Search Results (595)

Background: Radiation exposure in the cardiac catheterization laboratory remains a critical occupational hazard for interventional cardiologists and staff, contributing to orthopedic injuries, cataracts, and malignancy. In parallel, procedural complexity continues to increase, demanding both precision and safety. Robotic-assisted percutaneous coronary intervention (R-PCI), alongside advanced shielding systems and imaging integration, has emerged as a transformative strategy to minimize radiation and enhance operator ergonomics. Objective: This state-of-the-art review synthesizes the current clinical evidence and technological advances that support a radiation-reduction paradigm in percutaneous coronary intervention (PCI), with a particular focus on the role of R-PCI platforms, procedural modifications, and emerging shielding technologies. Methods: We reviewed published clinical trials, registries, and experimental studies evaluating robotic PCI platforms, contrast and radiation dose metrics, ergonomic implications, procedural efficiency, and radiation shielding systems. Emphasis was given to the integration of CT-based imaging (coronary computed tomography angiography—CCTA, fractional flow reserve computed tomography—FFR-CT) and low-dose acquisition protocols. Results: R-PCI demonstrated technical success rates of 81–100% and clinical success rates up to 100% in both standard and complex lesions, with significant reductions in operator radiation exposure (up to 95%) and procedural ergonomic burden. Advanced shielding technologies offer radiation dose reductions ranging from 86% to nearly 100%, while integration of (CCTA), (FFR-CT), and Artificial Intelligence (AI) -assisted procedural mapping facilitates further fluoroscopy minimization. Robotic workflows, however, remain limited by lack of device compatibility, absence of haptic feedback, and incomplete integration of physiology and imaging tools. Conclusions: R-PCI, in combination with shielding technologies and imaging integration, marks a shift towards safer, radiation-minimizing interventional strategies. This transition reflects not only a technical evolution but a philosophical redefinition of safety, precision, and sustainability in modern interventional cardiology. Full article

Matching drone images to satellite reference images is a critical step for achieving UAV self-localization. Existing drone visual localization datasets mainly focus on target localization, where each drone image is paired with a corresponding satellite image slice, typically with identical coverage. However, this one-to-one approach does not reflect real-world UAV self-localization needs as it cannot guarantee exact matches between drone images and satellite tiles nor reliably identify the correct satellite slice. To bridge this gap, we propose a one-to-many matching method between drone images and satellite reference tiles. First, we enhance the UAV-VisLoc dataset, making it the first in the field tailored for one-to-many imperfect matching in UAV self-localization. Second, we introduce a novel loss function, Incomp-NPair Loss, which better reflects real-world imperfect matching scenarios than traditional methods. Finally, to address challenges such as limited dataset size, training instability, and large-scale differences between drone images and satellite tiles, we adopt a Vision Transformer (ViT) baseline and integrate CNN-extracted features into its patch embedding layer. Full article

Citrus diseases severely impact fruit yield and quality. To facilitate in-field, non-destructive, and rapid detection of citrus leaf physiological and pathological conditions, this study proposes a classification method for citrus leaf physiological and pathological statuses that integrates visible/near-infrared multispectral technology with deep learning. First, a handheld spectrometer was employed to acquire spectral images of five sample categories—Healthy, Huanglongbing, Yellow Vein Disease, Magnesium Deficiency and Manganese Deficiency. Mean spectral data were extracted from regions of interest within the 350–2500 nm wavelength range, and various preprocessing techniques were evaluated. The Standard Normal Variate (SNV) transformation, which demonstrated optimal performance, was selected for data preprocessing. Next, we innovatively introduced an adaptive spectral positional encoding mechanism into the Transformer framework. A lightweight, learnable network dynamically optimizes positional biases, yielding the ASTransformer (Adaptive Spectral Transformer) model, which more effectively captures complex dependencies among spectral features and identifies critical wavelength bands, thereby significantly enhancing the model’s adaptive representation of discriminative bands. Finally, the preprocessed spectra were fed into three deep learning architectures (1D-CNN, 1D-ResNet, and ASTransformer) for comparative evaluation. The results indicate that ASTransformer achieves the best classification performance: an overall accuracy of 97.7%, underscoring its excellent global classification capability; a Macro Average of 97.5%, reflecting balanced performance across categories; a Weighted Average of 97.8%, indicating superior performance in classes with larger sample sizes; an average precision of 97.5%, demonstrating high predictive accuracy; an average recall of 97.7%, showing effective detection of most affected samples; and an average F1-score of 97.6%, confirming a well-balanced trade-off between precision and recall. Furthermore, interpretability analysis via Integrated Gradients quantitatively assesses the contribution of each wavelength to the classification decisions. These findings validate the feasibility of combining a handheld spectrometer with the ASTransformer model for effective citrus leaf physiological and pathological detection, enabling efficient classification and feature visualization, and offer a valuable reference for disease detection of physiological and pathological conditions in other fruit crops. Full article

