New AI Can Accurately Detect Health Conditions By Just Looking At Your Tongue

Hello!

Medical researchers are claiming that their new artificial intelligence algorithm can detect with near-perfect accuracy what ails you — all by looking at your tongue.

Modernizing an Ancient Practice

The proposed research system automates the ancient medical practice of tongue diagnosis with the help of modern machine learning.

As detailed in a new paper published in the journal Technologies, teaching hospitals in the Middle East supplied the researchers with 60 photos of tongues to feed into their algorithm. Each image came from patients with various health conditions, ranging from gastrointestinal issues to COVID-19.

In more than 98 percent of cases, the AI was able to accurately detect what condition each patient was suffering from based solely on the color of their tongues.

Training the Algorithms

To create the system, the researchers fed more than 5,200 tongue photos and corresponding diagnoses into six separate ML algorithms to see how accurately they could “predict tongue color under any lighting condition,” as the paper notes. The least accurate of these systems, known as the naïve Bayes methodology, still achieved a more than 91 percent rate of accuracy. Meanwhile, the open-source XGBoost framework correctly identified the diagnoses 98.71 percent of the time.

Main author Ali Al-Naji, an adjunct professor at Baghdad’s Middle Technical University and at the University of South Australia, said in a statement that he and his colleagues were inspired by the 2,000-year-old traditional Chinese medicine practice that to this day holds up as a pretty accurate gauge of diagnosis.

What the Tongue Reveals

“The color, shape, and thickness of the tongue can reveal a litany of health conditions,” Al-Naji said.

Now that they have tested out the impressive accuracy of this tongue-forward algorithm, the international team of researchers plans to create a smartphone app that utilizes it so people can get predictive diagnostics at home.

Javaan Chahl, another UniSA professor who co-authored the study, noted that this “computerized tongue analysis” is not only safe and effective but also “backs up modern methods with a centuries-old practice.”

Sticking your tongue out and saying “ahh” has always been a routine part of a doctor’s visit — but soon, that tongue-peeping doctor may well be an AI.

Also read:

• Trading Strategies For Cryptocurrency For Beginners
• Top Smart Home Technology to Consider When Moving to a New Place
• Create Best Promotional Videos For Your Product

Thank you!
Join us on social media!
See you!