The invention discloses a model for predicating lung cancer risks for people suffering from pulmonary nodules among the China urban population on the basis of CT (computed tomography) images and biomarker spectrums. The model comprises biomarkers, enzyme-labeled antibodies of the biomarkers, a carbonate buffer solution with the pH value being 9.6, a phosphate buffer solution with a pH value being 7.4, a serum protein diluent, a stop solution, a tetramethyl benzidine substrate solution, serum of normal people and positive control serum; the biomarkers are four biomarkers including pro-gastrin-releasing peptides, carcino-embryonic antigens, fragments of cytokeratin 19 and squamous-cell carcinoma antigens and are combined with CT imageology parameters and other clinical indexes. The model can assist in screening of lung cancer.