Search Results (2,265)

The growing use of remote-sensing technologies has placed greater demands on object-detection algorithms, which still face challenges. This study proposes a hierarchical adaptive feature aggregation network (HAF-YOLO) to improve detection precision in remote-sensing images. It addresses issues such as small object size, complex backgrounds, scale variation, and dense object distributions by incorporating three core modules: dynamic-cooperative multimodal fusion architecture (DyCoMF-Arch), multiscale wavelet-enhanced aggregation network (MWA-Net), and spatial-deformable dynamic enhancement module (SDDE-Module). DyCoMF-Arch builds a hierarchical feature pyramid using multistage spatial compression and expansion, with dynamic weight allocation to extract salient features. MWA-Net applies wavelet-transform-based convolution to decompose features, preserving high-frequency detail and enhancing representation of small-scale objects. SDDE-Module integrates spatial coordinate encoding and multidirectional convolution to reduce localization interference and overcome fixed sampling limitations for geometric deformations. Experiments on the NWPU VHR-10 and DIOR datasets show that HAF-YOLO achieved mAP50 scores of 85.0% and 78.1%, improving on YOLOv8 by 4.8% and 3.1%, respectively. HAF-YOLO also maintained a low computational cost of 11.8 GFLOPs, outperforming other YOLO models. Ablation studies validated the effectiveness of each module and their combined optimization. This study presents a novel approach for remote-sensing object detection, with theoretical and practical value. Full article

To tackle the challenges of detecting complex cracks on large stone slabs with noisy textures, this paper presents the first domain-optimized framework for stone slab cracks, an improved semantic segmentation model (YOLOv8-Seg) synergistically integrating U-NetV2, DSConv, and DySample. The network uses the lightweight U-NetV2 backbone combined with dynamic feature recalibration and multi-scale refinement to better capture fine crack details. The dynamic up-sampling module (DySample) helps to adaptively reconstruct curved boundaries. In addition, the dynamic snake convolution head (DSConv) improves the model’s ability to follow irregular crack shapes. Experiments on the custom-built ST stone crack dataset show that YOLOv8-Seg achieves an mAP@0.5 of 0.856 and an mAP@0.5–0.95 of 0.479. The model also reaches a mean intersection over union (MIoU) of 79.17%, outperforming both baseline and mainstream segmentation models. Ablation studies confirm the value of each module. Comparative tests and industrial validation demonstrate stable performance across different stone materials and textures and a 30% false-positive reduction in real production environments. Overall, YOLOv8-Seg greatly improves segmentation accuracy and robustness in industrial crack detection on natural stone slabs, offering a strong solution for intelligent visual inspection in real-world applications. Full article

This work investigated the ion exchange technique for selective separation of rare earth elements (REE) from acid mine drainage (AMD), using different column systems, pH values, and eluent concentrations. Systematic analysis of pH and eluent concentration showed that an initial pH of 6.0 and 0.02 mol L⁻¹ NH₄EDTA are the optimal conditions, achieving 98.4% heavy REE purity in the initial stage (0 to 10 bed volumes). This represents a 32-fold increase compared to the original AMD (6.7% heavy REE). The speciation of REE and impurities was determined by Visual Minteq 4.0 software using pH 2.0, which corresponds to the pH at the inlet of the fractionation column. Under this condition, La and Nd and the impurities (Ca, Mg, and Mn) remained in the fractionation column, while Al was partially retained. In addition, the heavy REE (Y and Dy) were mainly in the form of REE-EDTA complexes and not as free cations, which made fractionation more feasible. The fractionation column minimized impurities, retaining 100% of Ca and 67% of Al, generating a liquor concentrated in heavy REE. This sustainable approach adopted herein meets the critical needs for scalable recovery of REE from diluted effluents, representing a circular economy strategy for critical metals. Full article

