Search Results (243)

With the increasing demand for enhanced security and surveillance, the integration of multimodal AI has shown significant promise. We developed and fine-tuned a transformer-based model, the Large Language and Vision Assistant–OneVision, tailored for human event and action recognition. By utilizing a multimodal approach, we identified specific human actions, including eating, running, fighting, sitting, and sleeping, within diverse real-world settings. Through knowledge distillation and Low-Rank Adaptation, the model’s performance was optimized in demonstrating substantial improvements in context-aware recognition and response generation. Evaluation results showed recall-oriented understudy for obtaining evaluation (ROUGE)-1 score of 0.6844, ROUGE-2 score of 0.5751, ROUGE-L score of 0.6520, and the bilingual evaluation understudy score of 68.20, demonstrating significant gains in accuracy and interpretability. The model’s success highlights its potential for real-time applications in surveillance, healthcare, and interactive AI systems, providing reliable, efficient, and context-sensitive human action detection. Full article

This review focuses on electromagnetic imaging methods widely used in urban geophysics and civil engineering. The rapid growth of the urban population and the increase in the frequency of extreme events related to climate change make novel approaches to the geophysical monitoring of urban areas and civil infrastructures essential in the context of programs for the sustainability and resilience of cities. In this scenario, there is a growing interest in using ground-based electromagnetic methods to investigate strategic infrastructures such as bridges, tunnels, dam embankments, power plants, energy plants and pipelines in a non-invasive way. The development of cost-effective, user-friendly sensor arrays, robust methodologies for tomographic data inversion, and AI-based and machine learning techniques has rapidly transformed these methods. This review critically analyzes the results relating to the application of ground-based electromagnetic methods in infrastructure monitoring and surveillance over the past 20 years by presenting a selection of best practice examples and studies planned to support programs for the resilience and maintenance of engineering infrastructures. The analysis reveals that these methods are highly effective in addressing a broad spectrum of monitoring issues in view of effective maintenance of civil infrastructures. In fact, these methods are essential for detecting the geometry of buried objects (e.g., bars and voids), enabling the early detection of degradation phenomena, and mapping water infiltration processes inside structures, as well as many other challenging applications. Finally, prospectives for development are identified in terms of using soft robot technologies, miniaturized sensors, and AI-based methods to acquire, process and interpret data as well as to design smart operational guidelines for infrastructure management. Full article

Robust vehicle detection in real-world traffic surveillance remains challenging due to degraded imagery caused by motion blur, adverse weather, and low illumination, which significantly increases detector sensitivity to hyperparameter configurations. This study proposes a “Frugal AI” distributed multi-GPU framework that optimizes hyperparameters via a stochastic simplex-based search coupled with five-fold cross-validation. Utilizing three low-cost NVIDIA GTX 1050 Ti GPUs, the framework performs parallel candidate exploration with an asynchronous model-level exchange mechanism to escape local optima without the overhead of gradient synchronization. Seven CNN backbones—VGG16, VGG19, GoogLeNet, MobileNetV2, ResNet18, ResNet50, and ResNet101—were evaluated within YOLOv2 and Faster R-CNN detectors. To address memory constraints (4 GB VRAM), YOLOv2 was selected for extensive benchmarking. Performance was measured using a harmonic precision–recall-based cost metric to strictly penalize imbalanced outcomes. Experimental results demonstrate that under identical wall-clock time budgets, the proposed framework achieves an average 1.38% reduction in aggregated cost across all models, with the highly sensitive VGG19 backbone showing a 4.00% improvement. Benchmarking against Bayesian optimization, genetic algorithms, and random search confirms that our method achieves superior optimization quality with statistical significance ($p < 0.05$). Under a rigorous IoU = 0.75 threshold, the optimized models consistently yielded F1-scores 0.8444 $\pm$ 0.0346. Ablation studies further validate that the collaborative model exchange is essential for accelerating convergence in rugged loss landscapes. This research offers a practical, scalable, and cost-efficient solution for deploying robust AI surveillance in resource-constrained smart city infrastructure. Full article

