Search Results (2,430)

Addressing the challenges of dense conceptual content and intricate knowledge relations in computer science textbooks, where traditional pipeline-based information extraction suffers from error propagation and semantic decoupling, this paper proposes a concept-driven joint extraction method termed CDE (Concept-Driven Extraction).First, the model’s ability to focus on domain-specific terminology is enhanced through conceptual priors and attention re-weighting. This is integrated with a predefined schema and structured instruction templates to achieve normalized output for both entities and relations. Second, efficient domain knowledge transfer for computer science textbooks is realized by performing Low-Rank Adaptation (LoRA) fine-tuning on the Qwen3-4B large language model. Finally, the construction of the computer science textbook knowledge graph is accomplished using the Neo4j graph database. On a self-constructed instruction dataset of computer science textbooks, CDE achieves an F1 score of 81.83%, representing an improvement of approximately 2.47 percentage points over the LKD-KGC baseline. This performance significantly surpasses that of traditional pipeline models and existing joint extraction approaches. Experimental results demonstrate that CDE can effectively improve knowledge extraction accuracy in the textbook domain, thereby providing a novel research avenue for the rapid construction of knowledge graphs for computer science educational materials. Full article

Heat shock protein 90 (HSP90) is a molecular chaperone essential for maintaining the stability of many oncogenic client proteins. Although several HSP90 inhibitors (HSP90i) have entered clinical trials, their use has been limited by toxicity and resistance, underscoring the need for improved therapeutic strategies. In this study, we assessed the therapeutic potential of a new HSP90i, SP11, in T-cell acute lymphoblastic leukemia (T-ALL) in vitro and in the DLA mouse model in vivo, using single-cell transcriptomic profiling. Single-cell RNA sequencing showed that SP11 treatment reduces key oncogenic drivers, including MYC, BCL2, and stemness-related genes, consistent with impaired leukemic survival programs. In the DLA mouse model, SP11-mediated HSP90 inhibition was associated with alterations in the tumor microenvironment, including increased immune cell representation and enrichment of cytokine- and antigen-presentation-related transcriptional pathways. Despite these antitumor effects, a distinct subpopulation of cells continued to express or re-express MYC and BCL2, suggesting the development of early adaptive resistance. Consistent with these findings, an SP11-resistant MOLT4 cell line maintained high levels of MYC and BCL2 at both the transcript and protein levels, maintained CD44 expression, and exhibited altered inflammatory cytokine signaling. Functional studies confirmed that pharmacological inhibition of BCL2 notably increased SP11 sensitivity, supporting a rational combination strategy. Collectively, our results show that SP11 may exert both tumor-intrinsic and immune-modulating effects and reveal transcriptionally defined adaptive cellular states linked to resistance. This study provides mechanistic in sights into responses to HSP90 inhibition and supports combination approaches for improving therapeutic outcomes in T-ALL. Full article

This study proposes a novel intrusion detection and response architecture that formulates network traffic analysis as a sequential decision-making problem rather than a static classification task. The architecture integrates a Transformer Encoder for temporal feature extraction with a Dueling Double Deep Q-Network (DDQN) to enable autonomous and risk-aware security decisions. Network flows are modeled within a Markov Decision Process, where the agent learns an optimal policy over a hierarchical action space consisting of IGNORE, LOG, ESCALATE, and BLOCK actions. To evaluate generalization capability, a transfer learning-based cross-domain adaptation strategy was employed. The CICIDS2018 and CICIoT2023 datasets were re-partitioned using a stratified 70/15/15 train/validation/test split. The proposed model achieved high detection performance on these datasets with F1-scores of 99.48% and 99.13%, respectively. After transfer learning to the AWID3 dataset, the model preserved strong generalization capability with F1-scores of 96.76% and 96.61%, demonstrating its robustness across wired, IoT, and wireless network environments. A risk-aware reward function is designed to balance detection accuracy and operational cost, while Integrated Gradients-based explainability is incorporated to analyze decision behavior. Experimental results further show that the proposed Transformer–DDQN framework achieves more stable learning, lower optimization loss, and more consistent action policies compared to alternative reinforcement learning-based approaches. The model operates with high computational efficiency while maintaining real-time processing capability in high-throughput network environments. Full article

