Multimodal Data Fusion for Lung Cancer Diagnosis: A Review Study

Lung cancer is the world’s leading cause of cancer related deaths, so the early-stage diagnosis is crucial to improve the survival rate. Multimodality fusion of diverse data modalities, such as organized clinical records, molecular profiles (genomics, proteomics), and radiological imaging (CT/PET), is applied for lung cancer diagnosis. It may enhance prognostication, therapeutic decision-making, and diagnostic accuracy. Recent developments (2020–2025) in multimodal data fusion techniques for lung cancer diagnosis are summarized in this review, along with important multimodal public datasets, representative models and architectures, a thorough comparison table of significant studies, research gaps, and technical difficulties that should direct future investigations, are taken care of. This study suggests specific research avenues with a focus on clinical translation, interpretability, and robustness.