Search Results (5,074)

To address the inefficiency of traditional concrete crack detection methods and the heavy reliance of supervised learning on extensive labeled data, in this study, an intelligent assessment method of concrete damage based on pseudo-label semi-supervised learning and fractal geometry theory is proposed to solve two core tasks: one is binary classification of pixel-level cracks, and the other is multi-category assessment of damage state based on crack morphology. Using three-channel RGB images as input, a dual-path collaborative training framework based on U-Net encoder–decoder architecture is constructed, and a binary segmentation mask of the same size is output to achieve the accurate segmentation of cracks at the pixel level. By constructing a dual-path collaborative training framework and employing a dynamic pseudo-label refinement mechanism, the model achieves an F1-score of 0.883 using only 50% labeled data—a mere 1.3% decrease compared to the fully supervised benchmark DeepCrack (F1 = 0.896)—while reducing manual annotation costs by over 60%. Furthermore, a quantitative correlation model between crack fractal characteristics and structural damage severity is established by combining a U-Net segmentation network with the differential box-counting algorithm. The experimental results demonstrate that under a cyclic loading of 147.6–221.4 kN, the fractal dimension monotonically increases from 1.073 (moderate damage) to 1.189 (failure), with 100% accuracy in damage state identification, closely aligning with the degradation trend of macroscopic mechanical properties. In complex crack scenarios, the model attains a recall rate (Re = 0.882), surpassing U-Net by 13.9%, with significantly enhanced edge reconstruction precision. Compared with the mainstream models, this method effectively alleviates the problem of data annotation dependence through a semi-supervised strategy while maintaining high accuracy. It provides an efficient structural health monitoring solution for engineering practice, which is of great value to promote the application of intelligent detection technology in infrastructure operation and maintenance. Full article

 Background/Objectives: Identifying and classifying abnormal lung sounds is essential for diagnosing patients with respiratory disorders. In particular, the simultaneous recording of auscultation signals from multiple clinically relevant positions offers greater diagnostic potential compared to traditional single-channel measurements. This study aims to improve the accuracy of respiratory sound classification by leveraging multichannel signals and capturing positional characteristics from multiple sites in the same patient. Methods: We evaluated the performance of respiratory sound classification using multichannel lung sound data with a deep learning model that combines a convolutional neural network (CNN) and long short-term memory (LSTM), based on mel-frequency cepstral coefficients (MFCCs). We analyzed the impact of the number and placement of channels on classification performance. Results: The results demonstrated that using four-channel recordings improved accuracy, sensitivity, specificity, precision, and F1-score by approximately 1.11, 1.15, 1.05, 1.08, and 1.13 times, respectively, compared to using three, two, or single-channel recordings. Conclusion: This study confirms that multichannel data capture a richer set of features corresponding to various respiratory sound characteristics, leading to significantly improved classification performance. The proposed method holds promise for enhancing sound classification accuracy not only in clinical applications but also in broader domains such as speech and audio processing.  Full article

Synthetic aperture radar (SAR), as an active microwave imaging system, has the capability of all-weather and all-time observation. In response to the challenges of aircraft detection in SAR images due to the complex background interference caused by the continuous scattering of airport buildings and the demand for real-time processing, this paper proposes a YOLOv7-MTI recognition model that combines the attention mechanism and involution. By integrating the MTCN module and involution, performance is enhanced. The Multi-TASP-Conv network (MTCN) module aims to effectively extract low-level semantic and spatial information using a shared lightweight attention gate structure to achieve cross-dimensional interaction between “channels and space” with very few parameters, capturing the dependencies among multiple dimensions and improving feature representation ability. Involution helps the model adaptively adjust the weights of spatial positions through dynamic parameterized convolution kernels, strengthening the discrete strong scattering points specific to aircraft and suppressing the continuous scattering of the background, thereby alleviating the interference of complex backgrounds. Experiments on the SAR-AIRcraft-1.0 dataset, which includes seven categories such as A220, A320/321, A330, ARJ21, Boeing737, Boeing787, and others, show that the mAP and mRecall of YOLOv7-MTI reach 93.51% and 96.45%, respectively, outperforming Faster R-CNN, SSD, YOLOv5, YOLOv7, and YOLOv8. Compared with the basic YOLOv7, mAP is improved by 1.47%, mRecall by 1.64%, and FPS by 8.27%, achieving an effective balance between accuracy and speed, providing research ideas for SAR aircraft recognition. Full article