Effective port and coastal surveillance require persistent monitoring, flexible deployment, and reliable target detection in dynamic maritime environments. This paper presents a system- and deployment-oriented autonomous tethered drone architecture, integrated with AI-based perception, for persistent maritime surveillance in ports and coastal areas. Mounted on a moving maritime platform and powered through a tether, the drone provides a persistent elevated viewpoint without the endurance limitations of conventional battery-powered Unmanned Aerial Vehicles (UAVs). The system combines maritime platform integration, tethered flight operation, fail-safe and safety mechanisms, and a distributed Artificial Intelligence (AI) pipeline for real-time object detection and tracking. The perception module is based on YOLOv8m for vessel detection and BoT-SORT for multi-object tracking, enabling continuous monitoring of maritime targets in realistic operational scenarios. Field trials conducted from moving vessels in maritime environments demonstrate autonomous take-off and landing, stable surveillance operation under realistic wind and wave conditions, and effective vessel detection and tracking on real image sequences. The results show the potential of AI-enabled tethered drone surveillance as a persistent and operationally relevant tool for maritime monitoring and security. Full article

Background/Objectives: Intrapulmonary metastasis (IPM) after an initial single primary lung cancer (SPLC) is an adverse follow-up pattern; however, when studying population-based longitudinal records, the determinants remain unclear. We aimed to identify factors associated with subsequent IPM after initial SPLC using artificial intelligence (AI)-driven analytical approaches. Methods: We used Surveillance, Epidemiology, and End Results (SEER) lung cancer records from 2000 to 2019. Adults with at least two records were restricted to those with SPLC at the first record. Outcome at the second record was registry-classified IPM versus persistent SPLC. A machine learning framework based on random forest models was developed using baseline variables, first record characteristics, and the interval between records. Temporal validation was performed by training on cases from 2000 to 2013 and testing on cases from 2014 to 2019. A dynamic Bayesian network (DBN) supported simulated intervention (SI) analyses to estimate model-implied risk ratios (RRs) with 95% confidence intervals (CIs). Results: Among 3450 patients, 361 had registry-classified IPM at the second record. The random forest model achieved an area under the curve (AUC) of 0.852 in internal validation and 0.929 in temporal validation. Surgery and record timing were the leading predictors. The DBN retained surgery as the only direct parent and achieved an AUC of 0.779. SI analyses showed higher IPM probability for pleural invasion level (PL) 3 versus PL 0, RR 1.378 (95% CI, 1.080–1.657). Lobectomy with mediastinal lymph node dissection versus wedge resection lowered the IPM probability, RR 0.378 (95% CI, 0.219–0.636). Conclusions: AI-based time-sequence modeling integrating machine learning and a DBN allowed for the identification of surgery, pleural invasion, and record timing as key factors associated with subsequent IPM classification after initial SPLC. This framework demonstrates the potential of combining predictive and probabilistic dependency modeling to investigate registry-based disease classification patterns, and may support hypothesis generation for future prospective studies. Full article