This article re-examines scenes of anagnorisis (recognition) in the Fourth Gospel by expanding upon established literary and semiotic models. While existing scholarship successfully identifies these scenes as formal tools mapping Jesus’ identity, it leaves crucial textual variations and the underlying cognitive mechanisms unresolved. To address these gaps, this study proposes a revised theoretical framework based on three integrated criteria: narrative criticism, hierarchy of compositional models, and interpretive semiotics. This threefold approach is applied to the representative analysis of John 1:19–34 and 20:1–10. The study demonstrates that textual variations from the standard type scene are deliberate adaptations driven by Johannine theological and narrative demands. Furthermore, this paper argues that Johannine anagnorisis is not of a simplistic or material kind in response to a sign. Instead, it is a profoundly relational event and a moment of mutual self-disclosure between the revealing God and the receptive interpreter. Full article

The development of supramolecular materials has opened up unprecedented opportunities for smart, responsive systems. Yet, their practical application in extreme environments—deep space, deep sea, polar regions, high-temperature and high-pressure reservoirs—is fundamentally challenged by the inherent trade-off between structural stability and dynamic adaptability. This review addresses this core issue by presenting a comprehensive framework for understanding and overcoming the stability–dynamism mismatch under harsh condition. We systematically analyze the molecular mechanisms by which severe factors disrupt non-covalent networks. Based on these insights, we outline four universal molecular design strategies that re-establish the balance, and summarize engineering applications across aerospace, marine, energy, and polar exploration. Beyond offering a comprehensive roadmap for rational material design, this review highlights persistent challenges—including multi-field coupling failure mechanisms, industrialization barriers, and the limitations of current systems—and outlines future directions. By bridging fundamental chemistry with extreme environment engineering, this work aims to guide the next generation of supramolecular materials that can reliably serve in the most demanding operational scenarios. Full article

RGB-D salient object detection (SOD) aims to segment the most prominent objects from the background with a pair of given RGB and depth images. Existing RGB-D methods usually rely on heavy backbones to achieve high accuracy, while current lightweight methods struggle to maintain competitive performance. To break this intractable trade-off between effectiveness and model complexity, we propose a Lightweight Spatial-Frequency Collaborative Interaction Network (SFCINet), a unified and highly efficient framework. The core of SFCINet resides in the synergy between spatial-domain features and frequency-domain global priors. Specifically, we introduce the Spatial-Frequency Synergy (SFS) module, which shifts the perspective to a joint complex Fourier domain. By adaptively learning and optimizing the decoupled amplitude and phase components, it effectively isolates clutter to yield a purified global frequency-synergized prior, which modulates the spatial branches to eliminate cross-modal discrepancies for subsequent feature fusion while supplementing global information during decoding. To alleviate the interference caused by cross-modal representation discrepancies, we design the Cross-Guidance Interaction (CMGI) module, which employs a reciprocal anchoring mechanism. It guides the counterpart to mutually filter irrelevant noise and select task-relevant information, achieving fusion in an efficient manner. Finally, we present a Calibrated Hierarchical Decoder (CHD), which injects frequency-synergized global priors into the hierarchical decoding process. It re-establishes the connection between the frequency and spatial domains, ultimately achieving global-local consistency. Extensive experiments demonstrate that SFCINet delivers superior performance over state-of-the-art methods. Full article