The magnetic flux leakage (MFL) method is widely acknowledged as a highly effective non-destructive evaluation (NDE) technique for detecting local damage in ferromagnetic structures such as steel wire ropes. In this study, a multi-channel MFL sensor module was developed, incorporating a purpose-designed Hall sensor array and magnetic yokes specifically shaped for steel cables. To validate the proposed damage detection method, artificial damages of varying degrees were inflicted on wire rope specimens through experimental testing. The MFL sensor module facilitated the scanning of the damaged specimens and measurement of the corresponding MFL signals. In order to improve the signal-to-noise ratio, a comprehensive set of signal processing steps, including channel equalization and normalization, was implemented. Subsequently, the detected MFL distribution surrounding wire rope defects was transformed into MFL images. These images were then analyzed and processed utilizing an object detection method, specifically employing the YOLOv9 network, which enables accurate identification and localization of defects. Furthermore, a quantitative defect detection method based on image size was introduced, which is effective for quantifying defects using the dimensions of the anchor frame. The experimental results demonstrated the effectiveness of the proposed approach in detecting and quantifying defects in steel cables, which combines deep learning-based analysis of MFL images with the non-destructive inspection of steel cables. Full article

Many speech coding strategies have been developed over the years, but comparing them has been convoluted due to the difficulty in disentangling brand-specific and patient-specific factors from strategy-specific factors that contribute to speech understanding. Here, we present a comparison with a ‘virtual’ patient, by comparing two strategies from two different manufacturers, Advanced Combination Encoder (ACE) versus HiResolution Fidelity 120 (F120), running on two different implant systems in a computational model with the same anatomy and neural properties. We fitted both strategies to an expected T-level and C- or M-level based on the spike rate for each electrode contact’s allocated frequency (center electrode frequency) of the respective array. This paper highlights neural and electrical differences due to brand-specific characteristics such as pulse rate/channel, recruitment of adjacent electrodes, and presence of subthreshold pulses or interphase gaps. These differences lead to considerably different recruitment patterns of nerve fibers, while achieving the same total spike rates, i.e., loudness percepts. Also, loudness growth curves differ significantly between brands. The model is able to demonstrate considerable electrical and neural differences in the way loudness growth is achieved in CIs from different manufacturers. Full article

A multi-objective genetic algorithm is tailored to optimize the design of a wavelength interleaver/deinterleaver device. An interleaver combines data streams from two physical channels into one. The deinterleaver does the opposite job. The WDM requirements for this device include channel spacing of 50 GHz, channel bandwidth of 20 GHz, free spectral range of 100 GHz, maximum channel dispersion of 30 ps/nm, and maximum crosstalk of −23 dB. The challenges for the optimization process include the lack of a closed-form expression for the device performance and the trade-off between the conflicting performance parameters. So, for this multi-objective problem, the proposed approach maneuvers to find a compromise between the performance parameters within a few minutes, saving the designer the laborious design process previously proposed in the literature, which relies on visually inspecting the Z-plane for the dynamics of the transmission poles and zeros. Designs of better performance are achieved, with fewer ring resonators, a channel dispersion as low as 1.6 ps/nm, and crosstalk as low as −30 dB. Full article

Sonar is the primary tool used for detecting small targets at long distances underwater. Due to the influence of the underwater environment and imaging mechanisms, sonar images face challenges such as a small number of target pixels, insufficient data samples, and uneven category distribution. Existing target detection methods are unable to effectively extract features from sonar images, leading to high false positive rates and affecting the accuracy of target detection models. To counter these challenges, this paper presents a novel sonar small-target detection framework named DCE-Net that refines the YOLOv8 architecture. The Detail Enhancement Attention Block (DEAB) utilizes multi-scale residual structures and channel attention mechanism (AM) to achieve image defogging and small-target structure completion. The lightweight spatial variation convolution module (CoordGate) reduces false detections in complex backgrounds through dynamic position-aware convolution kernels. The improved efficient multi-scale AM (MH-EMA) performs scale-adaptive feature reweighting and combines cross-dimensional interaction strategies to enhance pixel-level feature representation. Experiments on a self-built sonar small-target detection dataset show that DCE-Net achieves an mAP@0.5 of 87.3% and an mAP@0.5:0.95 of 41.6%, representing improvements of 5.5% and 7.7%, respectively, over the baseline YOLOv8. This demonstrates that DCE-Net provides an efficient solution for underwater detection tasks. Full article

