Shun-Ping Wang

Background: Calcaneal fractures are challenging to diagnose on routine radiographs and may result in missed diagnoses. Objective: This study compares Convolutional Neural Networks (CNNs) and Vision Transformer (ViT) architectures for detecting calcaneal fractures separately in three standard radiographic views: Anteroposterior (AP), Harris (axial), and Lateral. Methods: A balanced retrospective dataset of 178 patients with complete three-view examinations (89 with displaced calcaneal fractures and 89 controls) was used to evaluate five deep learning models: ResNet50, InceptionV3, InceptionResNetV2, XCiT, and PiT. After image-quality screening, each view underwent an approximately 8:2 image-level training–validation split; no independent test set was used. Images were preprocessed using data augmentation and Contrast Limited Adaptive Histogram Equalization (CLAHE), while Gradient-weighted Class Activation Mapping (Grad-CAM) was employed to visualize model activation. Results: InceptionV3 achieved 100% sensitivity in the AP view, while ResNet50 achieved 100% specificity in both AP and Lateral views. XCiT matched ResNet50 across all five classification metrics in the AP view and matched ResNet50 and InceptionV3 in the Lateral view. PiT achieved 100% accuracy in the Harris validation set (95% CI, 89.0–100.0%). Grad-CAM visualizations showed activation over the calcaneal region in representative fracture images. Conclusions: This study demonstrates the feasibility of deep learning for identifying displaced calcaneal fractures through separate evaluations of AP, Harris, and Lateral radiographs. The model-specific performance patterns provide a foundation for further development of AI-assisted calcaneal fracture detection and future investigation of multi-view integration.

Background: Tibial plateau fractures are complex knee injuries where timely and accurate diagnosis is critical to preventing long-term disability. In high-pressure emergency settings, the risk of missed fractures (false negatives) remains a significant challenge. Objective: This study aims to develop a robust, clinically safe automated detection model using advanced deep learning architectures. Methods: We utilized a dataset of 1489 real-world clinical X-ray images, annotated by orthopedic surgeons, to train and evaluate five versions of the You Only Look Once (YOLO) algorithm (v8, v9, v10, v11, and v12). A rigorous two-stage evaluation process was implemented. First, an initial screening excluded YOLOv8 and YOLOv10 due to critical detection failures (“background errors”), in which the models failed to detect any object in the target region. Second, a comprehensive performance analysis identified YOLOv11 as the optimal architecture. Based on these results, the YOLOv11 model was integrated into a user-friendly, web-based diagnostic system using the Python Flask framework. Results: The YOLOv11 model achieved the highest Mean Average Precision (mAP) of 99.3% and an Accuracy of 98.32%. Crucially for clinical safety, YOLOv11 demonstrated superior sensitivity (96.58%) with the lowest false negative rate, attributed to its enhanced feature aggregation capabilities which effectively distinguish subtle fracture lines from trabecular bone patterns. Independent web interface validation (n = 109 real-world cases) confirmed 98.17% accuracy, 98.00% sensitivity, 98.31% specificity, 98.00% Positive Predictive Value (PPV), and 98.31% Negative Predictive Value (NPV). Conclusions: This system is designed to support clinical workflows by providing real-time, highly accurate second opinions, thereby reducing diagnostic errors and alleviating radiologist workload.

Jones fractures are Zone 2 fractures of the fifth metatarsal. Biomechanical comparisons of fixation strategies for Jones fractures remain limited by the lack of standardized, head-to-head evaluations across major fixation methods. The purpose of this study was to perform a standardized biomechanical comparison of six fixation configurations representing the three primary surgical techniques for Jones fractures and to examine the mechanical factors underlying differences in early construct stability. A synthetic fifth metatarsal model with a simulated Zone 2 fracture was stabilized using lateral plate fixation with different screw configurations, Kirschner wire fixation with or without tension-band wiring, or intramedullary headless screw fixation. All constructs were tested under displacement-controlled cantilever bending, and the force required to reach 1 mm of fracture site displacement was obtained and construct stiffness was calculated. Plate-based fixation demonstrated the highest resistance to bending deformation, followed by intramedullary screw fixation, whereas Kirschner wire-based constructs exhibited the lowest stability. These differences were explained by variations in load-sharing pathways and effective working length among fixation constructs. The addition of tension-band wiring did not result in a measurable improvement in stability compared with Kirschner wire fixation alone, consistent with the dependence of tension-band mechanisms on active muscle loading not represented in the experimental model. These findings provide a unified biomechanical comparison of commonly used fixation constructs for Jones fractures and clarify the mechanical basis for differences in early construct stability.