Robotic sorting-and-filling tasks in flexible manufacturing require robots to reproduce specified in-box arrangements while adapting to variations in container poses, object availability, sensing conditions, and external interventions. This paper proposes a box-centric one-shot teaching framework for robotic packing tasks with relative pose constraints. In the teaching stage, a human operator demonstrates the desired packing layout only once. The system uses reference-prompted SAM-based contour refinement to extract box and in-box object contours, object categories, quantities, and relative position and orientation constraints. These constraints are then converted from pixel-plane measurements into box-local pose constraints, forming a reusable box-centric packing template that preserves both translational and angular layout information. During execution, the recorded template is transferred to detected box instances with different global poses, and executable pick-and-place commands are generated through a task-level perception-to-command pipeline. A mechanism for continuous assignment and state updates is further introduced to maintain residual target slots, update object-to-slot allocation, and report missing or redundant objects across execution rounds. Single-box template transfer experiments achieved mean placement errors of 7.16 mm and 7.57 mm for two recorded templates, while representative post-execution images further showed that the relative object orientations were visually preserved with respect to the taught template footprints. Multi-box experiments demonstrated that unfinished residual slots could be preserved and completed after scene updates without re-teaching. Additional validation with different container types and object shapes showed the feasibility of extending the framework beyond cube-only cases. Ablation tests under nine exposure settings further showed that SAM refinement improved template-acquisition robustness compared with the previous recognition method. These results verify that the proposed framework enables one-shot template acquisition, box-centric layout transfer, relative pose preservation, and persistent task-level execution for constrained robotic packing tasks. Full article

Background/Objectives: Brain magnetic resonance imaging (MRI) is an important imaging modality for assessing neurological disorders. However, automatic multi-class MRI classification remains challenging because of visual similarity between disease categories, heterogeneous pathological patterns, class imbalance, and the need for reliable confidence estimation. This study aims to develop a comprehensive and well-calibrated deep learning framework for image-level brain MRI classification across multiple neurological categories. Methods: This paper introduces a new deep learning framework, MCND-ComputeNet++, for brain MRI classification into eight image-level categories using the MCND dataset, which comprises 16,400 two-dimensional brain MRI images belonging to eight diagnostic categories: AD-MildDemented, AD-ModerateDemented, AD-VeryMildDemented, BT-glioma, BT-meningioma, BT-pituitary, MS, and Normal. The proposed model uses a single pretrained EfficientNetV2-S backbone to extract hierarchical feature maps from three intermediate stages. These multi-level features are projected into a common latent space, spatially aligned, adaptively fused through learnable gated multi-scale fusion, further refined using convolutional processing, and aggregated using spatial attention pooling before classification. The training strategy combines class-balanced focal loss with label smoothing, MixUp/CutMix regularization, exponential moving average weight smoothing, warmup cosine learning-rate scheduling, temperature scaling, and test-time augmentation to improve generalization and calibration. The framework was evaluated using accuracy, precision, recall, macro-F1, macro-AUC, macro-average precision, expected calibration error, Brier score, bootstrap confidence intervals, ablation analysis, McNemar testing, and comparisons against standard pretrained baseline models. Results: MCND-ComputeNet++ achieved mean accuracy, macro-F1, macro-AUC, and macro-average precision values of 0.9738, 0.9771, 0.9993, and 0.9971, respectively, with narrow bootstrap confidence intervals indicating stable image-level performance. These findings should be interpreted as image-level/slice-level performance on MCND, because patient-level identifiers and subject-wise splitting were not available. These results outperformed most evaluated baselines, including ResNet50, DenseNet121, EfficientNetB0, EfficientNetV2-S with a standard classifier, Swin-Tiny, and ConvNeXt-Tiny, across several discrimination and calibration metrics. Compared with ConvNeXt-Tiny, the proposed model achieved higher macro-AUC and macro-average precision, together with a lower ECE and Brier score, suggesting improved image-level discrimination and confidence reliability. Compared with the EfficientNetV2-S standard classifier, accuracy increased from 0.9308 to 0.9738, while the Brier score decreased from 0.1045 to 0.0400. Conclusions: The results suggest that MCND-ComputeNet++ is a promising image-level brain MRI classification framework for the eight MCND categories. The proposed model integrates hierarchical feature extraction, shared latent projection, gated multi-scale fusion, convolutional refinement, spatial attention pooling, and calibrated inference within a unified architecture. However, because the current evaluation was conducted at the image/slice level without available patient-level identifiers, the findings should not be interpreted as patient-level clinical diagnostic validation. Further studies using subject-wise splitting, external multi-center datasets, 3D volumetric modeling, and multimodal clinical information are required to assess generalizability and potential clinical decision-support applicability. Full article

