Optimizing glenoid component positioning is a key determinant of stability, function and implant survival in total shoulder arthroplasty. Postoperative computed tomography (CT) is considered the reference method for assessing guide-pin or implant alignment, but it remains a retrospective evaluation and is often difficult to obtain in cadaveric research settings. Optical navigation systems such as the NDI Polaris provide immediate intraoperative geometric measurements and may represent a pragmatic alternative. This cadaveric validation study aimed to evaluate the agreement between NDI Polaris optical-navigation measurements and postoperative CT analysis for glenoid guide-pin positioning. Eight cadaveric shoulders underwent augmented-reality-guided placement of a glenoid guide-pin, followed by postoperative CT acquisition. Intraoperative optical-navigation measurements were recorded using an NDI Polaris tracking station. Two accuracy domains were analyzed relative to CT: (1) linear entry-point deviation (mm) and (2) angular deviation in version and inclination (degrees). Agreement between both measurement modalities was assessed using Bland-Altman analysis, complemented by descriptive accuracy statistics. Sixteen paired measurements were available for comparison. The mean absolute NDI-CT difference was 0.98 mm (SD 0.55; range 0.15-2.17 mm) for entry-point localization and 1.58° (SD 0.97; range 0.03-3.28°) for trajectory orientation. Bland-Altman analysis demonstrated a small systematic bias for both linear (-0.15 mm) and angular (-0.75°) parameters, with most discrepancies contained within approximately ±2 mm and ±4°, respectively. These results suggest that, under standardized acquisition conditions, optical navigation provides geometric measurements that closely approximate CT-based reference values and may serve as a practical surrogate in cadaveric research when CT access is limited. NDI Polaris optical tracking demonstrated good agreement with postoperative CT for evaluating glenoid pin positioning and may represent a practical alternative for accuracy assessment in cadaveric studies. These preliminary findings require confirmation in larger validation studies.
Meniscus informatics is a growing subject of study in the healthcare industry. One of the major hindrances to the healthcare system’s transformation is obtaining knowledge and meaningful information from complicated, high-dimensional and diverse sources. Modern biomedical research, for instance, has seen an increase in the use of complex, dissimilar, poorly documented, and generally unstructured electronic health records, imaging, sensor data and text, even after many current techniques have been used to extract more robust and useful elements from the data for analysis. New efficient standards for building end-to-end learning models from complex data are therefore needed. Therefore, the current study aims to examine the most recent research on the use of deep learning techniques for diagnosing meniscus tears and recommend creating comprehensive and meaningful interpretable structures that might benefit the healthcare industry. We also draw attention to shortcomings and the need for better technique development, and we provide new perspectives about this exciting new development in the field.
Robot-assisted surgery (RAS) is increasingly utilized in arthroplasty, with emerging evidence suggesting potential benefits, however, the overall evidence base remains immature. Given the heterogeneity of available systems, device-specific evaluation is required. The VELYS Robotic-Assisted Solution (VRAS; Johnson & Johnson MedTech) is an imageless, semi-active robotic platform for knee arthroplasty. This study provides a structured strategic evaluation of VRAS aligned with the National Institute for Health and Care Excellence Early Value Assessment (EVA) and proposes a roadmap for future evidence generation. 16 relevant studies were identified in a systematic literature search, predominantly retrospective and lacking randomized controlled trials. Findings were synthesized using a strengths, weaknesses, opportunities, and threats framework, with subsequent analysis to inform strategic recommendations. Key strengths of VRAS include avoidance of pre-operative computed tomography imaging, integration with established implant systems, a relatively short learning curve, and early signals of improved short-term outcomes. Weaknesses center on the limited quality and maturity of the evidence base, absence of long-term outcome data, and system-specific design considerations such as lack of haptic feedback boundaries. Opportunities include reduced resource burden with imageless workflows and potential optimization of patient outcomes, while threats involve high capital costs, competition from more established RAS platforms, and the risk of unmet evidence requirements within defined timelines. Across EVA domains, evidence for patient-reported outcomes, cost-effectiveness, and long-term clinical benefit remains insufficient. High-quality, adequately powered studies, comprehensive health-economic analyses, and registry-linked longitudinal studies are essential to define the clinical and economic value of VRAS in light of the EVA.
2D X-ray imaging is widely employed in image-guided interventions to provide real-time visualization of patient anatomy and interventional devices. However, lacking depth information and soft-tissue contrast, intraoperative X-rays are often complemented with high-resolution preoperative 3D computed tomography (CT) scans. Clinicians must mentally register the 3D preoperative information onto the 2D visualization, increasing cognitive workload and motivating the need for automated solutions. Traditionally, this registration task is formulated as estimating the position of the X-ray source relatively to the CT scan. State-of-the-art 3D/2D registration approaches are trained using synthetic CT-to-X-ray projections, but these methods still require manual annotations and employ a time-consuming optimization that limits their deployment for live image guidance. In this paper, we propose LXPose (live X-ray pose estimation), a fast multi-stage 3D/2D registration framework for real-time image guidance. Specifically, we introduce an efficient two-stage CNN that effectively bypasses slow optimization for fast inference. Critically, we also eliminate the need for any manual annotations by introducing an automated strategy for landmark extraction. Finally, we train LXPose using a projection loss for high accuracy, and apply extensive data augmentation in a first attempt to reduce the domain gap between the synthetic training X-rays and real testing data. We demonstrate LXPose on two datasets from different anatomical regions, where it yields results comparable to the state-of-the-art, while reducing inference time by two orders of magnitude, from several seconds to 20 ms. Overall, our results demonstrate the potential of LXPose for real-time clinical deployment. Our code is available at https://github.com/fedefacente/LXPose.
Lumbar plexus block (LPB) is a regional anesthesia technique widely used for hip and knee surgeries. However, despite the assistance of ultrasound guidance, the complex anatomical structure of the lumbar plexus poses significant challenges for anesthesiologists during the procedure. To accurately identify the lumbar plexus located in the posterior third of the psoas major in the Shamrock view, a deep learning-based segmentation model named DMRNet was proposed. This model is designed to precisely delineate muscles, nerves, and bony structures in Shamrock view ultrasound images. DMRNet integrates several innovative modules, including Adaptive Multi-Scale Dilated (AMD) Module that enhances the model's ability to capture multi-scale features; Dense Attention Residual (DAR) Module that adaptively selects salient feature regions; Attention-Enhanced Hybrid (AEH) Module that emphasizes critical features while suppressing irrelevant ones; and two attention mechanisms, Boundary-Aware Spatial Attention Mechanism (BASA) and the Enhanced Residual Multi-Head Attention Mechanism (ER-MHA), that improve the model's capacity to recognize complex contextual patterns. Experimental results demonstrated that DMRNet achieved a mean Intersection over Union of 0.863 and a mean Dice coefficient of 0.926 across all target structures, outperforming other state-of-the-art models. These findings suggest that DMRNet may serve as a useful assistive tool for sonoanatomical recognition in Shamrock view ultrasound images and may provide educational support for ultrasound-guided LPB training.
