Search Results (6,839)

Background and Objectives: Epilepsy is a chronic neurological disorder characterized by recurrent seizures caused by abnormal brain activity. Reliable near-real-time seizure detection is essential for preventing injuries, enabling early interventions, and improving the quality of life for patients with drug-resistant epilepsy. This study presents a near-real-time epileptic seizure detection framework designed for low-latency operation, focusing on improving both clinical reliability and patient comfort through electrode reduction. Method: The framework integrates bidirectional long short-term memory (BiLSTM) networks with wavelet-based feature extraction using Electroencephalogram (EEG) recordings from the EPILEPSIAE dataset. EEG signals from 161 patients comprising 1,032 seizures were analyzed. Wavelet features were combined with raw EEG data to enhance temporal and spectral representation. Furthermore, electrode reduction experiments were conducted to determine the minimum number of strategically positioned electrodes required to maintain performance. Results: The optimized BiLSTM model achieved 86.9% accuracy, 86.1% recall, and an average detection delay of 1.05 s, with a total processing time of 0.065 s per 0.5 s EEG window. Results demonstrated that reliable detection is achievable with as few as six electrodes, maintaining comparable performance to the full configuration. Conclusions: These findings demonstrate that the proposed BiLSTM-wavelet approach provides a clinically viable, computationally efficient, and wearable-friendly solution for near-real-time epileptic seizure detection using reduced EEG channels. Full article

Modern engineering systems require reliable and timely Fault Detection and Diagnosis (FDD) to ensure operational safety and resilience. Traditional model-based and rule-based approaches, although interpretable, exhibit limited scalability and adaptability in complex, data-intensive environments. This survey provides a systematic overview of recent studies exploring Machine Learning (ML) and Artificial Intelligence (AI) techniques for FDD across industrial, energy, Cyber-Physical Systems (CPS)/Internet of Things (IoT), and cybersecurity domains. Deep architectures such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Transformers, and Graph Neural Networks (GNNs) are compared with unsupervised, hybrid, and physics-informed frameworks, emphasizing their respective strengths in adaptability, robustness, and interpretability. Quantitative synthesis and radar-based assessments suggest that AI-driven FDD approaches offer increased adaptability, scalability, and early fault detection capabilities compared to classical methods, while also introducing new challenges related to interpretability, robustness, and deployment. Emerging research directions include the development of foundation and multimodal models, federated learning (FL), and privacy-preserving learning, as well as physics-guided trustworthy AI. These trends indicate a paradigm shift toward self-adaptive, interpretable, and collaborative FDD systems capable of sustaining reliability, transparency, and autonomy across critical infrastructures. Full article

Flooding is a recurrent problem in many Brazilian cities, resulting in significant losses that affect health, assets, finance, and the environment. The uncertainty regarding extreme rainfall events due to climate change makes this challenge even more severe, compounded by inadequate urban planning and the occupation of risk areas, particularly for the municipality of Catu, in the state of Bahia, which also suffers from recurrent floods. Critical hotspots include the Santa Rita neighborhood and its surroundings, the main supply center, and the city center—the municipality’s commercial hub. The focus of this research is the unprecedented quantification of the socioeconomic impact of these floods on the low-income population and the region’s informal sector (street vendors). This research focused on analyzing and modeling the destructive potential of intense rainfall in the Santa Rita region (Supply Center) of Catu, Bahia, and its effects on the local economy across different recurrence intervals. A hydrological simulation software suite based on computational and geoprocessing technologies—specifically HEC-RAS 6.4, HEC-HMS 4.11, and QGIS— 3.16 was utilized. Two-dimensional (2D) modeling was applied to assess the flood-prone areas. For the socioeconomic impact assessment, a loss procedure based on linear regression was developed, which correlated the different return periods of extreme events with the potential losses. This methodology, which utilizes validated, indirect data, establishes a replicable framework adaptable to other regions facing similar socioeconomic and drainage challenges. The results revealed that the area becomes impassable during flood events, preventing commercial activities and causing significant economic losses, particularly for local market vendors. The total financial damage for the 100-year extreme event is approximately US $30,000, with the loss model achieving an R² of 0.98. The research concludes that urgent measures are necessary to mitigate flood impacts, particularly as climate change reduces the return period of extreme events. The implementation of adequate infrastructure, informed by the presented risk modeling, and public awareness are essential for reducing vulnerability. Full article

