Abstract: Non-small cell lung cancer (NSCLC) is the most common type of lung cancer, early diagnosis and effective treatment are crucial for improving patient prognosis. ¹⁸F-fluorodeoxyglucose (FDG) PET/CT is an integrated imaging technique that combines functional and anatomical information. By detecting the glucose metabolic activity of tumors, it provides important assistance in the diagnosis, staging, treatment response evaluation, and survival prediction of NSCLC. In recent years, deep learning (DL) has advanced rapidly in the field of medical image analysis. The application of DL techniques to ¹⁸F-FDG PET/CT image analysis offers new opportunities for the diagnosis and treatment of NSCLC. The authors review the current status of DL applications in ¹⁸F-FDG PET/CT imaging for NSCLC, focusing on their value in improving accurate staging, predicting patient prognosis, and segmenting tumor regions, while also discussing existing challenges and future prospects.