Search Results (169)

In autonomous-driving real-time semantic segmentation, simultaneously maximizing accuracy, minimizing model size, and sustaining high inference speed remains challenging. This tripartite demand poses significant constraints on the design of lightweight neural networks, as conventional frameworks often suffer from a trade-off between computational efficiency and feature representation capability, thereby limiting their practical deployment in resource-constrained autonomous driving systems. We introduce ISANet, an information supplementary aggregation framework that markedly elevates segmentation accuracy without sacrificing frame rate. ISANet integrates three key components: (i) the Spatial-Supplementary Lightweight Bottleneck Unit (SLBU) that splits channels and employs compensatory branches to extract highly expressive features with minimal parameters; (ii) the Missing Spatial Information Recovery Branch (MSIRB) that recovers spatial details lost during feature extraction; and (iii) the Object Boundary Feature Attention Module (OBFAM) that fuses multi-stage features and strengthens inter-layer information interaction. Evaluated on Cityscapes and CamVid, ISANet attains 76.7% and 73.8% mIoU, respectively, while delivering 58 FPS and 90 FPS with only 1.37 million parameters. Full article

Background: The accurate classification of brain tumor subtypes from MRI scans is critical for timely diagnosis, yet the manual annotation of large datasets remains prohibitively labor-intensive. Method: We present SSPLNet (Semi-Supervised Pseudo-Labeling Network), a dual-branch deep learning framework that synergizes confidence-guided iterative pseudo-labelling with deep feature fusion to enable robust MRI-based tumor classification in data-constrained clinical environments. SSPLNet integrates a custom convolutional neural network (CNN) and a pretrained ResNet50 model, trained semi-supervised using adaptive confidence thresholds ($\tau = 0.98 \to 0.95 \to 0.90$) to iteratively refine pseudo-labels for unlabelled MRI scans. Feature representations from both branches are fused via a dense network, combining localized texture patterns with hierarchical deep features. Results: SSPLNet achieves state-of-the-art accuracy across labelled–unlabelled data splits ($90:10$ to $10:90$), outperforming supervised baselines in extreme low-label regimes (10:90) by up to 5.34% from Custom CNN and 5.58% from ResNet50. The framework reduces annotation dependence and with 40% unlabeled data maintains 98.17% diagnostic accuracy, demonstrating its viability for scalable deployment in resource-limited healthcare settings. Conclusions: Statistical Evaluation and Robustness Analysis of SSPLNet Performance confirms that SSPLNet’s lower error rate is not due to chance. The bootstrap results also confirm that SSPLNet’s reported accuracy falls well within the 95% CI of the sampling distribution. Full article

We propose a life-expectancy pay-off function (LEP) for determining optimal cancer treatment within a control theory framework. The LEP averages life expectancy over all future outcomes, outcomes that are determined by key events during therapy such as tumour elimination (cure) and patient death (including treatment related mortality). We analyse this optimisation problem for tumours treated with chemotherapy using tumour growth models based on ordinary differential equations. To incorporate tumour elimination we draw on branching processes to compute the probability distribution of tumour population extinction. To demonstrate the approach, we apply the LEP framework to simplified one-compartment models of tumour growth that include three possible outcomes: cure, relapse, or death during treatment. Using Pontryagin’s maximum principle (PMP) we show that the best treatment strategies fall into three categories: (i) continuous treatment at the maximum tolerated dose (MTD), (ii) no treatment, or (iii) treat-and-stop therapy, where the drug is given at the MTD and then halted before the treatment (time) horizon. Optimal treatment strategies are independent of the time horizon unless the time horizon is too short to accommodate the most effective (treat-and-stop) therapy. For sufficiently long horizons, the optimal solution is either no treatment (when treatment yields no benefit) or treat-and-stop. Patients, thus, split into an untreatable class and a treatable class, with patient demographics, tumour size, tumour response, and drug toxicity determining whether a patient benefits from treatment. The LEP is in principle parametrisable from data, requiring estimation of the rates of each event and the associated life expectancy under that event. This makes the approach suitable for personalising cancer therapy based on tumour characteristics and patient-specific risk profiles. Full article

