Artificial Intelligence for Radiogenomic Discovery: Linking Medical Imaging with Tumor Genomics

Abstract

Cancer remains one of the leading causes of morbidity and mortality worldwide despite remarkable advances in molecular diagnostics, targeted therapeutics, immunotherapy, and precision medicine. The biological heterogeneity of malignant tumors arises from complex genomic, epigenomic, transcriptomic, proteomic, metabolic, and immune interactions that influence disease progression, therapeutic response, and clinical outcomes. Radiogenomics has emerged as a transformative discipline that seeks to establish relationships between quantitative imaging phenotypes and underlying molecular characteristics, enabling non-invasive prediction of tumor biology through routine medical imaging. Recent advances in artificial intelligence (AI) have substantially accelerated radiogenomic research by enabling automated integration of radiological imaging, digital pathology, genomic sequencing, transcriptomics, proteomics, laboratory biomarkers, and longitudinal clinical information into unified computational frameworks. Machine learning, deep learning, transformer architectures, graph neural networks, multimodal foundation models, self-supervised learning, and generative AI have demonstrated remarkable capabilities in biomarker discovery, molecular subtype classification, treatment response prediction, prognostic assessment, immunotherapy selection, and precision therapeutics. AI-driven radiogenomics has the potential to reduce dependence on invasive tissue biopsies while enabling continuous molecular monitoring throughout disease progression. Nevertheless, important challenges remain regarding multimodal data harmonization, computational scalability, explainability, interoperability, regulatory validation, and equitable clinical implementation. This review provides a comprehensiveoverview of artificial intelligence in radiogenomic discovery, highlighting computational principles, current clinical applications, implementation challenges, and future perspectives for linking medical imaging with tumor genomics in precision oncology.[1]