Droughts are becoming more frequent, severe, and impactful across the globe. Agroecosystems, which are human-made ecosystems with high water demand that provide essential ecosystem services, are vulnerable to extreme droughts. Although water use efficiency in agriculture has increased in rec ent decades, drought management should be based on long-term, proactive strategies rather than crisis management. The AgrHyS network of sites in French Brittany collects high-resolution soil moisture data from agronomic stations and catchments to improve understanding of temporal soil moisture dynamics and enhance water use efficiency. Frequent mapping of soil moisture and plant water stress is crucial for assessing water stress risk in the context of global warming. Although satellite remote sensing provides reliable, periodic global data on surface soil moisture, it does so at a very coarse spatial resolution. The intrinsic spatial heterogeneity of surface soil moisture requires a higher spatial resolution in order to address upcoming challenges on a local scale. Drones are an excellent tool for upscaling point measurements to catchment level using different onboard cameras. In this study, we evaluated the potential of multispectral images, thermal images and LiDAR data captured in several concurrent drone flights for high-resolution mapping of soil moisture spatial variability, using in situ point measurements of soil water content and plant water stress in both agricultural areas and natural ecosystems. Statistical models were fitted to map soil water content in two areas: a natural marshland and a grassland-covered agricultural field. Our results demonstrate the statistical significance of topography, land surface temperature and red band reflectance in the natural area for retrieving soil water content. In contrast, the grasslands were best predicted by the transformed normalised difference vegetation index (TNDVI). Full article

The Zhag Cave in western Tibet, dated to the 11th to 12th centuries, features four walls fully adorned with images of the Thousand Buddhas of the Bhadrakalpa (Fortunate Aeon). According to the Tibetan inscriptions, the arrangement of the Thousand Buddhas creates a circumambulatory space for worship and confession, enabling practitioners to purify their sins. Four aspects of the Zhag Cave are comparable to those of Dunhuang. First, among the inscriptions is the Pratītyasamutpāda-gāthā, elaborated in the Śālistamba Sūtra, the Tibetan manuscripts of which have been unearthed in both western Tibet and Dunhuang. Second, the way of depicting Thousand Buddhas on four walls inside the cave could be found in earlier caves from the 5th to 6th centuries at Dunhuang. Third, the specific practice of only depicting the Bhadrakalpa Thousand Buddhas on the walls parallels similar caves from the mid-10th to early 13th centuries at Dunhuang. Fourth, the motifs depicted along the wall edges correspond with the prevalent themes found in the Bhadrakalpa Thousand Buddhas transformation tableaux during the 9th to 13th centuries, reflecting the apogee of Bhadrakalpa Thousand Buddhas devotion. These connections prompt us to think about the ways in which Western Tibet was part of the Silk Road network. I argue that this shared iconographic and ritual framework embodies the intertwined religious practices of the Dharma-ending Age (Mofa 末法) thought and Buddhist revival movements along the Silk Road, explaining these complex interconnections between the Zhag Cave and the Dunhuang relics within the broader context of religious beliefs and socio-cultural patterns. Full article

