Search Results (459)

Building on advances in promptable segmentation models, this work introduces a framework that integrates Large Vision-Language Model (LVLM) bounding box priors with prototype-based region of interest (ROI) selection to improve zero-shot medical image segmentation. Unlike prior methods such as SaLIP, which often misidentify regions due to reliance on text–image CLIP similarity, the proposed approach leverages visual prototypes to mitigate language bias and enhance ROI ranking, resulting in more accurate segmentation. Bounding box estimation is further strengthened through systematic prompt engineering to optimize LVLM performance across diverse datasets and imaging modalities. Evaluation was conducted on three publicly available benchmark datasets—CC359 (brain MRI), HC18 (fetal head ultrasound), and CXRMAL (chest X-ray)—without any task-specific fine-tuning. The proposed method achieved substantial improvements over prior approaches. On CC359, it reached a Dice score of 0.95 ± 0.06 and a mean Intersection-over-Union (mIoU) of 0.91 ± 0.10. On HC18, it attained a Dice score of 0.82 ± 0.20 and mIoU of 0.74 ± 0.22. On CXRMAL, the model achieved a Dice score of 0.90 ± 0.08 and mIoU of 0.83 ± 0.12. These standard deviations reflect variability across test images within each dataset, indicating the robustness of the proposed zero-shot framework. These results demonstrate that integrating LVLM-derived bounding box priors with prototype-based selection substantially advances zero-shot medical image segmentation. Full article

Background and Clinical Significance: Psoriasis, a chronic immune-mediated inflammatory disease, can affect musculoskeletal structures, including the Achilles tendon. Achilles pain in psoriasis patients may arise from tendinitis or neuropathic pain due to peripheral nerve dysfunction, such as sural nerve (SN) involvement, a condition frequently misdiagnosed due to limitations in conventional diagnostics. Fascial tissues are critical in nerve compression syndromes. This case explores the application of a novel quantitative Sonoguide Digital Palpation (SDP) protocol and ultrasound (US)-guided hydrodissection (HD) for SN dysfunction mimicking Achilles tendinopathy in a psoriasis patient. Case Presentation: A 41-year-old male with psoriasis presented with acute onset of right heel stiffness and paresthesia. Physical examination, radiographs, and ultrasound were performed. SDP, employing a validated four-criterion diagnostic framework (including fascial mobility quantification and concordant pain provocation), identified crural fascia restriction affecting SN and reproduced patient’s concordant Achilles pain. High-resolution ultrasonography provided key morphological evidence, revealing a 2.6-fold enlargement of the sural nerve’s cross-sectional area (CSA) on the affected side (13 mm²) compared to the asymptomatic side (5 mm²). Notably, a positive Tinel’s sign was elicited over the psoriatic plaque. US-guided HD was performed using 50 cc of 5% dextrose in water (D5W) without local anesthetic below the psoriatic lesion. Post-HD, the patient reported immediate and significant pain relief (Numeric Pain Rating Scale (NPRS) score reduction from 8 to 2), confirming the prompt correction of a clinically important fascial restriction, associated with improved SN mobility, objectively verified by a post-procedure SDP assessment. At 24-month follow-up, sustained symptom relief and complete functional recovery were reported. Conclusions: This case highlights SDP’s ability to objectively visualize and confirm fascial restriction as a cause of nerve dysfunction by quantitatively reproducing concordant pain. The objective finding of nerve swelling provides sonographic substantiation for the functional diagnosis of nerve dysfunction. This integrated diagnostic approach, combining dynamic functional assessment with morphological confirmation, offers a novel paradigm for evaluating peripheral nerve disorders. US-guided HD of the SN with D5W without local anesthetic shows promise as both a diagnostic confirmatory tool and therapeutic intervention for neuropathic Achilles pain in psoriasis patients with SN involvement, aligning with its efficacy in other peripheral neuropathies. The significant nerve swelling (13 mm²) provides robust morphological corroboration of the functional impairment diagnosed by SDP, offering a more comprehensive diagnostic paradigm. Full article

