On October 23rd, the reporter learned from Guangzhou Jinyu Medical Laboratory Group Co., Ltd. (hereinafter referred to as "Jinyu Medical") that the cancer pathology gene large-scale model DeepGEM has recently been deployed in the company. Based on this large model, the company will collaborate with Tencent and the First Affiliated Hospital of Guangzhou Medical University to develop a multi-modal pathological gene model, enabling accurate, timely, and universal gene mutation prediction technology to benefit more cancer patients. In cancer diagnosis and treatment, pathological diagnosis is the basis for diagnosing cancer, and gene sequencing is a necessary prerequisite for precise diagnosis and treatment. However, conventional genetic testing methods are difficult to widely apply in clinical practice due to their complex technology, long time consumption, and high cost, especially in resource scarce areas. To address this pain point, the research team of the First Affiliated Hospital of Guangzhou Medical University, in collaboration with Tencent Life Science Laboratory, has developed DeepGEM. Liang Wenhua, Director of the Chest Tumor Comprehensive Ward at the First Affiliated Hospital of Guangzhou Medical University, introduced that DeepGEM can provide accurate, timely, and economical prediction of gene mutations and their spatial distribution. It can complete the prediction of multiple common lung cancer driver gene mutations in just one minute, with an accuracy of 78% -99%, achieving the prediction of lung cancer gene mutations using conventional histopathological images. To further validate the clinical value of DeepGEM, in 2025, the R&D team collaborated with Jinyu Medical to conduct a larger scale validation. The results showed that DeepGEM performed particularly well in identifying common driver gene mutations in lung cancer, and its key performance indicators have reached the reference level for clinical auxiliary diagnosis. The reporter learned that with the successful completion of verification, the three parties will promote the clinical application of DeepGEM in predicting lung cancer gene mutations. At the same time, relevant parties will further develop a multi-modal pathological gene model, which can achieve AI assisted diagnosis of multiple sites, cancer types, and omics by integrating pathological morphology with proteomics, metabolomics, and other information. He Jianxing, Director of the National Respiratory Medicine Center and Dean of the Guangzhou Respiratory Health Research Institute at the First Affiliated Hospital of Guangzhou Medical University, said that he hopes that DeepGEM's deployment in Jinyu Medicine can provide a demonstration of achievement transformation and cooperation for clinical research, promote the true transformation and implementation of clinical research, and benefit the people. Liang Yaoming, Chairman and CEO of Jinyu Medical, expressed his hope to collaborate with more partners to launch more intelligent and inclusive clinical diagnostic solutions. (New Society)