인공 고관절 치환술에서 비구방향을 평가하기 위한 전방 골반면을 정의하는 반자동 알고리즘

Abstract

The anterior pelvic plane (APP) is commonly used as a reference plane to assess acetabular cup orientation. Conventional methods for determining the APP may not be accurate and are prone to user variability. To overcome these issues, we developed a robust method to accurately extract the anterior pelvic plane independent of pelvic pose in three-dimensional pelvic computed tomography (CT) volumes. Twenty-eight pelvic CT studies for individuals imaged for suspected non-musculoskeletal conditions were obtained from our institution’s database. The APP was determined by the left and right anterior superior iliac spines (ASISs) and pubic tubercles that were automatically extracted from user-defined regions of interest (ROIs) in CT images with compensation of pelvic pose. After ASISs and pubic tubercles were extracted, the APP was determined using the least squares method. The APP defined from these landmarks was quantitatively compared to the APPs determined by an expert and a novice. Intra-observer reliability was measured to evaluate time-interval variability. Finally, we evaluate the robustness of this method to patient posture using an arbitrarily rotated volume. Intra-class correlation coefficients (ICCs) were calculated to determine inter-observer and intra-observer reliabilities. The ICC values for the four landmarks and the APP were similar between the semi-automated method and expert determination (ICC > 0.937). The ICC values for intra-observer reliability over time for our method were all 1, demonstrating perfect reliability. Furthermore, the agreement between parameters determined from the original volume and the rotated volume was nearly perfect based on ICC values. In conclusion, we developed a robust method for measuring APP using ROI boxes. This method is likely to be a useful measurement tool for the APP as it was time-invariant, and the results were similar to an experienced surgeon’s determination. Furthermore, it was less dependent on the direction of the CT slice and more robust than a measurement by a novice.