The convergence of Internet of Things (IoT) and Artificial Intelligence (AI) has enabled personalized sports coaching, yet a significant gap remains: translating low-level sensor data into high-level, contextualized feedback. Large Language Models (LLMs) excel at reasoning and instruction but lack a native understanding of physical kinematics. This paper introduces BoxingPro, a novel framework that bridges this semantic gap by fusing wearable sensor data with LLMs for automated boxing coaching. Our core contribution is a dedicated translation methodology that converts multi-modal time-series data (IMU) and visual data (video) into structured linguistic prompts, enabling off-the-shelf LLMs to perform sophisticated biomechanical reasoning without extensive retraining. Our evaluation with professional boxers showed that the generated feedback achieved an average expert rating of over 4.0/5.0 on key criteria like biomechanical correctness and actionability. This work establishes a new paradigm for integrating sensor-based systems with LLMs, with potential applications extending far beyond boxing to any domain requiring physical skill assessment. Full article

Objective: Medical Visual Question Answering (Med-VQA), a key technology that integrates computer vision and natural language processing to assist in clinical diagnosis, possesses significant potential for enhancing diagnostic efficiency and accuracy. However, its development is constrained by two major bottlenecks: weak few-shot generalization capability stemming from the scarcity of high-quality annotated data and the problem of catastrophic forgetting when continually learning new knowledge. Existing research has largely addressed these two challenges in isolation, lacking a unified framework. Methods: To bridge this gap, this paper proposes a novel Evolvable Clinical-Semantic Alignment (ECSA) framework, designed to synergistically solve these two challenges within a single architecture. ECSA is built upon powerful pre-trained vision (BiomedCLIP) and language (Flan-T5) models, with two innovative modules at its core. First, we design a Clinical-Semantic Disambiguation Module (CSDM), which employs a novel debiased hard negative mining strategy for contrastive learning. This enables the precise discrimination of “hard negatives” that are visually similar but clinically distinct, thereby significantly enhancing the model’s representation ability in few-shot and long-tail scenarios. Second, we introduce a Prompt-based Knowledge Consolidation Module (PKC), which acts as a rehearsal-free non-parametric knowledge store. It consolidates historical knowledge by dynamically accumulating and retrieving task-specific “soft prompts,” thus effectively circumventing catastrophic forgetting without relying on past data. Results: Extensive experimental results on four public benchmark datasets, VQA-RAD, SLAKE, PathVQA, and VQA-Med-2019, demonstrate ECSA’s state-of-the-art or highly competitive performance. Specifically, ECSA achieves excellent overall accuracies of 80.15% on VQA-RAD and 85.10% on SLAKE, while also showing strong generalization with 64.57% on PathVQA and 82.23% on VQA-Med-2019. More critically, in continual learning scenarios, the framework achieves a low forgetting rate of just 13.50%, showcasing its significant advantages in knowledge retention. Conclusions: These findings validate the framework’s substantial potential for building robust and evolvable clinical decision support systems. Full article

High-quality image restoration (IR) is a fundamental task in computer vision, aiming to recover a clear image from its degraded version. Prevailing methods typically employ a static inference pipeline, neglecting the spatial variability of image content and degradation, which makes it difficult for them to adaptively handle complex and diverse restoration scenarios. To address this issue, we propose a novel adaptive image restoration framework named Hybrid Convolutional Transformer with Dynamic Prompting (HCTDP). Our approach introduces two key architectural innovations: a Spatially Aware Dynamic Prompt Head Attention (SADPHA) module, which performs fine-grained local restoration by generating spatially variant prompts through real-time analysis of image content and a Gated Skip-Connection (GSC) module that refines multi-scale feature flow using efficient channel attention. To guide the network in generating more visually plausible results, the framework is optimized with a hybrid objective function that combines a pixel-wise L1 loss and a feature-level perceptual loss. Extensive experiments on multiple public benchmarks, including image deraining, dehazing, and denoising, demonstrate that our proposed HCTDP exhibits superior performance in both quantitative and qualitative evaluations, validating the effectiveness of the adaptive restoration framework while utilizing fewer parameters than key competitors. Full article

