Search Results (103)

The integration of advanced computer vision and artificial intelligence (AI) techniques into collaborative robotic systems holds the potential to revolutionize human–robot interaction, productivity, and safety. Despite substantial research activity, a systematic synthesis of how vision and AI are jointly enabling context-aware, adaptive cobot capabilities across perception, planning, and decision-making remains lacking (especially in recent years). Addressing this gap, our review unifies the latest advances in visual recognition, deep learning, and semantic mapping within a structured taxonomy tailored to collaborative robotics. We examine foundational technologies such as object detection, human pose estimation, and environmental modeling, as well as emerging trends including multimodal sensor fusion, explainable AI, and ethically guided autonomy. Unlike prior surveys that focus narrowly on either vision or AI, this review uniquely analyzes their integrated use for real-world human–robot collaboration. Highlighting industrial and service applications, we distill the best practices, identify critical challenges, and present key performance metrics to guide future research. We conclude by proposing strategic directions—from scalable training methods to interoperability standards—to foster safe, robust, and proactive human–robot partnerships in the years ahead. Full article

Three-dimensional (3D) reconstruction has become a fundamental technology in applications ranging from cultural heritage preservation and robotics to forensics and virtual reality. As these applications grow in complexity and realism, the quality of the reconstructed models becomes increasingly critical. Among the many factors that influence reconstruction accuracy, the lighting conditions at capture time remain one of the most influential, yet widely neglected, variables. This review provides a comprehensive survey of classical and modern 3D reconstruction techniques, including Structure from Motion (SfM), Multi-View Stereo (MVS), Photometric Stereo, and recent neural rendering approaches such as Neural Radiance Fields (NeRFs) and 3D Gaussian Splatting (3DGS), while critically evaluating their performance under varying illumination conditions. We describe how lighting-induced artifacts such as shadows, reflections, and exposure imbalances compromise the reconstruction quality and how different approaches attempt to mitigate these effects. Furthermore, we uncover fundamental gaps in current research, including the lack of standardized lighting-aware benchmarks and the limited robustness of state-of-the-art algorithms in uncontrolled environments. By synthesizing knowledge across fields, this review aims to gain a deeper understanding of the interplay between lighting and reconstruction and provides research directions for the future that emphasize the need for adaptive, lighting-robust solutions in 3D vision systems. Full article

From the point of view of the intelligent operation and maintenance of high-speed train tracks, this paper examines the research status of high-speed train rail damage detection technology in the field of high-speed train track operation and maintenance detection in recent years, summarizes the damage detection methods for high-speed trains, and compares and analyzes different detection technologies and application research results. The analysis results show that the detection methods for high-speed train rail damage mainly focus on the research and application of non-destructive testing technology and methods, as well as testing platform equipment. Detection platforms and equipment include a new type of vortex meter, integrated track recording vehicles, laser rangefinders, thermal sensors, laser vision systems, LiDAR, new ultrasonic detectors, rail detection vehicles, rail detection robots, laser on-board rail detection systems, track recorders, self-moving trolleys, etc. The main research and application methods include electromagnetic detection, optical detection, ultrasonic guided wave detection, acoustic emission detection, ray detection, vortex detection, and vibration detection. In recent years, the most widely studied and applied methods have been rail detection based on LiDAR detection, ultrasonic detection, eddy current detection, and optical detection. The most important optical detection method is machine vision detection. Ultrasonic detection can detect internal damage of the rail. LiDAR detection can detect dirt around the rail and the surface, but the cost of this kind of equipment is very high. And the application cost is also very high. In the future, for high-speed railway rail damage detection, the damage standards must be followed first. In terms of rail geometric parameters, the domestic standard (TB 10754-2018) requires a gauge deviation of ±1 mm, a track direction deviation of 0.3 mm/10 m, and a height deviation of 0.5 mm/10 m, and some indicators are stricter than European standard EN-13848. In terms of damage detection, domestic flaw detection vehicles have achieved millimeter-level accuracy in crack detection in rail heads, rail waists, and other parts, with a damage detection rate of over 85%. The accuracy of identifying track components by the drone detection system is 93.6%, and the identification rate of potential safety hazards is 81.8%. There is a certain gap with international standards, and standards such as EN 13848 have stricter requirements for testing cycles and data storage, especially in quantifying damage detection requirements, real-time damage data, and safety, which will be the key research and development contents and directions in the future. Full article

