This study breaks through the limitations of a single biomarker, for the first time integrating inflammation, immunity, and metabolism dimensions, proposing a rapid calculation, non-imaging hematology prediction system, providing a low-cost, high-efficiency risk stratification tool for grassroots hospitals; the constructed nomogram model has been internally validated (C-index=0.954) and confirmed by clinical decision analysis to have good practicality, which is expected to promote the transition of AP from empirical intervention to individualized dynamic management, and improve the early warning ability and treatment efficiency of severe cases.