Background: Reliable platelet (PLT) measurement is crucial for the accurate diagnosis of thrombocytopenia. Several methods exist for automated PLT counting, including the impedance method (PLT-I), as well as optical and fluorescence methods (PLT-F). The impedance method is cost-effective but susceptible to interference from small red blood cells and schistocytes. In contrast, fluorescent assessment offers higher specificity but is more expensive, as it requires additional dyes and detectors. Hybrid platelet counting (PLT-H) combines impedance with measurements from the leukocyte differentiation channel and is available without additional cost. Aim: The aim of this study was to evaluate the accuracy of hybrid PLT counting in anemic samples. Methods: In this retrospective study, PLT counts from 583 unselected anemic samples were analyzed using two different analyzers: the Sysmex XN3500, equipped with fluorescent PLT-F technology, and the Mindray BC6200, which uses both impedance (PLT-I) and hybrid (PLT-H) technologies. Agreement between PLT-I and PLT-F, as well as between PLT-H and PLT-F, was assessed using Bland–Altman plots. Correlation between the methods was evaluated using the Pearson correlation coefficient. Results: The hybrid method demonstrated better accuracy in PLT counting compared to the impedance method. Correlation between PLT-H and PLT-F was excellent, ranging from 0.991 to 0.999. In thrombocytopenic samples (PLT < 50 G/L), the hybrid method also provided more reliable PLT counts than the impedance method, reducing the number of falsely elevated PLT results by nearly fivefold. Conclusions: Hybrid platelet counting yields more accurate results than the impedance method in anemic samples and shows excellent correlation with the fluorescence method. Full article

Geographic object-based image analysis (GEOBIA) has demonstrated substantial utility in remote sensing tasks. However, its integration with deep learning remains largely confined to image-level classification. This is primarily due to the irregular shapes and fragmented boundaries of segmented objects, which limit its applicability in semantic segmentation. While convolutional neural networks (CNNs) excel at local feature extraction, they inherently struggle to capture long-range dependencies. In contrast, Transformer-based models are well suited for global context modeling but often lack fine-grained local detail. To overcome these limitations, we propose GOFENet (Geo-Object Feature Enhanced Network)—a hybrid semantic segmentation architecture that effectively fuses object-level priors into deep feature representations. GOFENet employs a dual-encoder design combining CNN and Swin Transformer architectures, enabling multi-scale feature fusion through skip connections to preserve both local and global semantics. An auxiliary branch incorporating cascaded atrous convolutions is introduced to inject information of segmented objects into the learning process. Furthermore, we develop a cross-channel selection module (CSM) for refined channel-wise attention, a feature enhancement module (FEM) to merge global and local representations, and a shallow–deep feature fusion module (SDFM) to integrate pixel- and object-level cues across scales. Experimental results on the GID and LoveDA datasets demonstrate that GOFENet achieves superior segmentation performance, with 66.02% mIoU and 51.92% mIoU, respectively. The model exhibits strong capability in delineating large-scale land cover features, producing sharper object boundaries and reducing classification noise, while preserving the integrity and discriminability of land cover categories. Full article

