Search Results (689)

For the automation of Agrobacterium-mediated genetic transformation of tissue-cultured plantlets, accurate leaf vein segmentation is essential. The thin, low-contrast structure of leaf veins frequently leads to fragmented segmentation outputs, despite the proposal of various methodologies for vein segmentation. To address this issue, we propose Direction-Aware U-Net (DA-UNet), an improved U-Net architecture that incorporates a Direction-Aware Context Pooling (DACPool) module and Topology-aware Segmentation loss (TopoSeg loss). The DACPool module explicitly exploits vein orientation to aggregate directional contextual information, while the TopoSeg loss jointly optimizes pixel-level accuracy and topological continuity. DA-UNet achieves efficient leaf vein segmentation with improved continuity and structural integrity, according to evaluations on the self-constructed Tissue-Cultured Plantlet Vein Dataset 2025 (TCPVD2025). Comparative experiment results show that the improved model outperforms PSPNet, DeepLabV3+, U-Net, TransUNet, Swin-UNet, CCNet, and SegNeXt, as evidenced by Recall, Dice, and CONNECT scores of 71.35%, 69.08%, and −2.25, while maintaining competitive Precision of 66.98%. Ablation experiment results provide further evidence for the efficacy of the TopoSeg loss and the DACPool module. The results demonstrate the effectiveness of the proposed vein segmentation framework for generating outputs that are both accurate and structurally consistent, thus enabling reliable automated processes for plant genetic transformation. Full article

Accurate brain tumor segmentation from magnetic resonance imaging (MRI) is crucial for brain tumor diagnosis, clinical treatment decisions, and advancing research. CNNs and Transformers have dominated this area, but CNNs struggle with long-range modeling, whereas Transformers are limited by the high computational costs of self-attention. Recently, Mamba has garnered significant attention due to its remarkable performance in long sequence modeling. However, the original Mamba architecture, designed primarily for 1D sequence modeling, fails to effectively capture the spatial and structural relationships essential for brain tumor segmentation. In this paper, we propose FocusMamba, a Mamba-based model inspired by human visual observation patterns, which jointly enhances local detail modeling and global contextual understanding. FocusMamba consists of three components: (i) a novel hierarchical and tri-directional Mamba unit that elevates attention from the global to the window level, reinforcing local semantic feature extraction, while simultaneously achieving window-level interactions to maintain broader global awareness, (ii) a large kernel convolution unit that captures long-range dependencies within whole-volume features, overcoming the limitations of Mamba’s single-scale context modeling, and (iii) a fusion unit that enhances the overall feature representation by fusing information from different levels. Extensive experiments on the BraTS 2023 and BraTS 2020 datasets demonstrate that FocusMamba achieves superior segmentation performance compared with several advanced methods. Full article

Detecting unnatural pauses is a critical component of automated quality assessment (AQA) in interpreter training, as pause patterns directly reflect an interpreter’s cognitive load and fluency. Traditional pause detection methods rely on static temporal thresholds (e.g., 1.0 s), which often fail to account for segment-specific speech rate variability and individual speaking styles. This study proposes a context-adaptive pause detection framework that integrates unsupervised anomaly detection using Isolation Forest (iForest) with a sliding window technique. To enhance pedagogical validity, we specifically focused on intra-sentential pauses by delineating sentence boundaries using a specialized segmentation model. The proposed model was evaluated against ground-truth labels annotated by professional interpreting experts. Our results demonstrate that the sliding window–based contextual anomaly detection model significantly outperforms the conventional static baseline, particularly in terms of recall and Cohen’s kappa. Furthermore, by applying a weighted F3-score and the “Recognition-over-Recall” principle, we confirmed that the proposed model substantially reduces the instructor’s total operational burden by shifting the workload from de novo annotation creation to more efficient corrective pruning. These findings suggest that speech-adaptive modeling provides a more reliable and labor-saving framework for automated interpreting assessment and feedback. Specifically, this study makes three main contributions: (1) the proposal of a context-adaptive pause detection framework using anomaly detection, (2) the integration of sliding window–based local contextual modeling for speech-rate–aware analysis, and (3) the introduction of an evaluation strategy based on the Recognition-over-Recall principle to reduce instructor workload in interpreter training. Full article