Agricultural pest detection through image analysis is a key technology in automated pest-monitoring systems. However, some existing pest detection models face excessive model complexity. This study proposes YOLO-LCE, a lightweight model based on the YOLOv8 architecture for agricultural pest detection. Firstly, a Lightweight Complementary Residual (LCR) module is proposed to extract complementary features through a dual-branch structure. It enhances detection performance and reduces model complexity. Additionally, Efficient Partial Convolution (EPConv) is proposed as a downsampling operator. It adopts an asymmetric channel splitting strategy to efficiently utilize features. Furthermore, the Ghost module is introduced to the detection head to reduce computational overhead. Finally, WIoUv3 is used to improve detection performance further. YOLO-LCE is evaluated on the Pest24 dataset. Compared to the baseline model, YOLO-LCE achieves mAP50 improvement of 1.7 percentage points, mAP50-95 improvement of 0.4 percentage points, and precision improvement of 0.5 percentage points. For computational efficiency, parameters are reduced by 43.9%, and GFLOPs are reduced by 33.3%. These metrics demonstrate that YOLO-LCE improves detection accuracy while reducing computational complexity, providing an effective solution for lightweight pest detection. Full article

In this paper, we present an extremely compact eleven-state microwave filter with four concentric split-ring resonators (SRRs). Reconfigurability is achieved by switching off either single or multiple SRRs, thereby obtaining different triple-band, dual-band, and single-band configurations from the initial quad-band topology. Switches are placed on the vertical branches of SRRs in order to minimize the additional insertion loss. As switching elements, we first use traditional RF switches—PIN diodes—and then examine the integration of non-volatile RF switches—memristors—into filter design. Memristors’ ability to remember previous electrical states makes them a main building block for designing circuits that are both energy-efficient and adaptive, opening a new era in electronics and artificial intelligence. As RF memristors are not commercially available, PIN diodes are used for experimental filter verification. Afterwards, we compare the filter characteristics realized with PIN diodes and memristors to present capabilities of memristor technology. Memristors require no bias, and their parasitic effects are modeled with low resistance for the ON state and low capacitance for the OFF state. Measured performances of all obtained configurations are in good agreement with the simulations. The filter footprint area is 26 mm × 29 mm on DiClad substrate. Full article

Despite its value as a culinary, medicinal, and essential oil crop, caraway struggles to grow and develop its biochemical quality in drought-prone sandy soils. To tackle this challenge, we conducted two field trials under drip irrigation, testing four rates of organic manure (0, 5, 10, and 15 ton/hectare (ha) and three foliar biostimulants: vitamin B₁ (50 and 100 mg L⁻¹), vitamin E (50 and 100 mg L⁻¹), and active yeast (100 and 150 mL L⁻¹). We used a randomized split-plot design with three replicates, assigning manure rates to main plots and biostimulants to subplots. We measured plant height, stem diameter, branch number, dry biomass, umbels per plant, 1000-seed weight, seed yield (per plant and per ha), essential oil content, chlorophyll a and b, carotenoids, and leaf N, P, and K. All treatments outperformed the unfertilized control. Applying 15 ton/ha of manure alone increased mean plant height by 185.3 cm, stem diameter by 2.93 mm, branch number by 14.5, and herbal weight by 91.97 g across both seasons—a gain of about 11–15%. Foliar application of vitamin B₁ at 100 mg L⁻¹ (without manure) achieved even larger gains: mean plant height improved by 176.5 cm, stem diameter by 2.6 mm, branches number by 15.1, and herbal biomass by 103.95 g (20–36% growth increases). It also boosted essential oil yield by 1.89 mL per plant (16–50%) and enhanced nutrient uptake. The most pronounced synergy emerged when combining 15 ton/ha of manure with 100 mg L⁻¹ vitamin B₁, raising seed yield to 1698.8 kg/ha (35%), plant height to 184.7 cm (52%), number of branches to 17.4 per plant (56%), umbels to 38.1 per plant (42%), 1000-seed weight to 16.9 g (48%), and essential oil yield to 2.3 mL per plant (115%), compared to the control. Chlorophyll a increased by 50%, chlorophyll b by 33%, carotenoids by 35%, and leaf N, P, and K by 43%, 90%, and 76%, respectively. Manure combined with vitamin E or yeast delivered moderate improvements. These findings demonstrate that integrating organic manure with targeted foliar biostimulants—especially vitamin B₁—under drip irrigation, is a sustainable strategy to maximize caraway yield, oil content, and nutritional quality on marginal sandy soils. Full article