This review explores the advantages of ivabradine in the management of cardiac surgery patients, particularly highlighting its heart rate (HR)-reducing properties, its role in minimizing the impact of atrial fibrillation, and its contributions to improving left ventricular diastolic function, as well as reducing pain, stress, and anxiety. In parallel, studies provide evidence that ivabradine influences endothelial inflammatory responses through mechanisms such as biomechanical modulation. Unlike traditional beta-blockers that may induce hypotension, ivabradine selectively inhibits hyperpolarization-activated cyclic nucleotide-gated (HCN) channels, allowing for effective HR reduction without compromising blood pressure stability. This characteristic is particularly beneficial for patients at risk of atrial fibrillation post-surgery, where HR control is crucial for cardiovascular stability. This is an area in which ivabradine appears to play a role prophylactically, possibly in combination with beta-blockers. Furthermore, ivabradine has been associated with enhanced diastolic parameters in left ventricular function, reflecting its potential to improve surgical outcomes in patients with compromised heart function. In addition to its cardiovascular benefits, it appears to alleviate psychological stress and anxiety, common in postoperative settings, by moderating the neuroendocrine response to stress, thereby reducing stress-induced hormone levels. Furthermore, it has notable analgesic properties, contributing to pain management through its action on HCN channels in both the peripheral and central nervous systems. Collectively, these findings indicate that ivabradine may serve as a valuable therapeutic agent in the perioperative care of cardiac surgery patients, addressing both physiological and psychological challenges during recovery. Full article

The conductors of high-speed railway OCSs (Overhead Contact Systems) are susceptible to conductor galloping due to the impact of natural elements such as strong winds, rain, and snow, resulting in conductor fatigue damage and significantly compromising train operational safety. Consequently, monitoring the galloping status of conductors is crucial, and instance segmentation techniques, by delineating the pixel-level contours of each conductor, can significantly aid in the identification and study of galloping phenomena. This work expands upon the YOLO11-seg model and introduces an instance segmentation approach for galloping video and image sensor data of OCS conductors. The algorithm, designed for the stripe-like distribution of OCS conductors in the data, employs four-direction Sobel filters to extract edge features in horizontal, vertical, and diagonal orientations. These features are subsequently integrated with the original convolutional branch to form the FDSE (Four Direction Sobel Enhancement) module. It integrates the ECA (Efficient Channel Attention) mechanism for the adaptive augmentation of conductor characteristics and utilizes the FL (Focal Loss) function to mitigate the class-imbalance issue between positive and negative samples, hence enhancing the model’s sensitivity to conductors. Consequently, segmentation outcomes from neighboring frames are utilized, and mask-difference analysis is performed to autonomously detect conductor galloping locations, emphasizing their contours for the clear depiction of galloping characteristics. Experimental results demonstrate that the enhanced YOLO11-seg model achieves 85.38% precision, 77.30% recall, 84.25% AP@0.5, 81.14% F1-score, and a real-time processing speed of 44.78 FPS. When combined with the galloping visualization module, it can issue real-time alerts of conductor galloping anomalies, providing robust technical support for railway OCS safety monitoring. Full article

Multi-organ medical image segmentation is essential for accurate clinical diagnosis, effective treatment planning, and reliable prognosis, yet it remains challenging due to complex backgrounds, irrelevant noise, unclear organ boundaries, and wide variations in organ size. To address these challenges, the boundary feature learning and enhancement network is proposed. This model integrates a dedicated boundary learning module combined with an auxiliary loss function to strengthen the semantic correlations between boundary pixels and regional features, thus reducing category mis-segmentation. Additionally, channel and positional compound attention mechanisms are employed to selectively filter features and minimize background interference. To further enhance multi-scale representation capabilities, the dynamic scale-aware context module dynamically selects and fuses multi-scale features, significantly improving the model’s adaptability. The model achieves average Dice similarity coefficients of 81.67% on synapse and 90.55% on ACDC datasets, outperforming state-of-the-art methods. This network significantly improves segmentation by emphasizing boundary accuracy, noise reduction, and multi-scale adaptability, enhancing clinical diagnostics and treatment planning. Full article