Background: In recent decades, dermatological diagnostics have undergone a profound transformation, driven by the integration of new technologies alongside traditional methods. Classic techniques such as the Tzanck smear, potassium hydroxide (KOH) preparation, and Wood’s lamp examination remain fundamental in everyday clinical practice due to their simplicity, speed, and accessibility. At the same time, the development of non-invasive imaging technologies and the application of artificial intelligence (AI) have opened new frontiers in the early detection and monitoring of both neoplastic and inflammatory skin diseases. Methods: This review aims to provide a comprehensive overview of how conventional and emerging diagnostic tools can be integrated into dermatologic practice. Results: We examined a broad spectrum of diagnostic methods currently used in dermatology, ranging from traditional techniques to advanced approaches such as digital dermoscopy, reflectance confocal microscopy (RCM), optical coherence tomography (OCT), line-field confocal OCT (LC-OCT), 3D total body imaging systems with AI integration, mobile applications, electrical impedance spectroscopy (EIS), and multispectral imaging. Each method is discussed in terms of diagnostic accuracy, clinical applications, and potential limitations. While traditional methods continue to play a crucial role—especially in resource-limited settings or for immediate bedside decision-making—modern tools significantly enhance diagnostic precision. Dermoscopy and its digital evolution have improved the accuracy of melanoma and basal cell carcinoma detection. RCM and LC-OCT allow near-histological visualization of skin structures, reducing the need for invasive procedures. AI-powered platforms support lesion tracking and risk stratification, though their routine implementation requires further clinical validation and regulatory oversight. Tools like EIS and multispectral imaging may offer additional value in diagnostically challenging cases. An effective diagnostic approach in dermatology should rely on a thoughtful combination of methods, selected based on clinical suspicion and guided by Bayesian reasoning. Conclusions: Rather than replacing traditional tools, advanced technologies should complement them—optimizing diagnostic accuracy, improving patient outcomes, and supporting more individualized, evidence-based care. Full article

Climate change has intensified the frequency and severity of forest disturbances globally, including windthrow, which poses substantial risks for both forest productivity and ecosystem stability. Rapid and precise assessment of wind-induced tree damage is essential for effective management, yet many injuries remain visually undetectable in the early stages. This study employed drone-based multispectral imaging and a simulated wind stress experiment (static pulling) on Norway spruce (Picea abies (L.) Karst.) to investigate the detectability of physiological and structural changes over four years. Multispectral data were collected at multiple time points (2023–2024), and a suite of vegetation indices (the Normalised Difference Vegetation Index (NDVI), the Structure Insensitive Pigment Index (SIPI), the Difference Vegetation Index (DVI), and Red Edge-based indices) were calculated and analysed using mixed-effects models. Our results demonstrate that trees subjected to mechanical bending (“Bent”) exhibit substantial reductions in the near-infrared (NIR)-based indices, while healthy trees maintain higher and more stable index values. Structure- and pigment-sensitive indices (e.g., the Modified Chlorophyll Absorption Ratio Index (MCARI 2), the Transformed Chlorophyll Absorption in Reflectance Index/Optimised Soil-Adjusted Vegetation Index (TCARI/OSAVI), and RDVI) showed the highest diagnostic value for differentiating between damaged and healthy trees. We found the clear identification of group- and season-specific patterns, revealing that the most pronounced physiological decline in Bent trees emerged only several seasons after the disturbance. Full article

