Search Results (7,900)

Distinguishing Active from Inactive Tuberculosis (TB) on Chest X-rays presents a clinical challenge due to overlapping radiological signs. This study introduces Vision Mamba CGSM, a deep learning framework integrating a State Space Model (SSM) backbone with a Context-Guided Slot Mixing (CGSM) module. The SSM captures global anatomical context, while the CGSM module isolates subtle pathological features by applying localized spatial attention. We validated the model using a hierarchical diagnostic scheme covering Normal, Pneumonia, Active TB, and Inactive TB. Experimental evaluations demonstrate an accuracy of 92.96% and a Youden Index of 79.55% on the independent test set. In the specific binary classification of Active vs. Inactive TB, the model recorded a specificity of 97.04%, outperforming standard baseline architectures including ResNet152 and ViT-B. Additional validations on external datasets confirm the consistent generalization of the proposed feature extraction mechanism. Full article

Against the global carbon neutrality target, predictive maintenance (PdM) of automotive engines represents a core technical strategy to advance the sustainable development of the automotive industry. Conventional single-modal diagnostic approaches for engine abnormal sound defects suffer from low accuracy and weak anti-interference capability. Existing multi-modal fusion methods fail to deeply mine the physical coupling between cross-modal features and often entail excessive model complexity, hindering deployment on resource-constrained on-board edge devices. To resolve these limitations, this study proposes a Physical Prior-Embedded Cross-Modal Attention (PPE-CMA) mechanism for lightweight multi-modal fusion diagnosis of engine abnormal sound defects. First, wavelet packet decomposition (WPD) and mel-frequency cepstral coefficients (MFCC) are integrated to extract time-frequency features from engine audio signals, while a channel-pruned ResNet18 is employed to extract spatial features from engine thermal imaging and vibration visualization images. Second, the PPE-CMA module is designed to adaptively assign attention weights to audio and image features by exploiting the physical coupling between engine fault acoustic and visual characteristics, enabling efficient cross-modal feature fusion with redundant information suppression. A rigorous theoretical derivation is provided to link cosine similarity with the physical correlation of engine fault acoustic-visual features, justifying the attention weight constraint (β = 1 − α) from the perspective of fault feature physical coupling. Third, an improved lightweight XGBoost classifier is constructed for fault classification, and a hybrid data augmentation strategy customized for engine multi-modal data is proposed to address the small-sample challenge in industrial applications. Ablation experiments on ResNet18 pruning ratios verify the optimal trade-off between diagnostic performance and computational efficiency, while feature distribution analysis validates the authenticity and effectiveness of the hybrid augmentation strategy. Experimental results on a self-constructed multi-modal dataset show that the proposed method achieves 98.7% diagnostic accuracy and a 98.2% F1-score, retaining 96.5% accuracy under 90 dB high-level environmental noise, with an end-to-end inference speed of 0.8 ms per sample (including preprocessing, feature extraction, and classification). Cross-engine and cross-domain validation on a 2.0T diesel engine small-sample dataset and the open-source SEMFault-2024 dataset yield average accuracies of 94.8% and 95.2%, respectively, demonstrating strong generalization. This method effectively enhances the accuracy and robustness of engine abnormal sound defect diagnosis, offering a lightweight technical solution for on-board real-time fault diagnosis and in-plant online quality inspection. By reducing engine fault-induced energy loss and spare parts waste, it further promotes energy conservation and emission reduction in the automotive industry. Quantified experimental data on fuel efficiency improvement and carbon emission reduction are provided to substantiate the ecological benefits of the proposed framework. Full article

The electrical characteristics of distribution networks (DNs) are drastically changing, which is mainly due to widespread adoption of small-scale distributed energy resources (DERs) by end-users. In these cases, conventional planning models may lead to overinvestment choices. This paper presents a planning model for utility companies that explicitly incorporates a model of end-users’ energy-related decisions, considering a neighborhood energy trading scheme (NETS). The model is formulated based on the Stackelberg game (SG) approach, which guarantees the optimality of the final solution for each user and the utility. The proposed mixed-integer second-order cone programming (MISOCP) problem finds the optimal investment plan for transformers, lines, distributed generators (DGs), and energy storage systems (ESSs) for the utility, considering the scenarios of end-users’ investments in rooftop photovoltaic (PV) and battery systems that maximize their benefits. Additionally, a dynamic network charge (NC) scheme is designed to rationalize the network use. Also, Benders decomposition (BD) is used to improve the convergence of the solution algorithm. The numerical studies on a real 23-bus low voltage (LV) network in Perth, Australia, using real-world data reveals that the proposed planning model offers the lowest total cost and the highest penetration of DERs in comparison with conventional models. Full article