Background: Current artificial intelligence (AI) systems in gastrointestinal (GI) endoscopy primarily emphasize binary detection or static classification, providing limited support for the graded assessment of malignant potential that underpins clinical decision-making. We developed GastroMalign, a transformer-based framework designed to stratify GI lesions according to ordinal disease severity while maintaining clinical interpretability, addressing this unmet need in endoscopic risk assessment. Methods: This retrospective development and validation study used the publicly available GastroVision dataset, comprising 8000 de-identified endoscopic still images from the upper and lower gastrointestinal tract, including the esophagus, stomach, duodenum, colon, rectum, and terminal ileum. GastroMalign integrates a Vision Transformer (ViT) encoder with a Sequential Feature Learner that explicitly models ordinal disease severity along a benign-to-malignant spectrum. The framework produces both categorical risk classification and a continuous malignancy risk score. Images were stratified into training (80%), validation (10%), and test (10%) sets. Performance was compared with convolutional neural network (CNN) baselines and a Swin Transformer. Interpretability was assessed using Score-CAM visualizations reviewed by blinded expert endoscopists. Results: On the held-out test set (n = 800 images), GastroMalign achieved an overall accuracy of 80.06%, precision of 79.65%, recall of 80.06%, and F1-score of 79.17%, with a micro-averaged AUC of 0.98. In comparison, ResNet-50 and DenseNet-121 achieved accuracies of 32.42% and 36.77%, respectively, while the Swin Transformer achieved 60.56% accuracy (AUC = 0.93). Ablation analyses demonstrated a 17% absolute reduction in High-Risk lesion recall when the progression-aware module was removed. Continuous malignancy risk scores increased monotonically across ordinal classes, with mean values < 0.18 for Benign and >0.72 for High-Risk/Malignant lesions. Score-CAM visualizations demonstrated 92% overlap with clinician-annotated lesion regions. Conclusions: GastroMalign delivers an interpretable, progression-aware AI framework for GI lesion risk stratification that outperforms existing CNN- and transformer-based models. Clinically, GastroMalign is intended as an adjunct decision-support tool during endoscopic review to standardize lesion risk stratification (benign to malignant spectrum), support management decisions (biopsy vs. resection vs. surveillance), and reduce operator-dependent variability by pairing ordinal risk outputs with interpretable visual explanations. Full article

Accurate identification of disease vectors is crucial for public health, yet distinguishing morphologically similar species demands significant taxonomic expertise and data resources. This study proposes MVP-Net, an AI-driven framework designed to extract a minimal sufficient set of diagnostic anatomical views from multi-view imagery for efficient identification. The framework was evaluated on regionally collected datasets of Calyptratae (8 views) and Culicidae (11 views) from routine surveillance in Shanghai. Under all-view fusion, MVP-Net achieved Top-1 accuracies of 87.04% for Calyptratae and 100% for Culicidae. After Pareto-based view optimization, the required input was reduced to 5 views for Calyptratae and 2 views for Culicidae, lowering computational cost by 37.49% and 81.82%, respectively, while retaining comparable classification performance (86.11% for the recommended Calyptratae configuration and 100% for the recommended Culicidae configuration). These results show that MVP-Net can reduce view redundancy while preserving comparable identification performance within the current Shanghai surveillance setting, providing a practical approach for optimizing regional multi-view auxiliary identification workflows. Full article

Modeling the complex spatio-temporal dynamics of infectious diseases presents a significant computational challenge due to heterogeneous regional interactions, high-dimensional multimodal data streams, and the critical need for algorithmic fairness. This paper proposes a novel computational framework that unifies graph-based spatio-temporal forecasting, anomaly detection, and retrieval-augmented generation (RAG) into a single mathematical architecture. The predictive backbone employs a graph-augmented Temporal Fusion Transformer to capture non-linear temporal dependencies and spatial disease propagation. By formalizing regional topology and mobility flows as a weighted mathematical graph, the model systematically integrates structured epidemiological counts, continuous environmental covariates, and digital trace signals. To address algorithmic bias, we formulate a fairness-aware optimization problem by embedding a specific regularization term into the training objective, which mathematically penalizes disparities in true-positive rates across diverse socio-demographic strata. Furthermore, the numerical outputs and anomaly scores are processed by a large language model equipped with hybrid dense and sparse retrieval to generate interpretable, computationally grounded decision support. Extensive experiments on a longitudinal dataset comprising 62 administrative regions over 260 weeks validate the mathematical robustness of the proposed framework. The graph-augmented architecture improved forecasting accuracy by up to 24% and anomaly detection F1 scores by over 6% compared to state-of-the-art deep learning baselines, while the fairness-regularized loss function reduced the maximum subgroup recall gap by more than 50%. These findings demonstrate that predictive accuracy and algorithmic fairness can be jointly optimized, providing a rigorous computational methodology for spatio-temporal epidemic modeling and AI-driven surveillance. Full article