Background/Objectives: Tibial plateau fractures are difficult to detect using X-ray imaging due to limited three-dimensional visibility. This study evaluated the performance of four You Only Look Once (YOLO) deep learning models trained on different radiographic views for fracture detection. Methods: A total of 1489 knee X-rays were collected from a tertiary referral hospital, comprising 727 fracture images and 762 non-fracture images. YOLOv4, YOLOv5, YOLOv8, and YOLOv9 were each trained using anteroposterior (AP), lateral, and combined views. Results: YOLO models trained on AP views consistently outperformed those trained on other views. YOLOv9 trained on AP images achieved the highest accuracy, specificity, precision, F1-score, and area under the curve (AUC) of 0.99, with both sensitivity and negative predictive value (NPV) at 1.00. YOLOv8 trained on AP views reached 0.97 across all metrics with an AUC of 0.98. YOLOv5 trained on AP images achieved an accuracy and F1-score of 0.98, a sensitivity and NPV of 0.97, and an AUC of 1.00. YOLOv4 trained on AP views showed slightly lower performance, with an accuracy and F1-score of 0.96 and an AUC of 1.00. External validation confirmed the strong generalizability of AP-trained models, particularly YOLOv9, which reached an accuracy of 0.87, a sensitivity of 1.00, a specificity of 0.75, a precision of 0.80, an NPV of 1.00, an F1-score of 0.88, and an AUC of 0.93. Artificial intelligence-assisted YOLO models showed strong potential in detecting tibial plateau fractures. Conclusions: Models trained on AP views consistently achieved better diagnostic accuracy than those using other views. Among all, YOLOv9 delivered the best results, highlighting the benefits of newer deep learning architectures.

Background/Objectives: Mueller–Weiss syndrome (MWS) is a rare condition characterized by spontaneous adult-onset osteonecrosis of the navicular bone. This study aimed to assess the diagnostic value of quantitative and semi-quantitative standardized uptake value (SUV) measurements on Tc-99m MDP SPECT/CT for differentiating MWS from other foot pathologies. Methods: We retrospectively reviewed 21 MWS patients who underwent SPECT/CT and compared them with 10 feet from 5 non-MWS patients as controls. MWS severity was staged using the Maceira classification. Volumes of interest (VOIs) were defined in the lateral navicular and distal tibia. SUV_max values were measured for the navicular bone (N), tibial metaphysis (Tm), and diaphysis (Td). Uptake ratios (N/Tm and N/Td) were calculated for semi-quantitative comparison. Results: MWS patients showed significantly higher SUV_max in the navicular compared with controls (9.2 vs. 1.5, p < 0.001). Both N/Tm and N/Td ratios were also significantly elevated (p < 0.001). SUV_max and uptake ratios positively correlated with Maceira stage and visual navicular uptake intensity. Diagnostic thresholds of N SUV_max > 3.77 (AUC = 0.93), N/Tm > 1.139 (AUC = 0.95), and N/Td > 0.93 (AUC = 0.93) effectively distinguished MWS from non-MWS cases. Conclusions: Quantitative and semi-quantitative SUV analysis on SPECT/CT offers a reliable tool for diagnosing MWS and evaluating disease severity. Semi-quantitative ratios, by normalizing metabolic variability, provide a practical and reproducible alternative to absolute SUV measurements for early detection and treatment planning in MWS.

