Artificial Intelligence–Driven Three‐Dimensional Reconstruction in Lung Cancer Surgery: Current Status and Future Perspectives

ABSTRACT Background Three‐dimensional (3D) reconstruction is an important adjunct in lung cancer surgery for visualizing pulmonary anatomy, supporting preoperative planning, and improving operative precision. Recent advances in artificial intelligence (AI) have accelerated modeling and expanded clinical and educational applications. This review summarizes current evidence and future directions of AI‐driven 3D reconstruction in thoracic surgical practice. Methods This narrative review synthesized recent studies on 3D reconstruction for lung cancer surgery, focusing on the evolution from manual to AI‐driven techniques. Representative data on preoperative assessment, intraoperative orientation, lymph node evaluation, and training applications were examined to illustrate clinical performance, workflow impact, and remaining gaps. Results High‐quality manual reconstruction remains the reference standard for detailed bronchovascular mapping but is time‐consuming and operator‐dependent. AI‐driven 3D reconstruction can generate patient‐specific models within 5–10 min, improving identification of anatomical variation, lesion localization, margin planning, and surgical decision‐making. Clinical studies report gains in planning efficiency, more accurate recognition of segmental anatomy, reduced operative complexity, and high surgeon satisfaction when AI‐derived models are incorporated into routine workflows. However, performance decreases with suboptimal CT quality or distorted post‐treatment anatomy, and broader adoption is limited by data‐security concerns, interoperability issues, and limited model interpretability. Conclusion AI‐driven 3D reconstruction represents a major step toward precision lung cancer surgery and structured thoracic training. Priority areas include developing interactive AI–human modeling platforms, establishing standardized quality‐control metrics, extending validation to complex and post‐treatment cases, and integrating multimodal clinical and imaging data to support personalized surgical decision‐making and perioperative management.