Getting your Trinity Audio player ready...
|
Faculty members from the Faculty of Engineering and Faculty of Medicine at Chulalongkorn University have collaboratively developed DeepGI, an innovative AI technology for gastrointestinal endoscopy. This cutting-edge tool is designed for accurate cancer detection and is currently being utilised at King Chulalongkorn Memorial Hospital.
Prof Dr Rungsun Rerknimitr, Acting Assistant to the President of Chulalongkorn University for Innovation Affairs and a lecturer in the Department of Internal Medicine, played a pivotal role in showcasing DeepGI as a remarkable advancement in medical technology. He stressed the significance of this innovation in enhancing early detection and treatment of gastrointestinal cancers, thereby improving patient outcomes. Dr Rungsun highlighted that DeepGI leverages deep learning algorithms to analyse endoscopic images with high precision, which helps identify malignancies at an earlier, more treatable stage.
The Division of Gastroenterology at Chulalongkorn University was one of three centres globally selected to perform live endoscopy transmissions. One of the four cases presented utilised DeepGI technology to demonstrate its efficacy in detecting abnormal gastric mucosa before cancer development. The demonstration was well received, with participants from India and Italy expressing interest in future research collaborations to further enhance DeepGI for international use.
“DeepGI represents a significant advancement in digital health technology, combining expertise from the Faculty of Engineering and the Faculty of Medicine at Chulalongkorn University,” said Prof Rungsun. Prof Dr Peerapon Vateekul and Assistant Prof Dr Natawut Nupairoj from the Department of Computer Engineering, Prof Dr Rungsun and a research team, DeepGI, leverage advanced AI techniques to improve gastrointestinal endoscopy outcomes.
The system’s development also received support from a technology company that distributes DeepGI. It is operational at King Chulalongkorn Memorial Hospital and is being tested in several other hospitals, enhancing its real-world applicability and performance.
DeepGI offers four distinctive features that set it apart from conventional endoscopy systems:
- Deep Technology: It leverages deep learning algorithms for real-time detection and precise characterization of abnormalities. Its compatibility with various brands and models of gastroscopes ensures flexibility in its application.
- Future-Ready: It has the potential for future enhancements to detect abnormalities in additional body areas, such as the bile duct and stomach.
- Cost-Effective: Its affordability makes it more accessible to a wider range of hospitals and government agencies, enhancing diagnostic efficiency in provincial and rural regions.
Assoc Prof Dr Peerapon Vateekul explained that DeepGI has recently expanded its capabilities to detect gastrointestinal metaplasia (GIM), an early stage of stomach cancer. This innovation is particularly crucial as stomach cancer incidence rises, often linked to dietary habits. DeepGI assists medical staff by processing images from endoscopy videos, analysing abnormalities using AI deep learning, and analysing real-time alerts with over 90% accuracy.
Research indicates that DeepGI significantly enhances polyp detection efficiency by 16%, increases polyp detection rates by 1%, and reduces colorectal cancer risk by 3%. This improvement is vital for early cancer detection, particularly in the elderly population, who are at higher risk for gastrointestinal cancers. Early detection allows for timely treatment, reducing morbidity and mortality rates.
At Technology Readiness Level 5 (TRL5), DeepGI has been tested in real-world settings and benchmarked against competitors. It effectively detects large intestine and stomach abnormalities, preventing colorectal and stomach cancers. Future developments aim to expand its capabilities to include bile duct abnormalities, further broadening its diagnostic potential.
DeepGI represents a significant step forward in integrating digital technology with medical diagnostics. By combining AI and deep learning, it offers a powerful tool for early cancer detection, improving patient outcomes. As it continues to develop and expand its capabilities, DeepGI holds promise for transforming gastrointestinal endoscopy and setting new standards in digital health solutions.