Purpose: Early‐onset degeneration of the knee is linked to genetics, overload, injury, and potentially, knee morphology. The purpose of this study is to explore the characteristics of the small medial femoral condyle, as a distinct knee morphotype, by means of a landmark‐based three‐dimensional (3D) analysis and statistical parametric mapping.Methods: Sixteen knees with a small medial femoral condyle (SMC) were selected from a database of patients with distinct knee joint anatomy and 16 gender‐matched knees were selected from a control group database. 3D models were generated from the medical imaging. After normalization for size, a set of pre‐defined landmark‐based parameters was analysed for the femur and tibia. Local shape differences were evaluated by matching all bone surfaces onto each other and comparing the distances to the mean control group bone shape.Results: The small medial condyle group showed a significant association with medial compartment degeneration and had a 4% and 13% smaller medial condyle anteroposteriorly and mediolaterally, whereas the distal femur was 3% wider mediolaterally. The lateral condyle was 2% smaller anteroposteriorly and 8% wider mediolaterally. The complete tibial plateau was 3% smaller mediolaterally and the medial tibial plateau was 6% smaller.Conclusion: A new knee morphotype demonstrated an increased risk for medial compartment degeneration and was differentiated from a healthy control group based on the following morphological characteristics: a smaller medial femoral condyle and medial tibial plateau, a wider lateral femoral condyle and a wider distal femur on a smaller tibial plateau. This pilot study suggests a role for the SMC knee morphotype in the multifactorial process of medial compartment degeneration.Level of evidence: III

Manual anatomical landmarking for morphometric knee bone characterization in orthopedics is highly time-consuming and shows high operator variability. Therefore, automation could be a substantial improvement for diagnostics and personalized treatments relying on landmark-based methods. Applications include implant sizing and planning, meniscal allograft sizing, and morphological risk factor assessment. For twenty MRI-based 3D bone and cartilage models, anatomical landmarks were manually applied by three experts, and morphometric measurements for 3D characterization of the distal femur and proximal tibia were calculated from all observations. One expert performed the landmark annotations three times. Intra- and inter-observer variations were assessed for landmark position and measurements. The mean of the three expert annotations served as the ground truth. Next, automated landmark annotation was performed by elastic deformation of a template shape, followed by landmark optimization at extreme positions (highest/lowest/most medial/lateral point). The results of our automated annotation method were compared with ground truth, and percentages of landmarks and measurements adhering to different tolerances were calculated. Reliability was evaluated by the intraclass correlation coefficient (ICC). For the manual annotations, the inter-observer absolute difference was 1.53 ± 1.22 mm (mean ± SD) for the landmark positions and 0.56 ± 0.55 mm (mean ± SD) for the morphometric measurements. Automated versus manual landmark extraction differed by an average of 2.05 mm. The automated measurements demonstrated an absolute difference of 0.78 ± 0.60 mm (mean ± SD) from their manual counterparts. Overall, 92% of the automated landmarks were within 4 mm of the expert mean position, and 95% of all morphometric measurements were within 2 mm of the expert mean measurements. The ICC (manual versus automated) for automated morphometric measurements was between 0.926 and 1. Manual annotations required on average 18 min of operator interaction time, while automated annotations only needed 7 min of operator-independent computing time. Considering the time consumption and variability among observers, there is a clear need for a more efficient, standardized, and operator-independent algorithm. Our automated method demonstrated excellent accuracy and reliability for landmark positioning and morphometric measurements. Above all, this automated method will lead to a faster, scalable, and operator-independent morphometric analysis of the knee.

Purpose: The study aims to identify differences in tibiofemoral joint morphology between responders (R group, no pain) to arthroscopic partial medial meniscectomy (APMM) versus medial postmeniscectomy syndrome patients (MPMS group, recurrent pain at 2 years postmeniscectomy) in a clinically neutrally aligned patient population. The second aim was to build a morphology‐based predictive algorithm for response to treatment (RTT) in APMM.Methods: Two patient groups were identified from a large multicentre database of meniscectomy patients at 2 years of follow‐up: the R group included 120 patients with a KOOS pain score > 75, and the MPMS group included 120 patients with a KOOS pain score ≤ 75. Statistical shape models (SSMs) of distal femur, proximal tibia and tibiofemoral joint were used to compare knee morphology. Finally, a predictive model was developed to predict RTT, with the SSM‐derived morphologic variables as predictors.Results: No differences were found between the R and MPMS groups for patient age, sex, height, weight or cartilage status. Knees in the MPMS group were significantly smaller, had a wider femoral notch and a smaller medial femoral condyle. A morphology‐based predictive model was able to predict MPMS at 2 years follow‐up with a sensitivity of 74.9% (95% confidence interval [CI]: 74.4%–75.4%) and a specificity of 81.0% (95% CI: 80.6%–81.5%).Conclusion: A smaller tibiofemoral joint, a wider intercondylar notch and smaller medial femoral condyle were observed shape variations related to medial postmeniscectomy syndrome. These promising results are a first step towards a knee morphology‐based clinical decision support tool for meniscus treatment.Study Design: Case–control study.Level of Evidence: Level IIIb.

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.

Innovations in biomedical imaging and artificial intelligence (AI) computer vision algorithms support the automation and scalability of 3D anatomical modeling. Furthermore, automated landmarking methods enable precise and transparent morphometric analysis of the knee. Unique morphological traits, such as a narrow medial femoral condyle, appear to be linked to medial knee degeneration. Additionally, size discrepancies in meniscal allograft transplantation between donor supply (larger medial meniscal size) and patient demand (smaller medial meniscal size) highlight challenges in donor-recipient matching. Statistical shape analysis allows for comparison of the main modes of shape variation between responders to arthroscopic partial medial meniscectomy (APMM) and patients with medial post-meniscectomy syndrome (MPMS), a clinical diagnosis characterized by recurrent pain related to meniscal deficiency. In this PhD thesis, a prognostic model was developed for APMM to differentiate—prior to surgery—between patients likely to benefit from the procedure and those at risk of developing MPMS. By integrating knee bone morphology with clinical and demographic data into an end-to-end machine learning pipeline, we aim to optimize patient selection for APMM in populations without any of the already described risk factors for APMM failure. This data-driven approach has the potential to improve surgical success rates and reduce the occurrence of MPMS by identifying patients who may benefit from alternative treatment options.