Abstract:Colorectal cancer (CRC) is the third most common malignant tumor in the world, and its precise preoperative staging is crucial for treatment decisions. Currently, imaging techniques have their own limitations in staging applications; for example, computed tomography (CT) has high sensitivity for detecting distant metastasis but insufficient accuracy in distinguishing early T-stage, while magnetic resonance imaging (MRI) is highly accurate in Tstaging but has low sensitivity for detecting small lymph node metastases. This article reviews the boundaries of clinical applications of CT, MRI, ultrasound (US), and positron emission tomography/computed tomography (PET/CT), and focuses on innovative breakthroughs in imaging histology and artificial intelligence (AI), including the application of AI in staging automation and efficiency enhancement, as well as advances in multimodal image fusion and model interpretability research. These technological advances provide a new paradigm for individualized diagnosis and treatment of CRC and need to be validated by multicenter data to promote clinical translation in the future.
