This prospective single-center pilot study evaluates the clinical feasibility, localization accuracy, and safety of the LungVision system-an AI-augmented fluoroscopic navigation platform-for real-time intraoperative localization of small pulmonary nodules during video-assisted thoracoscopic surgery (VATS). All enrolled patients underwent dual localization: preoperative marking via CT-guided dye injection or virtual bronchoscopic navigation (Broncus Archimedes), followed by intraoperative localization using the LungVision system. The primary outcome was localization success rate, defined as the proportion of patients achieving fluoroscopic tool-in-lesion confirmation. Secondary outcomes included complete resection rate, navigation time, total operative time, and perioperative complication rate.

Lung cancer is the leading cause of cancer-related mortality worldwide, and accurate intraoperative localization of small pulmonary nodules is critical for guiding thoracoscopic resection. Conventional localization methods, including CT-guided percutaneous dye injection and virtual bronchoscopic navigation, carry limitations such as radiation exposure, pneumothorax risk, and dependence on pre-procedural planning without real-time intraoperative feedback.

The LungVision system (Body Vision Medical, Israel) is an AI-augmented fluoroscopic bronchoscopic navigation platform that integrates preoperative CT imaging with real-time C-arm fluoroscopy via machine learning-based image fusion. The system provides continuous intraoperative tool-tip tracking and lesion overlay without additional radiation beyond standard fluoroscopy, enabling real-time navigational feedback during bronchoscopic procedures.

This prospective, single-center, single-arm pilot study was conducted at the Department of Thoracic Surgery, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan. A total of 14 patients with small pulmonary nodules requiring preoperative localization prior to video-assisted thoracoscopic surgery (VATS) were enrolled between January and December 2024. Each patient underwent dual localization: preoperative marking via CT-guided percutaneous dye injection or virtual bronchoscopic navigation (Broncus Archimedes Software System), followed by intraoperative localization using the LungVision system with ICG dye injection under fluoroscopic guidance.

The primary outcome was localization success rate, defined as the proportion of patients achieving fluoroscopic tool-in-lesion confirmation using the LungVision system. Secondary outcomes included complete resection rate (confirmed by pathological examination), navigation time, total operative time, and perioperative complication rate assessed up to 30 days postoperatively. Thoracoscopic resection was performed based on the combined localization findings, and all resected specimens were submitted for histopathological analysis.

This study was approved by the Institutional Review Board of Tri-Service General Hospital (IRB No.: A202303004) and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrollment.

Inclusion Criteria:

• Age 18-65 years
• Scheduled to undergo thoracoscopic resection for solitary, subsolid, or ground-glass pulmonary nodules requiring localization
• Lesion characteristics and CT imaging indicating need for preoperative localization, as determined by the attending surgeon
• Able and willing to complete the study protocol
• Able to provide written informed consent

Exclusion Criteria:

• Age >65 years
• Pregnancy
• Bleeding tendency or current use of anticoagulants
• Impaired consciousness or inability to communicate
• Cognitive impairment precluding autonomous consent

Individual participant data (IPD) will not be shared publicly. This decision is based on the small sample size (n=14), potential re-identification risk of participants, and institutional data governance policies of Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan. Aggregated, de-identified summary data supporting the findings of this study are available from the corresponding author upon reasonable request.