(Yicai) Oct. 12 -- Chinese researchers have launched an artificial intelligence model that could slash lung cancer genetic testing costs by predicting gene mutations from pathology slides in as little as one minute.

The multimodal large model, called DeepGEM 2.0 and jointly developed by the National Center for Respiratory Medicine and lab operator KingMed Diagnostics, can pre-screen for genetic mutations associated with all targeted therapies approved for lung cancer in China, helping identify patients who may benefit from more focused genetic testing, Yicai learned.

The model analyzes routine digital pathology slides without requiring additional tissue samples, with prediction accuracy reaching as high as 90 percent, according to Liang Wenhua, deputy director of the NCRM.

Patients flagged by the AI model as having a high probability of specific genetic mutations can undergo targeted genetic testing, potentially cutting costs from several thousand yuan, or as much as CNY20,000 (USD2,970), to a few hundred yuan while shortening waiting times, Liang from the respiratory disease hub in Guangdong province said.

Targeted therapies can significantly extend the survival of lung cancer patients, but genetic testing is a prerequisite for receiving such treatments. Conventional genetic testing is often expensive and time-consuming, may require multiple tissue sampling procedures, and has limited coverage, Liang told Yicai.

“Traditional large-scale full-gene scanning leads to resource waste and high testing costs, imposing heavy financial pressure on patients,” Liang said.

By identifying likely mutations in advance, DeepGEM 2.0 allows doctors to prioritize smaller genetic testing panels covering a few or dozens of genes, rather than conducting extensive testing, reducing unnecessary expenses and helping patients obtain confirmation sooner, Liang explained.

Editors: Tang Shihua and Emmi Laine