Search Results (698)

Traditional GrabCut algorithms are limited by their reliance on manual intervention, often resulting in segmentation errors and missed detections, particularly against complex backgrounds. This study addresses these limitations by introducing the Auto Trademark Cut (AT-Cut), an advanced automated trademark image-segmentation method built upon an enhanced GrabCut framework. The proposed approach achieves superior performance through three key innovations: Firstly, histogram equalization is applied to the entire input image to mitigate noise induced by illumination variations and other environmental factors. Secondly, state-of-the-art object detection techniques are utilized to precisely identify and extract the foreground target, with dynamic region definition based on detection outcomes to ensure heightened segmentation accuracy. Thirdly, morphological erosion and dilation operations are employed to refine the contours of the segmented target, leading to significantly improved edge segmentation quality. Experimental results indicate that AT-Cut enables efficient, fully automated trademark segmentation while minimizing the necessity for labor-intensive manual labeling. Evaluation on the public Real-world Logos dataset demonstrates that the proposed method surpasses conventional GrabCut algorithms in both boundary localization accuracy and overall segmentation quality, achieving a mean accuracy of 90.5%. Full article

Parkinson’s disease (PD) is characterized by motor symptoms, with key diagnostic features, such as rigidity, traditionally assessed through subjective clinical scales. This study proposes a novel hybrid framework integrating model-driven biomechanical features (damping ratio, decay rate) and data-driven statistical features (maximum detail coefficient) from wearable sensor data during a modified pendulum test to quantify shoulder girdle rigidity objectively. Using weak supervision, these features were unified to generate robust labels from limited data, achieving a 10% improvement in PD/healthy control classification accuracy (0.71 vs. 0.64) over data-driven methods and matching model-driven performance (0.70). The damping ratio and decay rate, aligning with Wartenberg pendulum test metrics like relaxation index, revealed velocity-dependent aspects of rigidity, challenging its clinical characterization as velocity-independent. Outputs correlated strongly with UPDRS rigidity scores (r = 0.78, p < 0.001), validating their clinical utility as novel biomechanical biomarkers. This framework enhances interpretability and scalability, enabling remote, objective rigidity assessment for early diagnosis and telemedicine, advancing PD management through innovative sensor-based neurotechnology. Full article

Coffee silverskin (CS), the principal solid by-product from coffee roasting, is a promising raw material for sustainable food applications aligned with circular economy principles. Due to its high flammability at roasting temperatures, effective management of CS is not only an environmental but also a safety concern in coffee processing facilities. To the best of our knowledge, this is the first study evaluating the chemical composition, bioactivity, safety, and environmental impact of torrefacto (CT) and natural (CN) coffee silverskin. CT (from Arabica–Robusta blends subjected to sugar-glazing) and CN (from 100% Arabica) were characterized in terms of composition and function. Oven-dried CT showed higher levels of caffeine (13.2 ± 0.6 mg/g vs. 8.7 ± 0.7 mg/g for CN), chlorogenic acid (1.34 ± 0.08 mg/g vs. 0.92 ± 0.06 mg/g), protein (18.1 ± 0.2% vs. 16.7 ± 0.2%), and melanoidins (14.9 ± 0.3 mg/g vs. 9.6 ± 0.2 mg/g), but CN yielded more total phenolics (13.8 ± 0.6 mg GAE/g). Both types exhibited strong antioxidant capacity (ABTS: 48.9–59.2 µmol TE/g), and all oven-dried samples met food safety criteria (microbial loads below 10² CFU/g, moisture 7.9%). Oven drying was identified as the most industrially viable, ensuring preservation of bioactives and resulting in a 19% lower greenhouse gas emissions impact compared to freeze-drying. Sun drying was less reliable microbiologically. The valorization of oven-dried CT as a clean-label, antioxidant-rich colorant offers clear potential for food reformulation and waste reduction. Renewable energy use during drying is recommended to further enhance sustainability. This study provides scientific evidence to support the safe use of coffee silverskin as a novel food, contributing to regulatory assessment and sustainable food innovation aligned with SDGs 9, 12, and 13. Full article