Historically, pharmacological interventions aimed at platelets have targeted their canonical hemostatic and thrombotic roles. The therapeutic vision, however, has minimally embraced alternate mechanisms by which anucleate platelets, their parent cells, megakaryocytes, and cellular derivatives may be utilized to yield novel and effective therapies. Platelets contain storage granules rich in a wide variety of proteins, chemicals, growth factors, and lipid particles that can modulate the fate and activity of diverse cell types, and impact diseases not previously thought to have a platelet component. In this article, we will address unconventional platelet contributions to health and disease development. Recent studies indicate extensive platelet roles in neurodegeneration, insulin secretion, and bone marrow fibrosis, along with a recognition of platelets as immune cells in their own right, partially based on the presence of surface MHC, Toll-like receptors, and stored immunomodulatory molecules. Recent technological advances have produced iPS-derived gene-editable megakaryocytes (MKs) that have been differentiated to clinical-grade platelets for transfusion; however, such successes are still rare. Continued improvements in the standardization of cell isolation, iPS differentiation protocols, technology for the utilization of platelet derivatives, and platelet Omics will expand our understanding of underlying platelet and MK heterogeneity and direct novel therapeutic applications. Furthermore, additional roles for these cells as microniche sensors that monitor systemic pathology by endocytosing shed particles as they circulate through the vasculature will be explored. Taken together, novel insights into the many exciting potential uses of platelets outside of their canonical roles are on the horizon, and continued amelioration of existing protocols and enhanced understanding of communication pathways between platelets and specific cells will help expand opportunities for platelet-related clinical trials to yield improved health outcomes. Full article

Some estimates suggest that one in seven good readers and the majority of children with reading difficulties suffer from oculomotor dysfunction (OMD), an umbrella term for abnormalities in comfortable and accurate fixations, pursuits, and saccades. However, national vision evaluation programs worldwide are often limited to distance visual acuity (dVA), not testing for OMD despite its high prevalence and the ease of detecting it in brief optometric evaluations. We hypothesized that reading acquisition is dependent on good oculomotor functions, and therefore inadequate oculomotor control will be associated with reading difficulties. We retrospectively examined and compared oculomotor evaluations (using DEM and NSUCO) and reading assessments (using standardized national reading norms) of a normative class (28 first graders (6–7 yr. olds)) that were independently obtained while blind to the other assessment. Better oculomotor performance as estimated by DEM was associated with better reading performance, and almost a third (29.6%) of the children were categorized by DEM as having OMD-related difficulties. Control analysis revealed dVA was not positively associated with reading performance. Linear regression analyses further corroborated these findings. Since this study is based on a small cohort and since there are studies suggesting that DEM may actually reflect visual processing speed or cognitive factors rather than oculomotor function, replications are needed to substantiate the direct contribution of oculomotor functions to reading acquisition. Young children struggling with reading may benefit from a comprehensive visual evaluation, including oculomotor testing, to provide a more thorough assessment of their learning-related difficulties. Full article

The absence of standardized evaluation methodologies for single-layer dimensional accuracy significantly hinders the broader implementation of direct ink writing (DIW) technology. Addressing the critical need for precision non-contact assessment in DIW fabrication, this study develops a novel machine vision-based framework for dimensional accuracy evaluation. The methodology encompasses three key phases: (1) establishment of an optimized hardware configuration with integrated image processing algorithms; (2) comprehensive investigation of camera calibration protocols, advanced image preprocessing techniques, and high-precision contour extraction methods; and (3) development of an iterative closest point (ICP) algorithm-enhanced evaluation system. The experimental results demonstrate that our machine vision system achieves 0.04 mm × 0.04 mm spatial resolution with the ICP convergence threshold optimized to 0.001 mm. The proposed method shows an 80% improvement in measurement accuracy (0.001 mm) compared to conventional approaches. Process parameter optimization experiments validated the system’s effectiveness, showing at least 76.3% enhancement in printed layer dimensional accuracy. This non-contact evaluation solution establishes a robust framework for quantitative quality control in DIW applications, providing critical insights for process optimization and standardization efforts in additive manufacturing. Full article