Deepfakes threaten sensor-based authentication systems, including biometric sensors, surveillance cameras, and IoT edge devices. Unimodal detectors remain vulnerable to modality-specific attacks. We propose a multimodal deepfake detection framework optimized for resource-constrained edge devices, featuring a novel cross-modal attention fusion mechanism with adaptive gating. The architecture combines enhanced Res2Net for audio, temporal 3D CNN with SE attention for video, and bidirectional cross-modal attention with quality-based gates. On our benchmark (5472 audio + 1842 video samples), the fusion model achieves 96.7% accuracy, 96.6% F1-score, 0.988 AUC-ROC, and 3.3% EER. Adversarial testing shows 92.3% accuracy under the Fast Gradient Sign Method (FGSM) attack. The model has a 30.3 MB footprint and runs at 20 FPS on edge hardware. Modality contribution analysis reveals adaptive weighting (72% audio for TTS forgery, 78% video for lip-synced attacks). Cross-dataset evaluation on FakeAVCeleb achieves 92.3% overall accuracy, confirming generalization. Full article

Background: Cataract diagnosis and management pose a significant global health challenge, contributing to 17 million cases of blindness and over 83 million cases of vision impairment worldwide in 2020. This issue is particularly acute in regions lacking adequate ophthalmological services, where a shortage of eye care clinicians and specialized equipment like slit-lamp cameras leads to late diagnoses. To address this accessibility gap, we developed a computer-assisted cataract grading system using smartphone-acquired external eye photographs. This approach utilizes image processing and deep learning on a standard, hardware-free smartphone, offering a low-cost and portable alternative to traditional equipment. Methods: The study introduces a new advanced algorithm to stratify cataract severity into three distinct stages: normal, pre-mature, and mature. The methodology was developed using a combined dataset of 799 images sourced from the Cataract v01 Computer Vision Project and the Indian Institute of Technology, Delhi. A key step is isolating the iris and lens using a region of interest (ROI) extraction procedure powered by the open-source MediaPipe framework. Key to the algorithm’s efficacy is the use of transfer learning, adapting four customized ResNet architectures (ResNet-18, ResNet-34, ResNet-50, and ResNet-101) to address medical image analysis intricacies. These models were fine-tuned with specific modifications, including dropout layers and the Adam optimizer, for analyzing the digital periocular images. Results: Evaluation of the models shows varied performance across the various architectures when classifying cataract stages. While the simpler ResNet-18 model exhibited the lowest performance, the deeper models showed significant improvement. The ResNet-50 architecture achieved the highest accuracy of 94%. This model also demonstrated excellent precision (94%), recall (95%), and an F1-score of 95% in multi-class classification, outperforming the other tested models. Its depth enables precise cataract classification, positioning it as a robust and reliable tool for potential medical diagnostic deployment. Conclusions: Deep learning-based analysis of smartphone-acquired external eye images demonstrated feasibility for cataract detection in this study. This method could be a scalable and easy-to-use addition to screening, especially in places where resources are limited. Further work is needed to expand the dataset and to validate the algorithm against established clinical grading systems before broader clinical implementation. Full article

Point cloud classification and segmentation are key technologies for 3D perception and scene understanding, whose accuracy and efficiency directly affect the performance of high-level applications such as 3D modeling, object recognition, and intelligent interaction. Existing methods still exhibit obvious deficiencies in local feature representation, computational efficiency, and scene applicability. To address these issues, this paper proposes a lightweight point cloud classification and segmentation network based on adaptive feature extraction, referred to as AFE-PointNet. Firstly, an element-wise weighting set abstraction module based on the Hadamard product is designed. It leverages geometric topology learning to achieve adaptive feature enhancement, effectively improving the representation capability of local geometric structures. Meanwhile, a cascaded structure of feature aggregation and an inverted residual multi-layer perceptron (InvResMLP) is adopted for deep feature mining to achieve high-accuracy and high-efficiency point cloud classification and segmentation. Experimental results show that AFE-PointNet achieves an overall accuracy (OA) of 93.6% on the ModelNet40 dataset and 84.5% on the ScanObjectNN dataset, and attains a class mean intersection over union (Cls.mIoU) of 83.6% on the ShapeNetPart part segmentation dataset, yielding significant performance improvements over the PointNet++ model. The proposed adaptive feature enhancement and lightweight deep mining strategies effectively improve point cloud representation capability, providing a high-precision and efficient solution for 3D vision tasks. Full article