The accelerating pace of urbanization and population growth has increasingly disrupted the relationship between the built environment and nature, contributing to a decline in individuals’ psychological and spiritual well-being. Religious buildings—particularly mosques—have the potential to serve as restorative environments that support spiritual reflection and emotional balance. Within this context, the integration of biophilic design principles into sacred architecture has emerged as a significant area of contemporary research. This study investigates how biophilic elements in mosque interiors influence individuals’ perceptions of spirituality and the restorative quality of the space. Mosque interior images were generated using artificial intelligence tools (Midjourney V6.1 and DALL·E 3), informed by biophilic design theory. Grounded in Attention Restoration Theory (ART), the study utilized a quantitative research framework to explore how biophilic elements influence spiritual and restorative perceptions in sacred architectural spaces. Data were collected from 390 individuals in Türkiye with prior worship experience in Republican-era mosques—structures marked by ideological and spatial transformations. Two instruments were used, the Perceived Restorativeness Scale (PRS) and the newly developed Biophilic Spiritual Perception Scale (BSPS), which demonstrated high internal consistency (Cronbach’s α = 0.981). Analyses included exploratory and confirmatory factor analyses, as well as parametric and non-parametric statistical tests. Findings suggest that biophilic design in mosque interiors positively influences both spiritual experience and perceived environmental restorativeness. These results support the view that biophilic design should not be seen merely as an esthetic or ecological approach but as a multidimensional strategy that enhances the emotional and spiritual quality of sacred spaces. Implications for future mosque design in Türkiye are discussed. Full article

The Karnak Temples are considered one of Egypt’s most significant archaeological sites, dating back to the Middle Kingdom (c. 2000–1700 BC) and were continuously expanded until the Ptolemaic period (305–30 BC). As the second most visited UNESCO World Heritage archaeological site in Egypt after the Giza Pyramids, Karnak faces severe deterioration processes due to prolonged exposure to environmental impacts, mechanical damage, and historical interventions. This study employs a multidisciplinary approach integrating non-destructive testing (NDT) methods to assess the physical and mechanical condition and degradation mechanisms of scattered sandstone blocks at the site. Advanced documentation techniques, including Reflectance Transformation Imaging (RTI), photogrammetry, and Infrared Thermography (IRT), were used to analyze surface morphology, thermal stress effects, and weathering patterns. Ultrasonic Pulse Velocity (UPV) testing provided internal structural assessments, while spectral and gloss analysis quantified chromatic alterations and surface roughness. Additionally, the Karsten Tube test determined the water absorption behavior of the sandstone, highlighting variations in porosity and susceptibility to salt crystallization. In this sense, the results indicate that climatic factors such as extreme temperature fluctuations, wind erosion, and groundwater infiltration contributed to sandstone deterioration. Thermal cycling leads to microcracking and granular disintegration, while high capillary water absorption accelerates chemical weathering processes. UPV analyses showed substantial internal decay, with low-velocity zones correlating with fractures and differential cementation loss. Finally, an interventive conservation plan was proposed. Full article

The retina offers a unique window into both ocular and systemic health, motivating the development of AI-based tools for disease screening and risk assessment. In this study, we present a comprehensive evaluation of six state-of-the-art deep neural networks, including convolutional neural networks and vision transformer architectures, on the Brazilian Multilabel Ophthalmological Dataset (BRSET), comprising 16,266 fundus images annotated for multiple clinical and demographic labels. We explored seven classification tasks: Diabetes, Diabetic Retinopathy (2-class), Diabetic Retinopathy (3-class), Hypertension, Hypertensive Retinopathy, Drusen, and Sex classification. Models were evaluated using precision, recall, F1-score, accuracy, and AUC. Among all models, the Swin-L generally delivered the best performance across scenarios for Diabetes (AUC = 0.88, weighted F1-score = 0.86), Diabetic Retinopathy (2-class) (AUC = 0.98, weighted F1-score = 0.95), Diabetic Retinopathy (3-class) (macro AUC = 0.98, weighted F1-score = 0.95), Hypertension (AUC = 0.85, weighted F1-score = 0.79), Hypertensive Retinopathy (AUC = 0.81, weighted F1-score = 0.97), Drusen detection (AUC = 0.93, weighted F1-score = 0.90), and Sex classification (AUC = 0.87, weighted F1-score = 0.80). These results reflect excellent to outstanding diagnostic performance. We also employed gradient-based saliency maps to enhance explainability and visualize decision-relevant retinal features. Our findings underscore the potential of deep learning, particularly vision transformer models, to deliver accurate, interpretable, and clinically meaningful screening tools for retinal and systemic disease detection. Full article