Improving the catalytic performance of the oxygen evolution reaction (OER) for water splitting in acidic media is crucial for the production of clean and renewable hydrogen energy. Herein, we study the OER electrocatalytic properties of various active sites on four exposed (110) and ($\overline{1}$10) surfaces of Sn-doped RuO₂ (Sn/RuO₂) with antiferromagnetic arrangements in acidic environments. The Sn/RuO₂ bulk structure with the Cm space group exhibits favorable thermodynamic stability. The coordinatively unsaturated metal (M_cus) sites distributed on the right branch of the volcano plot are generally more active than the bridge-bonded lattice oxygen (O_br) sites located on the left. Different from the conventional knowledge that the most active site is located in the nearest neighbor of the doped atom, it has a lower OER overpotential when the active site is 3.6 Å away from the doped Sn atom. Among the sites studied, the 46-Ru_cus site exhibits the optimal OER catalytic performance. The inherent factors affecting the OER activity of each site on the Sn/RuO₂ surface are further analyzed, including the center of the d/p band at the active sites, the average electrostatic potential of the ions, and the number of transferred electrons. This work provides a reminder for the selection of active sites used to evaluate catalytic performance, which will benefit the development of efficient OER electrocatalysts. Full article

As renewable energy technologies advance, identifying efficient photocatalytic materials for water splitting to produce hydrogen has become an important research focus in materials science. This study presents a multi-task regression model (MTRM) designed to predict the conduction band minimum (CBM), valence band maximum (VBM), and solar-to-hydrogen efficiency (STH) of inorganic materials. Utilizing crystallographic and band gap data from over 15,000 materials in the SNUMAT database, machine-learning methods are applied to predict CBM and VBM, which are subsequently used as additional features to estimate STH. A deep neural network framework with a multi-branch, multi-task regression structure is employed to address the issue of error propagation in traditional cascading models by enabling feature sharing and joint optimization of the tasks. The calculated results show that, while traditional tree-based models perform well in single-task predictions, MTRM achieves superior performance in the multi-task setting, particularly for STH prediction, with an MSE of 0.0001 and an R² of 0.8265, significantly outperforming cascading approaches. This research provides a new approach to predicting photocatalytic material performance and demonstrates the potential of multi-task learning in materials science. Full article

Background: Renal tumors, encompassing benign, malignant, and normal variants, represent a significant diagnostic challenge in radiology due to their overlapping visual characteristics on computed tomography (CT) scans. Manual interpretation is time consuming and susceptible to inter-observer variability, emphasizing the need for automated, reliable classification systems to support early and accurate diagnosis. Method and Materials: We propose KidneyNeXt, a custom convolutional neural network (CNN) architecture designed for the multi-class classification of renal tumors using CT imaging. The model integrates multi-branch convolutional pathways, grouped convolutions, and hierarchical feature extraction blocks to enhance representational capacity. Transfer learning with ImageNet 1K pretraining and fine tuning was employed to improve generalization across diverse datasets. Performance was evaluated on three CT datasets: a clinically curated retrospective dataset (3199 images), the Kaggle CT KIDNEY dataset (12,446 images), and the KAUH: Jordan dataset (7770 images). All images were preprocessed to 224 × 224 resolution without data augmentation and split into training, validation, and test subsets. Results: Across all datasets, KidneyNeXt demonstrated outstanding classification performance. On the clinical dataset, the model achieved 99.76% accuracy and a macro-averaged F1 score of 99.71%. On the Kaggle CT KIDNEY dataset, it reached 99.96% accuracy and a 99.94% F1 score. Finally, evaluation on the KAUH dataset yielded 99.74% accuracy and a 99.72% F1 score. The model showed strong robustness against class imbalance and inter-class similarity, with minimal misclassification rates and stable learning dynamics throughout training. Conclusions: The KidneyNeXt architecture offers a lightweight yet highly effective solution for the classification of renal tumors from CT images. Its consistently high performance across multiple datasets highlights its potential for real-world clinical deployment as a reliable decision support tool. Future work may explore the integration of clinical metadata and multimodal imaging to further enhance diagnostic precision and interpretability. Additionally, interpretability was addressed using Grad-CAM visualizations, which provided class-specific attention maps to highlight the regions contributing to the model’s predictions. Full article

