Currently implemented accuracy metrics in open-source libraries for segmentation by supervised machine learning are typically one-dimensional scores [1]. While extremely relevant to evaluate applicability in clinics, anatomical location of segmentation errors is often neglected. This study aims to include the three-dimensional (3D) spatial information in the development of a novel framework for segmentation accuracy evaluation and comparison between different methods. Predicted and ground truth (manually segmented) segmentation masks are meshed into 3D surfaces. A template mesh of the same anatomical structure is then registered to all ground truth 3D surfaces. This ensures all surface points on the ground truth meshes to be in the same anatomically homologous order. Next, point-wise surface deviations between the registered ground truth mesh and the meshed segmentation prediction are calculated and allow for color plotting of point-wise descriptive statistics. Statistical parametric mapping includes point-wise false discovery rate (FDR) adjusted p-values (also referred to as q-values). The framework reads volumetric image data containing the segmentation masks of both ground truth and segmentation prediction. 3D color plots containing descriptive statistics (mean absolute value, maximal value,…) on point-wise segmentation errors are rendered. As an example, we compared segmentation results of nnUNet [2], UNet++ [3] and UNETR [4] by visualizing the mean absolute error (surface deviation from ground truth) as a color plot on the 3D model of bone and cartilage of the mean distal femur. A novel framework to evaluate segmentation accuracy is presented. Output includes anatomical information on the segmentation errors, as well as point-wise comparative statistics on different segmentation algorithms. Clearly, this allows for a better informed decision-making process when selecting the best algorithm for a specific clinical application.
The effect of pelvic tilt (PT) and sagittal balance in hips with pincer-type femoroacetabular impingement (FAI) with acetabular retroversion (AR) is controversial. It is unclear if patients with AR have a rotational abnormality of the iliac wing. Therefore, we asked: are parameters for sagittal balance, and is rotation of the iliac wing, different in patients with AR compared to a control group?; and is there a correlation between iliac rotation and acetabular version? A retrospective, review board-approved, controlled study was performed including 120 hips in 86 consecutive patients with symptomatic FAI or hip dysplasia. Pelvic CT scans were reviewed to calculate parameters for sagittal balance (pelvic incidence (PI), PT, and sacral slope), anterior pelvic plane angle, pelvic inclination, and external rotation of the iliac wing and were compared to a control group (48 hips). The 120 hips were allocated to the following groups: AR (41 hips), hip dysplasia (47 hips) and cam FAI with normal acetabular morphology (32 hips). Subgroups of total AR (15 hips) and high acetabular anteversion (20 hips) were analyzed. Statistical analysis was performed using analysis of variance with Bonferroni correction.Aims
Methods
