ABSTRACT
Radio genomics has emerged as one of the most promising disciplines in precision oncology by establishing biological relationships between quantitative imaging phenotypes and underlying genomic alterations. Conventional radio genomic approaches have demonstrated considerable potential for non-invasive molecular characterization of tumours but remain constrained by handcrafted imaging features, limited multimodal integration, and inadequate generalizability across diverse clinical populations. Recent advances in foundation artificial intelligence (AI) models have fundamentally transformed radio genomics through large-scale self-supervised learning, multimodal transformer architectures, graph neural networks, and generative AI capable of learning generalized biomedical representations from radiological imaging, digital pathology, genomics, transcriptomics, proteomics, metabolomics, epigenomics, spatial biology, laboratory biomarkers, electronic health records, and longitudinal clinical outcomes. These intelligent computational ecosystems enable comprehensive integration of imaging phenotypes with molecular biology to support biomarker discovery, molecular subtype classification, therapeutic response prediction, digital twin simulation, adaptive treatment planning, and intelligent clinical decision support. Emerging technologies including multimodal large language models, federated learning, reinforcement learning, retrieval-augmented generation, explainable artificial intelligence, and agentic AI further strengthen radio genomic intelligence by enabling collaborative, privacy-preserving, and continuously adaptive computational oncology. Despite remarkable technological advances, important scientific, technical, ethical, and regulatory challenges remain regarding multimodal data harmonization, computational scalability, interpretability, interoperability, cybersecurity, clinical validation, and equitable implementation. This review provides a comprehensive overview of intelligent radio genomics, emphasizing foundation AI models for imaging-genomic integration as a transformative paradigm in precision cancer medicine.[1]
Keywords: Radio genomics, Foundation models, Artificial intelligence, Precision oncology, medical imaging, Genomics, Radiomics, Multi-omics, Computational oncology, Personalized medicine.