Carbapenem-resistant Gram-negative infections have become one of the most formidable challenges in intensive care units (ICUs). Critically ill patients—often exposed to invasive procedures, prolonged hospitalization, and broad-spectrum antibiotics—are highly susceptible to infections by carbapenem-resistant Enterobacterales (CRE), Pseudomonas aeruginosa (CRPA), and Acinetobacter baumannii (CRAB). These pathogens are associated with mortality exceeding 40%, prolonged ICU stays, and increased healthcare costs. Therapeutic advances have reshaped management in recent years. New β-lactam/β-lactamase inhibitor combinations—ceftazidime–avibactam, meropenem–vaborbactam, imipenem–relebactam, and sulbactam–durlobactam—along with cefiderocol, have provided safer and more effective alternatives to previously used regimens. Yet, none are universally effective, particularly against carbapenemase-producing organisms, especially metallo-β-lactamase (MBL) producers, and resistance may still emerge during treatment. Rapid molecular and phenotypic diagnostics, when integrated into antimicrobial stewardship, have improved early therapy alignment and reduced unnecessary broad-spectrum use. Beyond antibiotics, colonization surveillance and infection control remain pivotal, as colonization often precedes invasive infection. Biofilm formation on devices such as endotracheal tubes and catheters further promotes persistence and relapse. Strategies targeting biofilm disruption, improved dosing guided by pharmacokinetic/pharmacodynamic optimization, and therapeutic drug monitoring are crucial in ICU practice. The future of managing these infections will depend on integrating precision tools—rapid diagnostics, mechanism-based therapy, and stewardship-guided decisions—with emerging treatments and adjunctive options such as immunomodulators, bacteriophages, and AI-driven decision support. Continued research in ICU-specific populations, especially regarding pharmacokinetics in patients on ECMO or CRRT, is urgently needed. In summary, while the therapeutic landscape for carbapenem-resistant Gram-negative infections has evolved substantially, sustained success will rely on a multifaceted strategy combining innovation, precision, and prevention to improve outcomes for the most vulnerable patients. Full article

Unmanned Aerial Vehicles (UAVs) are increasingly integrated into critical applications ranging from logistics and agriculture to defence and security operations, surveillance and emergency response. At the core of these systems lies the communication link between the UAV and its ground control station (GCS), which serves as the backbone for command, control and data exchange. However, communications links remain highly vulnerable to cyber-threats, including eavesdropping, signal falsification, radio frequency interference (RFI) and hijacking. These risks highlight the urgent need for secure communication protocols and effective defence mechanisms capable of protecting data confidentiality, integrity, availability and authentication. This study performs a comprehensive survey of secure UAV-GCS communication protocols and artificial intelligence (AI)-driven intrusion detection techniques. Initially, we review widely used communication protocols, examining their security features, vulnerabilities and existing countermeasures. Accordingly, a taxonomy of UAV-GCS security threats is proposed, structured around confidentiality, integrity, availability and authentication and map these threats to relevant attacks and defences. In parallel, our study examines state-of-the-art intrusion detection systems for UAVs, while particular emphasis is placed on emerging methods such as deep learning, federated learning, tiny machine learning and explainable AI, which hold promise for lightweight and real-time threat detection. The survey concludes by identifying open challenges, including resource constraints, lack of standardised secure protocols, scarcity of UAV-specific datasets and the evolving sophistication of attackers. Finally, we outline research directions for next-generation UAV architectures that integrate secure communication protocols with AI-based anomaly detection to achieve resilient and intelligent drone ecosystems. Full article