This paper presents an extended investigation into deep learning-based cryptanalysis of block ciphers by introducing and evaluating a multi-server attack environment. Building upon our prior work in centralized settings, we explore the practicality and scalability of deploying such attacks across multiple distributed edge servers. We assess the vulnerability of five representative block ciphers—DES, SDES, AES-128, SAES, and SPECK32/64—under two neural attack models: Encryption Emulation (EE) and Plaintext Recovery (PR), using both fully connected neural networks and Recurrent Neural Networks (RNNs) based on bidirectional Long Short-Term Memory (BiLSTM). Our experimental results show that the proposed federated learning-based cryptanalysis framework achieves performance nearly identical to that of centralized attacks, particularly for ciphers with low round complexity. Even as the number of edge servers increases to 32, the attack models maintain high accuracy in reduced-round settings. We validate our security assessments through formal statistical significance testing using two-tailed binomial tests with 99% confidence intervals. Additionally, our scalability analysis demonstrates that aggregation times remain negligible (<0.01% of total training time), confirming the computational efficiency of the federated framework. Overall, this work provides both a scalable cryptanalysis framework and valuable insights into the design of cryptographic algorithms that are resilient to distributed, deep learning-based threats. Full article

Triple-negative breast cancer (TNBC), which is characterized by a lack of the estrogen receptor, the progesterone receptor, and HER2 expression, is the most aggressive breast cancer subtype and has a poor prognosis and high recurrence rates because of frequent chemotherapy resistance. As a crucial epigenetic regulator, DNA methylation modulates gene expression through aberrant methylation patterns, contributing to tumor progression and therapeutic resistance. Early diagnosis and treatment of TNBC are vital for its prognosis. The development of DNA methylation testing technology and the application of liquid biopsy provide technological support for early diagnosis and treatment. Additionally, preclinical and early-phase clinical studies suggest that epigenetic therapies targeting DNA methylation may hold promise for TNBC treatment, pending larger clinical trials. Furthermore, research on DNA methylation-based prognostic models enables personalized precision treatment for patients, helping to reduce unnecessary therapies and improve overall survival. The emerging role of DNA methylation patterns in predicting the therapeutic response and overcoming drug resistance is highlighted. In this narrative review, we integrate current research findings and clinical perspectives. We propose that DNA methylation presents promising research prospects for the diagnosis, treatment and prognosis prediction of TNBC. Future efforts should focus on translating methylation-driven insights into clinically actionable strategies, ultimately advancing precision oncology for this challenging disease. Full article

Recurrent brain tumors—including high-grade gliomas, brain metastases, and aggressive meningiomas—continue to carry a poor prognosis, with high mortality despite therapeutic advances. The aim of this narrative review is to summarize and critically discuss the current evidence on the role of intra-arterial radioligand therapy (RLT) in the treatment of recurrent brain tumors. RLT, a targeted form of radionuclide therapy, has gained increasing attention for its potential theranostic applications in neuro-oncology. A literature search was conducted using PubMed and Scopus, including clinical studies evaluating intra-arterial radioligand delivery in central nervous system tumors. Recent research has explored intra-arterial administration of radioligands targeting somatostatin receptors and prostate-specific membrane antigen (PSMA). Somatostatin receptors are overexpressed in meningiomas, while PSMA is highly expressed in the neovasculature of glioblastomas and brain metastases; both targets can be addressed using lutetium-177 (¹⁷⁷Lu)- or actinium-225 (²²⁵Ac)-labeled radiopharmaceuticals, traditionally delivered intravenously. Available evidence indicates that the intra-arterial route achieves markedly higher radionuclide uptake on ⁶⁸Ga-PSMA-11 and ⁶⁸Ga-DOTATOC PET, as well as increased absorbed doses in dosimetric models. Dosimetric analyses consistently show greater tracer accumulation compared with intravenous administration, without evidence of significant peri-procedural toxicity. Uptake in healthy brain tissue is minimal, and no relevant differences have been reported in liver or salivary gland accumulation between intra-arterial and intravenous RLT. Although based on heterogeneous and limited data, intra-arterial RLT appears to be a promising therapeutic strategy for recurrent brain tumors. Future research should focus on improving radioligand delivery beyond the blood–brain barrier and enhancing effective tumor targeting. Full article