Frequency-domain electromagnetic (FDEM) measurements for shallow subsurface exploration are frequently corrupted by noise, which masks weak secondary-field responses and degrades interpretation. We propose an electromagnetic diffusion CNN (EMDiC) for 1D multi-frequency FDEM denoising, where denoising is formulated as conditional diffusion-based generation. EMDiC combines an analytic frequency–spatial encoder, a Feature-wise Linear Modulation (FiLM)-conditioned convolutional hourglass backbone, and a physics-informed composite loss built on velocity loss to improve waveform reconstruction under severe noise. A reproducible synthetic dataset is constructed through layered-earth forward modeling with concentric Transmitter–Receiver (TX–RX) geometry, multiple target categories, and mixed noise waveforms. On synthetic benchmarks covering multiple noise levels and material types, EMDiC achieves the best overall performance in Root Mean Square Error (RMSE), Signal-to-Noise Ratio (SNR), and Normalized cross-correlation (NCC) among 1D U-Net, diffusion-based variants, and representative neural baselines, with the clearest gains under medium-to-strong noise and for targets with pronounced induction responses. Ablation experiments verify the complementary contributions of electromagnetic positional encoding (EMPE), FiLM conditioning, and the composite loss. Field data validation with a self-developed GEM-3 system further shows that EMDiC improves cross-frequency coherence and suppresses oscillations while preserving the main response characteristics. Full article

Melanoma is characterized by a high mutational burden making it an established model for studying tumor neoantigens and developing strategies for personalized immunotherapy. In this study, a reproducible bioinformatics pipeline was developed and implemented for the identification and prioritization of candidate neoantigens derived from non-synonymous somatic mutations in melanoma, using genomic data from the MSK-IMPACT cohort (mel-mskimpact-2020; n = 696) and comparative reference information from TCGA-SKCM. From the somatic mutation annotation file (MAF), 16,311 non-synonymous mutations were filtered, from which 50,480 mutant 8–11-mer peptides were generated using a sliding-window approach centered on the mutated position. Peptide–HLA class I binding affinity was predicted using MHCflurry 2.0 across six representative alleles (HLA-A*02:01, HLA-A*24:02, HLA-B*35:01, HLA-B*39:05, HLA-C*04:01, and HLA-C*07:02). Candidate prioritization was initially based on predicted binding percentile (rank ≤ 2), identifying 12,209 peptide–HLA combinations with high predicted binding affinity. To refine candidate selection, additional computational analyses were incorporated, including proteasomal cleavage prediction using NetChop 3.1 and estimation of T-cell epitope immunogenicity using the Immune Epitope Database (IEDB) immunogenicity predictor. Furthermore, a direct comparison between mutant (MUT) and corresponding wild-type (WT) peptides was performed using Δaffinity and Δrank metrics to evaluate the predicted impact of somatic mutations on HLA binding. The analysis revealed a predominance of peptides associated with the HLA-B locus, particularly the allele HLA-B*35:01, among the interactions with the lowest predicted binding percentiles. Several high-ranking peptide candidates were derived from genes with known roles in melanoma biology, including PLCG2, GATA3, AKT1, PTEN, PTCH1, and SMO. Overall, the integrative computational framework implemented in this study enables the systematic prioritization of candidate neoantigens derived from non-synonymous mutations in melanoma. This pipeline provides a reproducible strategy for exploring tumor neoantigen repertoires and may serve as a foundation for subsequent experimental validation and for studies related to neoantigen-based immunotherapies and immunopeptidomics. Full article

Spatiotemporal fusion aims to generate remote sensing images with both high spatial and high temporal resolution by integrating multi-source observations. However, significant spectral inconsistencies often arise when fusing images acquired at different spatial scales, which severely degrade the radiometric fidelity and temporal reliability of the fused results. Most existing methods focus on enhancing spatial details or temporal consistency, while the cross-scale spectral discrepancy between coarse- and fine-resolution images has not been sufficiently addressed. To tackle this issue, we propose a cross-scale spectral calibration framework for spatiotemporal fusion (XSC-Net), which explicitly models and corrects spectral responses across different spatial scales. The proposed method introduces a spatial feature refinement block to enhance spatially discriminative structures and a hierarchical spectral refinement block to adaptively calibrate channel-wise spectral representations. By jointly exploiting spatial and spectral correlations, the proposed framework effectively suppresses spectral distortion while preserving fine spatial details. Extensive experiments on the public CIA and LGC datasets indicate that XSC-Net compares favorably with state-of-the-art methods, demonstrating superior performance over established baselines. Furthermore, ablation studies verify the efficacy and contribution of the proposed architectural components. Full article

