Objective To establish a mouse model of lung adenocarcinoma brain metastasis with human luc⁺-PC-9 cells stably expressing luciferase and to compare the evaluation values of bioluminescence imaging and¹⁸F-FDG (¹⁸F-fluorodeoxyglucose) SPECT/CT in these models. Methods Suspension of luc⁺-PC-9 cells was injected into the left ventricle of BALB/c nude mice to establish a mouse model of brain metastasis from lung cancer. Bioluminescence imaging and¹⁸F-FDG SPECT/CT were used to evaluate the metastasis of tumors as compared with HE-staining pathology as a golden standard. Results The success rate of brain metastases was 85% through injecting luc⁺-PC-9 cells into the left ventricle. The number of tumor cells was positively related to the intensity of light, with a linear correlation (R²=0.96). Fluorescence was observed in the brain, spine and femur by bioluminescence imaging, and the metastases were confirmed by H & E pathological examination.¹⁸F-FDG SPECT/CT observed abnormal density collective foci in the spine or femur but not in the brain. Conclusions Injection of tumor cell suspension into the mouse left ventricle is a good method to establish a brain metastasis of lung cancer. Bioluminescence has a higher sensitivity and specificity in detecting brain metastasis and bone metastasis, with advantages of real-time, dynamical and non-invasive detection of tumor metastasis growth.¹⁸F-FDG SPECT/CT does not have superiority in detection of brain metastases but is suitable for detecting bone metastasis.