Abstract: The incidence and mortality of lung cancer are extremely high, and the five-year survival rate of patients after conforming lung cancer is less than 20%, so early screening and intervention are particularly important. However, doctors are overwhelmed with a large workload in a short period of time and need to ensure that the rate of missed diagnosis and misdiagnosis is low. Artificial intelligence can assist doctors in early diagnosis and precise treatment of lung cancer in different ways, but there are also some problems and limitations. This paper reviews the application of artificial intelligence in the prediction of benign and malignant lung nodules, pathology, gene mutation, intervention scheme, treatment response and prognosis, and the challenges it faces.