Accurate detection of maize tassels plays a crucial role in yield estimation of maize in precision agriculture. Recently, UAV and deep learning technologies have been widely introduced in various applications of field monitoring. However, complex field backgrounds pose multiple challenges against the precision detection of maize tassels, including maize tassel multi-scale variations caused by varietal differences and growth stage variations, intra-class occlusion, and background interference. To achieve accurate maize tassel detection in UAV images under complex field backgrounds, this study proposes an MSMT-RTDETR detection model. The Faster-RPE Block is first designed to enhance multi-scale feature extraction while reducing model Params and FLOPs. To improve detection performance for multi-scale targets in complex field backgrounds, a Dynamic Cross-Scale Feature Fusion Module (Dy-CCFM) is constructed by upgrading the CCFM through dynamic sampling strategies and multi-branch architecture. Furthermore, the MPCC3 module is built via re-parameterization methods, and further strengthens cross-channel information extraction capability and model stability to deal with intra-class occlusion. Experimental results on the MTDC-UAV dataset demonstrate that the MSMT-RTDETR significantly outperforms the baseline in detecting maize tassels under complex field backgrounds, where a precision of 84.2% was achieved. Compared with Deformable DETR and YOLOv10m, improvements of 2.8% and 2.0% were achieved, respectively, in the mAP₅₀ for UAV images. This study proposes an innovative solution for accurate maize tassel detection, establishing a reliable technical foundation for maize yield estimation. Full article

This study investigates the parameters influencing the compatibility between cement and polycarboxylate ether (PCE) admixtures in cements produced with various types and dosages of grinding aids (GAs). A total of 29 cement types (including a control) were prepared using seven different GAs at four dosage levels, and 87 paste mixtures were produced with three PCE dosages. Rheological behavior was evaluated via the Herschel–Bulkley model, focusing on dynamic yield stress (DYS) and viscosity. The data were modeled using CNN, Random Forest (RF), and Neural Classification and Regression Tree (NCART), and each model was enhanced with synthetic data generated by Large Language Models (LLMs), resulting in CNN-LLM, RF-LLM, and NCART-LLM variants. All six variants were evaluated using R-squared, Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Logcosh. This study is among the first to use LLMs for synthetic data augmentation. It augmented the experimental dataset synthetically and analyzed the effects on the study results. Among the baseline methods, NCART achieved the best performance for both viscosity (MAE = 1.04, RMSE = 1.33, R² = 0.84, Logcosh = 0.57) and DYS (MAE = 8.73, RMSE = 11.50, R² = 0.77, Logcosh = 8.09). Among baseline models, NCART performed best, while LLM augmentation significantly improved all models’ predictive accuracy. It was also observed that cements produced with GA exhibited higher DYS and viscosity than the control, likely due to finer particle size distribution. Overall, the study highlights the potential of LLM-based synthetic augmentation in modeling cement admixture compatibility. Full article

High-quality end-of-life care (EoLC) is a critical yet often underemphasised component of oncology care. Several shortcomings in the delivery of EoLC for oncology patients in our centre during the COVID-19 pandemic were identified in our initial 2021 audit. In 2022, we introduced a care of dying patients proforma, an EoLC quality checklist, targeted education and training for staff, and an expanded end-of-life (EoL) committee. This re-audit aimed to review how these changes impacted on the care received by patients in a tertiary cancer centre. A second retrospective re-audit of patients who died between 11 July 2022 and 30 April 2023 was performed to assess quality of EoLC using the Oxford Quality indicators. A total of 72 deaths occurred over the audit period. Quality of EoLC improved significantly when compared to the initial audit (χ² (3, n = 138) = 9.75, p = 0.021). Exploration of patients’ wishes was documented in 48.8% and referral to pastoral care was documented in 68.3%, from 24.2% and 10.6%, respectively. The proportion of patients receiving poor EoLC reduced from 21.2% to 8.3%. Our study demonstrates the benefits of simple interventions, the importance of re-audit, and the role of ongoing interdisciplinary commitment to improving EoLC for our patients. Full article

The vast majority of photocatalysts find it difficult to consistently and stably exhibit high performance due to the variability of sunlight intensity within a day, as well as the high energy consumption of artificial light sources. In this study, mechanoluminescent Sr₂MgSi₂O₇:Eu,Dy phosphors is combined with NiS@g-C₃N₄ composite to construct a ternary heterogeneous photocatalytic system, denoted as NCS. In addition to the enhanced separation efficiency of photogenerated charge carriers by the formation of a heterojunction, the introduction of Sr₂MgSi₂O₇:Eu,Dy provides an ultra-driving force for the photocatalytic reactions owing to its mechanoluminescence-induced excitation. Results show that the degradation rate of RhB increased significantly in comparison with pristine g-C₃N₄ and NiS@g-C₃N₄, indicating the obvious advantages of the ternary system for charge separation and migration. Moreover, the additional photocatalytic activity of NCS under ultrasound stimulation makes it a promising all-weather photocatalyst even in dark environments. This novel strategy opens up new horizons for the synergistic combination of light-driven and ultrasound-driven heterogeneous photocatalytic systems, and it also has important reference significance for the design and application of high-performance photocatalysts. Full article

