目的基于解剖标志建立股骨三维坐标系,实现股骨近端解剖参数的自动化测量,辅助术前规划及个性化股骨柄设计。方法自主研发专用于股骨解剖测量的Femeter软件,可读取显示股骨STL模型,手工选取目标解剖标志点。运用半球形搜索迭代方法实现髓腔快速三维建模,运用经典最小二乘拟合等算法实现股骨头半径、颈干角、前倾角等相关16个测量参数的自动计算。结果通过30例股骨STL模型进行测量实验,平均单次测量的算法执行时间为(0.95±0.16)s,9个解剖测量项的评定者内和评定者间可靠性较高(组内相关性系数为0.907~0.999)。结论 Femeter软件的自动建模和测量算法执行速度较快,测量可重复性较好,操作简易,适用于临床。自动化三维股骨近端解剖测量可为人工全髋关节置换术前规划、股骨柄选择和个性化股骨柄设计提供可靠数据。 Objective To establish a three-dimensional coordinate system of femur based on anatomical landmarks to realize automatic measurement of anatomic parameters of the proximal femur, to assist preoperative planning and design of individual femoral stems. Methods The Femeter software, which is specially designed for anatomical measurement of femur, is developed independently. The STL model of the femur can be read and selected, and the target anatomical landmark points can be manually selected. The hemispherical search iterative method is used to realize rapid three-dimensional modeling of the medullary canal. The classical least squares fitting algorithm is used to automatically calculate the 16 measurement parameters such as the femoral head radius, the neck angle and the anteversion angle. Results The experiment was performed in 30 femoral STL models. The average single-time measurement was 0.95 ± 0.16 s, and the reliability of the nine anatomical items was higher among assessors and assessors (intra-group correlation coefficient 0.907 ~ 0.999). Conclusion Femeter software automatic modeling and measurement algorithms perform faster, better measurement repeatability, simple operation, suitable for clinical. Automated 3D proximal femur anatomic measurements provide reliable data for preoperative total hip arthroplasty planning, stem selection, and individualized femoral stem design.