Robotic-assisted unicompartmental knee arthroplasty (UKA) has gained popularity for its potential to improve implant precision and reduce surgical errors. However, comparative evidence on short-term outcomes versus conventional UKA is lacking. Thus, the purpose of this study was to compare the short-term outcomes of robotic-assisted versus conventional UKA using a nationally representative database. The Nationwide Readmissions Database 2016-2020 was retrospectively examined to identify adult patients who received an elective UKA. After applying exclusion criteria and 1:2 propensity score matching (PSM), 8310 patients were included in the analysis. Outcomes included in-hospital complications, implant malposition or failure, perioperative fracture, length of hospital stay (LOS), hospital costs, and 30- and 90-day readmission rates. Multivariable regression analyses were performed to adjust for residual confounding factors. Robotic-assisted UKA was associated with significantly lower complication rates compared with conventional UKA (3.7% versus 13.2%, p < 0.001). Specifically, robotic-assisted procedures had reduced risks of implant malposition or failure (odds ratio [OR] = 0.08; 95% confidence interval [CI]: 0.05-0.13; p < 0.001) and perioperative fracture (OR = 0.18; 95% CI 0.04-0.76; p = 0.020). No significant differences were observed in LOS, total hospital costs, or readmission rates at 30 and 90 days. Robotic-assisted UKA is associated with improved short-term surgical safety, including fewer complications, particularly, reduced implant malposition and perioperative fractures. However, broader hospital metrics such as LOS, cost, and readmissions were comparable between the two approaches. Further prospective studies are needed to validate these findings and assess long-term outcomes and cost-effectiveness. Level III. Not applicable.

Various surgical techniques for reconstruction of chronic Achilles tendinosis have been developed, including free semitendinosus tendon autograft and allograft for larger Achilles tendon defects. Flexor hallucis longus (FHL) transfer has emerged as an effective treatment option for patients with insertional Achilles tendinopathy (IAT). During FHL transfer, FHL tendon is harvested at its distal part and rerouted to augment or replace the diseased portion of the Achilles tendon, providing improved structural integrity and function, but the necessity of reattachment of the Achilles stump in FHL transfer remains controversial. Patients who underwent FHL transfer surgery for IAT between 2013 and 2022 were enrolled in this retrospective study. Evaluating parameters included clinical evaluations, complications, a satisfaction questionnaire, preoperative and postoperative visual analog scale (VAS) as well as outcomes scores including American Orthopaedic Foot and Ankle Society (AOFAS), Achilles Tendon Total Rupture Score (ATRS), Foot Function Index (FFI). Furthermore, active range of motion (AROM) and torque of strength in both plantarflexion(PF) and dorsiflexion(DF) for both healthy and operated ankles of all patients were measured. The follow-up time was more than 12 months for all patients. A total of twenty-nine patients, with a median age of 63.0 years old and an average follow-up time of 28.1 months were enrolled. Thirteen patients were assigned to the Non-attached group and sixteen patients were assigned to the Reattached group. The functional outcomes such as AOFAS, ATRS, FFI, and VAS scores, as well as ankle strength in both groups were significantly improved after surgeries. However, no significant differences between groups in preoperative and postoperative VAS scores and functional outcomes were found. AROM and ankle torque were comparable, except for less plantarflexion torque loss at 30°/sec in the reattached group (p = 0.046). All evaluating parameters significantly improved without notable complications in patients after FHL transfer surgery whether with or without reattachment. There were no significant differences in the improvement of most parameters among subgroups. While reattachment of the Achilles tendon stump to the FHL transfer for IAT may not be crucial for superior outcomes, the torque improvement seen in the reattached group indicates possible advantages that require additional investigation.