The exponential increase in electronic waste (e-waste) from end-of-life electrical and electronic equipment presents a growing environmental challenge. E-waste contains high concentrations of rare earth elements (REEs), which are classified as critical raw materials (CRMs). Their removal and recovery from contaminated systems not only mitigate pollution but also support resource sustainability within a circular economy framework. The present study proposed the use of hazelnut shells as a biosorbent to reduce water contamination and recover REEs. The sorption capabilities of this lignocellulosic material were assessed and optimized using the response surface methodology (RSM) combined with a Box–Behnken Design (three factors, three levels). Factors such as pH (4 to 8), salinity (0 to 30), and biosorbent dose (0.25 to 0.75 g/L) were evaluated in a complex mixture containing 9 REEs (Y, La, Ce, Pr, Nd, Eu, Gd, Tb and Dy; equimolar concentration of 1 µmol/L). Salinity was found to be the factor with greater significance for REEs sorption efficiency, followed by water pH and biosorbent dose. At a pH of 7, salinity of 0, biosorbent dose of 0.75 g/L, and a contact time of 48 h, optimal conditions were observed, achieving removals of 100% for Gd and Eu and between 81 and 99% for other REEs. Optimized conditions were also predicted to maximize the REEs concentration in the biosorbent, which allowed us to obtain values (total REEs content of 2.69 mg/g) higher than those in some ores. These results underscore the high potential of this agricultural waste with no relevant commercial value to improve water quality while providing an alternative source of elements of interest for reuse (circular economy). Full article

Weed infestation contributes significantly to global agricultural yield loss and increases the reliance on herbicides, raising both economic and environmental concerns. Effective weed detection in agriculture requires high accuracy and architectural efficiency. This is particularly important under challenging field conditions, including densely clustered targets, small weed instances, and low visual contrast between vegetation and soil. In this study, we propose GTDR-YOLOv12, an improved object detection framework based on YOLOv12, tailored for real-time weed identification in complex agricultural environments. The model is evaluated on the publicly available Weeds Detection dataset, which contains a wide range of weed species and challenging visual scenarios. To achieve better accuracy and efficiency, GTDR-YOLOv12 introduces several targeted structural enhancements. The backbone incorporates GDR-Conv, which integrates Ghost convolution and Dynamic ReLU (DyReLU) to improve early-stage feature representation while reducing redundancy. The GTDR-C3 module combines GDR-Conv with Task-Dependent Attention Mechanisms (TDAMs), allowing the network to adaptively refine spatial features critical for accurate weed identification and localization. In addition, the Lookahead optimizer is employed during training to improve convergence efficiency and reduce computational overhead, thereby contributing to the model’s lightweight design. GTDR-YOLOv12 outperforms several representative detectors, including YOLOv7, YOLOv9, YOLOv10, YOLOv11, YOLOv12, ATSS, RTMDet and Double-Head. Compared with YOLOv12, GTDR-YOLOv12 achieves notable improvements across multiple evaluation metrics. Precision increases from 85.0% to 88.0%, recall from 79.7% to 83.9%, and F1-score from 82.3% to 85.9%. In terms of detection accuracy, mAP:0.5 improves from 87.0% to 90.0%, while mAP:0.5:0.95 rises from 58.0% to 63.8%. Furthermore, the model reduces computational complexity. GFLOPs drop from 5.8 to 4.8, and the number of parameters is reduced from 2.51 M to 2.23 M. These reductions reflect a more efficient network design that not only lowers model complexity but also enhances detection performance. With a throughput of 58 FPS on the NVIDIA Jetson AGX Xavier, GTDR-YOLOv12 proves both resource-efficient and deployable for practical, real-time weeding tasks in agricultural settings. Full article

