Scientific Reports最新文献

Anterolateral lumbar interbody fusion (LIF) surgery has been shown to be a successful treatment option for patients with degenerative lumbar disease. However, the effect of paraspinal muscle (PM) degeneration on clinical outcomes and risk factors for nonunion and cage subsidence (CS) in LIF remains unclear. The study enrolled 127 patients who underwent ALIF or OLIF. Preoperative MRI was used to measure PM cross-sectional area (PMCSA) and the fatty infiltration (PMFI) at the superior and inferior levels of the fusion level. Fusion status and CS were assessed 2 years postoperatively. Smoking history, body mass index, and PMFI of the adjacent levels showed significant elevation in the nonunion and CS groups. PMFI at the inferior level emerged as a risk factor for nonunion and cage subsidence (p = 0.026, odds ratio [OR] = 1.12), particularly within the L1-L4 cohort. While multivariable analysis confirmed this association for L1-L4 levels, the correlation was less prominent in the L5-S1 subgroup. Nonunion of bone was observed when the PMFI rate surpassed 56.4% at the inferior level. Qualitative muscle assessment is a more reliable predictor of fusion outcomes than quantitative size measurements, especially for L1-L4 levels. Preoperative PMFI evaluation should be integrated into surgical planning to improve patient selection and risk management.

In clinical practice, medical inter-modality imaging results can assist doctors in making better decisions, as different modalities imaging results can provide complementary information. Traditionally, obtaining these imaging results requires using various medical devices to scan patients, which can be time-consuming, costly, and potentially harmful to the patient. Motivated by the need to address these limitations, we propose an alternative method that facilitates the conversion of volume CT into volume MRI. The method is based on a Diffusion model and incorporates a post-processing approach to enhance the model's output. To validate our approach, we conduct experiments and achieve good results on brain and pelvic datasets obtained from clinical practice, despite approximately 6% of the slices being incompletely paired. We also compare our method with state-of-the-art techniques, both qualitatively and quantitatively. Our experimental results show that our method outperforms state-of-the-art techniques, including MedSynthesisV1, CycleGAN, Pix2Pix and Diffusion, when using ground truth as a reference. Finally, we conduct an experiment to select the optimal hyperparameters, including the number of epochs and the parameters [Formula: see text] and [Formula: see text].

Coal bumps pose a significant threat to the safe operation of coal mines. This study examines the coal bumps incident at Tangshan Mine, focusing on the influences of various disaster-causing factors, including structural stress, rock mass impact tendencies, and support and bearing capacity. To facilitate the safe resumption of production at the 0250 working face, coal seam blasting and borehole pressure relief drilling were conducted to dissipate the accumulated elastic energy within the coal seam, thereby effectively mitigating the risk of bursting. Concurrently, the vibration and displacement of the roof were monitored. The relevant monitoring data indicate that roof activity remains within a controllable range, suggesting a relatively low likelihood of coal bump incidents.

This study investigates the effect of SiO₂ (silicon dioxide) powder as a reinforcement material in AA8011 aluminium matrix composites fabricated using the stir casting technique. The composites were produced with varying SiO₂ contents of 0, 3, 6, and 9 wt% to evaluate their mechanical and tribological performance. The incorporation of SiO₂ particles resulted in significant improvements in mechanical properties and wear resistance while maintaining a lightweight structure. The ultimate tensile strength increased from 156 MPa for the unreinforced alloy to 210 MPa for the composite containing 9 wt% SiO₂, indicating substantial strengthening due to effective load transfer and grain refinement. Similarly, microhardness showed a noticeable improvement with increasing reinforcement content. However, the impact strength decreased by 11.42%, 22.85%, and 34.28% for composites containing 3 wt%, 6 wt%, and 9 wt% SiO₂, respectively, compared with the base alloy. Wear tests conducted using a pin-on-disc apparatus demonstrated a considerable reduction in wear rate with increasing SiO₂ reinforcement, particularly under higher loading conditions, due to the load-bearing capability and hardness of SiO₂ particles. Microstructural analysis confirmed a relatively uniform distribution of reinforcement within the aluminium matrix, although minor particle agglomeration was observed at higher reinforcement levels. Overall, the results indicate that SiO₂-reinforced AA8011 composites fabricated through stir casting offer improved mechanical strength and enhanced wear resistance, making them suitable for lightweight structural and tribological applications in automotive, aerospace, and engineering sectors.