The relationship between insertional Achilles tendinopathy (IAT) and Haglund’s deformity has not been fully elucidated. This study aimed to determine the optimal cutoff value of MRI-based Achilles thickness and intratendinous degeneration for IAT and further explored the correlations of IAT and Haglund’s deformity. Thirty heels with symptomatic IAT (IAT group) underwent surgeries and 38 healthy participants without heel pain (control group) were enrolled. Five different radiographic parameters for diagnosing Haglund’s deformity were measured. The IAT changes including tendon thickness and intratendinous degenerative ratio (ITD ratio) were obtained on MRI. The correlations of IAT and different radiographic parameters of Haglund’s deformity were evaluated. A total of 68 feet were enrolled. The optimal thresholds of Achilles tendon thickness and ITD ratio were 6.1 mm and 48.8%, respectively. Both tendon thickness and ITD ratio were positively correlated with bump height and bump-calcaneus ratio (p < 0.001), but Fowler–Phillip angle, Heneghan–Pavlov parallel pitch lines and X-Y ratio were not. Achilles tendon thickness of more than 6.1 mm and intratendinous abnormality of over 48.8% were confirmed to be diagnostic cutoff values for IAT. Haglund’s deformity was determined as a risk factor of IAT requiring surgery. Both radiographic parameters of Haglund’s deformity using the bump height and the bump-calcaneus ratio were useful in predicting the presence of IAT on MRI. Level III, Retrospective cohort study.

Postoperative early mobilization after total knee replacement (TKR) is essential for preventing complications and offers numerous benefits. The purpose of this study was to evaluate the factors determining early mobilization (EM) after primary TKR and the effects on risk of falls and length of stay (LOS). This retrospective study recruited elder participants undergoing primary TKR. All patients were classified as EM(+), early mobilization (EM) within 24 h, or EM(-), delayed mobilization over 24 h. Demographic data, pain Visual Analog Scale (VAS), American Society of Anesthesiologists (ASA) score, perioperative blood pressure, postoperative pain control mode, medical catheters, dizziness or nausea/vomiting (PONV), falls during admission, and the LOS were collected. A total of 1759 elder participants undergoing primary TKR were enrolled. Mean age was 73.00 ± 8.11 years. Among them, there were 1260 EM(-) and 499 EM(+) cases, with no significant differences in sex, BMI, PONV, postoperative Patient Controlled Analgesia (PCA) mode, or the use of medical catheters, but age, higher ASA score, VAS, muscle strength, postoperative nausea, and substantial changes of blood pressure were significantly different between the two groups. A greater reduction in systolic blood pressure, when compared to the preoperative baseline blood pressure, was found in EM(-) patients than EM(+) patients on the operative day (13 vs. 4 mmHg, P < 0.001) and first day post-TKR (20 vs. 17 mmHg, P = 0.005). LOS in EM(-) patients was significant longer than that in EM(+) patients, but risk of falls was not significantly different. Multiple factors influence patients’ willingness to mobilize or ambulate earlier. However, more blood pressure reduction impeding early mobilization after TKR should be addressed. Maintaining post-TKR blood pressure without significant decreases is recommended for the successful mobilization after surgery. Early mobilization within 24 h after TKR may be beneficial in reducing the LOS and did not increase risk of falls.

Introduction: A spiral clavicle plate has been accepted for its superior multidirectional compatibility in the treatment of midshaft clavicle fractures from a biomechanical perspective. However, the influence of the sextant angle (spiral level) definition on biomechanical performance has not been clarified. A conceptual finite element analysis was conducted to identify the advantages and drawbacks of spiral clavicle plates with various sextant angle definitions. Methods: Conventional superior and three different conceptual spiral plates with sextant angle definitions ranging from 45 to 135 degrees were constructed to restore an OTA 15-B1.3 midshaft clavicle fracture model. Three major loading scenarios (cantilever downward bending, axial compression, and axial torsion) were simulated to evaluate the reconstructed structural stiffness and the stress on the clavicle plate and bone screws. Results: The spiral clavicle plate demonstrated greater capability in resisting cantilever downward bending with an increase in sextant angle and showed comparable structural stiffness and implant stress compared to the superior clavicle plate. However, weakened resistance to axial compression load was noted for the spiral clavicle plate, with lowered stiffness and increased stress on the clavicle plate and screws as the spiral level increased. Conclusion: The spiral clavicle plate has been reported to offer multidirectional compatibility for the treatment of midshaft clavicle fractures, as well as geometric advantages in anatomical matching and reduced skin prominence after surgery. The current study supports that remarkable cantilever bending strength can be achieved with this plate. However, users must consider the potential drawback of lowered axial compression resistance in safety considerations.