Search

Background: To address issues in current deep learning models for detecting defects on industrial bearing surfaces, such as large parameter sizes and low precision in identifying small defects, we propose a lightweight detection algorithm for small-sized bearing appearance defects. Methods: First, we introduce a large separable convolution attention module on the spatial pyramid pooling fusion module. The deep convolutional layer with large convolutional kernels effectively captures more extensive context information of small-sized bearing defects while reducing the computation burden and learns attention weights to adaptively select the importance of input features. Secondly, we integrate the SimAM (simple attention mechanism) into the model without increasing the original network parameters, thereby augmenting the capacity to extract small-sized features and enhancing the model's feature fusion capability. Finally, utilizing SIoU (Scylla IoU) as the regression loss and Soft-NMS (soft non-max suppression) for handling redundant boxes strengthens the model's capacity to identify overlapping areas. Results: Experimental results demonstrate that our improved YOLOv8n model, sized at 6.5 MB, outperforms the baseline in terms of precision, recall, and mAP (mean average precision), with FPS (frames per second) of 146.7 (f/s), significantly enhancing bearing defect recognition for industrial applications. [ABSTRACT FROM AUTHOR]

Purpose: The aim of the study was to compare the different internal fixations between elastic stable intramedullary nailing and Kirschner wires in treatment of angulated radial neck fractures. Methods: We retrospectively reviewed the patients with radial neck fracture without associated injuries who underwent surgery approach in our department during April 2011–March 2020. There were 62 patients meeting all the criteria with complete clinical data, with median age of 7.5 (IQR 5.8–9.5) years, 34 males and 28 females. The preoperative fracture pattern was assessed according to the Judet classification system. Depending on the materials implanted and fixation strategy, the patients could be divided into a Kirschner wire group and an elastic stable intramedullary nailing group. Final functional outcomes of patients were assessed by the Mayo Elbow Performance Score and Tibone–Stoltz functional evaluation classification. Results: The Kirschner wire group included 37 patients, with 4.8 years median follow-up. The elastic stable intramedullary nailing group included 25 patients with 5.9 years median follow-up. There were no significant differences in gender, age, Judet classification, average operative time, Mayo Elbow Performance Score, Tibone–Stoltz classification, or length of hospital stay between groups. However, the time to union in the Kirschner wire group was significantly shorter than that in the elastic stable intramedullary nailing group (p < 0.05). Both groups achieved satisfactory functional and cosmetic results. Conclusion: In the management of pediatric radial neck fractures, both elastic stable intramedullary nailing and Kirschner wire internal fixation have shown equivalent therapeutic results, leading to satisfactory functional outcomes. The selection of the internal fixation approach can be influenced by the patient's fracture characteristics and the surgeon's preferences. Level of evidence: Level III; Retrospective Comparison; Treatment Study. [ABSTRACT FROM AUTHOR]