Lung cancer is among the most prevalent cancers and has the highest mortality rate globally. The diagnosis, pathohistological classification, and molecular testing of lung cancer primarily rely on tissue biopsy or surgical resection. These methods are invasive and associated with limitations, including sample quantity and quality, as well as patient tolerance. Radiomics, an emerging technology, enables the extraction of high-throughput quantitative information from medical images, providing radiomic features applicable to clinical diagnosis and treatment. Significant advancements have been made in the application of radiomics to the diagnosis, molecular detection, efficacy prediction, and prognosis of lung cancer. This review examines the progress in radiomics for individualized and precise diagnosis and treatment of lung cancer in recent years.