Artificial Intelligence Model Achieves 98% Accuracy in Detecting Diseases Based on Tongue Color
Researchers in Iraq and Australia have developed an advanced computer algorithm capable of analyzing the color of an individual’s tongue to detect various medical conditions with an impressive 98% accuracy rate. Senior study author Ali Al-Naji, a professor at Middle Technical University in Baghdad and the University of South Australia, explained that different tongue colors can indicate specific health issues. For instance, individuals with diabetes often exhibit a yellow tongue, while cancer patients may have a purple tongue with a thick greasy coating. Al-Naji further noted that acute stroke patients typically present with an unusually shaped red tongue, while a white tongue could signal anemia. Moreover, people with severe cases of COVID-19 are likely to have a deep red tongue, while an indigo or violet-colored tongue could indicate vascular and gastrointestinal problems or asthma.
This innovative imaging system is inspired by the traditional Chinese medicine practice of examining the tongue for signs of disease. To train the artificial intelligence model, researchers utilized 5,260 images to identify tongue color and the corresponding medical conditions. Subsequently, the technology was tested with 60 tongue images from two teaching hospitals in the Middle East. The program, operated via a webcam-equipped laptop, successfully determined the disease in almost all cases.
In a study published in the journal Technologies, co-author Javaan Chahl, a professor at the University of South Australia, emphasized the potential of this technology. He envisioned the development of a smartphone app that could diagnose a range of conditions, including diabetes, stroke, anemia, asthma, liver and gallbladder issues, COVID-19, and more. Chahl asserted that computerized tongue analysis offers a secure, efficient, user-friendly, and affordable method for disease screening, incorporating centuries-old wisdom into modern healthcare practices.
Despite its promising implications, there are challenges that researchers must address, such as patient reluctance to provide data and potential camera reflections affecting the algorithm’s accuracy. Additionally, a 2023 review highlighted the necessity of comprehensive data sets for AI tongue image analyses. Nevertheless, the technology’s potential for enhancing disease diagnosis and treatment is deemed invaluable by the scientific community.