Background: Artificial intelligence (AI) is rapidly transforming thoracic surgery by enhancing diagnostic accuracy, surgical precision, intraoperative guidance, and postoperative management. AI-driven technologies, including machine learning (ML), deep learning, computer vision, and robotic-assisted surgery, have the potential to optimize clinical workflows and improve patient outcomes. However, challenges such as data integration, ethical concerns, and regulatory barriers must be addressed to ensure AI’s safe and effective implementation. This review aims to analyze the current applications, benefits, limitations, and future directions of AI in thoracic surgery. Methods: This review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A comprehensive literature search was performed using PubMed, Scopus, Web of Science, and Cochrane Library for studies published up to January 2025. Relevant articles were selected based on predefined inclusion and exclusion criteria, focusing on AI applications in thoracic surgery, including diagnostics, robotic-assisted surgery, intraoperative guidance, and postoperative care. A risk of bias assessment was conducted using the Cochrane Risk of Bias Tool and ROBINS-I for non-randomized studies. Results: Out of 279 identified studies, 36 met the inclusion criteria for qualitative synthesis, highlighting AI’s growing role in diagnostic accuracy, surgical precision, intraoperative guidance, and postoperative care in thoracic surgery. AI-driven imaging analysis and radiomics have improved pulmonary nodule detection, lung cancer classification, and lymph node metastasis prediction, while robotic-assisted thoracic surgery (RATS) has enhanced surgical accuracy, reduced operative times, and improved recovery rates. Intraoperatively, AI-powered image-guided navigation, augmented reality (AR), and real-time decision-support systems have optimized surgical planning and safety. Postoperatively, AI-driven predictive models and wearable monitoring devices have enabled early complication detection and improved patient follow-up. However, challenges remain, including algorithmic biases, a lack of multicenter validation, high implementation costs, and ethical concerns regarding data security and clinical accountability. Despite these limitations, AI has shown significant potential to enhance surgical outcomes, requiring further research and standardized validation for widespread adoption. Conclusions: AI is poised to revolutionize thoracic surgery by enhancing decision-making, improving patient outcomes, and optimizing surgical workflows. However, widespread adoption requires addressing key limitations through multicenter validation studies, standardized AI frameworks, and ethical AI governance. Future research should focus on digital twin technology, federated learning, and explainable AI (XAI) to improve AI interpretability, reliability, and accessibility. With continued advancements and responsible integration, AI will play a pivotal role in shaping the next generation of precision thoracic surgery. Full article

Background: Cerebral vasospasm after craniotomy tumor (CVACT) is a rare complication that can occur following tumor craniotomy and significantly affects the outcome of patients. Unfortunately, it is not well understood, leading to delayed and ineffective management. This study aims to investigate CVACT by examining the factors contributing to its occurrence, its underlying mechanisms, diagnostic approaches, management strategies, and outcomes. The goal is to identify the characteristics and risk factors associated with CVACT, its clinical symptoms, diagnostic methods, management options, and potential outcomes. Methods: A systematic search used relevant keywords to identify cases of “cerebral vasospasm” after tumor resection in PubMed and Science Direct databases. Relevant cross-references were added by manually searching the references of all retrieved articles. Result: We included 60 inclusion patients from 14 case reports and 13 case series with 33 (55%) females and 27 (45%) males with a mean age of 44.05 ± 16.8 years. The most common tumors were pituitary adenomas, which were found in 22 (36.66%), the most common tumor location was the middle cranial fossa (75%), and the most common surgery technique used was transsphenoidal surgery (50%). Most of those who experience vasospasm have a craniotomy with the TSS technique (50%) with complications of intraoperative bleeding. The range of onset of VS symptoms postoperatively was 0–30 days (mean 6.59 d). The symptoms included asymptomatic, headache, loss of vision, hemiparesis, diplopia, etc. The vascular involvement was mainly anterior circulation (78.33%). The diagnostic tools most commonly used were angiography and transcranial doppler (TCD). The most common management of VS from the included studies was pharmacology. The survival rate was 61.66%. We found the tumor location and vascular-affected vasospasm were significantly correlated with mortality rates: p = 0.015 and p = 0.02. Conclusions: Cerebral vasospasm after craniotomy tumor removal (CVACT) frequently arises in tumors situated in the medial cranial fossa, predominantly pituitary adenomas and meningiomas. The minimally invasive surgical approach of TSS may contribute to the mechanism of CVACT incidence. The existence of preoperative vascular pathology, as encasement or narrowing, appears to be a predictor alongside the incidence of intra- or postoperative hemorrhage. The vascular structures most susceptible to vasospasm are located in the anterior circulation of the Willis circle, which appears to correlate with the vascular problems that typically undergo preoperative encasement of the internal carotid artery (ICA). The most reliable and real time diagnostic instrument employed is TCD, while imaging continues to be the gold standard. Nimodipine treatment continues to be a viable therapeutic option that can enhance patient outcomes. Full article