To identify optimal strategies for high-yield and high-efficiency cultivation under a “short-season direct-seeded cotton with once-only irrigation” regime, we conducted two-year field experiments (2022 and 2023) using a split-plot factorial design with three planting densities (30,000 (D1), 45,000 (D2), and 60,000 (D3) plants·ha⁻¹) and three nitrogen application rates (150 (N1), 180 (N2), and 210 (N3) kg·ha⁻¹). Our study systematically examined how these treatment combinations influenced canopy architecture, physiological traits, yield components, and fiber quality. The results showed that increased planting density significantly enhanced plant height, the leaf area index (LAI), and the number of fruiting branches, with the highest density (D3) contributing to a more compact and efficient canopy. Moderate nitrogen input (N2) significantly increased peroxidase (POD) activity, reduced malondialdehyde (MDA) accumulation, delayed functional leaf senescence, and prolonged the canopy’s photosynthetic performance. A significant interaction between planting density and nitrogen application was observed. The D3N2 treatment (high density with moderate nitrogen) consistently achieved the highest fruiting branch count, boll number per plant, and yields of both seed cotton and lint in both years, while maintaining stable fiber quality. This indicates its strong capacity to balance high yield with quality and maintain physiological resilience. By contrast, the D1N1 treatment (low density and low nitrogen) exhibited a loose canopy, premature photosynthetic decline, and the lowest yield. The D3N3 treatment (high density and high nitrogen) promoted vigorous early growth but reduced stress tolerance during later growth stages, leading to yield instability. These findings demonstrate that moderately increasing planting density while maintaining appropriate nitrogen levels can effectively optimize canopy structure, improve stress resilience, and enhance yield under short-season direct-seeded cotton systems with once-only irrigation. This provides both theoretical underpinning and practical guidance for achieving stable and efficient cotton production under such systems. Full article

Three-dimensional object detection has emerged as a focal point of increasing interest among researchers, driven by advancements in and widespread adoption of autonomous driving technologies. However, this field still faces inherent challenges in single-modal approaches that rely solely on point cloud data for 3D object detection, such as the difficulty in effectively extracting features from sparse point clouds and the lack of critical texture information in the captured representations. To overcome these limitations, we introduce PomageNet, a fusion approach that combines point cloud and image data for 3D object detection. First, initial detection results from the two different kinds of data were applied as inputs, and joint encoding was performed. The encoded joint tensor is then fed into fusion layers. In the fusion stage, multiple 1 × 1 2D convolutions are employed to extract joint high-dimensional features. To enhance feature extraction, a parallel dual-branch framework was designed, and a multidimensional joint encoding mechanism tailored to the network was proposed to better capture contextual information. Experiments show that the capacity of our model is comparable to state-of-the-art (SOTA) methods on KITTI, which was achieved by the proposed network. Results were delivered by the method in detecting small objects, a key challenge in 3D object detection. An average precision (AP) of 67.87% and 60.40% was reached on the cyclist and pedestrian splits of KITTI. Compared to CLOCs, significant improvements were achieved by PomageNet, with 1.28%, 8.40%, and 3.64% increases in the result of detection achieved on the car, cycle, and pedestrian splits of the KITTI dataset. Full article

Terahertz (THz) biosensing faces critical challenges in balancing high sensitivity and broadband spectral coverage, particularly under miniaturized device constraints. Conventional quasi-bound states in the continuum (QBIC) metasurfaces achieve high quality factor (Q) but suffer from narrow bandwidth, while angle-scanning strategies for broadband detection require complex large-angle illumination. Here, we propose a symmetry-engineered, all-dielectric metasurface that leverages multipolar interference coupling to overcome this limitation. By introducing angular perturbation, the metasurface transforms the original magnetic dipole (MD)-dominated QBIC resonance into hybridized, multipolar modes. It arises from the interference coupling between MD, toroidal dipole (TD), and magnetic quadrupole (MQ). This mechanism induces dual counter-directional, frequency-shifted, resonance branches within angular variations below 16°, achieving simultaneous 0.42 THz broadband coverage and high Q of 499. Furthermore, a derived analytical model based on Maxwell equations and mode coupling theory rigorously validates the linear relationship between frequency splitting interval and incident angle with the Relative Root Mean Square Error (RRMSE) of 1.4% and the coefficient of determination ($R^2$) of 0.99. This work establishes a paradigm for miniaturized THz biosensors, advancing applications in practical molecular diagnostics and multi-analyte screening. Full article

