Abstract: Non-small cell lung cancer (NSCLC) is the leading cause of cancer-related morbidity and mortality worldwide and is the most prevalent thoracic malignancy in China. With regards to the diagnosis and treatment of NSCLC, advances in radiomics have enabled the high-throughput extraction and quantification of intratumoral spatial heterogeneity. ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT), which integrates high-resolution anatomical details with functional metabolic data, has been widely used in oncologic imaging. By synergistically combining ¹⁸F-FDG PET/CT with radiomics to construct machine learning models, the clinical utility of this modality has markedly expanded, and it has been rapidly integrated into the comprehensive management of lung cancer. In the diagnostic setting, these integrated models can differentiate benign from malignant pulmonary nodules with high reliability, thereby enhancing the accuracy of early stage NSCLC detection and enabling a non-invasive prediction of histopathologic subtypes. Radiomics signatures can also provide robust support for precision oncology by facilitating accurate TNM staging, predicting actionable genomic alterations, and estimating immune checkpoint expression levels. Moreover, machine learning-based radiomic analyses are valuable in treatment response monitoring and prognostication, effectively forecasting therapeutic efficacy and patient survival outcomes. In this review, we summarize recent advances in the application of ¹⁸F-FDG PET/CT radiomics across the entire continuum of NSCLC care.