The generation of images from scene graphs is an important area in computer vision, where structured object relationships are used to create detailed visual representations. While recent methods, such as generative adversarial networks (GANs), transformers, and diffusion models, have improved image quality, they still face challenges, like scalability issues, difficulty in generating complex scenes, and a lack of clear evaluation standards. Despite various approaches being proposed, there is still no unified way to compare their effectiveness, making it difficult to determine the best techniques for real-world applications. This review provides a detailed assessment of scene-graph-based image generation by organizing current methods into different categories and examining their advantages and limitations. We also discuss the datasets used for training, the evaluation measures applied to assess model performance, and the key challenges that remain, such as ensuring consistency in scene structure, handling object interactions, and reducing computational costs. Finally, we outline future directions in this field, highlighting the need for more efficient, scalable, and semantically accurate models. This review serves as a useful reference for researchers and practitioners, helping them understand current trends and identify areas for further improvement in scene-graph-based image generation. Full article

This paper presents the experimental evaluation and analyzes the results of the first edition of the pedestrian attribute recognition (PAR) contest, the international competition which focused on smart visual sensors based on multi-task computer vision methods for the recognition of binary and multi-class pedestrian attributes from images. The participant teams designed intelligent sensors based on vision-language models, transformers and convolutional neural networks that address the multi-label recognition problem leveraging task interdependencies to enhance model efficiency and effectiveness. Participants were provided with the MIVIA PAR Dataset, containing 105,244 annotated pedestrian images for training and validation, and their methods were evaluated on a private test set of over 20,000 images. In the paper, we analyze the smart visual sensors proposed by the participating teams, examining the results in terms of accuracy, standard deviation and confusion matrices and highlighting the correlations between design choices and performance. The results of this experimental evaluation, conducted in a challenging and realistic framework, suggest possible directions for future improvements in these smart sensors that are thoroughly discussed in the paper. Full article

The paper presents business models (BMs) for blockchain-based businesses. The paper is a study of IT-aligned BMs categorized by the concepts and possibilities of blockchain business applications. The research aimed to recognize and analyze the extent and directions in which blockchain architectures influence the means of conducting businesses. A set of almost 40,000 decentralized applications is examined to justify the rationale behind the presented analysis. This is an argumentative study that uses the design-oriented approach, as it is suitable for addressing real-world problems, like analyzing business models, while ensuring that artifacts are created and evaluated under methodological standards. Firstly, the concept of a business model is analyzed. Then, a theoretical analysis of different business models is made to identify the ones that are well aligned with the decentralized vision of business and the ones that are obsolete or inoperative from the blockchain business-conducting perspective. In the end, the outcome is applied to examples of existing business startups. Fifteen identified BMs in 7 business sector groups are recognized and 55 cases are detected. Full article

Diabetic retinopathy (DR) is one of the most prevalent complications of diabetes, affecting nearly one-third of patients with diabetes mellitus and remaining a leading cause of blindness worldwide. Among the various diabetes-induced complications, DR is of particular importance due to its direct impact on vision and the irreversible damage to the retina. DR is characterized by multiple pathological processes, primarily a hyperglycemia-induced inflammatory response and oxidative stress. Current gold standard therapies, such as anti-VEGF injections and photocoagulation, have shown efficacy in slowing disease progression. However, challenges such as drug resistance, partial therapeutic responses, and the reliance on direct eye injections—which often result in low patient compliance—remain unresolved. This review provides a comprehensive overview of the underlying molecular mechanisms in DR, the current therapies, and their unmet needs for DR treatment. Additionally, emerging therapeutic strategies for improving DR treatment outcomes are discussed. Full article

