The intricate interplay between systemic immunity and tumors profoundly influences not only the onset and progression of tumors but also serves as a crucial window into understanding tumor evolution and treatment status. This project aims to establish a large, multi-center pancreatic cancer cohort and a standardized clinical sample repository, capturing multimodal immunity big data on pancreatic cancer occurrence, progression, and treatment response across the spatial dimension of peripheral versus local and the temporal dimension of tumor evolution/pre- and post-treatment. By integrating patient imaging and clinical information, the investigators will develop a technical platform for intelligent extraction and fusion analysis of cross-scale, multimodal data, thereby mapping the immunity landscape of pancreatic cancer evolution, identifying characteristic changes in immunity parameters, and devising an AI model for early pancreatic cancer diagnosis with high accuracy and robust generalization capabilities, while also exploring the model's interpretability. Additionally, the investigators will focus on specific immune cell subsets associated with pancreatic cancer evolution and treatment response, elucidating their roles and mechanisms in the occurrence, development, and drug resistance of pancreatic cancer. This project will establish a high-quality, standardized, and shareable pancreatic cancer immunity sample and data repository, refine the framework for multi-dimensional immunity data fusion and analysis, and create a time-space atlas of local and systemic immunity in pancreatic cancer. This will facilitate a deeper understanding of the dynamics of systemic and local immunity in early-stage pancreatic cancer and at various stages of its evolution, offering novel insights and approaches for the diagnosis and treatment of pancreatic cancer.