This study investigated the incorporation of Salicornia perennans powder as a natural antioxidant and functional ingredient in cooked sausages, with the aim of improving product quality and promoting sustainable production strategies. The inclusion of 3% Salicornia perennans resulted in a nutritionally favorable shift in the fatty acid profile, with a 1.5-fold increase in α-linolenic acid ALA and the presence of long-chain ω-3 fatty acids EPA and DHA, along with improved PUFA/SFA and ω-6/ω-3 ratios. Lipid and protein oxidation were significantly suppressed during refrigerated storage, as evidenced by the reduced peroxide value of 10.6 vs. 12.8 meq/kg, thiobarbituric acid-reactive substance value of 0.158 vs. 0.210 mg MDA/kg, acid value of 4.6 vs. 5.5 mg KOH/g, and carbonyl compound value of 101.9 vs. 112.3 nmol/mg protein compared to the control. Color stability was enhanced, with ΔE* values remaining below perceptible thresholds in Salicornia perennans-supplemented sausages, highlighting its role in preserving visual quality. Antioxidant capacity was markedly higher, with FRAP values of 14.5 mg GAE/g undetected in the control and improved DPPH radical-scavenging activity of 22.6% vs. 12.5%. These findings demonstrate that Salicornia perennans not only enriches meat products with bioactive compounds and health-promoting lipids but also reduces oxidative spoilage, thereby extending shelf life. The results emphasize the potential of halophyte-based ingredients to support technological innovation, environmental impact reduction, and the development of clean-label functional meat products aligned with sustainable production strategies. Full article

Progress in clinical diagnosis increasingly relies on innovative technologies and advanced disease biomarker detection methods. While cell labeling remains a well-established technique, label-free approaches offer significant advantages, including reduced workload, minimal sample damage, cost-effectiveness, and simplified chip integration. These approaches focus on the morpho-biophysical properties of cells, eliminating the need for labeling and thus reducing false results while enhancing data reliability and reproducibility. Current label-free methods span conventional and advanced technologies, including phase-contrast microscopy, holographic microscopy, varied cytometries, microfluidics, dynamic light scattering, atomic force microscopy, and electrical impedance spectroscopy. Their integration with artificial intelligence further enhances their utility, enabling rapid, non-invasive cell identification, dynamic cellular interaction monitoring, and electro-mechanical and morphological cue analysis, making them particularly valuable for cancer diagnostics, monitoring, and prognosis. This review compiles recent label-free cancer cell detection developments within clinical and biotechnological laboratory contexts, emphasizing biophysical alterations pertinent to liquid biopsy applications. It highlights interdisciplinary innovations that allow the characterization and potential identification of cancer cells without labeling. Furthermore, a comparative analysis addresses throughput, resolution, and detection capabilities, thereby guiding their effective deployment in biomedical research and clinical oncology settings. Full article

In order to enhance leakage detection accuracy in water distribution networks (WDNs) while reducing sensor deployment costs, an intelligent algorithm for the optimal deployment of water network monitoring sensors based on the automatic labelling and graph neural network (ALGN) was proposed for the optimal deployment of WDN monitoring sensors. The research aims to develop a data-driven, topology-aware sensor deployment strategy that achieves high leakage detection performance with minimal hardware requirements. The methodology consisted of three main steps: first, the dung beetle optimization algorithm (DBO) was employed to automatically determine optimal parameters for the DBSCAN clustering algorithm, which generated initial cluster labels; second, a customized graph neural network architecture was used to perform topology-aware node clustering, integrating network structure information; finally, optimal pressure sensor locations were selected based on minimum distance criteria within identified clusters. The key innovation lies in the integration of metaheuristic optimization with graph-based learning to fully automate the sensor placement process while explicitly incorporating the hydraulic network topology. The proposed approach was validated on real-world WDN infrastructure, demonstrating superior performance with 93% node coverage and 99.77% leakage detection accuracy, surpassing state-of-the-art methods by 2% and 0.7%, respectively. These results indicate that the ALGN framework provides municipal water utilities with a robust, automated solution for designing efficient pressure monitoring systems that balance detection performance with implementation cost. Full article

Rolling bearings are essential for modern mechanical equipment and serve in various operational environments. This paper addresses the challenge of vibration data discrepancies in bearings across different operating conditions, which often results in inaccurate fault diagnosis. To tackle this related limitation, a novel lightweight multi-scale attention-based joint adaptive adversarial transfer network, termed MAJATNet, is developed. The proposed network integrates a feature extraction network innovation module with an improved loss function, namely IJA loss. The feature extraction module employs a one-dimensional multi-scale attention residual structure to derive characteristics from monitoring data of source and target domains. IJA loss evaluates the joint distribution discrepancy of high-dimensional features and labels between these domains. IJA loss integrates a joint maximum mean discrepancy (JMMD) loss with a domain adversarial learning loss, which directs the model’s focus toward categorical features while minimizing domain-specific features. The performance and advantages of MAJATNet are demonstrated through cross-domain fault diagnosis experiments using bearing datasets. Experimental results show that the proposed method can significantly improve the accuracy of cross-domain fault diagnosis for bearings. Full article