Object detection algorithms have evolved from two-stage to single-stage architectures, with foundation models achieving sustained improvements in accuracy. However, in intelligent retail scenarios, small object detection and occlusion issues still lead to significant performance degradation. To address these challenges, this paper proposes an improved model based on YOLOv11, focusing on resolving insufficient multi-scale feature coupling and occlusion sensitivity. First, a multi-scale feature extraction network (MFENet) is designed. It splits input feature maps into dual branches along the channel dimension: the upper branch performs local detail extraction and global semantic enhancement through secondary partitioning, while the lower branch integrates CARAFE (content-aware reassembly of features) upsampling and SENet (squeeze-and-excitation network) channel weight matrices to achieve adaptive feature enhancement. The three feature streams are fused to output multi-scale feature maps, significantly improving small object detail retention. Second, a convolutional block attention module (CBAM) is introduced during feature fusion, dynamically focusing on critical regions through channel–spatial dual attention mechanisms. A fuseModule is designed to aggregate multi-level features, enhancing contextual modeling for occluded objects. Additionally, the extreme-IoU (XIoU) loss function replaces the traditional complete-IoU (CIoU), combined with XIoU-NMS (extreme-IoU non-maximum suppression) to suppress redundant detections, optimizing convergence speed and localization accuracy. Experiments demonstrate that the improved model achieves a mean average precision (mAP50) of 0.997 (0.2% improvement) and mAP50-95 of 0.895 (3.5% improvement) on the RPC product dataset and the 6th Product Recognition Challenge dataset. The recall rate increases to 0.996 (0.6% improvement over baseline). Although frames per second (FPS) decreased compared to the original model, the improved model still meets real-time requirements for retail scenarios. The model exhibits stable noise resistance in challenging environments and achieves 84% mAP in cross-dataset testing, validating its generalization capability and engineering applicability. Video streams were captured using a Zhongweiaoke camera operating at 60 fps, satisfying real-time detection requirements for intelligent retail applications. Full article

Background: The brachial plexus is a network of nerves responsible for the motor and sensory innervation of the upper limb. Variations in the formation and course of the brachial plexus are well documented, though combinations of multiple unilateral abnormalities are rare. The complex pathology of this structure nerve may result in clinical consequences. We present a unique set of brachial plexus abnormalities involving the C4–C6 nerve roots, superior and middle trunks, additional communicating branches, and delayed median nerve union. Case Presentation: During the routine dissection of a 70-year-old female cadaver, several unique variations in the brachial plexus anatomy were identified. The C4 root contributed to C5 before the superior trunk formed, resulting in a superior trunk composed of C4–C6. The C5 root was located anterior to the anterior scalene muscle, whereas C6 maintained its usual posterior position. Additionally, an anterior communicating branch from the middle trunk to the posterior cord was observed. A communicating branch between the lateral and medial cords split into two terminal branches: one merged with the ulnar nerve, and the other joined the medial contribution of the median nerve. The median nerve contributions from the lateral and medial cords merged approximately two inches above the elbow. Conclusions: This rare combination of brachial plexus anomalies has not been previously described in the literature and is of significant clinical relevance. The additional anterior communicating branch from the middle trunk may suggest potential flexor muscle innervation by the posterior cord, which typically innervates extensor muscles. Additionally, the delayed convergence of the median nerve may provide a protective mechanism in cases of midshaft humeral fracture. Awareness of these peripheral nerve abnormalities is important for diagnostic imaging, surgery, or peripheral nerve blocks. Knowledge of such variations is critical for clinicians managing upper limb pathologies. Full article

As the solar share in energy generation is expanding globally, solar nowcasting is becoming increasingly important for the efficient and economical management of the power grid. This study leveraged the spatial context provided by all-sky images (ASI) in addition to the meteorological records for improved nowcasting of global, direct, and diffuse irradiance components. The proposed methodology consists of two branches for processing the multimodal data of ASIs and meteorological data. Due to its capability of understanding the overall characteristics of the image through self-attention, a vision transformer is utilized for the image branch while normal dense layers process the tabular meteorological data. The proposed architecture is compared against the baselines of the Ineichen clear sky model, a feedforward neural network (FFNN) where cloud coverage is computed from the ASIs by a simple color-channel threshold algorithm, and a hybrid of FFNN and U-Net model, which replaces the color threshold algorithm with fully convolutional layers for cloud segmentation. The models are trained, validated, and tested using the quality-assured ground-truth data collected in Ulaanbaatar, Mongolia, from May to August 2024, under one-minute intervals with a random split of 70%, 15%, and 15%. Our approach exhibits superior performance to baselines with a significantly lower mean absolute error (MAE) of 15–33 W/m² and root mean square error (RMSE) of 26–72 W/m², thus potentially aiding grid operators’ decision-making in real-time. Full article