Channels with three standard symmetrical sections and one asymmetric section are mounted as cantilever beams with the web oriented vertically. A classical solution to the analysis of stress in each thin-walled cantilever channel is provided using the principle of wall shear flow superposition. The latter is coupled with a further superposition between axial stress arising from bending and from the constraint placed on free warping imposed at the fixed end. Closed solutions for design are tabulated for the net shear stress and the net axial stress at points around any section within the length. Stress distributions thus derived serve as a benchmark structure for alternative numerical solutions and for experimental investigations. The conversion of the transverse free end-loading applied to a thin-walled cantilever channel into the shear and axial stress that it must bear is outlined. It is shown that the point at which this loading is applied within the cross-section is crucial to this stress conversion. When a single force is applied to an arbitrary point at the free-end section, three loading effects arise generally: bending, flexural shear and torsion. The analysis of each effect requires that this force’s components are resolved to align with the section’s principal axes. These forces are then considered in reference to its centroid and to its shear centre. This shows that axial stress arises directly from bending and from the constraint imposed on free warping at the fixed end. Shear stress arises from flexural shear and also from torsion with a load offset from the shear centre. When the three actions are combined, the net stresses of each action are considered within the ability of the structure to resist collapse from plasticity and buckling. The novelty herein refers to the presentation of the shear flow calculations within a thin wall as they arise from an end load offset from the shear centre. It is shown how the principle of superposition can be applied to individual shear flow and axial stress distributions arising from flexural bending, shear and torsion. Therein, the new concept of a ‘trans-moment’ appears from the transfer in moments from their axes through centroid G to parallel axes through shear centre E. The trans-moment complements the static equilibrium condition, in which a shift in transverse force components from G to E is accompanied by torsion and bending about the flexural axis through E. Full article

The lacustrine to deltaic depositional systems of the Upper Jurassic Qigu Formation in the Yongjin area constitute a significant petroleum reservoir in the central Junggar Basin, China. Based on core observations, petrology analyses, paleoenvironment indicators and modern sedimentary analyses, sequence stratigraphy, lithofacies associations, sedimentary environment, evolution, and models were investigated. The Qigu Formation can be divided into a third-order sequence consisting of a lowstand systems tract (LST) and a transgressive systems tract (TST), which is further subdivided into six fourth-order sequences. Thirteen lithofacies and five lithofacies associations were identified, corresponding to shallow-water delta-front deposits. The paleoenvironment of the Qigu Formation is generally characterized by an arid freshwater environment, with a dysoxic to oxic environment. During the LST depositional period (SQ1–SQ3), the water depth was relatively shallow with abundant sediment supply, resulting in a widespread distribution of channel and mouth bar deposits. During the TST depositional period (SQ4–SQ6), the rapid rise in base level, combined with reduced sediment supply, resulted in swift delta retrogradation and widespread lacustrine sedimentation. Combined with modern sedimentary analysis, the shallow-water delta in the study area primarily comprises a composite system of single main channels and distributary channel-mouth bar complexes. The channel-bar complex eventually forms radially distributed bar assemblages with lateral incision and stacking. The distributary channel could incise a mouth bar deeply or shallowly, typically forming architectural patterns of going over or in the mouth bar. Reservoir test data suggest that the mouth bar sandstones are favorable targets for lithological reservoir exploration in shallow-water deltas. Full article

Reaction bonding (RB) using Al powder is an effective method for preparing porous ceramics with low shrinkage, high porosity, and high strength. However, it remains challenging to optimize mechanical strength and oil conductivity simultaneously for atomizer applications. Herein, aiming at addressing this issue, porous Al₂O₃ ceramics with ultra-high mechanical strength and oil conductivity were fabricated via the RB process using polymethyl methacrylate (PMMA) microspheres as the pore-forming agent. The pore structure was gradually optimized by regulating the additive amount, particle size, and particle gradation of PMMA microspheres. The bimodal pores, formed by Al oxidation-induced hollow structures (enhancing bonding force) and burnout of large-sized PMMA microspheres, significantly improved mechanical strength; meanwhile, three-dimensional interconnected pores derived from particle gradation increased the diversity and quantity of oil-conduction channels, boosting oil conductivity. Consequently, under an open porosity of 58.2 ± 0.1%, a high compressive strength of 7.9 ± 0.3 MPa (a 54.7% improvement) and an excellent oil conductivity of 2.1 ± 0.0 mg·s⁻¹ (a 46.5% improvement) were achieved. This superior performance combination, overcoming the trade-off between strength and oil conductivity, demonstrates substantial application potential in atomizers. Full article