Chulalongkorn University Launches AI for Gastrointestinal Cancer Detection

Researchers at Chulalongkorn University in Bangkok have developed an innovative artificial intelligence (AI) system called Deep GI, aimed at improving the detection of gastrointestinal cancers. This groundbreaking tool has demonstrated performance levels comparable to expert specialists, which is particularly significant in a country facing a shortage of trained endoscopy professionals.

Approved by the Thai FDA, Deep GI has been trained on hundreds of thousands of endoscopic images and is poised for national deployment and commercial startup. Colorectal cancer is the third most prevalent cancer in Thailand, particularly affecting individuals aged 50 and older. With only about 1,000 endoscopy specialists available in the country, the need for efficient screening solutions is critical.

Closing the Screening Gap

Deep GI acts as a “co-pilot” during endoscopic procedures, identifying abnormal lesions, particularly polyps, in real time. This capability allows healthcare professionals to make confident diagnoses more efficiently. The first phase of Deep GI, completed in 2022, concentrated on detecting colorectal cancer. Following its success, the second phase launched in June 2025, expanding its reach to include gastric and bile duct cancers, thus becoming the first AI system globally to identify abnormalities in all three gastrointestinal organs.

Gastric and bile duct cancers present unique diagnostic challenges due to their often subtle or flat lesions, which can easily be overlooked. Deep GI leverages AI learning capabilities to enhance the accuracy, consistency, and speed of diagnoses. In testing, the system achieved accuracy rates as high as 97% in colorectal cancer detection, matching the proficiency of experienced physicians.

Transforming Cancer Screening in Thailand

New diagnostic features, referred to as CADx, are currently in development to assist in classifying polyps as benign or precancerous. This advancement marks a significant milestone in computer-assisted diagnosis. According to Prof. Dr. Rungsun Rerknimitr, MD, Assistant to the President for Innovation and a specialist in gastrointestinal health, “AI trained on Thai medical data is more accurate for Thai patients. With more Deep GI systems in hospitals, we can make nationwide screening faster, broader, and more efficient.”

Supported by the Board of Investment, Chulalongkorn University plans to deploy 35 Deep GI units across multiple hospitals in Thailand. This initiative is expected to enhance access to early cancer detection, reduce healthcare costs, and ultimately improve survival rates. Prof. Dr. Rerknimitr added, “We hope Deep GI will serve as a powerful tool for early GI cancer screening — preventing severe illness, improving quality of life, and reducing cancer-related deaths and costs in Thailand.”

This innovative approach to cancer detection underscores the potential of AI in healthcare, addressing both immediate clinical needs and long-term public health objectives in Thailand. For further details, visit Chulalongkorn University’s official page.