Historic urban alleys encapsulate cultural identity and collective memory but are increasingly threatened by commercialization and context-insensitive redevelopment. Preserving their authenticity while enhancing environmental resilience requires design strategies that integrate both heritage and ecological values. This study explores the potential of generative artificial intelligence (AI) to support biophilic design in historic alleys, focusing on Daegu, South Korea. Four alley typologies—path, stairs, edge, and node—were identified through fieldwork and analyzed across cognitive, emotional, and physical dimensions of place identity. A Flux-based diffusion model was fine-tuned using low-rank adaptation (LoRA) with site-specific images, while a structured biophilic design prompt (BDP) framework was developed to embed ecological attributes into generative simulations. The outputs were evaluated through perceptual and statistical similarity indices and expert reviews (n = 8). Results showed that LoRA training significantly improved alignment with ground-truth images compared to prompt-only generation, capturing both material realism and symbolic cues. Expert evaluations confirmed the contextual authenticity and biophilic effectiveness of AI-generated designs, revealing typology-specific strengths: the path enhanced spatial legibility and continuity; the stairs supported immersive sequential experiences; the edge transformed rigid boundaries into ecological transitions; and the node reinforced communal symbolism. Emotional identity was more difficult to reproduce, highlighting the need for multimodal and interactive approaches. This study demonstrates that generative AI can serve not only as a visualization tool but also as a methodological platform for participatory design and heritage-sensitive urban regeneration. Future research will expand the dataset and adopt multimodal and dynamic simulation approaches to further generalize and validate the framework across diverse urban contexts. Full article

In the context of deep AI–design integration, traditional methods struggle to translate multi-source requirements into sustainable engineering solutions while balancing innovation with practicality. This study proposes AQTA, an intelligent design framework that integrates Analytic Hierarchy Process (AHP), Quality Function Deployment (QFD), Theory of Inventive Problem Solving (TRIZ), and AI-Generated Content (AIGC) to enable sustainable product development. AQTA employs a four-stage closed-loop process: requirement analysis, contradiction resolution, solution generation, and validation. QFD and AHP quantify user and sustainability requirements to identify key contradictions, TRIZ resolves technical conflicts and stimulates innovative solutions, while AIGC generates eco-efficient visual concepts through prompt engineering. Multi-criteria decision-making supports evaluation and optimization based on environmental and economic indicators. Empirical studies demonstrate that AQTA significantly enhances innovation quality, design efficiency, and sustainability performance. The framework provides a replicable, hybrid ‘theory-driven + AI-generated’ methodology, which is validated through the case study of urban fire trucks, contributing to sustainable manufacturing practices in the intelligent era. Full article

The integration of Large Language Models (LLMs), particularly Visual Language Models (VLMs), into robotics promises more intuitive human–robot interactions; however, challenges remain in efficiently translating high-level commands into precise physical actions. This paper presents a novel architecture for vision-based object manipulation that leverages a VLM’s reasoning capabilities while incorporating symmetry principles to enhance operational efficiency. Implemented on a Yahboom DOFBOT educational robot with a Jetson Nano platform, our system introduces a prompt-based framework that uniquely embeds symmetry-related cues to streamline feature extraction and object detection from visual data. This methodology, which utilizes few-shot learning, enables the VLM to generate more accurate and contextually relevant commands for manipulation tasks by efficiently interpreting the symmetric and asymmetric features of objects. The experimental results in controlled scenarios demonstrate that our symmetry-informed approach significantly improves the robot’s interaction efficiency and decision-making accuracy compared to generic prompting strategies. This work contributes a robust method for integrating fundamental vision principles into modern generative AI workflows for robotics. Furthermore, its implementation on an accessible educational platform shows its potential to simplify complex robotics concepts for engineering education and research. Full article

The accelerating complexity of urban environments has prompted cities to adopt digital technologies that improve efficiency, sustainability, and resilience. Among these, Urban Digital Twins (UDTw) have emerged as transformative tools for real-time representation, simulation, and management of urban systems. This Systematic Literature Review (SLR) examines the integration of Digital Twins (DTw), the Internet of Things (IoT), and Artificial Intelligence (AI) into the Smart City Development (SCD). Following the PSALSAR framework and PRISMA 2020 guidelines, 64 peer-reviewed articles from IEEE Xplore, Association for Computing Machinery (ACM), Scopus, and Web of Science (WoS) digital libraries were analyzed by using bibliometric and thematic methods via the Bibliometrix package in R. The review allowed identifying key technological trends, such as edge–cloud, architectures, 3D immersive visualization, Generative AI (GenAI), and blockchain, and classifies UDTw applications into five domains: traffic management, urban planning, environmental monitoring, energy systems, and public services. Persistent challenges have been also outlined, including semantic interoperability, predictive modeling, data privacy, and impact evaluation. This study synthesizes the current state of the field, by clearly identifying a thematic mapping, and proposes a research agenda to align technical innovation with measurable urban outcomes, offering strategic insights for researchers, policymakers, and planners. Full article

