Artificial intelligence (AI)–assisted three-dimensional (3D) modeling has expanded the role of medical imaging in surgical planning; however, its clinical value is often conflated with advanced visualization rather than true decision support. This Mini Review critically examines AI-driven 3D modeling as a precision tool for surgical planning, emphasizing the distinction between static anatomical reconstructions and dynamic, intelligence-driven systems capable of adapting to intraoperative conditions. Beyond classical convolutional neural networks, contemporary architectures such as Vision Transformers and diffusion-based models are discussed, highlighting their implications for generalizability, uncertainty estimation, and robustness. Attention is given to imaging standardization, algorithmic responsibility, economic thresholds for adoption, and the persistent gap between visualization and quantifiable surgical benefit.
References
Topol EJ et al. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019.
Clifton DA et al. Dynamic organ modeling using intraoperative data. IEEE Trans Med Imaging. 2024.
Grant MJ, Booth A. A typology of reviews: an analysis of 14 review types and associated methodologies. Health Info Libr J. 2009 Jun;26(2):91-108. doi: 10.1111/j.1471-1842.2009.00848.x. PMID: 19490148.
Ronneberger O et al. U-Net: Convolutional networks for biomedical image segmentation. MICCAI. 2015.
Dosovitskiy A et al. An image is worth 16×16 words: Transformers for image recognition. ICLR. 2021.
Ho J et al. Denoising diffusion probabilistic models. Med Image Anal. 2023.
McBee MP et al. Imaging standardization and AI bias in clinical deployment. Radiology. 2024.
Raza A et al. Clinical governance of artificial intelligence in surgery. Ann Surg. 2024.
Fritz J et al. Cost-effectiveness of AI-assisted surgical planning. Eur J Surg Oncol. 2025.