Background: Bladder cancer (BC) predominantly affects older patients, and their multidisciplinary treatment often includes surgical intervention. Frailty can influence treatment decisions and is associated with poorer outcomes. This study analyses trends in demographics, treatment patterns and frailty in a large, nationwide, real-world inpatient cohort in Germany. Methods: This retrospective observational study included a total of 49,139 consecutive patients, who received inpatient care for BC at all HELIOS hospitals in Germany between 2016 and 2022. Frailty was assessed using the Hospital Frailty Risk Score (HFRS) and categorised as low (<5), intermediate (5–15), or high (>15). Trends in HFRS, treatment modalities, and demographic variables were analysed using regression models and compared between the periods 2016–2019 and 2020–2022. Results: Of the 49,139 patients, 27,979 were treated between 2016–2019 and 21,160 between 2020–2022. Patients treated in the later period were slightly older but had a lower comorbidity index. The proportion of patients with low frailty increased (73.4% vs. 75.5%, p < 0.01), intermediate frailty decreased (23.5% vs. 21.5%, p < 0.01) and the proportion of highly frail patients remained stable at 3.0% (p = 0.95). Rates of transurethral resection declined over time, whereas rates of RC remained stable (p = 0.12). The use of systemic therapy increased (p = 0.003), particularly among low frailty elderly patients. Early intravesical chemotherapy following transurethral resection declined significantly in 2020–2022 (p < 0.001), particularly among elderly patients with high frailty. Mean length of hospital stay decreased by one day, while ICU admission rates and in-hospital mortality remained stable across time periods. Conclusions: This study shows frailty-specific changes in hospitalisation patterns and inpatient management of BC in Germany, underscoring the value of frailty assessment in population-based research. The proportion of patients classified as having low frailty increased over time. Significant changes in the use of intravesical chemotherapy and systemic therapy were associated with frailty. The decline in early intravesical chemotherapy may have implications for recurrence risk and downstream healthcare utilisation. Full article

In this article, a data-driven algorithm is developed to assess the natural frequency and damping ratio of a nonlinear oscillating single-degree-of-freedom (SDOF) system. The algorithm is based on hybrid convolutional–long short-term memory neural networks (CNN-LSTM) that process a short moving window belonging to a free-decay response and provide estimates of both parameters over time. The novelty of the study resides in the fact that the neural network is trained exclusively using synthetic data issued from linear SDOF models. Since the recurrent neural network (RNN) requires relatively small amounts of data to operate effectively, the nonlinear system locally behaves as a quasi-linear model, allowing each data segment to be processed under this assumption. The proposed RecuID tool is experimentally validated on a laboratory-scale nonlinear SDOF system. To demonstrate its effectiveness, it is compared to conventional identification algorithms. The experimental study yields a maximum mean absolute error (MAE) of 0.244 Hz for the natural frequency and 0.015 for the damping ratio. RecuID proves to be a faster and more robust methodology, capable of handling time-varying damping ratios up to 0.2 and a wide range of natural frequencies defined relative to the sampling rate. Full article

Background: Human noroviruses are the leading cause of foodborne gastroenteritis worldwide. Accumulating evidence suggests that food matrices containing fucosylated or histo-blood group antigen (HBGA)-like glycans may facilitate viral attachment and persistence, yet the molecular mechanisms underlying these interactions remain unclear. Methods: In this study, we performed a comparative computational analysis of norovirus–glycan interactions by integrating AlphaFold3-based structure prediction, molecular docking, and molecular dynamics simulations. A total of 182 P-domain models representing all genotypes across five human norovirus genogroups (GI, GII, GIV, GVIII, and GIX) were predicted and docked with a lettuce-derived core-fucosylated hexasaccharide (H₂N₂F₂) previously identified by our group. The three complexes exhibiting the most favorable docking energies were further examined using 40 ns molecular dynamics simulations, followed by MM/GBSA binding free energy calculations and per-residue decomposition analyses. Results: Docking results indicated that the majority of modeled P proteins were able to adopt energetically favorable interaction poses with H₂N₂F₂, with predicted binding energies ranging from −3.7 to −7.2 kcal·mol⁻¹. The most favorable docking energies were observed for GII.6_S9c_KC576910 (−7.2 kcal·mol⁻¹), GII.3_MX_U22498 (−7.1 kcal·mol⁻¹), and GII.4_CARGDS11182_OR700741 (−6.8 kcal·mol⁻¹). Molecular dynamics simulations suggested stable ligand engagement within canonical HBGA-binding pockets, with recurrent residues such as Asp374, Gln393, and Arg345 contributing to electrostatic and hydrophobic interactions, consistent with previously reported HBGA-binding motifs. MM/GBSA analyses revealed comparatively favorable binding tendencies among these complexes, particularly for globally prevalent genotypes including GII.3, GII.4, and GII.6. Conclusions: This work provides a large-scale structural and energetic assessment of the potential interactions between a naturally occurring lettuce-derived fucosylated hexasaccharide and human norovirus P domains. The results support the notion that core-fucosylated food-associated glycans can serve as interaction partners for diverse norovirus genotypes and offer comparative molecular insights into glycan recognition patterns relevant to foodborne transmission. The integrative AlphaFold3–docking–dynamics framework presented here may facilitate future investigations of virus–glycan interactions within food matrices. Full article