Camellia oleifera is an economically vital woody oil crop. Its productivity and oil quality are severely compromised by various diseases. Implementing pixel-level lesion segmentation within complex field environments is crucial for advancing precision plant protection. Despite recent progress, existing segmentation methods struggle with three primary challenges: semantic ambiguity arising from evolving pathological stages, blurred boundaries due to overlapping lesions, and the high omission rate of micro-lesions. To address these issues, this paper presents TB-DLossNet (Text-Conditioned Boundary-Aware Network with Dynamic Loss Reweighting), a novel segmentation framework based on semantic-visual multi-modal fusion. Leveraging VMamba as the visual backbone, the proposed model innovatively integrates BERT-encoded structured text as an auxiliary modality to resolve visual ambiguities through cross-modal semantic guidance. Furthermore, a boundary enhancement branch is incorporated alongside a multi-scale deep supervision strategy to mitigate boundary displacement and ensure the topological continuity of lesion structures. To tackle the detection of small-scale targets, we designed a dynamic weight loss function conditioned on lesion area, significantly bolstering the model’s sensitivity to minute pathological features. Additionally, to alleviate the scarcity of high-quality data, we curated a comprehensive multi-modal dataset encompassing seven typical diseases of Camellia oleifera. Experimental results demonstrate that TB-DLossNet achieves a Mean Intersection over Union (mIoU) of 87.02%, outperforming the state-of-the-art unimodal VMamba and multimodal Lvit by 4.9% and 2.59%, respectively. Qualitative evaluations confirm that our model exhibits lower false-negative rates and superior boundary-fitting precision in heterogeneous field scenarios. Finally, generalization tests on an apple disease dataset further validate the robustness and transferability of the proposed framework. Full article

Multi-UAV collaborative mapping in communication-constrained indoor environments is often hampered by a trade-off between overall map refinement and the timely completion of safety-relevant shared regions. In high-density or unmapped areas, network congestion can delay the updates that matter most for close-proximity coordination, because standard bandwidth allocation does not distinguish between general map refinement and hotspot-related spatial data. To address this issue, we propose a resilient scheduling framework that prioritizes globally useful map updates while improving safety-relevant hotspot completeness under unreliable links. At its core is a Safety Reserve allocation strategy for “hotspot” submaps—areas where UAV trajectories overlap or approach unknown frontiers. By enforcing this reserve, the system directs a limited uplink budget to hotspot-related updates earlier during congestion. To remain useful under packet loss, we implement a prefix-decodable hierarchical data structure over a lightweight stateless protocol, allowing immediate fusion of valid partial updates. The framework identifies hotspots using feedback from a Lambda-Field risk model and a truncated least squares solver with graduated non-convexity (TLS–GNC) pose-graph optimizer. Experiments on S3DIS and ScanNet under partition-based two-agent emulation show that the proposed method improves hotspot-band completeness and progressive mapping quality over the tested baselines, especially under packet loss. Full article

The detection of surface defects on smartphone components is a critical step in quality assurance for industrial manufacturing. However, existing deep learning-based methods struggle with the extreme variations in defect morphology and scale, while labeled training data remains scarce due to the high cost of expert annotation. To address these challenges, we propose a twofold solution. First, we introduce MSAdaNet, a Multi-Scale Adaptive Defect Detection Network, which integrates three novel modules: a Parallel Multi-Scale Feature Aggregation (PMSFA) backbone, a Focusing Diffusion Pyramid Network (FDPN) neck, and a Scale-Adaptive Shared Detection (SASD) head. Second, to combat data scarcity, we propose a novel data generation pipeline, creating the synthetic Smartphone Camera Bezel Dataset (SCBD) of 4936 images. Extensive experiments on both real-world and synthetic datasets validate our approach. On the challenging public SSGD, MSAdaNet achieves a state-of-the-art mAP@0.5 of 54.8%, outperforming prominent frameworks and improving upon the strong YOLOv11m baseline by +10.6 points in mAP@0.5 and +18.3 points in recall. Furthermore, on our synthetic SCBD, the model achieves an impressive 94.0% mAP@0.5, confirming the quality of our data generation pipeline and the robustness of our architecture across different data distributions. Ablation studies systematically confirm the significant contribution of each proposed module, validating MSAdaNet as an effective and efficient solution for industrial defect detection. Full article

