Artificial intelligence learns to diagnose diseases by examining human tongue

The ancient Chinese method helped AI establish diabetes and renal failure with 94% accuracy.

For millennia, Chinese herbalists have relied on tongue examination to diagnose diseases. Today, this ancient practice is being revolutionized by artificial intelligence and machine learning.

A recent collaboration between Iraqi and Australian researchers has demonstrated that a computer-assisted system for tongue diagnosis can accurately identify diseases such as diabetes and renal failure in 94% of cases.

Harnessing the wisdom of Chinese medicine, computer scientists are now employing machine learning and artificial intelligence to analyze tongue features for disease detection.

More to read:
Innovation: Treating nerve pain with optical fibers

This approach is gaining momentum globally, offering a promising and cost-effective solution for remote health monitoring, particularly in times of pandemic restrictions, according to a recent study by engineers from the Middle Technical University (MTU) in Baghdad and the University of South Australia (UniSA).

Using a simple setup of a USB web camera and computer, the researchers captured tongue images from 50 patients with various conditions, comparing them with a vast database. Through advanced image processing techniques, they achieved a remarkable 94% accuracy in diagnosing diseases, surpassing traditional laboratory methods.

Additionally, they developed a system to relay diagnosis information via text message, ensuring swift communication with patients or healthcare providers.

Associate Professor Ali Al-Naji and his team emphasize the potential of computerized tongue analysis in disease diagnosis, noting that by leveraging the color, shape, and texture of the tongue, this technology can detect a range of ailments remotely and efficiently.

In the context of the COVID-19 pandemic, where access to healthcare facilities is limited, remote diagnosis methods like these hold immense value, he stressed.

The future of tongue-based diagnosis looks promising, and the authors now intend to expand their research beyond the three diseases studied, by adding new features from the tongues of more patients, and refining image processing.

Perhaps there will be an app deployed in the future for disease detection and monitoring based on the findings of this research. People would scan their tongues to identify the disease before seeking for medical advice.

NewsCafe relies in its reporting on research papers that need to be cracked down to average understanding. Some even need to be paid for. Help us pay for science reports to get more interesting stories. Use PayPal: office[at] or