In the modern era, demand forecasting enhances the decision-making tasks of industries for controlling production planning and reducing inventory costs. However, the dynamic nature of the fashion and apparel retail industry necessitates precise demand forecasting to optimize supply chain operations and meet customer expectations. Consequently, this research proposes the Formicary Zebra Optimization-Based Distributed Attention-Guided Convolutional Recurrent Neural Network (FZ-DACR) model for improving the demand forecasting. In the proposed approach, the combination of the Formicary Zebra Optimization and Distributed Attention mechanism enabled deep learning architectures to assist in capturing the complex patterns of the retail sales data. Specifically, the neural networks, including convolutional neural networks (CNNs) and recurrent neural networks (RNNs), facilitate extracting the local features and temporal dependencies to analyze the volatile demand patterns. Furthermore, the proposed model integrates visual and textual data to enhance forecasting accuracy. By leveraging the adaptive optimization capabilities of the Formicary Zebra Algorithm, the proposed model effectively extracts features from product images and historical sales data while addressing the complexities of volatile demand patterns. Based on extensive experimental analysis of the proposed model using diverse datasets, the FZ-DACR model achieves superior performance, with minimum error values including MAE of 1.34, MSE of 4.7, RMS of 2.17, and R² of 93.3% using the DRESS dataset. Moreover, the findings highlight the ability of the proposed model in managing the fluctuating trends and supporting inventory and pricing strategies effectively. This innovative approach has significant implications for retailers, enabling more agile supply chains and improved decision making in a highly competitive market. Full article

Recurrent endometrial cancer (EC) has limited therapeutic options beyond platinum-based chemotherapy, highlighting the need to identify exploitable molecular vulnerabilities. Tumors with high genomic instability, including microsatellite instability-high (MSI-h) or copy-number-high (CNH) ECs, rely on the ATR-CHK1 signaling pathway to tolerate replication stress and maintain genome integrity, making this pathway an attractive therapeutic target. However, acquired resistance to ATR and CHK1 inhibitors (ATRi/CHK1i) often develops, and the transcriptomic basis of this resistance in EC remains unknown. Here, we established isogenic ATRi- and CHK1i-resistant cell line models from MSI-h (HEC1A) and CNH (ARK2) EC lineages and performed baseline transcriptomic profiling to characterize stable resistance-associated states. MSI-h-derived resistant clones adopted a unified transcriptional state enriched for epithelial-mesenchymal transition, cytokine signaling, and interferon responses, while ATRi-resistant models showing additional enrichment of developmental and KRAS/Notch-associated pathways. In contrast, CNH-derived resistant clones diverged by inhibitor class, with ATRi resistance preferentially enriching proliferation-associated pathways and CHK1i resistance inducing interferon signaling. Notably, THBS1, EDN1, and TENM2 were consistently upregulated across all resistant models relative to parental lines. Together, these findings demonstrate that acquired resistance to ATRi and CHK1i in EC is shaped by both lineage and inhibitor class and provide a transcriptomic framework that may inform future biomarker development and therapeutic strategies. Full article

Land abandonment and recurrent wildfires are major drivers of landscape transformation in Mediterranean Europe, creating favorable conditions for the spread of non-native invasive woody species. Among these, Australian wattles (genus Acacia) are particularly widespread and problematic in Portugal. This work analyzed the spatiotemporal dynamics of Acacia spp. in two municipalities of central Portugal (Sertã and Pedrógão-Grande) by combining multitemporal photointerpretation of aerial imagery (2004–2021), generalized additive models (GAMs), and local perception surveys. Results reveal a 417% increase in occupied area over the last two decades. Modeling outcomes indicate a temporal shift in invasion drivers: from an establishment phase (2004–2010), mainly constrained by altitude and proximity to primary introduction sites, to a disturbance-driven expansion phase (2010–2021), influenced by fire recurrence, slope, and land-use context. Spatial clustering persisted throughout, underscoring the role of founder populations. Surveys confirmed high public awareness of Acacia invasiveness and identified abandonment and wildfire as the main perceived triggers of spread. By integrating ecological and social dimensions, this study provides a socioecological perspective on Acacia spp. expansion in Mediterranean rural landscapes and highlights the urgent need for integrated, landscape-scale management strategies. Full article

