AI Tongue Analysis Model 98% Accurate in Detecting Diseases

Tongue color is a critical health indicator used to identify diseases and gauge their progression. Various characteristics of the tongue, such as its color, shape, and coating, can signal different health conditions. For instance, a yellow tongue often indicates diabetes, while a purple tongue with a thick coating might suggest cancer. Patients with acute strokes typically have unusually shaped, red tongues. A white tongue could mean anemia; a deep red tongue is frequently seen in severe COVID-19 cases; and indigo or violet tongues may point to vascular, gastrointestinal issues, or asthma. Leveraging this concept, artificial intelligence (AI) is now modernizing a 2000-year-old practice from traditional Chinese medicine that involves diagnosing health conditions by examining the tongue.

Researchers from Middle Technical University (MTU, Baghdad, Iraq) and the University of South Australia (UniSA, Adelaide, Australia) conducted experiments utilizing AI to analyze tongue color for diagnosing diseases. They trained machine learning algorithms using 5260 images and collected 60 additional tongue images from patients with various health conditions at two teaching hospitals in the Middle East. Their imaging system proposed in a new paper published in Technologies analyzes tongue color to offer immediate diagnostic insights, demonstrating AI's potential to significantly advance medical practice.

In their study, cameras positioned 20 centimeters from subjects captured images of their tongues, and the AI system assessed the health conditions in real-time. The AI model successfully correlated tongue colors with specific diseases in nearly all cases, achieving a 98% accuracy rate 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. The researchers anticipate that in the future, smartphones could be employed to perform similar diagnoses, enhancing accessibility and convenience in medical diagnostics.

“These results confirm that computerized tongue analysis is a secure, efficient, user-friendly and affordable method for disease screening that backs up modern methods with a centuries-old practice,” said co-author UniSA Professor Javaan Chahl.

Related Links:
MTU
UniSA

Visit expo >

Gold Member

Troponin T QC

Troponin T Quality Control

New

Gold Member

Strep Pneumoniae Rapid Test

Strep Pneumoniae (6503 – 6573)

New

Hematology Analyzer

BH-6180

New

Silver Member

Epstein-Barr Virus Test

ReQuest EB VCA IgM ELISA Kit