1Department of Orthopaedics & Traumatology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong.
2Orthopaedic Department and.
Neurosurg Focus. 2022 Apr;52(4):E10. doi: 10.3171/2022.1.FOCUS21721.
The purpose of this study was to evaluate the ability of a novel artificial intelligence (AI) model in identifying optimized transpedicular screw trajectories with higher bone mineral density (BMD) as well as higher pull-out force (POF) in osteoporotic patients.
An innovative pedicle screw trajectory planning system called Bone's Trajectory was developed using a 3D graphic search and an AI-based finite element analysis model. The preoperative CT scans of 21 elderly osteoporotic patients were analyzed retrospectively. The AI model automatically calculated the number of alternative transpedicular trajectories, the trajectory BMD, and the estimated POF of L3-5. The highest BMD and highest POF of optimized trajectories were recorded and compared with AO standard trajectories.
The average patient age and average BMD of the vertebral bodies were 69.6 ± 7.8 years and 55.9 ± 17.1 mg/ml, respectively. On both sides of L3-5, the optimized trajectories showed significantly higher BMD and POF than the AO standard trajectories (p < 0.05). On average, the POF of optimized trajectory screws showed at least a 2.0-fold increase compared with AO trajectory screws.
The novel AI model performs well in enabling the selection of optimized transpedicular trajectories with higher BMD and POF than the AO standard trajectories.
本研究旨在评估一种新型人工智能（AI）模型在识别具有更高骨密度（BMD）和更高拔出力（POF）的优化经椎弓根螺钉轨迹的能力，这些优化经椎弓根螺钉轨迹适用于骨质疏松患者。
使用 3D 图形搜索和基于 AI 的有限元分析模型开发了一种名为“Bone's Trajectory”的新型椎弓根螺钉轨迹规划系统。回顾性分析了 21 名老年骨质疏松症患者的术前 CT 扫描。AI 模型自动计算了替代经椎弓根轨迹的数量、轨迹 BMD 和 L3-5 的估计 POF。记录并比较了优化轨迹的最高 BMD 和最高 POF 与 AO 标准轨迹。
患者的平均年龄和椎体的平均 BMD 分别为 69.6 ± 7.8 岁和 55.9 ± 17.1 mg/ml。在 L3-5 的两侧，优化轨迹的 BMD 和 POF 明显高于 AO 标准轨迹（p < 0.05）。平均而言，优化轨迹螺钉的 POF 比 AO 轨迹螺钉至少增加了 2.0 倍。
新型 AI 模型在选择具有比 AO 标准轨迹更高的 BMD 和 POF 的优化经椎弓根螺钉轨迹方面表现良好。