The identification and detection of planting holes, combined with UAV technology, provides an effective solution to the challenges posed by manual counting, high labor costs, and low efficiency in large-scale planting operations. However, existing target detection algorithms face difficulties in identifying planting holes based on their edge features, particularly in complex environments. To address this issue, a target detection network named YOLO-PH was designed to efficiently and rapidly detect planting holes in complex environments. Compared to the YOLOv8 network, the proposed YOLO-PH network incorporates the C2f_DyGhostConv module as a replacement for the original C2f module in both the backbone network and neck network. Furthermore, the ATSS label allocation method is employed to optimize sample allocation and enhance detection effectiveness. Lastly, our proposed Siblings Detection Head reduces computational burden while significantly improving detection performance. Ablation experiments demonstrate that compared to baseline models, YOLO-PH exhibits notable improvements of 1.3% in mAP50 and 1.1% in mAP50:95 while simultaneously achieving a reduction of 48.8% in FLOPs and an impressive increase of 26.8 FPS (frames per second) in detection speed. In practical applications for detecting indistinct boundary planting holes within complex scenarios, our algorithm consistently outperforms other detection networks with exceptional precision (F1-score = 0.95), low computational cost, rapid detection speed, and robustness, thus laying a solid foundation for advancing precision agriculture. Full article

Recent advances in precision manufacturing and high-end equipment technologies have imposed ever more stringent requirements on the accuracy, real-time performance, and lightweight design of online steel strip surface defect detection systems. To reconcile the persistent trade-off between detection precision and inference efficiency in complex industrial environments, this study proposes StripSurface–YOLO, a novel real-time defect detection framework built upon YOLOv8n. The core architecture integrates an Efficient Cross-Stage Local Perception module (ResGSCSP), which synergistically combines GSConv lightweight convolutions with a one-shot aggregation strategy, thereby markedly reducing both model parameters and computational complexity. To further enhance multi-scale feature representation, this study introduces an Efficient Multi-Scale Attention (EMA) mechanism at the feature-fusion stage, enabling the network to more effectively attend to critical defect regions. Moreover, conventional nearest-neighbor upsampling is replaced by DySample, which produces deeper, high-resolution feature maps enriched with semantic content, improving both inference speed and fusion quality. To heighten sensitivity to small-scale and low-contrast defects, the model adopts Focal Loss, dynamically adjusting to sample difficulty. Extensive evaluations on the NEU-DET dataset demonstrate that StripSurface–YOLO reduces FLOPs by 11.6% and parameter count by 7.4% relative to the baseline YOLOv8n, while achieving respective improvements of 1.4%, 3.1%, 4.1%, and 3.0% in precision, recall, mAP₅₀, and mAP_50:95. Under adverse conditions—including contrast variations, brightness fluctuations, and Gaussian noise—SteelSurface-YOLO outperforms the baseline model, delivering improvements of 5.0% in mAP₅₀ and 4.7% in mAP_50:95, attesting to the model’s robust interference resistance. These findings underscore the potential of StripSurface–YOLO to meet the rigorous performance demands of real-time surface defect detection in the metal forging industry. Full article

Background/Objectives: Clinical stability within 24 h prior to discharge is a key metric for safe care transitions in hospitalised patients with community-acquired pneumonia (CAP). However, its association with post-discharge outcomes, particularly readmissions, remains underexplored. This study assessed whether clinical instability before discharge is associated with 30-day mortality, readmissions, or a composite of both in hospitalised CAP patients. Methods: This retrospective cohort study included adults (≥18 years) admitted with CAP to two tertiary Australian hospitals between 1 January 2020 and 31 December 2023. Clinical instability was defined as abnormal vital signs (temperature, heart rate, respiratory rate, blood pressure, or oxygen saturation) within 24 h before discharge. Pneumonia severity was assessed using the CURB-65 score and frailty using the Hospital Frailty Risk Score. Multilevel logistic regression models were used to evaluate associations with outcomes, adjusting for age, sex, comorbidities, frailty, disease severity, microbiological aetiology, antibiotics prescribed during admission, and prior healthcare use. Competing risk regression accounted for death when analysing readmissions. Results: Of 3984 patients, 20.4% had clinical instability within 24 h before discharge. The composite outcome occurred in 21.9% patients, with 15.8% readmitted and 6.1% dying within 30 days. Clinical instability was significantly associated with the composite outcome (adjusted odds ratio [aOR] 1.73, 95% CI 1.42–2.09, p < 0.001), primarily driven by increased mortality risk (aOR 3.70, 95% CI 2.73–5.00, p < 0.001). However, no significant association was found between clinical instability and readmissions (aOR 1.16, 95% CI 0.93–1.44, p > 0.05). Conclusions: Clinical instability within 24 h before discharge predicts worse outcomes in CAP patients, driven by increased mortality risk rather than readmissions. Full article

