Search Results (1,378)

With the increasing prevalence of brain tumors, it becomes crucial to ensure fast and reliable segmentation in MRI scans. Medical professionals struggle with manual tumor segmentation due to its exhausting and time-consuming nature. Automated segmentation speeds up decision-making and diagnosis; however, achieving an optimal balance between accuracy and computational cost remains a significant challenge. In many cases, current methods trade speed for accuracy, or vice versa, consuming substantial computing power and making them difficult to use on devices with limited resources. To address this issue, we present PRA-UNet, a lightweight deep learning model optimized for fast and accurate 2D brain tumor segmentation. Using a single 2D input, the architecture processes four types of MRI scans (FLAIR, T1, T1c, and T2). The encoder uses inverted residual blocks and bottleneck residual blocks to capture features at different scales effectively. The Convolutional Block Attention Module (CBAM) and the Spatial Attention Module (SAM) improve the bridge and skip connections by refining feature maps and making it easier to detect and localize brain tumors. The decoder uses depthwise separable convolutions, which significantly reduce computational costs without degrading accuracy. The BraTS2020 dataset shows that PRA-UNet achieves a Dice score of 95.71%, an accuracy of 99.61%, and a processing speed of 60 ms per image, enabling real-time analysis. PRA-UNet outperforms other models in segmentation while requiring less computing power, suggesting it could be suitable for deployment on lightweight edge devices in clinical settings. Its speed and reliability enable radiologists to diagnose tumors quickly and accurately, enhancing practical medical applications. Full article

The high-precision prediction of near-space atmospheric temperature holds significant importance for aerospace, national defense security, and climate change research. To address the deficiencies of extracting features in conventional convolutional neural networks, this paper designs a ConvLSTM hybrid model that combines the spatiotemporal feature extraction capability of 3D convolution with a residual attention mechanism, effectively capturing the dynamic evolution patterns of the near-space temperature field. The comparative analysis with various models, including GRU, shows that the proposed model demonstrates superior performance, achieving an RMSE of 2.433 K, a correlation coefficient R of 0.993, and an MRE of 0.76% on the test set. Seasonal error analysis reveals that the prediction stability is better in winter than in summer, with errors in the mesosphere primarily stemming from the complexity of atmospheric processes and limitations in data resolution. Compared to traditional CNNs and single time-series models, the proposed method significantly enhances prediction accuracy, providing a new technical approach for near-space environmental modeling. Full article

Objective: Our study evaluated the time to normalization of postvoid residual volume after radical hysterectomy and identified risk factors for postoperative bladder dysfunction. We also aimed to establish a predictive threshold for bladder dysfunction on the third postoperative day to develop a decision-making algorithm for postoperative voiding management. Methods: This retrospective, single-center study included early-stage cervical cancer patients undergoing type B1 or C1 radical hysterectomy. Factors associated with elevated postvoid residual volume were analyzed using logistic regression, and the threshold was determined using the Youden index. Results: 67 patients were included: 36 patients (53.7%) underwent C1 radical hysterectomy and 31 (46.3%) B1. At discharge, 13 (19.4%) patients required a catheter: 8 (61.5%) required intermittent catheterization, 5 (38.5%) had a Foley catheter. By postoperative day 3, 49 (73.1%) patients recovered their voiding function. The median time to postvoid residual volume recovery was 1 day (IQR: 1–2) for type B1 and 2.5 days (IQR: 2–5) for type C1 (p < 0.01). Compared with B1, C1 radicality was independently associated with a higher risk of postoperative voiding dysfunction (OR = 11.46; 95% CI: 1.75–75.24; p < 0.05). Based on these findings, we propose an algorithm for risk-adapted postoperative voiding management: B1 patients can safely have catheters removed on postoperative day 1 without a voiding trial, whereas C1 patients require one. C1 patients with postvoid residual volume ≥170 mL should have delayed catheter removal. Conclusions: Surgical radicality is a risk factor for postoperative bladder dysfunction. In type C1 radical hysterectomy, a postvoid residual volume ≥170 mL on the first postoperative day identifies patients at high risk of delayed recovery, supporting a tailored approach to postoperative voiding management. Full article