Background/Objectives: Traditional biomechanical models describe human locomotion as an articulated chain of rigid segments with constrained degrees of freedom, primarily focusing on kinematic descriptions of movement. While this approach facilitates modelling and teaching, it may limit the representation of internal force transmission and dynamic interactions, particularly during transitional phases such as gait initiation. The objective of this article is to propose a conceptual framework, Unconstrained Segmental Biomechanics (USB), to reinterpret locomotor mechanics beyond rigid joint assumptions. Methods: An exploratory analysis of recent PubMed-indexed publications (2024) and commonly adopted educational references in sport science institutions was conducted to examine how locomotion is conceptually represented and to identify possible models analogous to the framework. The aim was to situate the framework within current modelling approaches rather than to provide a systematic literature evaluation. Results: The exploratory analysis provided an exploratory contextual impression that kinematic representations were more readily identifiable than conceptually analogous models explicitly addressing dynamic intersegmental force transmission. USB is presented as a conceptual framework generating testable biomechanical hypotheses concerning the temporal organisation of intersegmental force transmission during locomotor transitions, including the expectation that during gait initiation gluteus maximus activation precedes observable segmental displacement, that early CoP/GRF changes precede the visible step, and that trunk activation actively contributes to intersegmental force regulation during the transition. Conclusions: USB offers a conceptual framework that enriches the interpretation of gait initiation and locomotor transitions. Future empirical investigations will be necessary to test the biomechanical hypotheses generated by this framework and to evaluate its potential contribution to biomechanics research, education, and applied movement sciences. Full article

Early and precise detection of plant diseases is essential for safeguarding crop yield and ensuring sustainable agricultural practices. In this study, we propose the Modified RefineNet with Attention based Fusion (MoRefNet-AF), a Modified RefineNet architecture enhanced with attention-based fusion for multi-class classification of corn (maize) and Pepper leaf diseases. Unlike the original RefineNet, which was segmentation-oriented and computationally heavy, MoRefNet-AF is redesigned for lightweight and discriminative classification. The modifications include replacing standard convolutions with depthwise separable convolutions for efficiency, adopting the Mish activation function for smoother gradient flow, redesigning the multi-resolution fusion module with concatenation and shared convolution for richer cross-scale integration, and incorporating Squeeze-and-Excitation (SE) blocks for adaptive channel recalibration. Additionally, Chained Residual Pooling (CRP) with atrous convolutions enhances contextual representation, while global average pooling with dense layers improves classification readiness. When evaluated on a curated six-class dataset combining PlantVillage and Mendeley leaf disease repositories, MoRefNet-AF achieved 99.88% accuracy, 99.74% precision, 99.73% recall, 99.95% F1-score, and 99.73% specificity. These results outperform strong baselines including ResNet152V2, DenseNet201, EfficientNet-B0, and ConvNeXt-Tiny, while maintaining only 0.3 M parameters. With its compact design and TensorFlow Lite (v2.13) compatibility, MoRefNet-AF offers a robust, lightweight, and real-time deployable solution for precision agriculture and smart plant disease monitoring. Full article

Radiomics may aid the noninvasive characterization of adrenal incidentalomas; however, reproducibility is limited by methodological heterogeneity. In this retrospective, single-center, exploratory study, we tested whether radiomic features from baseline non-enhanced computed tomography (NECT) discriminate benign from malignant/metastatic adrenal lesions and contextualized results with a PRISMA 2020 systematic review (PubMed/Scopus 2017–2025; PROSPERO CRD420251276627). Thirty-three patients (36 lesions: 12 lipid-rich adenomas, 9 lipid-poor adenomas, 6 pheochromocytomas, 7 malignant/metastatic lesions, 2 myelolipomas) were included; myelolipomas were excluded from primary comparisons. Two abdominal radiologists performed consensus 3D segmentation on NECT. Using LIFEx (v7.8.0) and IBSI definitions, 42 features were extracted and z-score standardized. LASSO selected four heterogeneity descriptors: First-order Entropy, gray-level co-occurrence matrix (GLCM) entropy, gray-level size zone matrix (GLSZM) non-uniformity, and neighboring gray tone difference matrix (NGTDM) busyness. Heterogeneity increased from lipid-rich adenomas to pheochromocytomas and malignant/metastatic lesions (Kruskal–Wallis, all p < 0.001. Pairwise separability, measured using the Vargha–Delaney A index (VDA) as a rank-based measure of separability, was highest for lipid-rich adenomas versus malignant/metastatic lesions (0.93), intermediate for lipid-poor adenomas versus pheochromocytomas (0.73), and lowest for lipid-rich versus lipid-poor adenomas (0.64). The review identified 18 eligible CT radiomics studies that consistently reported higher entropy/non-uniformity in pheochromocytomas and malignant lesions than in lipid-rich adenomas. Global heterogeneity metrics on NECT may complement conventional CT criteria in indeterminate lesions; external validation with robust reference standards is needed in larger, multicenter cohorts with harmonization. Full article

