Xiaoxia Duan, Yuxin Guo, Yanyan Yu, Sheng Li, Yujie Sun
Precise anatomical understanding and spatial cognition are fundamental to effective oncologic management. While traditional two-dimensional imaging serves as the diagnostic standard, it often lacks the intuitive depth required for complex decision-making. This review synthesizes the role of visualization-driven digital technologies in oncology-including AI-driven 3D reconstruction, augmented reality (AR), multimodal fusion, radiomics, and digital twins-as cognitive interfaces across the entire continuum of cancer care. We comprehensively review current applications, ranging from deep learning-enhanced early screening and segmentation in pre-treatment planning to AR-guided surgical navigation and adaptive radiotherapy during treatment. Furthermore, we highlight the emerging utility of radiomics in post-treatment surveillance and the humanistic value of immersive visualization in patient education for anxiety reduction. Despite these advancements, clinical integration remains impeded by challenges related to intraoperative precision stability, economic feasibility in resource-limited settings, and data privacy risks associated with biometric re-identification. Looking ahead, we propose that the future of oncologic visualization lies in the convergence of physics-informed deep learning for dynamic motion correction, cloud-based edge computing for cost democratization, and privacy-preserving collaborative learning strategies. Current advances in visualization technologies support the transition from passive anatomical viewing toward more interactive, data-driven clinical decision support, though the integration of these approaches into comprehensive predictive models requires further prospective validation.