Abstract: Recently, immunotherapy has become the firstline treatment for some patients with advanced lung cancer, offering a new therapeutic method for the patients with poor response to traditional treatment. Response evaluation criteria in solid tumors (RECIST) is still the standard for the therapy evaluation of solid tumor. However, due to the particularity of immunotherapy, traditional RECIST evaluation cannot accurately evaluate the therapeutic effect of tumor in time. Medical imaging has entered a new era of functional imaging, which can provide information on micromolecules and metabolic changes. Artificial intelligence and radiomics can provide more diagnostic information by mining more mathematical rules from the radiological data, therefore is expected to provide new methods and new ideas for the efficacy evaluation and prediction of immunotherapy for lung cancer. This article reviews the mechanism, current development status, and evaluation criteria of immunotherapy for advanced lung cancer, focusing on the imaging analysis of efficacy evaluation and prediction, and makes a prospect analysis.