Accurate cell and nuclei segmentation remains challenging due to the sensitivity of classical graph-cut methods to parameter tuning. While deep learning models like U-Net offer strong performance, they require large annotated datasets and substantial GPU resources. This work presents a cost-effective alternative: a genetic programming (GP) framework that jointly optimizes unary cost functions and regularization parameters for graph-cut segmentation, coupled with automatic seed selection. Evaluation is conducted under two distinct protocols: (1) oracle-guided per-image optimization, establishing upper-bound performance (mean Dice 0.822, IoU 0.733), and (2) true generalization via train/test split, where expressions learned on 50 images are applied to 50 unseen images (mean Dice 0.695, IoU 0.588). The fixed-model generalization still significantly outperforms the baseline graph cut ($+0.158$ Dice, $p<0.001$). Cross-dataset validation on MoNuSeg (H&E histopathology) achieves a Dice score of 0.823 with the fixed GP model, significantly outperforming the baseline (+0.272). This result uses a single fixed model—the best-performing expression from BBBC038 training—applied in a zero-shot manner to MoNuSeg without any retraining or domain adaptation. All 100 images showed non-negative improvement under oracle optimization in the experiments. The method requires no GPU training, runs in 550 s per image for oracle search, and offers interpretable symbolic cost functions. Code and annotations are provided to ensure reproducibility. This approach offers a practical, interpretable alternative in resource-constrained biomedical imaging settings. Full article

Accurate liver and tumor segmentation from abdominal computed tomography (CT) scans is essential for diagnosis and treatment planning; however, centralized deep learning approaches are often constrained by privacy regulations and inter-institution data-sharing limitations. To address these challenges, we propose a skip-free feature-forward collaborative segmentation framework called Feature-Forward Residual U-Net (FF-ResUNet), in which each institution executes the encoder locally and transmits only compact bottleneck representations to a central server. High-resolution encoder features and skip connections remain strictly within institutional boundaries, reducing privacy exposure and communication overhead. The server reconstructs segmentation masks using a multi-scale dilated residual decoder with progressive upsampling and returns lightweight updates for encoder refinement. FF-ResUNet is evaluated on the Liver Tumor Segmentation (LiTS) Challenge dataset, with cross-domain testing on 3D-IRCADb and AMOS-CT to assess robustness under distribution shifts and simulated multi-institution collaboration. On LiTS, the proposed framework achieves a liver Dice score of 0.952 ± 0.015 and a tumor Dice score of 0.737 ± 0.060, with a tumor HD95 of 10.9 ± 4.1 mm. Cross-domain experiments demonstrate stable generalization to unseen datasets, while multi-client simulations show improved performance as the number of participating institutions increases before saturation. Compared with skip-based collaborative U-Net architectures, FF-ResUNet reduces communication payload by 92–98% per training iteration while maintaining competitive segmentation accuracy. These results indicate that FF-ResUNet provides an effective balance between segmentation performance, communication efficiency, and privacy preservation evaluated under simulated multi-institution collaborative settings, supporting practical multi-center clinical deployment in bandwidth- and policy-constrained healthcare environments. Full article

Laser Powder Bed Fusion for metals (PBF-LB/M) is a complex additive manufacturing process in which metal powder is selectively melted layer-by-layer to fabricate 3D parts. Process parameters critically influence the resulting microstructure in nickel alloys, with features such as melt pool marks, grain size and orientation, porosity, and cracks serving as key process signatures. These features are typically analyzed post-process to identify suboptimal conditions. This research aims to develop automated post-process measurement and analysis techniques using image processing, pattern recognition, and statistical learning to correlate process parameters with part quality. Optical microscopy images of build surfaces are analyzed using machine learning algorithms to evaluate porosity, grain size, and relative density in fabricated test coupons. Effect plots are generated to identify trends related to increasing energy density. A novel deep learning approach based on Mask R-CNN is used to detect and segment melt pool regions in optical microscopy images. From the segmented regions, melt pool dimensions—such as width, depth, and area—are extracted using bounding geometry coordinates. Manually labeled images (Type I and Type II) are used to train the model. A comparison between ResNet-50 and ResNet-101 backbones shows that the ResNet-50-based model (Model 2) achieves superior performance, with lower training loss (0.1781 vs. 0.1907) and validation loss (8.6140 vs. 9.4228). Quantitative evaluation using the Jaccard index, precision, and recall metrics shows that the ResNet-101 backbone outperforms ResNet-50, achieving about 4% higher mean Intersection-over-Union, with values of 0.85 for Type I and 0.82 for Type II melt pools, where Type I is detected more accurately due to its more regular morphology and clearer boundaries. By extending Faster R-CNNs with a mask prediction branch, the method allows for precise melt pool measurements, providing valuable insights into process quality and dimensional accuracy, and aiding in the detection of defects in PBF-LB-fabricated parts. Full article

