In a groundbreaking development, researchers from Iraq and Australia have created an artificial intelligence model capable of diagnosing medical conditions by analyzing the color and characteristics of a person’s tongue. This innovative approach boasts an impressive 98% accuracy rate and may revolutionize how diseases are detected and monitored in real time.
According to senior researcher Dr. Ali Al-Naji, who teaches at both Middle Technical University in Baghdad and the University of South Australia, the method draws inspiration from traditional Chinese medicine. For centuries, practitioners have examined the tongue as a diagnostic tool, correlating its appearance with various health issues. “Typically, people with diabetes exhibit a yellow tongue, cancer patients may show a purple tongue with a thick, greasy coating, and acute stroke patients often present with a uniquely shaped red tongue,” Al-Naji explained.
Expanding on these insights, Al-Naji highlighted that a white tongue can indicate anemia, while severe COVID-19 cases often present with a deep-red tongue. Other findings suggest that an indigo or violet tongue may signify vascular, gastrointestinal, or respiratory conditions such as asthma.
The AI model is designed to emulate this ancient diagnostic practice using modern technology. For the study, patients sat approximately eight inches from a laptop equipped with a webcam. The device captured images of their tongues, which were then analyzed by the algorithm. Over 5,200 images were used to train the AI, ensuring its ability to identify tongue colors and patterns associated with specific health conditions. The system was later tested with images from 60 patients at two teaching hospitals in the Middle East, successfully diagnosing nearly all cases.
The research, published in the journal Technologies, has garnered significant attention for its potential applications. Co-author Professor Javaan Chahl of the University of South Australia envisions a future where this technology becomes widely accessible through a smartphone app. Such an app could screen for a variety of conditions, including diabetes, stroke, anemia, asthma, liver and gallbladder issues, and even COVID-19.
“This computerized tongue analysis is a secure, efficient, and cost-effective method for disease screening,” Chahl noted. “It bridges modern medicine with a centuries-old practice, offering an innovative tool for early diagnosis.”
Despite its promise, the technology faces some challenges. For instance, camera reflections can occasionally interfere with accurate readings, and patient hesitancy to provide personal data remains a concern. Moreover, the lack of a standardized dataset for AI tongue analysis has been identified as a barrier to broader implementation. A 2023 review of five years’ worth of similar studies highlighted the need for more comprehensive datasets to improve reliability and scalability.
Still, experts agree that AI-driven tongue analysis holds “enormous value” for the future of healthcare, offering a non-invasive, user-friendly solution to early disease detection. With further refinement, this technology could play a pivotal role in making accurate and affordable healthcare more accessible worldwide.
