Real-Time Detection of Gastric Cancer Tissue During Surgery... AI-Powered Technology Achieves 88% Accuracy

by Jo Inkyung

Published 06 May.2026 15:27(KST)

National Cancer Center and Kookmin University Research Team Enhances Surgical Precision by Identifying Tumor Location

A team of South Korean researchers has discovered a way to identify the extent of gastric cancer during surgery using artificial intelligence (AI).

Hongman Yoon, Director of the Gastric Cancer Center at the National Cancer Center

The National Cancer Center announced on May 6 that a research team led by Hongman Yoon, Director of the Gastric Cancer Center, in collaboration with Professor Hyungmin Kim’s team from the Department of Applied Chemistry at Kookmin University, has developed a technology capable of distinguishing gastric cancer tissue in real time during surgery.

This technology combines AI with a method that analyzes the faint light naturally emitted by tissue (autofluorescence spectroscopy), helping surgeons quickly and accurately determine the location and extent of tumors.

Gastric cancer is one of the leading causes of cancer-related deaths worldwide. Precisely identifying the boundaries of a tumor during surgery significantly impacts the success rate of the procedure. However, until now, immediate assessment has been challenging because the process required removing tissue and confirming it in a laboratory setting.

The research team focused on autofluorescence spectroscopy, which does not require any staining or preparation process. However, this method faces challenges, as measurement results can vary depending on the equipment, are affected by the surrounding environment, and it is difficult to distinguish signals when multiple light sources are mixed together.

To address this, the team developed a customized environmental control device that maintains constant temperature, humidity, and light. Data obtained in this controlled environment was then analyzed using deep learning-based AI, enabling the precise separation of complex light signals. As a result, the researchers successfully distinguished gastric cancer tissue from normal tissue with an accuracy of 88.1%, and were able to determine the extent of the tumor in near real time during surgery.

Director Hongman Yoon stated, "This is an innovative study that overcomes the limitations of traditional light analysis methods through AI and environmental control technology. It will be a significant aid to surgeons by allowing them to immediately identify tumor boundaries in the operating room."

This research was supported by the Public Interest Cancer Research Project at the National Cancer Center and the Basic Science Research Program of the Ministry of Science and ICT. The results were published in March in the international journal Analytical Chemistry (impact factor 6.7), which specializes in analytical chemistry.