Few-Shot Industrial Anomaly Detection (FSIAD) is an essential yet challenging problem in practical scenarios such as industrial quality inspection. Its objective is to identify previously unseen anomalous regions using only a limited number of normal support images from the same category. Recently, large pre-trained vision-language models (VLMs), such as CLIP, have exhibited remarkable few-shot image-text representation abilities across a range of visual tasks, including anomaly detection. Despite their promise, real-world industrial anomaly datasets often contain noisy labels, which can degrade prompt learning and detection performance. In this paper, we propose AnomalyNLP, a new Noisy-Label Prompt Learning approach designed to tackle the challenge of few-shot anomaly detection. This framework offers a simple and efficient approach that leverages the expressive representations and precise alignment capabilities of VLMs for industrial anomaly detection. First, we design a Noisy-Label Prompt Learning (NLPL) strategy. This strategy utilizes feature learning principles to suppress the influence of noisy samples via Mean Absolute Error (MAE) loss, thereby improving the signal-to-noise ratio and enhancing overall model robustness. Furthermore, we introduce a prompt-driven optimal transport feature purification method to accurately partition datasets into clean and noisy subsets. For both image-level and pixel-level anomaly detection, AnomalyNLP achieves state-of-the-art performance across various few-shot settings on the MVTecAD and VisA public datasets. Qualitative and quantitative results on two datasets demonstrate that our method achieves the largest average AUC improvement over baseline methods across 1-, 2-, and 4-shot settings, with gains of up to 10.60%, 10.11%, and 9.55% in practical anomaly detection scenarios. Full article

Background: Macular edema (ME) secondary to retinal vein occlusion (RVO) is a significant cause of vision impairment. Many patients show suboptimal responses to anti-vascular endothelial growth factor (anti-VEGF) monotherapy, prompting the exploration of alternative treatments. Faricimab is a bispecific antibody that targets VEGF-A and angiopoietin-2. We report 9-month real-world outcomes of switching to faricimab in therapy-resistant RVO-associated ME. Methods: In this retrospective study at a single tertiary center, patients with persistent or recurrent ME despite prior treatments (ranibizumab, aflibercept, or dexamethasone implant) were switched to faricimab. All eyes received a loading phase of four monthly faricimab injections, followed by a treat-and-extend regimen individualized per response. Key outcomes included best-corrected visual acuity (BCVA, logMAR), the central subfield thickness (CST, μm), and the intraretinal fluid (IRF) status on optical coherence tomography, assessed from the baseline (month 0, mo0) through the loading phase (mo1–mo3) and at month 9 (mo9). Results: Nineteen eyes (19 patients, mean age 64.8 years) were analyzed. The median BCVA improved from 0.20 to 0.00 logMAR by month 3 (p < 0.01) and was maintained at month 9. The median CST decreased from 325 μm at the baseline to 285 μm at month 3 (p < 0.01) and remained at 285 μm at month 9. IRF was present in 100% of eyes at the baseline, 26% at month 3, and 26% at month 9 (p < 0.01 for the baseline vs. month 9). Among eyes previously on anti-VEGF therapy (n = 14), the median treatment interval increased from 45.50 days at the baseline to 56.50 days at month 9 (p = 0.01; δ = 0.86). No intraocular inflammation or other adverse events were observed in this cohort over nine months. Conclusions: In this retrospective series, switching to faricimab was associated with improvements in vision and retinal anatomy that were maintained over 9 months; injection intervals were extended in a subset of eyes. These exploratory findings warrant confirmation in larger, controlled studies to define long-term effectiveness, safety, and dosing strategies. Full article

The term heart disease refers to a wide range of conditions that impact the heart and blood vessels. It continues to be a major global cause of morbidity and mortality. The narrowing or blockage of blood vessels, which can result in major medical events like heart attacks, angina (chest pain) or strokes, is a common issue linked to heart disease. In order to lower the risk of serious complications and facilitate prompt medical intervention, early diagnosis and prediction are essential. This study developed predictive models that can precisely identify people at risk by applying a variety of machine learning algorithms to a structured dataset on heart disease. Blood pressure, cholesterol, age, gender, and other health-related indicators are among the 13 essential characteristics that make up the dataset. Numerous machine learning models such as Naïve Bayes, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree, Random Forest, and others were trained using these features. Using the RapidMiner platform, which offered a visual environment for data preprocessing, model training, and performance analysis, all models were created and assessed. The best-performing model was the Naïve Bayes classifier which achieved an impressive accuracy rate of 90% after extensive testing and comparison of performance metrics like accuracy precision and recall. This outcome shows how well the model can predict heart disease in actual clinical settings. By supporting individualized health recommendations, enabling early diagnosis, and facilitating timely treatment, the effective application of such models can significantly benefit patients and healthcare professionals. Furthermore, heart disease incidence can be considerably decreased by identifying and addressing modifiable risk factors such as high blood pressure, elevated cholesterol, smoking, diabetes, and physical inactivity. In summary, machine learning has the potential to improve the identification and treatment of heart-related disorders. This study highlights the value of data-driven methods in healthcare and indicates that incorporating predictive models into standard medical procedures may enhance patient outcomes, lower healthcare expenses, and improve public health administration. Full article