To enhance the abrasion resistance of TC4 titanium alloy and meet the demand for wear-resistant and corrosion-resistant friction pair materials for water-hydraulic components of marine equipment, the tribological properties of the material subsequent to ultrasonic rolling extrusion surface strengthening under seawater-lubricated conditions were investigated. The process of ultrasonic rolling machining was simulated and analyzed by the finite element method. The influence of process parameters on surface residual stress and surface roughness of TC4 was studied, and the appropriate range of process parameters was determined. The effects of key process parameters such as rolling times, static pressure, amplitude, and rotational speed on the surface properties of TC4 were investigated by the single-factor test method. Based on the response surface methodology, a prediction model of surface hardness and roughness of TC4 was constructed, and the process parameters were optimized and analyzed. The friction coefficient, wear amount, and wear rate of TC4 and CFRPEEK under seawater lubrication before and after strengthening were studied by wear tests. The wear morphologies of the specimens prior to and subsequent to strengthening were analyzed, and the friction and wear mechanisms were explored in depth. The results indicate that ultrasonic rolling extrusion surface strengthening process facilitates grain refinement in the surface layer of TC4, enhances surface hardness, and optimizes surface roughness, thereby improving its wear resistance. This is of guiding significance to the design and use of hydraulic components in seawater and has a promoting effect on the development of marine equipment. Full article

Selective laser melting (SLM) is a 3D printing technology for precision manufacturing. Owing to its high forming accuracy, parts fabricated by SLM can often be used directly without secondary machining. Consequently, the stress field in the structure, especially local stress concentration in small regions, is of great importance. Building on our previous work, this study proposes an accurate and efficient thermo-mechanical analysis method that combines a computational fluid dynamics (CFD) model and a finite element method (FEM) model for stress prediction in micrometer-scale SLM. Compared with the conventional element birth–death method, the present model more faithfully reproduces the SLM process and the post-solidification morphology and stress distribution. Numerical simulation of a single-track TC4 scan shows that pronounced surface undulations and lack-of-fusion regions exhibit significant stress concentration: the local residual stress can reach approximately 900 MPa, whereas regions with relatively smooth surface geometry exhibit stresses of about 650 MPa. This indicates a clear positive correlation between surface quality and stress concentration. The results provide a new theoretical basis for understanding defect formation mechanisms, spatial stress distribution, and scan-path optimization in SLM components. Full article

Laser surface texturing (LST) is an effective method for enhancing surface functionality, but its effect on corrosion resistance highly depends on texture design and processing parameters. This study investigates the influence of two LST patterns—orthogonal ellipses and concentric octo-donuts—applied with 1 to 20 repetitions on the corrosion resistance of AISI 430 ferritic stainless steel. Corrosion behavior was evaluated using potentiodynamic polarization in a 3.5 wt.% NaCl solution at room temperature, complemented by SEM and EDS analysis. The results indicate that while a single laser pass can maintain good corrosion resistance, increasing the number of repetitions significantly degrades performance. This is attributed to the disruption of the protective oxide layer, the introduction of residual stress, and the creation of localized sites for galvanic corrosion. Consequently, the study concludes that a low number of laser repetitions is crucial for preserving the corrosion resistance of LST-processed AISI 430 steel. Full article

We develop a shared denominator Carathéodory–Fejér (CF) method for efficiently evaluating linear combinations of $\phi$-functions for matrices whose spectrum lies in the negative real axis, as required in exponential integrators for large stiff ODE systems. This entire family is approximated with a single set of poles (a common denominator). The shared pole set is obtained by assembling a stacked Hankel matrix from Chebyshev boundary data for all target functions and computing a single SVD; the zeros of the associated singular-vector polynomial, mapped via the standard CF slit transform, yield the poles. With the poles fixed, per-function residues and constants are recovered by a robust least squares fit on a suitable grid of the negative real axis. For any linear combination of resolvent operators applied to right-hand sides, the evaluation reduces to one shifted linear solve per pole with a single combined right-hand side, so the dominant cost matches that of computing a single $\phi$-function action. Numerical experiments indicate geometric convergence at a rate consistent withHalphen’s constant, and for highly stiff problems our algorithm outperforms existing Taylor and Krylov polynomial-based algorithms. Full article

This study proposes a low-cost indoor positioning system based on a single Light Detection and Ranging (LiDAR) sensor and several fixed reflective reference points. Distances are obtained by trigonometric measurement, and positions are computed by trilateration. In static tests the average error was 7.4 mm. When the target moves at walking speed, small survey errors of the reference points cause the average error to increase to 21.8 mm. Finally, the proposed Reference Point Update Method (RPUM) that continuously corrects reference point coordinates using a moving average of recent residuals reduces the dynamic error from 208.71 mm to 20.34 mm, which is about 90% improvement. The method used in this paper requires no additional hardware and runs in real time. Full article