Background and Objectives: Janus kinase (JAK) inhibitors have expanded the therapeutic options for moderate-to-severe atopic dermatitis (AD). The possibility of dose modulation with abrocitinib (100 and 200 mg) and upadacitinib (15 and 30 mg), both selective JAK1 inhibitors, represents a potential clinical advantage, allowing dose escalation in cases of insufficient response and dose de-escalation in the setting of poor tolerability or during maintenance treatment. However, real-world data remain limited, and the rationale for dose adjustment is not always standardized. This study aimed to describe, using a real-world database, the frequency, timing, and reasons for dose modifications of these agents. Materials and Methods: We retrospectively analyzed the clinical data of 212 patients with moderate-to-severe AD treated with abrocitinib (n = 47) or upadacitinib (n = 165). Dose adjustments, including dose escalation and dose de-escalation, were recorded together with their timing and clinical reasons. Results: In the abrocitinib group, 34/47 patients (72.3%) initiated treatment at 100 mg, whereas 13/47 patients (27.7%) started at 200 mg. Six patients (12.8%) underwent a dose adjustment. One patient (2.1%) switched from 200 to 100 mg because of complete AD remission and concomitant menstrual cycle alterations, whereas five patients (10.6%) underwent dose escalation from 100 to 200 mg because of incomplete disease control. Among the six abrocitinib-treated patients who underwent dose adjustment, achievement of IGA 0/1 after dose modification was documented in all cases. In the upadacitinib group, 93/165 patients (56.4%) started at 15 mg, whereas 72/165 patients (43.6%) started at 30 mg. Overall, 44/165 patients (26.7%) underwent at least one dose adjustment, accounting for a total of 50 dose modifications: 27 escalations from 15 to 30 mg and 23 de-escalations from 30 to 15 mg. Among patients initiating treatment at 15 mg, 23/93 patients (24.7%) increased the dose to 30 mg after a median of 33.1 weeks because of suboptimal disease control. Among those starting at 30 mg, 21/72 patients (29.2%) reduced the dose to 15 mg after a median of 44.3 weeks. Of these, 12/21 patients (57.1%) reduced the dose because of adverse events, including herpetic infections and acne, whereas the remaining patients de-escalated because of optimal disease control. Some patients underwent multiple dose modifications: four followed a 30→15→30 mg sequence, with re-escalation after 13.2 weeks because of suboptimal disease control, and two followed a 15→30→15 mg sequence, with dose reduction after approximately 26.7 weeks because of herpes zoster. Overall, 29/44 patients achieved IGA 0/1 within 16 weeks and 38/44 within 32 weeks after dose modification. Conclusions: In this real-world cohort, dose adjustments of selective JAK1 inhibitors were frequently performed in patients with moderate-to-severe AD, particularly among those treated with upadacitinib. Dose escalation was mainly used to address suboptimal disease control, whereas dose de-escalation was performed in the setting of adverse events or optimal disease control. The availability of two dosing regimens may allow treatment intensity to be adapted to individual disease severity, response, and tolerability, supporting a personalized approach to AD management. Full article

In response to the challenges of insufficient multi-scale feature representation and limited model adaptability in soybean leaf disease recognition from field images, a semantic residual Transformer (Sem-ResFormer) model is proposed for soybean leaf disease identification. The proposed model is constructed by integrating multi-scale residual feature extraction, Transformer-based global dependency modeling, and a semantic mapping mechanism, through which effective modeling and semantic representation of multi-scale visual information in lesion regions are achieved. A multimodal large model fine-tuning strategy combined with cross-architecture hyperparameter transfer is employed. The optimal hyperparameter configuration of the Vision Transformer, obtained via Bayesian optimization, is transferred to Qwen2.5-VL, and a progressive fine-tuning strategy is adopted, whereby the adaptability of the model to task-specific data is gradually enhanced. Experiments were conducted on a constructed five-class field-image soybean leaf disease dataset containing 3852 images, with 674 labeled images used in the initial few-shot fine-tuning stage. Under an input resolution of 720 × 720, the proposed method achieved an overall accuracy (OA) of 95.33%, surpassing the OA obtained with the default parameter configuration (93.64%) and the ResNet-50-based transfer method (93.43%). In the initial few-shot stage, the OA was improved from 74.05% under zero-shot conditions to 90.66%. These results demonstrate that the proposed method effectively improves soybean leaf disease recognition accuracy and model adaptability under the constructed field-image dataset with visual variability. Full article