Weakly Supervised Video Anomaly Detection (WSVAD) is a critical task in computer vision. It aims to localize and recognize abnormal behaviors using only video-level labels. Without frame-level annotations, it becomes significantly challenging to model temporal dependencies. Given the diversity of abnormal events, it is also difficult to model semantic representations. Recently, the cross-modal pre-trained model Contrastive Language-Image Pretraining (CLIP) has shown a strong ability to align visual and textual information. This provides new opportunities for video anomaly detection. Inspired by CLIP, WSVAD-CLIP is proposed as a framework that uses its cross-modal knowledge to bridge the semantic gap between text and vision. First, the Axial-Graph (AG) Module is introduced. It combines an Axial Transformer and Lite Graph Attention Networks (LiteGAT) to capture global temporal structures and local abnormal correlations. Second, a Text Prompt mechanism is designed. It fuses a learnable prompt with a knowledge-enhanced prompt to improve the semantic expressiveness of category embeddings. Third, the Abnormal Visual-Guided Text Prompt (AVGTP) mechanism is proposed to aggregate anomalous visual context for adaptively refining textual representations. Extensive experiments on UCF-Crime and XD-Violence datasets show that WSVAD-CLIP notably outperforms existing methods in coarse-grained anomaly detection. It also achieves superior performance in fine-grained anomaly recognition tasks, validating its effectiveness and generalizability. Full article

Smart contracts are central to blockchain ecosystems, yet their development remains technically demanding, error-prone, and tied to platform-specific programming languages. This paper introduces SCEditor-Web, a web-based modeling environment that combines model-driven engineering (MDE) with generative artificial intelligence (Gen-AI) to simplify contract design and code generation. Developers specify the structural and behavioral aspects of smart contracts through a domain-specific visual language grounded in a formal metamodel. The resulting contract model is exported as structured JSON and transformed into executable, platform-specific code using large language models (LLMs) guided by a tailored prompt engineering process. A prototype implementation was evaluated on Solidity contracts as a proof of concept, using representative use cases. Experiments with state-of-the-art LLMs assessed the generated contracts for compilability, semantic alignment with the contract model, and overall code quality. Results indicate that the visual-to-code workflow reduces manual effort, mitigates common programming errors, and supports developers with varying levels of expertise. The contributions include an abstract smart contract metamodel, a structured prompt generation pipeline, and a web-based platform that bridges high-level modeling with practical multi-language code synthesis. Together, these elements advance the integration of MDE and LLMs, demonstrating a step toward more accessible and reliable smart contract engineering. Full article

Farmland ecosystems present complex pest–predator co-occurrence patterns, posing significant challenges for image-based multitarget recognition and ecological modeling in sensor-driven computer vision tasks. To address these issues, this study introduces a tree-guided Transformer framework enhanced with a knowledge-augmented co-attention mechanism, enabling effective feature extraction from sensor-acquired images. A hierarchical ecological taxonomy (Phylum–Family Species) guides prompt-driven semantic reasoning, while an ecological knowledge graph enriches visual representations by embedding co-occurrence priors. A multimodal dataset containing 60 pest and predator categories with annotated images and semantic descriptions was constructed for evaluation. Experimental results demonstrate that the proposed method achieves 90.4% precision, 86.7% recall, and 88.5% F1-score in image classification, along with 82.3% hierarchical accuracy. In detection tasks, it attains 91.6% precision and 86.3% mAP@50, with 80.5% co-occurrence accuracy. For hierarchical reasoning and knowledge-enhanced tasks, F1-scores reach 88.5% and 89.7%, respectively. These results highlight the framework’s strong capability in extracting structured, semantically aligned image features under real-world sensor conditions, offering an interpretable and generalizable approach for intelligent agricultural monitoring. Full article