patients to capture the color of their tongues, and the system analyzed the data to predict their health condition in real time.

UniSA professor and co-author Javaan Chahl predicts that smartphones will diagnose diseases this way in the future.

“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,” Chahl said in the statement.

Tongue Diagnosis in Chinese Medicine

The above study builds on an ancient practice used in Chinese medicine. In this system, the tongue is a vital diagnostic tool that observes its different aspects, such as color, shape, coating, and moisture, to determine patterns, visualize the state of the internal organs, diagnose illness, and predict its prognosis.

Giovanni Maciocia, who died in 2018, was a highly respected author, lecturer, and practitioner of Chinese medicine. Throughout his career, he wrote numerous textbooks on the subject used by colleges worldwide—including one on tongue diagnosis.

“Observation of the tongue is a pillar of diagnosis because it provides clearly visible clues to the patient’s disharmony. Tongue diagnosis is remarkably reliable: whenever there are conflicting manifestations in a complicated condition, the tongue nearly always reflects the basic and underlying pattern,” he wrote in his book “The Foundations of Chinese Medicine.”

Final Thoughts

As for how soon we may expect to see the screening technique in clinics and hospitals, Al-Naji says he hopes that it will be in the next few years.

“While the technology is promising and shows great potential, further testing, validation, and regulatory approval are necessary before it can be widely adopted for clinical use,” he said.