In recent research, a computer algorithm has achieved an impressive 98% accuracy in predicting various diseases by analyzing the color of the human tongue. This innovative imaging system, developed collaboratively by researchers from Middle Technical University (MTU) in Iraq and the University of South Australia (UniSA), represents a significant leap forward in medical diagnostics.
The proposed system leverages advanced machine learning algorithms to diagnose a wide range of health conditions, including diabetes, stroke, anemia, asthma, liver and gallbladder issues, COVID-19, as well as vascular and gastrointestinal problems.
By analyzing 5260 images during training phases, the researchers were able to fine-tune their AI model to detect subtle variations in tongue color that correlate with specific diseases.
How was the system trained for diagnosis?
Two teaching hospitals in the Middle East contributed to this study by providing 60 tongue images from patients suffering from various health conditions. These images were integral to training the AI model. The artificial intelligence was then able to match tongue colors with corresponding diseases with remarkable accuracy.
One of the most exciting features of this system is its ability to provide real-time diagnosis. Cameras positioned 20 centimeters from a patient captured images of their tongues, which were then analyzed instantly by the AI model. This capability underscores the potential for rapid and non-invasive diagnostic procedures.
A Modern Twist on Ancient Practices
Interestingly, this modern technology draws inspiration from traditional Chinese medicine practices that date back over 2000 years. Senior author Ali Al-Naji, who holds positions at both MTU and UniSA as an Adjunct Associate Professor, highlights how traditional practitioners have long examined tongue characteristics—such as color, shape, and thickness—to diagnose ailments.
“The color, shape, and thickness of the tongue can reveal a litany of health conditions,” says Al-Naji. For example:
- Diabetes: Yellow tongue
- Cancer: Purple tongue with a thick greasy coating
- Acute Stroke: Unusually shaped red tongue
- Anemia: White tongue
- COVID-19: Deep red tongue
- Vascular/Gastrointestinal Issues or Asthma: Indigo or violet-colored tongue
Future Prospects
The research team envisions even more accessible applications for this technology in the future. Co-author Professor Javaan Chahl from UniSA suggests that soon smartphones could be equipped with this diagnostic capability. “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,” Prof Chahl states.
Such advancements could improve access to healthcare diagnostics globally by providing quick and accurate assessments using widely available technology like smartphones.
Source: Ali Raad Hassoon, Ali Al-Naji, Ghaidaa A. Khalid, Javaan Chahl. Tongue Disease Prediction Based on Machine Learning Algorithms. Technologies, 2024; 12 (7): 97 DOI: 10.3390/technologies12070097
Comments