Multimodal Artificial Intelligence in Precision Oncology: From Integrated Data to Clinical Implementation

The rapid expansion of biomedical data in oncology, including advanced medical imaging, genomic profiling, and large-scale clinical records, has created unprecedented opportunities to better understand tumor biology and personalize treatment strategies. However, these data sources are often analyzed separately, limiting their potential to reveal the complex biological interactions that drive cancer progression and therapeutic response.Recent advances in artificial intelligence (AI) provide powerful tools to integrate heterogeneous data modalities and extract meaningful patterns from high-dimensional datasets. In particular, multimodal AI approaches that combine imaging, genomics, and clinical information have shown promise in improving tumor characterization, predicting treatment outcomes, and supporting more personalized therapeutic strategies.Despite this progress, the translation of these technologies into routine clinical practice remains challenging. Many AI models remain confined to research settings, often functioning as opaque “black boxes” with limited interpretability and uncertain clinical applicability. Furthermore, their integration into hospital workflows raises important questions regarding usability, regulatory pathways, data governance, and the impact on healthcare systems.This Research Topic aims to explore how multimodal AI frameworks can move beyond proof of-concept studies and become clinically actionable tools capable of supporting decision-making in precision oncology.This Article Collection explores the integration of artificial intelligence, medical imaging, and genomics in precision oncology from multiple complementary perspectives. The convergence of these fields is rapidly reshaping the way biomedical data are analyzed and interpreted. Artificial intelligence is not only introducing powerful computational tools, but is also contributing to a deeper transformation in the scientific approach to complex biological systems. By enabling the simultaneous analysis of heterogeneous and high-dimensional datasets, AI offers a new analytical lens through which the multifaceted nature of cancer can be investigated, revealing patterns and relationships that remain largely inaccessible to traditional analytical frameworks.In this context, the integration of imaging, genomic, and clinical data represents one of the most promising directions for advancing precision oncology. However, the successful translation of these approaches into clinical practice requires a multidisciplinary perspective that goes beyond purely methodological innovation.For this reason, this Research Topic addresses the field from several complementary viewpoints. Contributions will explore the topic from clinical and scientific perspectives, while also examining the technological foundations, the ethical implications, and the organizational and economic impact that AI-driven approaches may have on healthcare systems and hospital management.By bringing together these different dimensions, the collection aims to provide a comprehensive perspective on how multimodal artificial intelligence can evolve from experimental research into clinically actionable tools capable of supporting the future of oncology care.We welcome contributions covering a broad spectrum of topics, including but not limited to the following areas:- Multimodal Data Integration in Oncology: methodological frameworks and artificial intelligence architectures designed to integrate imaging, genomic, and clinical data for improved tumor characterization, risk stratification, and personalized treatment strategies.- Radiomics, Radiogenomics, and Imaging-Derived Biomarkers: approaches combining quantitative imaging features with genomic and molecular information to better understand tumor heterogeneity, biological behavior, and spatial disease patterns.- AI Models for Prediction of Treatment Response and Disease Progression: machine learning models aimed at predicting therapeutic outcomes, disease recurrence, metastatic risk, or survival in cancer patients.- Explainable and Interpretable AI for Clinical Decision Support: methods that improve transparency, interpretability, and trustworthiness of AI systems used in oncology, enabling clinicians to better understand model predictions and integrate them into clinical decision-making.- Translational Frameworks and Clinical Validation of AI Models: studies addressing prospective validation, reproducibility, robustness across institutions, and regulatory considerations necessary for the clinical deployment of AI-driven tools.- Integration of AI Tools into Clinical Workflows: research investigating how AI systems can be effectively embedded into multidisciplinary tumor boards, radiology workflows, and clinical decision-making processes.- Data Infrastructure, Governance, and Ethical Considerations: challenges related to data sharing, privacy protection, interoperability, and ethical frameworks for the development and deployment of large-scale multimodal oncology datasets.- Impact of AI on Healthcare Systems and Hospital Management: studies exploring how AI-driven tools influence clinical efficiency, treatment planning, resource allocation, and the economic sustainability of healthcare systems and hospital operations.- Generative AI, Large Language Models, and Agentic AI in Oncology: exploration of how emerging AI paradigms, including generative models, large language models, and agentic AI systems, can support scientific discovery, clinical reasoning, and daily clinical practice.We welcome contributions from clinicians, scientists, engineers, and healthcare managers working at the intersection of oncology, artificial intelligence, and biomedical data science. Both original research articles and perspectives addressing methodological advances, clinical validation, and healthcare system implications are encouraged. By bringing together diverse expertise and viewpoints, this Research Topic aims to stimulate interdisciplinary dialogue and contribute to shaping the future of multimodal AI-driven precision oncology. Topic editors: Marco Diego Dominietto, Enrico Capobianco, Bradley Jay Edelman Source: https://www.frontiersin.org/research-topics/79162/multimodal-artificial-intelligence-in-precision-oncology-from-integrated-data-to-clinical-implementation