机器人外科学杂志（中英文）2026，Vol.7Issue(1)：63-69,7.DOI:10.12180/j.issn.2096-7721.2026.01.009
基于几何特征约束的机器人辅助骨科手术护理路径优化研究
Optimization of nursing path for robot-assisted orthopedic surgery based on geometric feature constraints
摘要
Abstract
Objective:An optimization model based on geometric feature constraints is proposed to address the problem of nursing path optimization in robot-assisted orthopedic surgery.Method:A three-layer neural network model is developed,where simulated annealing is used to initialize the network weights to avoid local minima,and backpropagation is employed to adjust the weights for path optimization.Results:The model can achieve optimal performance when the hidden layer nodes is 20 and the learning rate is 0.0008,with a mean absolute error of 0.214,a root mean square error(RMSE)of 0.286,and a mean absolute percentage error(MAPE)of 0.352.After three rounds of path correction,the final path accuracy reaches 0.005 mm for the left lamina and 0.006 mm for the right lamina,with a real-time correction response time of 1.12 seconds.Conclusion:The model proposed in this study is an effective approach to optimizing the nursing path in robot-assisted orthopedic surgery.关键词
机器人辅助骨科手术/护理路径优化/模拟退火算法/反向传播神经网络/几何特征Key words
Robot-assisted Orthopedic Surgery/Nursing Pathway Optimization/Simulated Annealing Algorithm/Backpropagation Neural Network/Geometric Feature分类
医药卫生引用本文复制引用
陈佳静,付静文,宫春利,魏红红,王茜..基于几何特征约束的机器人辅助骨科手术护理路径优化研究[J].机器人外科学杂志（中英文）,2026,7(1):63-69,7.基金项目
陕西省重点研发计划项目(2023-ZDLSF-14)Key R&D Plan Project of Shaanxi Province(2023-ZDLSF-14) （2023-ZDLSF-14）
