Segmentation of anatomical components is a major step in creating accurate and realistic 3D models of the human body, which are used in many clinical applications, including orthopedics. Recently, many deep learning approaches have been proposed to solve the problem of manual segmentation, which is time-consuming and operator-dependent. In the present study, SegResNet has been adapted from other domains, such as brain tumors, for knee joints, in particular, to segment the femoral bone from magnetic resonance images. This algorithm has been compared to the well-known U-Net in terms of evaluation metrics, such as the Dice similarity coefficient and Hausdorff distance. In the training phase, various combinations of hyperparameters, such as epochs and learning rates, have been tested to determine which combination produced the most accurate results. Based on their comparable results, both U-Net and SegResNet performed well in accurately segmenting the femur. Dice similarity coefficients of 0.94 and Hausdorff distances less than or equal to 1 mm indicate that both models are effective at capturing anatomical boundaries in the femur. According to the results of this study, SegResNet is a viable option for automating the creation of 3D femur models. In the future, the performance and applicability of SegResNet in real-world settings will be further validated and tested using a variety of datasets and clinical scenarios.
14 March 2024 in Archives of orthopaedic and trauma surgery
Introduction: Clinical gait analysis can be used to evaluate the recovery process of patients undergoing total hip arthroplasty (THA). The postoperative walking patterns of these patients can be significantly influenced by the choice of surgical approach, as each procedure alters distinct anatomical structures. The aim of this study is twofold. The first objective is to develop a gait model to describe the change in ambulation one week after THA. The secondary goal is to describe the differences associated with the surgical approach. Materials and methods: Thirty-six patients undergoing THA with lateral (n = 9), anterior (n = 15), and posterior (n = 12) approaches were included in the study. Walking before and 7 days after surgery was recorded using a markerless motion capture system. Exploratory Factor Analysis (EFA), a data reduction technique, condensed 21 spatiotemporal gait parameters to a smaller set of dominant variables. The EFA-derived gait domains were utilized to study post-surgical gait variations and to compare the post-surgical gait among the three groups. Results: Four distinct gait domains were identified. The most pronounced variation one week after surgery is in the Rhythm (gait cycle time: $$+32.9\mathrm{\%}$$ ), followed by Postural control (step width: $$+27.0\mathrm{\%}$$ ), Phases (stance time: $$+11.0\%$$ ), and Pace (stride length: − $$9.3\%$$ ). In postsurgical walking, Phases is statistically significantly different in patients operated with the posterior approach compared to lateral (p-value = 0.017) and anterior (p-value = 0.002) approaches. Furthermore, stance time in the posterior approach group is significantly lower than in healthy individuals (p-value < 0.001). Conclusions: This study identified a four-component gait model specific to THA patients. The results showed that patients after THA have longer stride time but shorter stride length, wider base of support, and longer stance time, although the posterior group had a statistically significant shorter stance time than the others. The findings of this research have the potential to simplify the reporting of gait outcomes, reduce redundancy, and inform targeted interventions in regards to specific gait domains.
The objective of this study was to develop a U-net capable of generating highly accurate 3D models of knee bones, in particular the femur. As part of the approach, a U-net was designed, trained, and validated. In order to achieve these goals, a novel architecture was proposed, including an architecture that reduces encoder parameters and incorporates transfer learning, in order to enhance the attention U-net. Additionally, an extra depth layer was added to extract more salient information. Moreover, the model includes a classifier unit to reduce false positives, as well as a Tversky focal loss function, which is an innovative loss function. The proposed architecture achieved a Dice coefficient of 98.05. By using these enhanced tools, clinicians can visualize and analyze knee structures more accurately, improve surgical intervention effectiveness, and improve patient care quality overall.
28 July 2021 in Computer-Aided Design and Applications
Since every structure in the human body can vary, customization is important to choose the most appropriate medical option according to the patient. Total knee arthroplasty (TKA) is a surgical procedure for the knee replacement that has a high rate of patient’s dissatisfaction. Indeed, conventional prostheses are based on anthropometric data that accommodate common knees. However, mismatch can occur due to anatomical variations among the individuals. Thanks to the advances in imaging techniques and 3D modeling, it is possible to create customized knee implants starting from medical images. In this context, the present research proposes a methodology to design a customized knee implant taking into account clinical (e.g., prosthesis alignment and surgical cuts) and technical parameters (e.g., materials) that have a direct impact on TKA performance and patient’s satisfaction. Changing these parameters, different scenarios have been modeled and simulated to understand the most suitable combination. Finite element analysis (FEA) has been employed to simulate and compare the proposed customized models, changing the different clinical and technical parameters. Stress induced by different combinations of the parameters has been evaluated to choose the optimal solution among the eight proposed scenarios. The optimum is reached with a physiological alignment, with six femoral facets and the ultra-high molecular weight polyethylene (UHMWPE) tibial insert. The implant design maintains the natural joint line and allows preserving more bone. The material is the parameter that mostly influences the stress distribution.
Anterior cruciate ligament (ACL) reconstruction is a widely performed procedure for ACL injury, but there are several factors which may lead to re-rupture or clinical failure. An intercondylar notch (or fossa) that is narrower may increase the likelihood of injury. Traditional two-dimensional assessments are limited, and three-dimensional (3D) volume analysis may offer more detailed insights. This study employs deep learning and statistical shape modeling (SSM) to enhance 3D modeling of the intercondylar notch, aiming to gain a deeper understanding of this complex 3D anatomical region. A methodology was developed to generate accurate 3D models of the intercondylar fossa within seconds. The variability of the intercondylar notch in ACL-injured samples was analyzed using SSM techniques, focusing on its principal components. Additionally, gender differences in notch volume were examined using t-tests. The best deep learning method for automatic segmentation of the notch was SegResNet, which achieved a Dice similarity coefficient of over 0.88 and a Hausdorff distance of 0.73 mm. The small volume-related relative error (0.06) illustrates the goodness of the result. Three principal components accounted for 72.59% of the variation, including notch volume, shape, width, and height. Females had statistically significant smaller notch compared with males with ACL injury (P < 0.001). By examining notch volume and its variability in ACL-injured patients, it is possible to understand the complex anatomy of the intercondylar notch and tailor ACL reconstructions accordingly.