Precision neurosurgery is rapidly evolving as a medical specialty by merging genomic medicine, multi-omics technologies, and artificial intelligence (AI) technology, while at the same time, society is shifting away from the traditional, anatomic model of care to consider a more precise, molecular model of care. The general purpose of this review is to contemporaneously reflect on how these advances will impact neurosurgical care by providing us with more precise diagnostic and treatment pathways. We hope to provide a relevant review of the recent advances in genomics and multi-omics in the context of clinical practice and highlight their transformational opportunities in the existing models of care, where improved molecular insights can support improvements in clinical care. More specifically, we will highlight how genomic profiling, CRISPR-Cas9, and multi-omics platforms (genomics, transcriptomics, proteomics, and metabolomics) are increasing our understanding of central nervous system (CNS) disorders. Achievements obtained with transformational technologies such as single-cell RNA sequencing and intraoperative mass spectrometry are exemplary of the molecular diagnostic possibilities in real-time molecular diagnostics to enable a more directed approach in surgical options. We will also explore how identifying specific biomarkers (e.g., IDH mutations and MGMT promoter methylation) became a tipping point in the care of glioblastoma and allowed for the establishment of a new taxonomy of tumors that became applicable for surgeons, where a change in practice enjoined a different surgical resection approach and subsequently stratified the adjuvant therapies undertaken after surgery. Furthermore, we reflect on how the novel genomic characterization of mutations like DEPDC5 and SCN1A transformed the pre-surgery selection of surgical candidates for refractory epilepsy when conventional imaging did not define an epileptogenic zone, thus reducing resective surgery occurring in clinical practice. While we are atop the crest of an exciting wave of advances, we recognize that we also must be diligent about the challenges we must navigate to implement genomic medicine in neurosurgery—including ethical and technical challenges that could arise when genomic mutation-based therapies require the concurrent application of multi-omics data collection to be realized in practice for the benefit of patients, as well as the constraints from the blood–brain barrier. The primary challenges also relate to the possible gene privacy implications around genomic medicine and equitable access to technology-based alternative practice disrupting interventions. We hope the contribution from this review will not just be situational consolidation and integration of knowledge but also a stimulus for new lines of research and clinical practice. We also hope to stimulate mindful discussions about future possibilities for conscientious and sustainable progress in our evolution toward a genomic model of precision neurosurgery. In the spirit of providing a critical perspective, we hope that we are also adding to the larger opportunity to embed molecular precision into neuroscience care, striving to promote better practice and better outcomes for patients in a global sense. Full article

Real-time quality monitoring using molten pool images is a critical focus in researching high-quality, intelligent automated welding. To address interference problems in molten pool images under complex welding scenarios (e.g., reflected laser spots from spatter misclassified as porosity defects) and the limited interpretability of deep learning models, this paper proposes a multi-granularity spatiotemporal representation learning algorithm based on the hybrid enhancement of handcrafted and deep learning features. A MobileNetV2 backbone network integrated with a Temporal Shift Module (TSM) is designed to progressively capture the short-term dynamic features of the molten pool and integrate temporal information across both low-level and high-level features. A multi-granularity attention-based feature aggregation module is developed to select key interference-free frames using cross-frame attention, generate multi-granularity features via grouped pooling, and apply the Convolutional Block Attention Module (CBAM) at each granularity level. Finally, these multi-granularity spatiotemporal features are adaptively fused. Meanwhile, an independent branch utilizes the Histogram of Oriented Gradient (HOG) and Scale-Invariant Feature Transform (SIFT) features to extract long-term spatial structural information from historical edge images, enhancing the model’s interpretability. The proposed method achieves an accuracy of 99.187% on a self-constructed dataset. Additionally, it attains a real-time inference speed of 20.983 ms per sample on a hardware platform equipped with an Intel i9-12900H CPU and an RTX 3060 GPU, thus effectively balancing accuracy, speed, and interpretability. Full article

