Purpose <p>Lung cancer continues to be a leading cause of cancer-related mortality worldwide, with late diagnosis and tumor heterogeneity contributing to poor patient outcomes. Early and accurate detection is therefore critical for improving survival rates. Over the last few years, artificial intelligence (AI) which includes machine learning (ML), deep learning (DL), and radiomics has become a powerful tool to perform automated lung cancer detection and classification.</p> Methods <p>This review systematically looks into 30 studies published from 2016 to 2025, utilizing ML, DL, and radiomics-based methods for lung cancer identification, tumor characterization, and prognosis. The evaluation of model performance is done using various metrics such as accuracy, area under the curve (AUC), sensitivity, specificity and F1-score. Alongside each approach, we present the strengths and weaknesses of the respective approaches.</p> Results <p>The machine learning models such as support vector machines, random forests, and ensemble classifiers have been very successful with structured clinical data whereas, the deep learning models, especially the convolutional neural networks, acquired the highest performance in automated imaging analysis and tumor segmentation. Radiomics was used to increase the predictive ability of the model by providing quantitative high-dimensional features of the medical images.</p> Conclusion <p>LungNet is a multimodal artificial intelligence system, combining clinical, imaging and genomic data based on hybrid deep learning and transformer modules, to address the challenges of data heterogeneity, interpretability and transferability. This review discusses the role of multimodality, the challenges ahead, and future directions for developing AI as a reproducible and clinically actionable diagnostic tool for lung cancer.</p>