Domain generalization (DG) has become a pivotal research area in machine learning, focusing on equipping models with the ability to generalize effectively to unseen test domains that differ from the training distribution. This capability is crucial, as real-world data frequently exhibit domain shifts that violate the assumption of independent and identically distributed (i.i.d.) data, resulting in significant declines in model performance. Among the various strategies to address domain generalization, data augmentation has garnered substantial attention as an effective approach for mitigating domain shifts and improving model robustness. In this survey, we examine the role of data augmentation in domain generalization, offering a comprehensive overview of its methods, applications, and challenges. We present a detailed taxonomy of data augmentation techniques, categorized along three dimensions: scope, nature, and training dependency. Additionally, we provide a comparative analysis of key methods, highlighting their strengths and limitations. Finally, we explore the domain-specific applications of data augmentation and analyze their effectiveness in enhancing generalization across various real-world tasks, including computer vision, NLP, speech, and robotics. We conclude by examining key challenges—such as computational cost and augmentation overfitting—and outline promising research directions, with a focus on advancing cross-modal augmentation techniques and developing standardized evaluation benchmarks. Full article

A positioning method for a roadheader based on fiber-optic strap-down inertial navigation and binocular vision is proposed to address the issue of low measurement accuracy of the mining machine position caused by single-sensor methods in underground coal mines. A vision system for the mining machine position is constructed based on the four-point target fixed on the body of the roadheader, and the position and attitude information of the roadheader are obtained by combining the inertial navigation on the body. To deal with the problem of position detection inaccuracies caused by the accumulation of errors in inertial navigation measurements over time and disturbances from body vibrations to the combined positioning system, an Adaptive Derivative Unscented Kalman Filtering (ADUKF) algorithm is proposed, which can suppress the impact of process variance uncertainties on the filtering. The simulation results demonstrate that, compared to the Unscented Kalman Filtering algorithm, the position errors in the three directions are reduced by 20%, 20.68%, and 28.57%, respectively. Experiments demonstrate that the method can compensate for the limitations of single-measurement methods and meet the positioning accuracy requirements for underground mining standards. Full article

Background: Artificial intelligence (AI) has emerged as a transformative technology in healthcare, with its integration into cardiac surgery offering significant advancements in precision, efficiency, and patient outcomes. However, a comprehensive understanding of AI’s applications, benefits, challenges, and future directions in cardiac surgery is needed to inform its safe and effective implementation. Methods: A systematic review was conducted following PRISMA guidelines. Literature searches were performed in PubMed, Scopus, Cochrane Library, Google Scholar, and Web of Science, covering publications from January 2000 to November 2024. Studies focusing on AI applications in cardiac surgery, including risk stratification, surgical planning, intraoperative guidance, and postoperative management, were included. Data extraction and quality assessment were conducted using standardized tools, and findings were synthesized narratively. Results: A total of 121 studies were included in this review. AI demonstrated superior predictive capabilities in risk stratification, with machine learning models outperforming traditional scoring systems in mortality and complication prediction. Robotic-assisted systems enhanced surgical precision and minimized trauma, while computer vision and augmented cognition improved intraoperative guidance. Postoperative AI applications showed potential in predicting complications, supporting patient monitoring, and reducing healthcare costs. However, challenges such as data quality, validation, ethical considerations, and integration into clinical workflows remain significant barriers to widespread adoption. Conclusions: AI has the potential to revolutionize cardiac surgery by enhancing decision making, surgical accuracy, and patient outcomes. Addressing limitations related to data quality, bias, validation, and regulatory frameworks is essential for its safe and effective implementation. Future research should focus on interdisciplinary collaboration, robust testing, and the development of ethical and transparent AI systems to ensure equitable and sustainable advancements in cardiac surgery. Full article