Human action recognition in close-contact sports is hindered by mutual occlusion, rapid pose changes, and distracting backgrounds. We study freestyle wrestling-a representative close-contact setting with sustained physical interaction-and present the Open FSW dataset of 210 trimmed clips covering seven techniques (30 clips per class), sourced from both controlled training sessions and broadcast footage. We introduce a foreground-aware RGB pipeline that segments athletes with a fine-tuned DeepLabV3+ model, extracts per-frame features using CNN backbones (VGG16, InceptionV3, EfficientNet-B7), and aggregates them with a bidirectional LSTM to produce clip-level predictions. Under a group-aware six-fold cross-validation protocol stratified by match/session ID to reduce train-test contamination across related sequences, the best configuration (DeepLabV3+ (foreground) + EfficientNet-B7 + Bi-LSTM) attains 82.9% top-1 accuracy. Ablation results quantify the added value of foregrounding, showing consistent gains for the strongest backbone and the largest improvements on high-occlusion techniques, at the cost of additional inference latency due to segmentation. Due to the modest dataset size, we mitigate overfitting via transfer learning and extensive augmentation, and we frame conclusions as domain-specific to freestyle wrestling. The dataset and code are released. To comply with copyright constraints, the controlled subset is provided as processed clips, while the broadcast subset is released as annotations and clip metadata to enable reconstruction.

The proliferation of technology-enhanced learning in higher education generates unprecedented student data, creating opportunities for learning analytics while raising critical privacy concerns. Current data sharing practices are severely constrained by privacy regulations and ethical considerations, impeding collaborative research and methodological advancement. This study addresses this fundamental tension by introducing SynEdu-HEDL, a comprehensive privacy-preserving synthetic dataset specifically designed for learning analytics in higher education.Developed through an innovative five-phase methodological framework integrating conditional tabular generative adversarial networks, temporal sequence generators, and differential privacy mechanisms, the dataset captures authentic educational patterns while ensuring robust privacy protection. SynEdu-HEDL comprises 20,000 synthetic student records encompassing 85 features across demographic characteristics, temporal learning interactions, engagement patterns, and academic performance metrics.A comprehensive three-dimensional validation framework evaluated SynEdu-HEDL across privacy protection, statistical fidelity, and analytical utility. Results demonstrate that the dataset provides strong privacy guarantees (membership inference AUC-ROC=0.512, statistically indistinguishable from random guessing), preserves essential statistical properties (average Wasserstein distance=0.043, correlation matrix similarity=94.1%), and supports diverse learning analytics tasks with models achieving performance within 1-5% of those trained on original data. Notably, transfer learning experiments show 24.4% performance improvement with only 10% real data, demonstrating practical value for resource-constrained settings.SynEdu-HEDL advances educational data science by providing a practical solution to data sharing barriers, supporting reproducible research, and establishing methodological standards for synthetic educational data validation. SynEdu-HEDL is openly available at https://github.com/drsanjayagal/SynEdu-HEDL and has been archived in Zenodo with the permanent DOI https://doi.org/10.5281/zenodo.18884938. The repository includes comprehensive documentation, fostering community engagement and accelerating progress in privacy-preserving learning analytics.

The ocean's biological carbon pump regulates climate by transferring a portion of surface-fixed CO₂ to the deep ocean through sinking particulate organic carbon (POC). Although this flux is strongly attenuated in the twilight zone, the seasonal controls on attenuation remain poorly understood. We examined seasonal variations in POC flux, the nitrogen isotope ratio of sinking particulate nitrogen (δ¹⁵N_sink), and mineral composition using sediment traps observations at subarctic (K2) and subtropical (S1) stations in the western North Pacific. POC sequestration efficiency at 500 m [Seq₍₅₀₀₎ = POC flux/NPP] was quantified, with net primary productivity (NPP) reconstructed from δ¹⁵N_sink using an empirical framework. Seq₍₅₀₀₎ remained nearly constant at K2 (7.4-8.1%) but varied substantially at S1 (3.4-6.5%). At K2, CaCO₃ and opal contents exhibited complementary seasonal patterns, whereas S1 showed pronounced variability primarily in CaCO₃. We propose that mineral composition modulates aggregate settling velocity and adhesive strength, thereby regulating POC attenuation through fragmentation processes. These findings indicate that seasonal shifts in surface ecosystem structure influence the physical properties of sinking aggregates and ultimately control the fate of NPP in the twilight zone.

Contemporary digital work is increasingly shaped by frequent multitasking, which places sustained demands on directed attention. Restorative design theories propose that nature-related cues can support attention recovery, yet it remains unclear whether and how such effects translate to immersive virtual environments. In response to this study, virtual environments with and without nature-related wooden elements were constructed, and 54 subjects were recruited to complete cognitive tasks, eye movement experiments, and questionnaires. The experimental results showed that: (1) the response time of the wood group in the Stroop test was shortened by 15.3%, and the correctness rate of the OSPAN test was increased by 11.3%; the wood elements in the virtual environment could optimize the cognitive processing process, significantly shorten the response time of the cognitive task, and increase the accuracy of working memory. (2) Eye-movement data showed that the wood elements could improve the stability of attention and thus enhance the ability of sustained concentration. The findings of this study provide a reference for the selection of materials based on the principle of psychological recovery for the future design of digital office space.