Historic streets, as living heritage environments, preserve everyday cultural practices while facing increasing digital mediation in tourism and daily life. This study examines how a digital sense of place is constructed online in the Ming–Qing Old Street of Songyang, China. User-generated text and image data were collected primarily from Weibo, supplemented by user reviews from major travel platforms, including Dianping, Fliggy, Mafengwo, and Ctrip, and analysed through a multimodal framework. BERTopic was applied to identify thematic narratives in textual content, and ResNet-50 was used to classify visual scene elements in shared images, enabling an integrated interpretation of textual and visual representations. The results reveal four dominant dimensions of digital place perception: local food culture, living handicrafts, historic spatial fabric, and everyday atmosphere. Textual narratives emphasise emotional attachment and experiential interpretation, while visual representations highlight photogenic, performative, and shareable street scenes. The integration of these modalities forms a layered digital sense of place grounded in cultural continuity and daily life. The study demonstrates the value of multimodal social media analysis in understanding how living heritage streets are digitally represented and perceived, offering implications for sustainable heritage conservation, community-centred revitalisation, and data-informed cultural tourism management. Full article

Deep learning has advanced rapidly in medical image segmentation, yet hepatopancreatic tumor delineation remains challenging due to low contrast, small lesion size, organ variability, and limited high-quality annotations. Existing reviews are outdated or overly broad, leaving recent architectural developments, training strategies, and dataset limitations insufficiently synthesized. To address this gap, we conducted a PRISMA 2020 systematic literature review of studies published between 2021 and 2026 on deep learning-based liver and pancreatic tumor segmentation. From 2307 records, 84 studies met inclusion criteria. U-Net variants continue to dominate, achieving strong liver segmentation but inconsistent tumor accuracy, while transformer-based and hybrid models improve global context modeling at higher computational cost. Attention mechanisms, boundary-refinement modules, and semi-supervised learning offer incremental gains, yet pancreatic tumor segmentation remains notably difficult. Persistent issues, including domain shift, class imbalance, and limited generalization across datasets, underscore the need for more robust architectures, standardized benchmarks, and clinically oriented evaluation. This review consolidates recent progress and highlights key challenges that must be addressed to advance reliable hepatopancreatic tumor segmentation. Full article

Fine-grained image classification under few-shot learning conditions remains a significant challenge due to limited labeled data and high intra-class similarity. This paper proposes a novel cross-modal framework that integrates Contrastive Language-Image Pretraining (CLIP) embeddings within a Siamese similarity network to enable robust and label-efficient classification. By leveraging the semantic alignment between textual class descriptions and visual representations, the model forms hybrid similarity pairs of image-to-image and image-to-text within a shared latent space, facilitating discriminative learning under low-shot scenarios. The architecture employs a dual-branch CLIP encoder and a contrastive loss function to optimize intra-class compactness and inter-class separability. Experiments conducted on benchmark datasets including miniImageNet and CUB-200-2011 demonstrate substantial improvements over zero-shot and few-shot baselines, achieving 70.32% accuracy, 71.15% F1-score, and 68.47% mAP on 5-way 1-shot and 78.41% accuracy, 79.02% F1-score, and 76.83% mAP on 5-way 5-shot tasks (averaged over 600 episodes with 95% confidence intervals on the CUB-200-2011 dataset). Extended evaluations under 10-way settings show similarly strong performance. Ablation studies further validate the critical roles of contrastive learning, normalization, and cross-modal embeddings in enhancing generalization. This work presents a scalable and interpretable paradigm for fine-grained classification in data-scarce domains. Full article