Temporal monitoring of bioprocesses is inherently complex because process variables do not evolve independently over time, and their interpretation changes as the reaction progresses. In many existing abnormality detection methods, sensor signals are analyzed at isolated time points or temporal characteristics are only weakly reflected through model structures. As a result, such approaches struggle to explain or detect abnormal behavior that emerges differently across reaction states. This study proposes TPP-TimeNet, a time-aware artificial intelligence framework developed to improve abnormality detection in bioprocess monitoring. Unlike conventional methods, the proposed framework explicitly incorporates reaction time as contextual information. Multivariate process signals are reorganized into sliding windows that reflect reaction-state transitions rather than uniform time segmentation. Temporal behavior inside each window is captured using a sequential encoding model, and reaction-state information is subsequently integrated to form state-dependent representations. Through this design, the model can distinguish between temporal patterns that are similar in shape but occur at different points in the reaction timeline. This capability leads to improved sensitivity to abnormal events that may otherwise remain undetected. Abnormality is evaluated at the window level using a probabilistic scoring scheme with a fixed threshold, enabling consistent and reproducible decision-making. The performance of TPP-TimeNet was evaluated using publicly available process control datasets from Kaggle. The datasets were reinterpreted in a bioprocess context by mapping variables such as temperature, pH, and pressure. Experimental results show that the proposed method outperforms traditional machine learning models as well as deep learning approaches that focus only on temporal features, achieving higher accuracy, sensitivity, and F1-score. These findings suggest that incorporating explicit reaction-state awareness is essential for effective abnormality detection in bioprocess monitoring systems. Full article

Adenosine triphosphate (ATP) serves as a central energy currency and signaling molecule in cellular processes, with ATP-binding sites in proteins playing critical roles in enzymatic catalysis, signal transduction, and gene regulation. The accurate identification of ATP-binding sites is essential for understanding protein function mechanisms and facilitating drug discovery, enzyme engineering, and disease pathway analysis. In this study, we present a novel hybrid deep learning framework that synergizes heterogeneous learning paradigms based on protein sequence information for accurate ATP-binding site prediction. Our approach integrates two complementary base classifiers. One is a Transformer-based model, which leverages high-level contextual embeddings generated by Evolutionary Scale Modeling 2 (ESM-2), a state-of-the-art protein language model, combined with a local–global dual-attention mechanism that enables the model to simultaneously characterize short-segment and long-range contextual dependencies across the entire protein sequence. The other is a deep Q-network (DQN)-inspired classifier that achieves residue-level prediction as a sequential decision-making process. The final predictions are generated using a weighted ensemble strategy, where optimal weights are determined via cross-validations to leverage the strengths of both models. The prediction results on benchmark independent testing sets indicate that our method achieves satisfactory performance on key metrics. Beyond predictive efficacy, this work uncovers the intrinsic biological mechanisms underlying protein–ATP interactions, including the synergistic roles of local structural motifs and global conformational constraints, as well as family-specific binding patterns, endowing the research with substantial biological significance. The research in this work offers a deeper understanding of the protein–ligand recognition mechanisms and supportive efforts on large-scale functional annotations that are critical for system biology and drug target discovery. Full article

Pedestrian safety at urban crosswalks remains a major public concern, as both vehicle speeds and roadway characteristics strongly influence drivers’ behaviour when approaching these locations. This study investigates driver behaviour patterns when approaching pedestrian crossings by integrating operating speed with key road-layout features derived from a naturalistic driving experiment conducted in Florence. A dataset of 401 observations was analysed using an unsupervised clustering framework specifically designed to handle mixed numerical and categorical variables. After preprocessing, the optimal number of clusters was identified using an elbow-based model selection applied to the K-Prototypes algorithm. The analysis produced four distinct clusters, primarily differentiated by operating speed and secondarily by contextual variables such as lane number, lane width, and acceleration behaviour. Lower-speed clusters were associated with single narrow-lane configurations, whereas higher-speed clusters were characterised by wider or multilane segments and more frequent acceleration near crossings. Information Gain analysis confirmed the dominant role of lane-related attributes, while the presence of crosswalks alone did not systematically reduce speeds. Complementary clustering excluding speed resulted in fewer clusters, indicating that speed adds essential granularity to behavioural segmentation. These findings highlight the interplay between road design and driver behaviour and provide evidence-based insights to support crosswalk configurations that mitigate high-speed conflicts in urban settings. Full article

