RESUMEN
Proximal femoral fracture segmentation in computed tomography (CT) is essential in the preoperative planning of orthopedic surgeons. Recently, numerous deep learning-based approaches have been proposed for segmenting various structures within CT scans. Nevertheless, distinguishing various attributes between fracture fragments and soft tissue regions in CT scans frequently poses challenges, which have received comparatively limited research attention. Besides, the cornerstone of contemporary deep learning methodologies is the availability of annotated data, while detailed CT annotations remain scarce. To address the challenge, we propose a novel weakly-supervised framework, namely Rough Turbo Net (RT-Net), for the segmentation of proximal femoral fractures. We emphasize the utilization of human resources to produce rough annotations on a substantial scale, as opposed to relying on limited fine-grained annotations that demand a substantial time to create. In RT-Net, rough annotations pose fractured-region constraints, which have demonstrated significant efficacy in enhancing the accuracy of the network. Conversely, the fine annotations can provide more details for recognizing edges and soft tissues. Besides, we design a spatial adaptive attention module (SAAM) that adapts to the spatial distribution of the fracture regions and align feature in each decoder. Moreover, we propose a fine-edge loss which is applied through an edge discrimination network to penalize the absence or imprecision edge features. Extensive quantitative and qualitative experiments demonstrate the superiority of RT-Net to state-of-the-art approaches. Furthermore, additional experiments show that RT-Net has the capability to produce pseudo labels for raw CT images that can further improve fracture segmentation performance and has the potential to improve segmentation performance on public datasets. The code is available at: https://github.com/zyairelu/RT-Net.
RESUMEN
Objective: Displaced intra-articular tongue-type fractures are often treated with surgical interventions, and there is a lack of consensus regarding the surgical approach. This retrospective cohort study aimed to compare percutaneous cannulated screw (PCS) fixation and plating with a minimally invasive longitudinal approach (MILA) after closed reduction for the treatment of tongue-type calcaneal fractures. Materials and Methods: A total of 77 patients with intra-articular tongue-type calcaneal fractures between September 2015 and July 2019 were included in this study. They were randomly allocated into two groups: PCS fixation (n = 32) and MILA (n = 45). The outcome measures included demographic variables, operation time (OT), preoperative time (POT), hospital stay time (HST), blood loss, visual analog scale (VAS) scores, American Orthopaedic Foot and Ankle Society (AOFAS) hindfoot scores, postoperative complications, and imaging parameters. The patients were clinically examined at 1, 3, 6, and 12 months, with a final follow-up period of 27 months. Results: No significant differences were observed during the follow-up in calcaneal length, height, Gissane's and Böhler's angles, VAS scores, AOFAS hindfoot scores, or complication rates between the two groups. However, the postoperative VAS scores in the PCS group were significantly lower than those in the MILA group (p < 0.05). Furthermore, the OT, POT, and HST in the PCS group were significantly shorter than those in the MILA group (p < 0.05). Blood loss was lower in the PCS group than those in the other group (p = 0.044). However, postoperative calcaneal widening was significantly smaller in the MILA group than that in the PCS group (p < 0.001). Conclusions: After closed reduction for the treatment of tongue-type calcaneal fractures, PCS fixation was superior to MILA in terms of OT, POT, HST, blood loss, pain, and degree of comfort. Meanwhile, MILA has the advantage of restoring the calcaneal width. Under the same rehabilitation program, the two approaches showed similar abilities in maintaining the closed reduction.
RESUMEN
Objective: To compare the clinical outcomes between use of sliding fixation (three cannulated screws, TCS) and non-sliding fixation (four cannulated screws, FCS) in the treatment of femoral neck fractures. Methods: We retrospectively analyzed 102 patients with fresh femoral neck fractures treated with TCS (60 cases) and FCS (42 cases) between January, 2018 and December, 2019. The demographic data, follow-up time, hospitalization time, operation time, blood loss, length of femoral neck shortening (LFNS), soft tissue irritation of the thigh (STIT), Harris hip score, and complications (such as internal fixation failure, non-union, and avascular necrosis of the femoral head) were also collected, recorded, and compared between the two groups. Results: A total of 102 patients with an average age of 60.9 (range, 18-86) years were analyzed. The median follow-up time was 25 (22 to 32) months. The LFNS in the FCS group (median 1.2 mm) was significantly lower than that in the TCS group (median 2.8 mm) (P < 0.05). In the Garden classification, the number of displaced fractures in the TCS group was significantly lower than that in the FCS group (P < 0.05). The median hospitalization time, operation time, blood loss, reduction quality, internal fixation failure rate (IFFR), STIT, and Harris hip score were not statistically different between the two groups (P > 0.05). However, in the subgroup analysis of displaced fractures, the LFNS (median 1.2 mm), STIT (2/22, 13.6%), and Harris hip score (median 91.5) of the FCS group at the last follow-up were significantly better than the LFNS (median 5.7 mm), STIT (7/16, 43.8%), and Harris hip score (median 89) of the TCS group (P < 0.05). No complications such as incision infection, deep infection, pulmonary embolism, or femoral head necrosis were found in either group. Conclusion: TCS and FCS are effective for treating femoral neck fractures. For non-displaced fractures, there was no significant difference in the clinical outcomes between the two groups. However, for displaced fractures, the LFNS of the FCS is significantly lower than that of the TCS, which may reduce the occurrence of STIT and improve the Harris hip score.
