1255-Multi-Omics Framework for Colorectal Cancer Prognosis

This research introduces a deep learning radiomics model (DLRM) designed to improve survival predictions for colorectal cancer (CRC) patients by analyzing CT imaging. By evaluating over one hundred machine learning combinations, researchers developed a risk stratification system that effectively categorizes patients into high- and low-risk groups across multiple clinical centers. Beyond imaging, the study integrates transcriptomic and metabolomic data to uncover the biological drivers behind these survival differences. Findings indicate that high-risk tumors are characterized by aggressive structural remodeling, while low-risk tumors show stronger immune system activity and specific metabolic protections. Ultimately, the authors combine these imaging signatures with clinical factors like CEA levels and N stage to create a highly accurate prognostic nomogram. This comprehensive framework offers a non-invasive method to guide individualized treatment and better understand the molecular landscape of the disease.References:Li Z, Cai R, Qin Y, et al. Integration of radiomics, deep learning, transcriptomics, and metabolomics reveals prognostic risk stratification and underlying biological mechanisms in colorectal cancer[J]. NPJ Precision Oncology, 2026.前往小宇宙评论区与主播互动