Community detection is a fundamental task for understanding network structures, crucial for identifying groups of nodes with close connections. However, existing methods generally treat all connections in networks as equally important, overlooking the inherent inequality of connection strengths in social networks, and often require large quantities of labeled data. To address these challenges, we propose a few-shot community detection framework, Strong Triadic Closure Community Detection with Prompt (STC-CDP), which combines the Strong Triadic Closure (STC) principle, Graph Neural Networks, and prompt learning. The STC principle, derived from social network theory, states that if two nodes share strong connections with a third node, they are likely to be connected with each other. By incorporating STC constraints during the pre-training phase, STC-CDP can differentiate between strong and weak connections in networks, thereby more accurately capturing community structures. We design an innovative prompt learning mechanism that enables the model to extract key features from a small number of labeled communities and transfer them to the identification of unlabeled communities. Experiments on multiple real-world datasets demonstrate that STC-CDP significantly outperforms existing state-of-the-art methods under few-shot conditions, achieving higher F1 scores and Jaccard similarity particularly on Facebook, Amazon, and DBLP datasets. Our approach not only improves the precision of community detection but also provides new insights into understanding connection inequality in social networks. Full article

The study investigates the use of powerful machine learning approaches to the real-time detection of phishing URLs, addressing a critical cybersecurity concern. The dataset we utilized in this research work was collected from the University of California Irvine (UCI) Machine Learning Repository. It has 235,795 instances with fifty-four distinct parameters. The label class is of binomial type and has only two target classes. We used a range of complex algorithms, including k-nearest neighbor, naive Bayes, decision trees, random forests, and random tree, to assess the discriminative characteristics retrieved from URLs. The random forest classifier beat the other classifiers, reaching the greatest accuracy of 99.99%. The study demonstrates that these models achieve superior accuracy in identifying phishing attempts, significantly outperforming traditional detection methodologies. The findings underscore the potential of machine learning to provide a scalable, efficient, and robust solution for real-time phishing detection. Implementing these innovative platforms to existing security solutions is going to play a critical role in sustaining the protective line against continuously evolving and persistent phishing schemes. Full article

Organisations are increasingly adopting the Chain of Custody (CoC) standards in the mining industry to enhance the traceability of minerals. It ensures that the minerals they have received are from credible sources and accompanied by verifiable information. However, unlikeother industries such as timber, where the effectiveness and benefits of CoC standards are mainly explored, this study subtly shifts the focus towards identifying strategic opportunities and innovation areas within the CoC standards that could extend beyond traceability. Four CoC standards were selected, and their provisions examined. It was found that implementing these requirements could not only enhance transparency but also support broader sustainability goals across the entire value chain. The study also identifies several challenges that could act as barriers to the CoC system, and these are seen as opportunities for innovative approaches to enhance the effectiveness of the standards. These are labelled as transformative innovation areas, and while they do include blockchains and analytical proof of origin technologies, this study also seeks to advocate for solutions that are more pragmatic and scalable. By identifying opportunities and areas of innovation, the findings will help improve the practical implementation of the standards and suggest areas for future evaluations of effectiveness that could consider aspects beyond traceability, such as sustainability and transparency. Full article

Wildfires pose a regularly increasing threat to ecosystems and critical infrastructure. The severity of this threat is steadily increasing. The growing threat necessitates the development of technologies for rapid and accurate early detection. However, the prevailing fire point detection algorithms, including several deep learning models, are generally constrained by the inherent hard threshold limitations in their decision-making logic. As a result, these methods lack adaptability and robustness in complex and dynamic real-world scenarios. To address this challenge, the present paper proposes an innovative two-stage, semi-supervised anomaly detection framework. The framework initially employs a Transformer-based autoencoder, which serves to transform raw fire-free time-series data derived from satellite imagery into a multidimensional deep anomaly feature vector. Self-supervised learning achieves this transformation by incorporating both reconstruction error and latent space distance. In the subsequent stage, a semi-supervised XGBoost classifier, trained using an iterative pseudo-labeling strategy, learns and constructs an adaptive nonlinear decision boundary in this high-dimensional anomaly feature space to achieve the final fire point judgment. In a thorough validation process involving multiple real-world fire cases in Yunnan Province, China, the framework attained an F1 score of 0.88, signifying a performance enhancement exceeding 30% in comparison to conventional deep learning baseline models that employ fixed thresholds. The experimental results demonstrate that by decoupling feature learning from classification decision-making and introducing an adaptive decision mechanism, this framework provides a more robust and scalable new paradigm for constructing next-generation high-precision, high-efficiency wildfire monitoring and early warning systems. Full article