Introduction: Artificial intelligence (AI) and robot-assisted platforms are increasingly influencing cardiothoracic surgery. AI enhances risk prediction, imaging interpretation, and early complication detection, while robotics improves visualization, dexterity, and minimally invasive access. This systematic review evaluates the current evidence supporting these technologies and their implications for clinical practice. Methods: A systematic literature search was conducted across PubMed, Embase, Scopus, Web of Science, and Google Scholar (January 2000–May 2025) following PRISMA 2020 guidelines. After screening and eligibility assessment, 67 studies met predefined inclusion criteria and were incorporated into the qualitative synthesis. Additional high-impact reviews and consensus documents were consulted for contextual interpretation. Results: Machine learning models demonstrated modest but consistent improvements in predictive performance compared with EuroSCORE II and STS scores, particularly in high-risk cohorts. Robot-assisted mitral and coronary procedures showed reduced postoperative pain, blood loss, ICU stay, and recovery time in experienced centers, though early learning phases were associated with longer operative, cross-clamp, and bypass times. AI-enabled intraoperative tools, such as video analysis, workflow recognition, and real-time anatomical segmentation, emerged as promising adjuncts for surgical precision. Structured robotic training programs, especially simulation-based and dual-console pathways, accelerated proficiency acquisition. Conclusions: AI and robotic systems act as augmentative technologies that enhance rather than replace the surgeon’s role. Their safe and effective adoption requires standardized training, transparent AI decision pathways, and clear ethical and medico-legal governance. Full article

Extracting road networks from Synthetic Aperture Radar (SAR) data represents a core challenge in remote sensing scene analysis, particularly for applications in traffic monitoring and emergency management. The task is complicated by several inherent limitations: speckle noise degrades image quality, geometric distortions arise from the side-looking acquisition geometry, and roads often exhibit weak radiometric separation from surrounding terrain. Traditional processing pipelines and recent single-branch deep learning frameworks have shown insufficient performance when global contextual reasoning and fine-scale spatial detail must both be addressed. This work presents DINOv3-PEFT, a parameter-efficient dual-encoder network designed specifically for SAR road segmentation. The architecture employs two complementary processing streams tailored to SAR characteristics: one stream utilizes adapter-based fine-tuning applied to pre-trained DINOv3 weights (kept frozen), which captures long-distance spatial relationships crucial for maintaining network connectivity despite speckle corruption. The second stream, based on convolutional operations, focuses on extracting localized geometric features that preserve the narrow, elongated structure and sharp boundaries typical of road infrastructure. Feature fusion occurs through the Topological-Geometric Feature Integration (TGFI) Module, which synthesizes multi-scale representations hierarchically. This mechanism proves effective at bridging fragmented road segments and recovering geometric accuracy in scenarios with heavy shadow casting or signal interference. Performance evaluation on the GF-3 satellite dataset across four spatial resolutions (1 m, 3 m, 5 m, and 10 m) demonstrates the proposed method achieves an 82.61% F1-score, a 76.51% IoU, and a 98.08% overall accuracy, all averaged across the four resolutions. When benchmarked against six state-of-the-art methods, DINOv3-PEFT demonstrates substantial improvements in road class segmentation quality and topological connectivity preservation, supporting its robustness for operational SAR road mapping tasks. Full article

The rapid growth of digital tourism platforms has intensified information overload and decision complexity for both locals and travelers, while operators struggle to differentiate their offerings and sustain profitable, data-driven e-commerce models. This paper presents Doroob, a big data and artificial intelligence (AI)-driven, context-aware recommendation system that integrates traditional recommender techniques with real-time facial emotion recognition (FER) to enable intelligent tourism commerce. Doroob combines three AI-based recommendation strategies: smart adaptive recommendation (SAR) collaborative filtering, a Vowpal Wabbit-based context-aware model, and a LightFM hybrid model. It trained on datasets built from the Google Places API and enriched with ratings adapted from MovieLens. FER, implemented with DeepFace and OpenCV, analyzes short video segments as users browse destination details, converts emotion scores into 1–5 satisfaction ratings, and stores this implicit feedback alongside explicit ratings to support adaptive, emotion-aware personalization. Experimental results show that the context-aware model achieves the strongest top-K ranking performance, the hybrid LightFM model yields the highest AUC of 0.95, and the SAR model provides the most accurate rating predictions, demonstrating that combining contextual modeling and FER-based implicit feedback can enhance personalization, mitigate cold-start, and support data-driven promotion of local tourist services in intelligent e-commerce ecosystems. Full article

