Journal of Clinical Medicine Research

The objective is to investigate the clinical value and safety of an AI-assisted implant planning system in complex maxillary sinus elevation surgery, and to provide evidence-based support for digital and intelligent transformation in posterior maxillary implant rehabilitation. A total of 46 patients with severe posterior maxillary bone insufficiency requiring complex maxillary sinus elevation admitted to our department from January 2023 to December 2024 were enrolled and randomly divided into a control group (23 cases, conventional CBCT manual planning) and an observation group (23 cases, AI-assisted implant planning). All patients underwent maxillary sinus elevation with simultaneous or delayed implant placement. The observation group used a deep learning-based AI system for automatic anatomical segmentation, 3D reconstruction, simulated osteotomy, implant position design, and risk prediction. The control group relied on surgeon experience for manual measurement and planning. Preoperative planning time, anatomical structure recognition accuracy, intraoperative implant placement accuracy, Schneiderian membrane perforation rate, operation duration, postoperative marginal bone loss, implant survival rate, and adverse events were compared between the two groups. Preoperative planning time in the observation group was significantly shorter than that in the control group (P<0.05). The AI system showed high accuracy in identifying maxillary sinus septa, alveolar antral artery, and residual bone height. Intraoperative deviations in three-dimensional implant position were significantly lower in the observation group (P<0.05). The Schneiderian membrane perforation rate was 4.35% in the observation group, significantly lower than 21.74% in the control group (P<0.05). Operation duration was shorter in the observation group (P<0.05). At 6 and 12 months postoperatively, marginal bone loss in the observation group was significantly less than that in the control group (P<0.05). The 12-month implant survival rate was 100% in the observation group and 95.65% in the control group, with no statistically significant difference in safety indicators between the two groups (P>0.05). In complex maxillary sinus elevation, the AI-assisted implant planning system significantly improves preoperative evaluation efficiency and surgical accuracy, reduces the risk of Schneiderian membrane perforation and other complications, minimizes marginal bone loss, and enhances the stability and long-term prognosis of implants. It exhibits reliable safety and is suitable for widespread clinical application in difficult maxillary implant cases.