Cancer stem cells (CSCs) represent a small but highly resilient tumor subpopulation responsible for sustained growth, metastasis, therapeutic resistance, and recurrence. Their survival is supported by aberrant activation of developmental and inflammatory pathways, including Wnt/β-catenin, Notch, Hedgehog, PI3K/Akt/mTOR, STAT3, and NF-κB, as well as epithelial–mesenchymal transition (EMT) programs and niche-driven cues. Increasing evidence shows that phytochemicals, naturally occurring bioactive compounds from medicinal plants, can disrupt these networks through multi-targeted mechanisms. This review synthesizes current findings on prominent phytochemicals such as curcumin, sulforaphane, resveratrol, EGCG, genistein, quercetin, parthenolide, berberine, and withaferin A. Collectively, these compounds suppress CSC self-renewal, reduce sphere-forming capacity, diminish ALDH⁺ and CD44⁺/CD24⁻ fractions, reverse EMT features, and interfere with key transcriptional regulators that maintain stemness. Many phytochemicals also sensitize CSCs to chemotherapeutic agents by downregulating drug-efflux transporters (e.g., ABCB1, ABCG2) and lowering survival thresholds, resulting in enhanced apoptosis and reduced tumor-initiating potential. This review further highlights the translational challenges associated with poor solubility, rapid metabolism, and limited bioavailability of free phytochemicals. Emerging nanotechnology-based delivery systems, including polymeric nanoparticles, lipid carriers, hybrid nanocapsules, and ligand-targeted formulations, show promise in improving stability, tumor accumulation, and CSC-specific targeting. These nanoformulations consistently enhance intracellular uptake and amplify anti-CSC effects in preclinical models. Overall, the consolidated evidence supports phytochemicals as potent modulators of CSC biology and underscores the need for optimized delivery strategies and evidence-based combination regimens to achieve meaningful clinical benefit. Full article

Glioblastoma (GB) is one of the most aggressive and invasive cancers. Current treatment protocols for GB include surgical resection, radiotherapy, and chemotherapy with temozolomide. However, despite these treatments, physicians still struggle to effectively image, diagnose, and treat GB. As such, patients frequently experience recurrence of GB, demanding innovative strategies for early detection and effective therapy. Bioconjugated quantum dots (QDs) have emerged as powerful nanoplatforms for precision imaging and targeted drug delivery due to their unique optical properties, tunable size, and surface versatility. Due to their extremely small size, QDs can cross the blood–brain barrier and be used for precision imaging of GB. This review explores the integration of QDs with click chemistry for robust bioconjugation, focusing on artificial intelligence (AI) to advance GB therapy, mechanistic insights into cellular uptake and signaling, and strategies for mitigating toxicity. Click chemistry enables site-specific and stable conjugation of targeting ligands, peptides, and therapeutic agents to QDs, enhancing selectivity and functionalization. Algorithms driven by AI may facilitate predictive modeling, image reconstruction, and personalized treatment planning, optimizing QD design and therapeutic outcomes. We discuss molecular mechanisms underlying interactions of QDs with GB, including receptor-mediated endocytosis and intracellular trafficking, which influence biodistribution and therapeutic efficacy. Use of QDs in photodynamic therapy, which uses reactive oxygen species to induce apoptotic cell death in GB cells, is an innovative therapy that is covered in this review. Finally, this review addresses concerns associated with the toxicity of metal-based QDs and highlights how QDs can be coupled with AI to develop new methods for precision imaging for detecting and treating GB for induction of apoptosis. By converging nanotechnology and computational intelligence, bioconjugated QDs represent a transformative platform for paving a safer path to smarter and more effective clinical interventions of GB. Full article

Generative Artificial Intelligence (GenAI) systems have introduced new classes of security incidents that traditional response frameworks were not designed to manage, ranging from model manipulation and data exfiltration to misinformation cascades and prompt-based privilege escalation. This study proposes a Practical Incident-Response Framework for Generative AI Systems (GenAI-IRF) that bridges established cybersecurity standards with emerging AI assurance principles. Using a Design Science Research (DSR) approach, this study identifies six recurrent incident archetypes and formalises a structured playbook aligned with NIST SP 800-61r3, NIST AI 600-1, MITRE ATLAS, and OWASP LLM Top-10. The artefact was evaluated in controlled scenarios using scenario-based simulations and expert reviews involving AI-security practitioners from academia, finance, and technology sectors. The results suggest high inter-rater reliability (κ = 0.88), strong usability (SUS = 86.4), and improved incident resolution times compared to baseline procedures. The findings demonstrate how traditional response models can be adapted to GenAI contexts using taxonomy-driven analysis, artefact-centred validation, and practitioner feedback. This framework provides a practical foundation for security teams seeking to operationalise AI incident response and contributes to the emerging body of work on trustworthy and resilient AI systems. Full article