AI model 98% accurate in detecting diseases — just by looking at your tongue
Researchers from Iraq and Australia have developed an artificial intelligence (AI) model capable of diagnosing various diseases with up to 98% accuracy by analyzing images of the human tongue. nypost.com
This innovative system integrates traditional Chinese medicine practices, which have long utilized tongue examination as a diagnostic tool, with advanced machine learning techniques.
Traditional Basis for Tongue Diagnosis
In traditional Chinese medicine, the tongue's appearance—including its color, shape, and coating—is believed to reflect the body's internal health. Specific tongue characteristics have been associated with particular health conditions:
Yellow Tongue: Often linked to diabetes. Purple Tongue with Thick Greasy Coating: May indicate cancer. Unusually Shaped Red Tongue: Commonly observed in acute stroke patients. White Tongue: Associated with anemia. Deep Red Tongue: Seen in severe COVID-19 cases. Indigo or Violet Tongue: Can point to vascular or gastrointestinal issues, or asthma.
These associations provided a foundational framework for the AI model's development.
nypost.com
Development and Training of the AI Model

The research team compiled a comprehensive dataset comprising over 5,200 tongue images. This dataset included both healthy tongues and those exhibiting signs of various diseases, such as mycotic infections, asthma, COVID-19, fungiform papillae, and anemia. The AI system was trained to recognize subtle differences in tongue color and texture associated with these conditions.

Popular Science Diagnostic Accuracy and Potential Applications
 In controlled experiments, the AI model demonstrated remarkable diagnostic accuracy:
98% Accuracy: In diagnosing a variety of conditions, including diabetes, stroke, anemia, asthma, liver and gallbladder diseases, COVID-19, and various vascular and gastrointestinal issues, by analyzing tongue color.
Labmedica The system's high accuracy suggests significant potential for integration into medical facilities, offering a secure, efficient, user-friendly, and cost-effective method for disease screening. Researchers envision that, in the future, smartphones equipped with appropriate applications could perform similar diagnostic functions, enhancing accessibility and convenience in medical diagnostics. Labmedica
Conclusion
By merging ancient diagnostic techniques with modern AI technology, this innovative system exemplifies how traditional medical practices can be revitalized to meet contemporary healthcare needs. The AI model's ability to accurately diagnose diseases through tongue analysis holds promise for early detection and improved patient outcomes, potentially transforming preventive healthcare and routine screenings.