Inherited cardiomyopathies associated with high-risk genotypes, are characterized by a disproportionate risk of malignant ventricular arrhythmias and sudden cardiac death, often independent of left ventricular systolic dysfunction or advanced structural remodeling. Traditional surveillance strategies based on intermittent electrocardiography and phenotype-driven risk assessment are insufficient to capture the dynamic and often silent progression of electrical instability in these populations. This narrative review evaluates the emerging role of artificial intelligence (AI)-enabled sensor technologies in remote arrhythmic monitoring of genetically defined cardiomyopathy cohorts. Wearable ECG devices, implantable cardiac monitors, multisensor cardiac implantable electronic device algorithms, pulmonary artery pressure sensors, and contact-free systems enable continuous acquisition of electrophysiological and hemodynamic data, generating digital biomarkers that may reflect early arrhythmic vulnerability and subclinical decompensation. AI-driven analytics enhance signal processing, automated event detection, and remote data triage, with the potential to reduce clinical workload while preserving diagnostic sensitivity. However, current evidence predominantly derives from heterogeneous heart failure or general arrhythmia populations, and prospective validation in genotype-specific cohorts remains limited. Key challenges include algorithm generalizability, signal quality in ambulatory environments, data governance, interpretability of AI models, and integration into structured remote-care pathways. The convergence of genotype-informed risk stratification and multimodal AI-enabled sensing represents a promising strategy to transition from reactive device-based protection to proactive, precision-guided arrhythmic prevention. Dedicated genotype-focused studies and standardized digital endpoints are required to support safe and effective implementation in inherited cardiomyopathies. Full article

Background: Endoscopic assessment is central to the management of inflammatory bowel disease (IBD), particularly within treat-to-target strategies. However, conventional high-definition white-light endoscopy (HD-WLE) is limited by interobserver variability and its inability to reliably reflect microscopic inflammation or predict long-term outcomes. Over the last decade, multiple technological innovations have reshaped the role of endoscopy in both disease activity monitoring and dysplasia surveillance. Methods: This narrative review provides a comprehensive and clinically oriented overview of emerging endoscopic technologies in IBD, including image-enhanced endoscopy, ultra-high-magnification techniques, artificial intelligence (AI), and molecular imaging. We discuss their diagnostic performance, prognostic implications, and potential integration into clinical practice. Results: Image-enhanced endoscopy improves visualization of subtle mucosal and vascular alterations and demonstrates stronger correlation with histological activity compared with HD-WLE alone. Confocal laser endomicroscopy and endocytoscopy enable in vivo microscopic assessment of epithelial architecture and barrier integrity, redefining remission beyond macroscopic healing. AI systems have shown expert-level performance in grading inflammatory severity in ulcerative colitis and high sensitivity in capsule endoscopy for Crohn’s disease, supporting objective and reproducible assessment. In surveillance, targeted high-definition inspection has replaced random biopsies, while adjunctive optical and AI-based tools enhance lesion detection and characterization. Molecular imaging introduces a predictive dimension by enabling visualization of drug–target engagement and dysplasia-specific pathways. Conclusions: Endoscopy in IBD is evolving from a descriptive modality toward a multimodal precision tool integrating enhanced imaging, AI-driven standardization, and molecular profiling. Although further validation and cost-effectiveness studies are required, these innovations have the potential to improve therapeutic stratification, surveillance strategies, and long-term patient outcomes. Full article