Background: Autism spectrum disorder (ASD) is characterised by atypical behavioural and cognitive diversity, yet the neural underpinnings linking brain activity and individual presentations remain poorly understood. In this study, we investigated the relationship between resting-state functional magnetic resonance imaging (fMRI) signal complexity and intelligence (full-scale intelligence quotient (FIQ); verbal intelligence quotient (VIQ); and performance intelligence quotient (PIQ)) in male adults with ASD (n = 14) and matched neurotypical controls (n = 15). Methods: We used three complexity-based metrics: Hurst exponent (H), fuzzy approximate entropy (fApEn), and fuzzy sample entropy (fSampEn) to characterise resting-state fMRI signal dynamics, and correlated these measures with standardised intelligence scores. Results: Using a whole-brain measure, ASD participants showed significant negative correlations between PIQ and both fApEn and fSampEn, suggesting that increased neural irregularity may relate to reduced cognitive–perceptual performance in autistic individuals. No significant associations between entropy (fApEn and fSampEn) and PIQ were found in the control group. Group differences in brain–behaviour associations were confirmed through formal interaction testing using Fisher’s r-to-z transformation, which showed significantly stronger correlations in the ASD group. Complementary regression analyses with interaction terms further demonstrated that the entropy (fApEn and fSampEn) and PIQ relationship was significantly moderated by group, reinforcing evidence for autism-specific neural mechanisms underlying cognitive function. Conclusions: These findings provide insight into how cognitive functions in autism may not only reflect deficits but also an alternative neural strategy, suggesting that distinct temporal patterns may be associated with intelligence in ASD. These preliminary findings could inform clinical practice and influence health and social care policies, particularly in autism diagnosis and personalised support planning. Full article

Traditional bathymetry measures require a large number of human hours, and many bathymetry records are obsolete or missing. Automated measures of bathymetry would reduce costs and increase accessibility for research and applications. In this paper, we optimized a recent machine learning model, named CatBoostOpt, to estimate bathymetry based on high-resolution WorldView-2 (WV-2) multi-spectral optical satellite images. CatBoostOpt was demonstrated across the Florida Big Bend coastline, where the model learned correlations between in situ sound Navigation and Ranging (Sonar) bathymetry measurements and the corresponding multi-spectral reflectance values in WV-2 images to map bathymetry. We evaluated three different feature transformations as inputs for bathymetry estimation, including raw reflectance, log-linear, and log-ratio transforms of the raw reflectance value in WV-2 images. In addition, we investigated the contribution of each spectral band and found that utilizing all eight spectral bands in WV-2 images offers the best solution for handling complex water quality conditions. We compared CatBoostOpt with linear regression (LR), support vector machine (SVM), random forest (RF), AdaBoost, gradient boosting, and deep convolutional neural network (DCNN). CatBoostOpt with log-ratio transformed reflectance achieved the best performance with an average root mean square error (RMSE) of 0.34 and coefficient of determination (${\mathrm{R}}^2$) of 0.87. Full article

The quality and freshness of fruits and vegetables are critical factors in consumer acceptance and are significantly affected during transport and storage. This study aimed to evaluate the quality of greenhouse-grown tomatoes stored for 24 days by combining non-destructive image analysis, spectrophotometric assays (including total phenolic content and antioxidant and antiradical activity assessments), and attenuated total reflectance–Fourier transform infrared (ATR-FTIR) spectroscopy. Additionally, water activity, moisture content, total soluble solids, texture, and color were evaluated. Most physicochemical changes occurred between days 14 and 17, without major impact on overall fruit quality. A progressive transition in peel hue from orange to dark orange, and increased surface irregularity of their textural image were noted. Moreover, the combined use of instrumental and image analyses results via multivariate analysis allowed the clear discrimination of tomatoes according to storage days. In this sense, tomato samples were effectively classified by ATR-FTIR spectral bands, linked to carotenoids, phenolics, and polysaccharides. Machine learning (ML) models, including Random Forest and Gradient Boosting, were trained on image-derived features and accurately predicted shelf life and quality traits, achieving R² values exceeding 0.9. The findings demonstrate the effectiveness of combining imaging, spectroscopy, and ML for non-invasive tomato quality monitoring and support the development of predictive tools to improve postharvest handling and reduce food waste. Full article