2024年 · 第58卷第01期

Expert Introduction: Professor Wang Dengbin of Xinhua Hospital affiliated to Shanghai Jiaotong University School of Medicine pointed out that in recent years, Professor Gu Yajia's team of Cancer Hospital affiliated to Fudan University has deeply cultivated the field of imagomics and deep learning around imaging research related to breast cancer diagnosis and treatment. They focus on the research of new technologies and methods of breast imaging and cross-scale fusion data mining. As an important member of the multidisciplinary breast team, the team continues to conduct in-depth exchanges, actively promotes cross-cooperation between multidisciplinary medicine and engineering, and has achieved a series of research results in accurate imaging diagnosis of breast cancer, helping the accurate diagnosis and treatment of tumor imaging in the era of big data, laying a solid foundation for further promoting clinical application transformation. Professor Gu Yajia's team focused on the Fudan classification of triple-negative breast cancer, realized the non-invasive classification diagnosis of MRI omics, and explored the biological explanation behind the model; Focusing on the visualization of key genes and molecules, realize non-invasive prediction through imagomics and correlate multi-omics data verification to explore potential therapeutic targets; For the accurate prediction of neoadjuvant efficacy, we innovatively fused imagomics and somatic mutation characteristics to build a predictive model of complete pathological remission of neoadjuvant chemotherapy, and explored the association between high-frequency gene mutations and drug resistance events. Professor Wang Peijun of Tongji Hospital, Tongji University School of Medicine, pointed out that cross-scale multi-omics research led by imaging, driven by multi-omics, provides imaging basis for the evolution and development of breast tumors, and is a new direction for breast tumor imaging research in the era of big data. Professor Gu Yajia's team explores the potential treatment strategies for imaging heterogeneity in tumors, explores the key features of visualization of molecular targets based on macroscopic images, and innovatively integrates popular MRI omics with mutation features of acute cells to construct the efficacy prediction model of neoadjuvant chemotherapy, which guides the accurate diagnosis and treatment of breast cancer on the basis of accurate typing and visualization of key molecules, providing new prospects for clinical translation.