The growing demand for high-fidelity, multi-parameter, distributed sensing in critical domains such as structural health monitoring, oil and gas exploration, and secure perimeter surveillance is pushing traditional optical fiber sensors (OFS) to their performance limits. Although conventional multiplexing techniques such as time-division and wavelength-division multiplexing (TDM, WDM) have been commercially successful, they are rapidly approaching fundamental bottlenecks in sensor density, spatial resolution, and data capacity. This review argues that the synergistic convergence of space-division multiplexing (SDM) and artificial intelligence (AI) represents a paradigm shift, enabling a new generation of intelligent, high-dimensional sensing networks. We comprehensively survey the state of the art in SDM-based OFS, detailing the operating principles and applications of multi-core fibers (MCFs) for ultra-dense sensor arrays and 3D shape sensing, as well as few-mode fibers (FMFs) for mode-division multiplexing and enhanced multi-parameter discrimination. However, the unprecedented spatial parallelism provided by SDM introduces significant challenges, including inter-channel crosstalk, complex signal demultiplexing, and massive data volumes. This paper systematically explores how AI, particularly machine learning (ML) and deep learning (DL), is being leveraged not merely as a tool but as an indispensable core technology to mitigate these impairments. We critically analyze AI’s role in digital crosstalk suppression, intelligent mode demultiplexing, signal denoising, and solving complex inverse problems for parameter estimation. Furthermore, we highlight how this AI–SDM synergy enables capabilities beyond the reach of either technology alone, such as super-resolution sensing and predictive analytics. The discussion is extended to include the critical supporting pillars of this ecosystem, such as advanced interrogation techniques and the associated data management challenges. Finally, we provide a forward-looking perspective on the trajectory of the field, outlining a path toward cognitive sensing networks that are self-calibrating, adaptive, and capable of autonomous decision-making. This review is intended to serve as a foundational reference for researchers and engineers at the intersection of photonics and intelligent systems, illuminating the pathway toward tomorrow’s intelligent sensing infrastructure. Full article

The rapid deployment of surveillance cameras in urban, industrial, and domestic environments has intensified the need for intelligent systems capable of analyzing video streams beyond the limitations of single-camera setups. Unlike traditional single-camera approaches, multi-camera systems expand spatial coverage, reduce blind spots, and enable consistent tracking of people and objects across non-overlapping views, thereby improving robustness against occlusions and viewpoint changes. This article presents a comprehensive review of multi-camera vision systems published between 2020 and 2025, covering application domains including public security and biometrics, intelligent transportation, smart cities and IoT, healthcare monitoring, precision agriculture, industry and robotics, pan–tilt–zoom (PTZ) camera networks, and emerging areas such as retail and forensic analysis. The review synthesizes predominant technical approaches, including deep-learning-based detection, multi-target multi-camera tracking (MTMCT), re-identification (Re-ID), spatiotemporal fusion, and edge computing architectures. Persistent challenges are identified, particularly in inter-camera data association, scalability, computational efficiency, privacy preservation, and dataset availability. Emerging trends such as distributed edge AI, cooperative camera networks, and active perception are discussed to outline future research directions toward scalable, privacy-aware, and intelligent multi-camera infrastructures. Full article

Hepatocellular carcinoma (HCC) remains a major cause of cancer-related mortality worldwide, and its management is limited by heterogeneous risk profiles, suboptimal surveillance performance, diagnostic uncertainty in chronically diseased livers, and difficulty individualizing prognosis after treatment. The aim of this narrative review was to critically evaluate artificial intelligence (AI) applications across the HCC care continuum, with emphasis on their intended clinical role, reported performance, evidence maturity, and barriers to implementation. A major strength of this review is that it moves beyond a descriptive catalog of models by structuring the literature around clinically relevant decision points and by explicitly distinguishing emerging proof-of-concept tools from applications with stronger translational potential. Across risk stratification, surveillance, imaging-based diagnosis, pathology, treatment-response prediction, and prognostication, we found that AI consistently demonstrates promise, particularly for identifying patients at higher future HCC risk, improving lesion detection and characterization on ultrasound, CT, MRI, and contrast-enhanced ultrasound, assisting histopathologic classification, and predicting outcomes such as microvascular invasion, recurrence, survival, and response to locoregional therapies. However, we also found that the evidence base remains highly uneven: many diagnostic studies are retrospective and lesion-enriched rather than embedded in true surveillance populations, many prognostic models lack robust external validation and calibration assessment, and reference standards, imaging protocols, and dataset composition vary substantially across studies. These findings are clinically relevant because they highlight both where AI may offer near-term value and why most published systems are not yet ready for routine use. Overall, AI in HCC should be viewed as a rapidly evolving but still transitional field. Its future impact will depend not only on higher-performing algorithms but on clearly defined clinical use cases, multicenter and prospective validation, transparent reporting, workflow-aware evaluation, and implementation strategies that support safe, equitable, and scalable adoption. Full article