High-resolution remote sensing (RS) images exhibit complex backgrounds, large intra-class variability, and low inter-class differences, posing substantial challenges for semantic segmentation. Although existing class-level contextual modeling methods partially alleviate these issues, they often overlook the importance of accurate and discriminative class representations and fail to effectively handle hard samples during training. To address these limitations, we propose CRECA-Net, a class representation-enhanced class-aware network designed from two complementary perspectives: class prototype refinement and difficulty-aware learning. Specifically, we introduce a class prototype refinement (CPR) module that improves class representations through pixel selection, confidence-aware contribution weighting, and an inter-class prototype separation loss, yielding more reliable and discriminative class centers. In addition, class-level context aggregation (CLCA) modules capture pixel-to-class prototype correlations via cross-attention to inject class-aware semantics into decoder features, thereby reducing interference from cluttered backgrounds and visually similar categories. Furthermore, a difficulty-aware (DA) loss dynamically estimates pixel-wise difficulty and redistributes the loss weights within each image, gradually shifting the learning focus from easy to hard samples while maintaining training stability. Extensive experiments on two benchmark RS segmentation datasets demonstrate that CRECA-Net consistently outperforms state-of-the-art methods across multiple evaluation metrics. Full article

Background/Objectives: Translating brain volumetric biomarkers to individual-level Alzheimer’s disease (AD) diagnosis remains challenging due to difficulty interpreting raw volumes without longitudinal monitoring or matched controls. We tested a classification model using population-referenced volumetric percentiles to distinguish AD from cognitively normal (CN) subjects and evaluated its generalization across independent cohorts. Methods: Brain volumes from 95 regions were extracted using an automated segmentation pipeline and converted to age and sex adjusted percentiles using a reference population (N = 1833). A logistic regression classifier was trained on ADNI subjects (N = 873; AD = 183, CN = 690) split into training (60%), validation (20%), and test (20%) sets. The model was evaluated on two independent validation datasets: the held-out ADNI validation set and an external Korean cohort (N = 72; AD = 36, CN = 36) acquired with different scanner protocols and demographic characteristics. Results: The model achieved excellent discrimination across all evaluation sets: ADNI validation (AUC = 0.963, accuracy = 90.3%), ADNI test (AUC = 0.960, accuracy = 89.7%), and Korean external validation (AUC = 0.981, accuracy = 87.5%). The minimal validation gap (0.018) demonstrated robust generalization. Positive coefficients for ventricular regions reflected AD-associated atrophy patterns, while negative coefficients for medial temporal structures indicated their contribution within multivariate patterns distinguishing AD from normal aging. Conclusions: Population-referenced brain volumetric percentiles enable accurate AD classification with robust generalization across populations and scanner protocols. By contextualizing individual brain structure relative to normative populations while accounting for age and sex, this approach demonstrates potential for clinical translation as an accessible neuroimaging-based diagnostic tool. Full article

Building extraction is a core task in the semantic segmentation of satellite remote sensing imagery. Conventional pixel-level segmentation methods often prioritize texture over geometric structure, resulting in suboptimal performance in complex scenes affected by illumination variations, shadows, and scale changes. In this article, an innovative object-level building extraction approach is introduced to better capture the geometric structure of buildings, which incorporates superpixel segmentation to represent images as a set of adjacent regions. The proposed model consists of a cascade multi-scale fusion module (CMSFM) that progressively integrates contextual information across different receptive fields, along with a boundary-assisted loss function designed to enhance edge delineation and improve object-level accuracy. The experimental results on the WHU building dataset and the Massachusetts Buildings Dataset show that the proposed method notably outperforms other representative semantic segmentation approaches, such as FCN, UNet, DeepLab V3, and SETR. On the WHU dataset, MRLNet achieves the largest MIoU of 90.14% and the highest F1 score of 92.47%. On the Massachusetts Buildings Dataset, MRLNet attains the best MIoU of 83.14% and the highest F1 score of 90.46%. In addition, our building extraction model achieves a substantial performance improvement after the addition of the CMSFM module and the boundary-assisted loss function, demonstrating the effectiveness of these two enhancements used in our proposed model. It is expected that this research can provide a promising tool for the accurate extraction of buildings using satellite remote sensing images, which is indispensable in urban planning, disaster assessment, and other fields. Full article