Background/Objectives: Psoriasis and age-related macular degeneration (AMD) may share immune-related pathophysiologic characteristics. However, few studies have investigated the relationship between psoriasis and AMD. We assessed the possible causal link between psoriasis and AMD in European populations. Methods: Single-nucleotide polymorphisms associated with psoriasis exposure were employed as instrumental variables (IVs) based on genome-wide significance (p < 5.0 ×${10}^{-8}$) in the FinnGen genome-wide association study (GWAS). The GWAS data for AMD were obtained from 11 studies performed by the International AMD Genomics Consortium. We performed a two-sample Mendelian randomisation (MR) study to estimate causal effects using the inverse-variance weighted, weighted median, and MR-Egger methods, as well as the MR-Pleiotropy Residual Sum and Outlier (MR-PRESSO) test. Results: We observed significant causal associations of psoriasis with AMD. Using the weighted median method, the odds ratio (OR) was 1.09 (95% CI = [1.03–1.16] and p = 0.005), and using the MR-PRESSO test, the OR was 1.04 (95% CI = [1.00–1.09] and p = 0.043). Conclusions: A potential causal association between psoriasis and AMD underscores the need to investigate inflammation as a risk factor for AMD. Full article

Modern weeding technologies include chemical weeding, non-contact methods such as laser weeding, and conventional mechanical inter-row cultivation characterized by soil loosening and weed uprooting. For maize, mechanical inter-row cultivation is key to cutting herbicide use and enhancing the soil–crop environment. This study developed a vision-guided intelligent inter-row cultivator with electric lateral shifting—its frame fabricated from Q235 low-carbon structural steel and assembled mainly via bolted and pinned joints—that computes real-time lateral deviation between the implement and crop rows through maize plant recognition and crop row fitting and uses delay compensation to command a servo-electric cylinder for precise ±15 cm inter-row adjustments corresponding to 30% of the 50 cm row spacing. To test the system’s dynamic response, 1–15 cm-commanded lateral displacements were evaluated at 0.31, 0.42, and 0.51 m/s to characterize the time-displacement response of the servo-electric shift mechanism; field tests were conducted at 0.51 m/s with three 30 m passes per maize growth stage to collect row-guidance error and root-injury data. Field results show that at an initial offset of 5 cm, the mean absolute error is 0.76–1.03 cm, and at 15 cm, the 95th percentile error is 7.5 cm. A root damage quantification method based on geometric overlap arc length was established, with rates rising with crop growth: 0.12% at the V2 to V3 stage, 1.46% at the V4 to V5 stage, and 9.61% at the V6 to V8 stage, making the V4 to V5 stage the optimal operating window. Compared with chemical weeding, the system requires no herbicide application, avoiding issues related to residues, drift, and resistance management. Compared with laser weeding, which requires high tool power density and has limited effective width, the tractor–implement system enables full-width weeding and shallow inter-row tillage in one pass, facilitating integration with existing mechanized operations. These results, obtained at a single forward speed of 0.51 m/s in one field and implement configuration, still require validation under higher speeds and broader field conditions; within this scope they support improving the precision of maize mechanical inter-row cultivation. Full article

The four serotypes of dengue virus (DENV), types 1 to 4 (DENV-1 to DENV-4), exhibit approximately 60% identity in the encoded amino acid residues of viral proteins. Reverse transcription of RNA extracted from patient serum specimens followed by PCR amplification with serotype-specific probes is the current standard technique for DENV serotyping. However, this method is time- and cost-consuming, and rapid detection systems with low cost are desirable. Previously, we developed a prototype serotype-specific immunochromatography system. That system was composed of four strips with four corresponding distinct sample buffers, each specifically detecting a single DENV serotype. In the present study, we improved this system by combining pairs of strips into one lateral-flow cassette each, providing DENV-1 and DENV-2 detection in one device and DENV-3 and DENV-4 detection in a second device; this strategy successfully reduced the required sample volume. Furthermore, we were able to adjust the composition of the sample buffers such that a single sample buffer sufficed for all four DENV serotype detection reactions, allowing much easier handling of the devices. Evaluation of this new device against laboratory and clinical DENV isolates and clinical specimens from DENV-infected individuals showed sensitivity that was comparable to that of our previous version, yielding serotype specificity of 100%. These new devices are expected to be of use in the clinical setting, accelerating both prospective and retrospective epidemiological studies. Full article

