AI-Assisted 3D Reconstruction for Lung Resection Planning

This prospective observational study aims to validate an artificial intelligence (AI)-based system for automated three-dimensional (3D) reconstruction of pulmonary anatomy from preoperative chest computed tomography (CT) scans in patients undergoing lung resection. The AI system will automatically identify and reconstruct relevant pulmonary anatomical structures, including pulmonary arteries, veins, bronchi, lobes, and segments. AI-generated 3D reconstructions will be compared with expert manual 3D reconstructions and with anatomical findings documented during surgery. The primary objective is to assess the anatomical accuracy of AI-generated 3D reconstructions in identifying surgically relevant bronchovascular anatomy and anatomical variants. Secondary objectives include evaluating accuracy for individual anatomical structures, discrepancies between automated and manual reconstructions, concordance with intraoperative findings, and the time required to generate and review the automated reconstruction. The study is observational and will not modify the planned surgical procedure or standard patient care.

Inclusion Criteria: * Age ≥18 years * Scheduled for anatomical lung resection (segmentectomy, lobectomy, bilobectomy, or pneumonectomy) * Availability of a preoperative contrast-enhanced chest CT scan suitable for 3D reconstruction, with slice thickness ≤1.5 mm. * Ability to provide informed consent. Exclusion Criteria: * Preoperative CT images of insufficient quality or technical characteristics for 3D reconstruction. * Cancellation of the planned surgical procedure before surgery. * Inability or unwillingness to provide informed consent.