As a type of distinctive pit on Mars, skylights are entrances to subsurface lava caves. They are very important for studying volcanic activity and potential preserved water ice, and are also considered as potential sites for human extraterrestrial bases in the future. Most skylights are manually identified, which has low efficiency and is highly subjective. Although deep learning methods have recently been used to identify skylights, they face challenges of few effective samples and low identification accuracy. In this article, 151 positive samples and 920 negative samples based on the MRO-HiRISE image data was used to create an initial skylight dataset, which contained few positive samples. To augment the initial dataset, StyleGAN2-ADA was selected to synthesize some positive samples and generated an augmented dataset with 896 samples. On the basis of the augmented skylight dataset, we proposed YOLOv9-Skylight for skylight identification by incorporating Inner-EIoU loss and DySample to enhance localization accuracy and feature extracting ability. Compared with YOLOv9, the P, R, and the F1 of YOLOv9-Skylight were improved by about 9.1%, 2.8%, and 5.6%, respectively. Compared with other mainstream models such as YOLOv5, YOLOv10, Faster R-CNN, Mask R-CNN, and DETR, YOLOv9-Skylight achieved the highest accuracy (F1 = 92.5%), which shows a strong performance in skylight identification. Full article

Papaya ripeness identification is a key task in orchard management. To achieve efficient deployment of this task on edge computing devices, this paper proposes a lightweight detection model, ABD-YOLO-ting, based on YOLOv8. First, the width factor of YOLOv8n is adjusted to construct a lightweight backbone network, YOLO-Ting. Second, a low-computation ADown module is introduced to replace the standard downsampling structure, aiming to enhance feature extraction efficiency. Third, an enhanced BiFPN is integrated into the neck structure to achieve efficient multi-scale feature fusion. Finally, to strengthen the model’s capability in identifying small objects, the dynamic detection head DyHead is introduced to improve ripeness recognition accuracy. On a self-constructed Japanese quince orchard dataset, ABD-YOLO-ting achieves a mAP50 of 94.7% and a mAP50–95 of 77.4%, with only 1.47 M parameters and 5.4 G FLOPs, significantly outperforming mainstream models such as YOLOv5, YOLOv8, and YOLOv11. On edge devices, the model achieves a well-balanced trade-off between detection speed and accuracy, demonstrating strong potential for practical applications in intelligent harvesting and orchard management. Full article

According to the World Health Organization (WHO), a healthy lifestyle is defined as a way of living that lowers the risk of becoming seriously ill or dying prematurely. Physical activity, as a well-known contributor to overall health, plays a vital role in supporting such a lifestyle. Exercise induces complex molecular responses that mediate both acute metabolic stress and long-term physiological adaptations. FGF21 (fibroblast growth factor 21) and GDF15 (growth differentiation factor 15) are recognized as metabolic stress markers, while their receptors play critical roles in cellular signaling. However, the differential gene expression patterns of these molecules in trained and untrained individuals following exhaustive exercise remain poorly understood. This study aimed to examine the transcriptional and protein-level responses in trained and untrained individuals performed a treadmill maximal exercise test to voluntary exhaustion. Blood samples were collected at six time points (pre-exercise, immediately post-exercise, and 0.5 h, 6 h, 24 h, and 48 h post-exercise). Gene expression of FGF21, GDF15, FGFR1 (fibroblast growth factor receptors), FGFR3, FGFR4, KLB (β-klotho), and GFRAL (glial cell line-derived neurotrophic factor receptor alpha-like) was analyzed using RT-qPCR, while plasma protein levels of FGF21 and GDF15 were quantified via ELISA. The results obtained were statistically analyzed by using Shapiro–Wilk, Mann–Whitney U, and Wilcoxon tests in Statistica 13 software. Untrained individuals demonstrated significant post-exercise upregulation of FGFR3, FGFR4, KLB, and GFRAL. FGF21 and GDF15 protein levels were consistently lower in trained individuals (p < 0.01), with no significant correlations between gene and protein expression. Trained individuals showed more stable expression of genes, while untrained individuals exhibited transient upregulation of genes after exercise. Full article