Background: Sentinel lymph node biopsy (SLNB) is the standard axillary staging procedure in clinically node-negative breast cancer but remains invasive, non-therapeutic and increasingly questioned in contemporary de-escalation algorithms. After neoadjuvant chemotherapy (NACT), however, the safety of omitting SLNB solely on the basis of a negative axillary ultrasound (AUS) is uncertain, particularly across molecular subtypes with heterogeneous chemosensitivity. This study evaluated the diagnostic performance of preoperative AUS after NACT and explored clinicopathological and biological factors associated with SLNB positivity in ultrasound-negative axillae. Methods: In this single-centre retrospective cohort, 135 women with invasive breast cancer who received NACT followed by surgery (2022–2024) were analysed. To avoid spectrum bias, 77 patients with clipped, cytologically or histologically proven node-positive disease at baseline were excluded from the main analysis. All patients underwent preoperative AUS and definitive axillary staging. Ninety-six women with ultrasound-negative axillae who proceeded to SLNB constituted the primary study population. Oestrogen receptor (ER), progesterone receptor (PR), HER2, Ki-67 and immunohistochemistry-based molecular subtype were recorded. Receiver operating characteristic (ROC) analysis and uni/multivariable logistic regression were used as exploratory tools to identify factors associated with SLNB positivity. Results: In the overall cohort, AUS sensitivity, specificity, negative predictive value and false-negative rate for axillary metastasis were 47.8%, 90.9%, 62.5% and 52.2%, respectively. Among ultrasound-negative axillae, SLNB was positive in 37.5%. Compared with SLNB-negative patients, those with SLNB metastases more frequently harboured an intratumoural ductal carcinoma in situ (DCIS) component, showed higher ER/PR expression and lower Ki-67, and were predominantly luminal A or luminal B/HER2−, whereas AUS performance appeared more favourable in HER2-enriched and triple-negative tumours. ROC-derived cut-offs for ER (82.5%), PR (25.0%) and Ki-67 (17.5%) provided only moderate discrimination (area under the curve 0.68–0.70). In multivariable analysis, absence of a DCIS component and low PR expression were independently associated with reduced odds of SLNB positivity, suggesting that DCIS and high PR may act as indicators of residual nodal risk in ultrasound-negative axillae. All estimates are limited by sample size and wide confidence intervals and should be interpreted as hypothesis-generating. Conclusions: Preoperative AUS alone cannot reliably exclude sentinel lymph node metastasis after NACT, particularly in luminal A and luminal B/HER2− tumours with strong hormone receptor expression and a low proliferative index. Until prospective, biology-stratified trials confirm the safety of omission, SLNB should not be withheld solely on the basis of a negative AUS in these subtypes. Axillary management after NACT should systematically integrate both imaging findings and tumour biology when considering further de-escalation of surgery. Full article

Recently, the convergence of advanced sensor technologies and innovations in artificial intelligence and robotics has highlighted facial emotion recognition (FER) as an essential component of human–computer interaction (HCI). Traditional FER studies based on handcrafted features and shallow machine learning have shown a limited performance, while convolutional neural networks (CNNs) have improved nonlinear emotion pattern analysis but have been constrained by local feature extraction. Vision transformers (ViTs) have addressed this by leveraging global correlations, yet both CNN- and ViT-based single networks often suffer from overfitting, single-network dependency, and information loss in ensemble operations. To overcome these limitations, we propose ArecaNet, an assembled residual enhanced cross-attention network that integrates multiple feature streams without information loss. The framework comprises (i) channel and spatial feature extraction via SCSESResNet, (ii) landmark feature extraction from specialized sub-networks, (iii) iterative fusion through residual enhanced cross-attention, (iv) final emotion classification from the fused representation. Our research introduces a novel approach by integrating pre-trained sub-networks specialized in facial recognition with an attention mechanism and our uniquely designed main network, which is optimized for size reduction and efficient feature extraction. The extracted features are fused through an iterative residual enhanced cross-attention mechanism, which minimizes information loss and preserves complementary representations across networks. This strategy overcomes the limitations of conventional ensemble methods, enabling seamless feature integration and robust recognition. The experimental results show that the proposed ArecaNet achieved accuracies of 97.0% and 97.8% using the public databases, FER-2013 and RAF-DB, which were 4.5% better than the existing state-of-the-art method, PAtt-Lite, for FER-2013 and 2.75% for RAF-DB, and achieved a new state-of-the-art accuracy for each database. Full article