The extensive use of chemical preservatives in the food industry has raised concerns over their association with gut microbiota imbalance, allergenic reactions, and potential carcinogenicity. Growing consumer demand for “clean label” products, coupled with regulatory pressures, has accelerated the search for safer and more sustainable alternatives. In this study, it is reported for the first time that the synthesis of AIEE-type Supra-CDs using p-phenylenediamine (p-PA) and thiourea (TU), a breakthrough that provides a new class of nanomaterials with superior optical and antimicrobial properties. More importantly, the study demonstrates a quantitative improvement of spectral overlap through controllable inner filter effect (IFE), establishing a reliable strategy to enhance detection sensitivity and broaden applicability in food safety monitoring. Beyond their intrinsic antimicrobial potential, these Supra-CDs integrate seamlessly with intelligent detection platforms such as biosensors, CRISPR-based assays, and AI-assisted analytics, enabling real-time evaluation of probiotic- and postbiotic-based preservation systems. By combining novel material synthesis with precision monitoring technologies, this work offers a dual innovation: reducing reliance on synthetic additives while providing scalable tools for sustainable food preservation. The findings not only advance the frontier of biopreservation research but also align with global initiatives for consumer health and environmental sustainability. Full article

Kinetoplastids are protozoa that possess a unique organelle called a kinetoplast. These include the parasites Trypanosoma cruzi, T. brucei and related African trypanosomes, and Leishmania spp. These parasites cause a variety of neglected tropical diseases in humans and livestock, with devastating consequences. In the absence of any vaccine, pharmaceutical interventions are the mainstay of control, but these have historically been underfunded, fragmented, and inadequately aligned with the complex zoonotic and ecological realities of the parasites’ transmission dynamics. In this review, the landscape of current and emerging drugs for treating leishmaniasis, Chagas disease, and African trypanosomiasis is critically evaluated across both veterinary and human contexts. It examines the challenges of legacy compounds, the pharmacological shortcomings in multi-host, multi-tropic and multi-stage disease systems, and the gaps in veterinary therapeutics, specifically for African animal trypanosomiasis and canine leishmaniasis but also the animal reservoir of T. cruzi. Emphasis is placed on pharmacokinetic divergence between species, the accompanying risks with the use of off-label human drugs in animals, and the ecological effects of environmental drug exposure. We propose a far-reaching One Health framework for pharmaceutical research and development, promoting dual-indication co-development, ecological pharmacology, regulatory harmonisation, and integrated delivery systems. In this context, we argue that the drug development pipeline must be rationalised as a transdisciplinary and ecologically embedded process, able to interrupt parasite transmission to human, animal, and vector interfaces. Our findings reveal that we can bridge age-old therapeutic gaps, advance towards sustainable control, and eventually eliminate the neglected diseases caused by kinetoplastid protozoan parasites by aligning pharmaceutical innovation with One Health principles. This article aims to promote future research and development of innovative drugs that are sustainable under the One Health framework. Full article

The increasing demand for sustainably produced food has intensified interest in fermented milk products, such as yoghurt, which combine nutritional value with environmental and ethical considerations. However, the authenticity of sustainability claims in this sector remains contested, raising concerns about consumer trust and regulatory clarity. This review examines the role of certification and labelling in verifying and communicating the sustainability of fermented milk products. The analysis covers regulatory frameworks, consumer perceptions, and the potential of digital tools to improve transparency. Findings highlight inconsistencies in defining key terms such as organic, probiotic, and carbon-neutral, which hinder certification harmonization. Complex labels and allergen declarations can reduce clarity and trust, while overlapping or vague eco-labels risk contributing to consumer confusion and skepticism. Despite this, credible certifications still enhance purchase intent. Modern technologies, including blockchain traceability, interactive QR codes, and digital product passports, offer new ways to reinforce trust, though implementation costs and regulatory gaps remain barriers. This review concludes that effective sustainability communication must integrate robust certification schemes with simplified, transparent messaging. Harmonized standards, improved label design, and consumer education are essential to support informed choices and foster trust in sustainable dairy. Full article

The growing global demand for rapid, sensitive, and cost-effective food safety monitoring has driven the development of nanozyme-based biosensors as alternatives to natural enzyme-based methods. Among various nanozymes, bimetallic gold–platinum (AuPt) nanozymes show superior catalytic performance compared to monometallic and other Au-based bimetallic hybrids. This is due to their synergistic colorimetric, catalytic, geometric, and ensemble properties. This review systematically evaluates AuPt nanozymes in food safety applications, focusing on their synthesis, structural design, and practical uses. Various structural types are highlighted, including plain, magnetic, porous nanomaterial-labeled, and flexible nanomaterial-loaded AuPt hybrids. Key synthesis methods such as seed-mediated growth and one-pot procedures with different reducing agents are summarized. Detection modes covered include colorimetric, electrochemical, and multimodal sensing, demonstrating efficient detection of important food contaminants. Key innovations include core–shell designs for enhanced catalytic activity, new synthesis strategies for improved structural control, and combined detection modes to increase reliability and reduce false positives. Challenges and future opportunities are discussed, such as standardizing synthesis protocols, scaling up production, and integration with advanced sensing platforms. This review aims to accelerate the translation of AuPt nanozyme technology into practical food safety monitoring solutions that improve food security and public health. Full article