AI Doctor Technology Transforms Tongue Diagnosis with Near-Perfect Accuracy

Researchers from Middle Technical University in Iraq and the University of South Australia have developed a groundbreaking machine learning system capable of diagnosing various health conditions through tongue analysis, New Atlas reported yesterday.

This innovative approach builds on centuries of traditional diagnostic practices, particularly in Chinese medicine.

The AI was trained on 5,260 images of tongues under different lighting conditions, including those from individuals with known medical issues such as asthma, diabetes, and anemia. In tests, the system achieved an impressive 98% accuracy when diagnosing conditions from a distance of 20 cm.

Professor Ali Al-Naji highlighted that tongue characteristics—such as color, shape, and thickness—can indicate numerous health problems. For instance, a yellow tongue may suggest diabetes, while a deep red tongue can signal severe COVID-19 cases. Future advancements could allow users to obtain tongue diagnoses simply by using their smartphone cameras, making health monitoring more accessible.