In hyperspectral image classification (HSIC), each pixel contains information across hundreds of contiguous spectral bands; therefore, the ability to perform long-distance modeling that stably captures and propagates these long-distance dependencies is critical. A selective structured state space model (SSM) named Mamba has shown strong capabilities for capturing cross-band long-distance dependencies and exhibits advantages in long-distance modeling. However, the inherently high spectral dimensionality, information redundancy, and spatial heterogeneity of hyperspectral images (HSI) pose challenges for Mamba in fully extracting spatial–spectral features and in maintaining computational efficiency. To address these issues, we propose S²GL-MambaResNet, a lightweight HSI classification network that tightly couples Mamba with progressive residuals to enable richer global, local, and multi-scale spatial–spectral feature extraction, thereby mitigating the negative effects of high dimensionality, redundancy, and spatial heterogeneity on long-distance modeling. To avoid fragmentation of spatial–spectral information caused by serialization and to enhance local discriminability, we design a preprocessing method applied to the features before they are input to Mamba, termed the Spatial–Spectral Gated Attention Aggregator (SS-GAA). SS-GAA uses spatial–spectral adaptive gated fusion to preserve and strengthen the continuity of the central pixel’s neighborhood and its local spatial–spectral representation. To compensate for a single global sequence network’s tendency to overlook local structures, we introduce a novel Mamba variant called the Global_Local Spatial_Spectral Mamba Encoder (GLS²ME). GLS²ME comprises a pixel-level global branch and a non-overlapping sliding-window local branch for modeling long-distance dependencies and patch-level spatial–spectral relations, respectively, jointly improving generalization stability under limited sample regimes. To ensure that spatial details and boundary integrity are maintained while capturing spectral patterns at multiple scales, we propose a multi-scale Mamba encoding scheme, the Hierarchical Spectral Mamba Encoder (HSME). HSME first extracts spectral responses via multi-scale 1D spectral convolutions, then groups spectral bands and feeds these groups into Mamba encoders to capture spectral pattern information at different scales. Finally, we design a Progressive Residual Fusion Block (PRFB) that integrates 3D residual recalibration units with Efficient Channel Attention (ECA) to fuse multi-kernel outputs within a global context. This enables ordered fusion of local multi-scale features under a global semantic context, improving information utilization efficiency while keeping computational overhead under control. Comparative experiments on four publicly available HSI datasets demonstrate that S²GL-MambaResNet achieves superior classification accuracy compared with several state-of-the-art methods, with particularly pronounced advantages under few-shot and class-imbalanced conditions. Full article

Accurate state of charge (SOC) estimation is crucial for the safety, reliability, and energy efficiency of lithium-ion battery systems. However, variations in battery parameters and the loss of historical information during the update steps of traditional unscented Kalman filters (UKFs) often lead to decreased estimation accuracy under dynamic operating conditions. To address these issues, this paper proposes a variable forgetting factor recursive least squares (VFFRLS) algorithm combined with a multi-innovation unscented Kalman filter (MIUKF) algorithm. First, a second-order RC equivalent circuit model is established, and the battery parameters are identified online using the VFFRLS method, enabling the model to dynamically adapt to changing operating conditions. Then, multi-innovation theory is incorporated into the standard UKF, extending the single-innovation matrix to a multi-innovation matrix, effectively enhancing the utilization of historical residuals and improving robustness to measurement noise and model uncertainty. Experimental validation under four typical dynamic operating conditions (FUDS, DST, BJDST, and US06) demonstrates that the proposed method significantly improves SOC estimation accuracy. Compared to the traditional UKF, MIUKF reduces MAE and RMSE by 25–30% while maintaining real-time performance, with single-step computation time reaching the microsecond level. Robustness tests under different initial SOC errors further validate MIUKF’s strong robustness to initial biases. In summary, the proposed method provides an effective solution for high-precision SOC estimation of batteries and has the potential for application in battery management systems. Full article