MRI-guided radiotherapy (MRIgRT) enables the real-time visualization of tumor motion, allowing adaptive radiation delivery based on dynamic anatomical changes. However, respiratory-induced tumor motion remains a major challenge, particularly for thoracic and abdominal tumors. Continuous tumor motion may reduce treatment accuracy and increase radiation exposure to surrounding healthy tissues. Therefore, reliable and efficient tumor tracking is essential for real-time motion management in MRI-guided radiotherapy. Recent advances in artificial intelligence have demonstrated significant potential for medical image analysis; however, many existing tumor tracking approaches rely on segmentation-based methods that require detailed annotations and complex processing, which can limit their use in real-time clinical environments. In this work, we propose a detection-based tumor tracking framework that integrates the YOLOv11 object detection model with an enhanced StrongSORT tracking algorithm (EA-StrongSORT). The proposed approach replaces the conventional re-identification backbone with a lightweight EfficientNetV2 architecture augmented with an Efficient Channel Attention (ECA) mechanism. The overall framework follows a tracking-by-detection concept, where tumor regions are first detected and subsequently associated across frames. The proposed framework is evaluated on the TrackRAD2025 dataset using multiple YOLOv11 variants to analyze the balance between performance and model complexity. Experimental results demonstrate that the lightweight YOLOv11n model achieves the best detection performance, with a precision of 0.912, recall of 0.607, mean Average Precision (mAP) of 0.771, and ${\mathrm{mAP}}_{50–95}$ of 0.608. Furthermore, the proposed tracking framework achieves stable temporal association, with Multiple-Object Tracking Accuracy (MOTA) scores exceeding 91% and Higher-Order Tracking Accuracy (HOTA) scores around 90%. The proposed framework demonstrates the potential of detection-based tumor localization and tracking for real-time MRI-guided radiotherapy applications. Full article

Person Re-identification (Person Re-ID) is essential in surveillance and security. Traditional image processing methods often struggle to identify individuals accurately due to the sensitivity to occlusions and limited discriminative capability of the global feature representation. To address these challenges, this study proposes a deep-learning architecture for Person Re-ID, termed Dynamic Part-Based Fusion (DPBF), which integrates the Salient Part Discrimination (SPD) and the Adaptive Feature Integration and Contextual Fusion (AFICF) frameworks within a unified pipeline. The SPD module enhances representation learning by emphasizing discriminative body regions through an attention-guided part-based mechanism guided by human parsing information. The AFICF component performs the correlation-aware integration of localized part-specific features and global contextual features, reducing redundancy and improving discriminative feature representation. The proposed framework coordinates part-level feature extraction and correlation-aware integration within a unified pipeline to improve robustness under occlusion and appearance variations. Additional analyses demonstrate a stable performance across independent training runs, competitive computational complexity, and robustness under severe occlusion conditions through adaptive local–global feature integration. The method was evaluated on several Person Re-ID datasets, including Occluded-ReID, Market-1501, DukeMTMC-ReID, Occluded-Duke, P-DukeMTMC-ReID, and CUHK03-Labeled. The experimental results demonstrate a competitive performance compared with existing methods, while additional reproducibility, computational-complexity, and occlusion-stability analyses further validate the robustness and practical applicability of the proposed framework. Specifically, DPBF achieves a 10.6% increase in Rank-1 accuracy and a 16% improvement in mAP over the closest competitor on the Occluded-ReID dataset. Full article