搜索结果：Distributed computing

This study proposes a novel framework for network security situational awareness and risk warning in cloud computing environments, integrating adaptive Machine Learning (ML), Hierarchical Multi-Label Classification (HMC), and a dynamic trust evaluation mechanism based on the cloud model. The complexity, diversity, and real-time nature of emerging cyberattacks-such as zero-day exploits, distributed denial-of-service (DDoS), and botnets-pose significant challenges to traditional rule-based and static detection methods. To address these challenges, we developed an effective SDN-based cloud architecture utilizing the Ryu OpenFlow controller and OpenFlow switches. This architecture enables real-time link information collection, dynamic scheduling, and scalable, reliable data transmission. The hierarchical classification framework suggested can break multiclass problems into binary tasks, alleviating the effect of sample imbalance and enhancing the recognition of low-frequency attacks, including User to Root (U2R). Ensemble learning techniques, including AdaBoost and Bagging, further enhance detection accuracy for fine-grained attack types. Experiments conducted on DDoS datasets, cloud traffic data, and simulations in Mininet and EstiNet demonstrate that the combined ML-HMC-trust approach significantly improves detection precision, reduces false positives, and enables real-time response. These results confirm that integrating adaptive learning, hierarchical classification, and dynamic trust evaluation provides a robust and scalable solution for securing large-scale cloud platforms.

Accurate modeling of ultrasound wave propagation is essential for high-fidelity simulation and imaging in ultrasonic testing. A primary challenge lies in characterizing the excitation source, particularly for transducers with large apertures relative to the acoustic wavelengths. In such cases, non-uniform excitation and spatial interference significantly affect the resulting radiation patterns. This paper proposes a distributed source inversion strategy to reconstruct an effective spatio-temporal transducer model that reproduces experimentally measured wavefields. The reconstructed source model captures aperture-dependent phase and amplitude variations without requiring detailed knowledge of the transducer structure. The approach is validated using directivity measurements on an aluminum half-cylinder, where simulations incorporating the reconstructed source model show close agreement with experimental directivity patterns and waveform shapes. Finally, synthetic studies on reverse time migration and full-waveform inversion demonstrate that accurate transducer modeling is critical for the success of simulation-based imaging and inversion workflows and significantly improves reconstruction quality.

Objective.Biological neural networks (BNNs) are composed of interconnected neurons, where a single neuron may connect to thousands of other neurons, and this interregional connectivity enables distributed computation across multiple nodes. Biocomputing exploits these inherent processing capabilities of neural tissue for computational tasks. However, mostin vitrobiocomputing systems rely on isolated neural cultures that may not be suitable for complex multi-network processing tasks. Building scalable biocomputing architectures requires reliable communication frameworks between physically separated biological processing units. Structured information exchange across distinct BNNs remains largely unexplored.Approach.virtual white matter (VWM) is a previously established platform that enables real-time functional connectivity between neural cultures in separate microelectrode array (MEA) dishes. In this study, we expanded the VWM platform to enable structured binary communication between dissociated cortical cells cultured in separate MEAs implemented on MED64 MEAs with Intan RHS stimulation/recording hardware and a custom real-time decoding pipeline. The system encodes 3-bit data words using spatiotemporal electrical stimulation patterns, with designated electrodes representing binary states; evoked responses from selected output electrodes are decoded in real time using machine learning, and parity-based error correction is applied to enhance transmission fidelity. We validated the framework through unidirectional and bidirectional communication experiments across multiple independent neural preparation pairs.Main Results.The VWM system successfully transmitted 3-bit data packets between physically separate neural cultures, achieving individual bit decoding accuracies of 75%-90% and aggregate word accuracy exceeding 52%, well above the 12.5% chance level and an optimal-threshold spike-count baseline. Parity-based error correction with confidence-based bit toggling enhanced transmission fidelity, enabling reliable word-level communication. In bidirectional experiments, end-to-end round-trip accuracy reached approximately 20%, remaining above chance.Significance.This work demonstrates that physically separated neural culturescan reliably exchange structured information through engineered communication frameworks. By establishing reliable structured communication between living neural networks, VWM provides a foundational framework for scalable, distributed biocomputing architectures that may eventually support multi-networkin vitromodels of neural computation.

Shared immersive environment sports venues, virtual classrooms, and collaborative workspaces require multiple users to stream 360° videos simultaneously over the same edge network, yet every existing adaptive bitrate system optimises each viewer in isolation. This self-interested behaviour triggers a bandwidth auction that chronically starves the most uncertain viewers: Jain's Fairness Index for ten independently optimised agents routinely falls below 0.85. We present FairEdge360, a hierarchical multi-agent reinforcement learning framework that reformulates multi-user 360° streaming as a Decentralised Partially Observable Markov Decision Process (Dec-POMDP) and proves, formally, that fairness and quality are complementary rather than competing objectives. Three tightly coupled innovations make this possible. First, a Lightweight Uncertainty Estimator (LUE) a compact 8385-parameter four-layer MLP evaluates per-device viewport prediction confidence cti=σ(w4⊤h3) in under approximately 2.1 ms on commodity smartphones (95th percentile, iPhone 12 A14 Bionic), enabling selective edge offloading that reduces device energy consumption by 38.9%. Second, a variational Graph Neural Network compresses each agent's 256-dimensional GRU state into a 32-byte INT8 latent, transmitted over a dynamic RTT-gated neighbourhood graph at 96 bytes per agent per 500 ms 75% less overhead than competing approaches. Third, the edge coordinator maximises the Nash social welfare objective NSW=(∏i=1NQi)1/N, whose gradient ∂NSW/∂Qi∝1/Qi automatically prioritises the most disadvantaged viewer; a formal proof guarantees that every Pareto-optimal policy satisfies Qi/∑jQj≥1/N. Counterfactual advantage estimation correctly attributes each agent's marginal contribution to the global reward, eliminating the credit-assignment ambiguity inherent in standard multi-agent baselines. Evaluated on 284 users, 52 omnidirectional videos, and 10,000 real network traces spanning 4G LTE, 5G mmWave, HSDPA, and campus WiFi, FairEdge360 raises Jain's Fairness Index from 0.934 to 0.976 (+4.5%), improves worst-case user quality-of-experience from MOS 2.54 to MOS 3.21 (+26.4%), and halves rebuffering rate from 2.1% to 1.1%, all within a 20 ms motion-to-photon budget on a commodity smartphone.

The modern power system is characterized by large-scale networks, diverse types of sources and loads, and complex grid structures. Virtual Power Plants (VPPs) are proposed to address the operation problem after the integration of Distributed Energy Resources (DERs). Optimization problems in the VPP operation are predominantly mixed-integer programming (MIP) problems belonging to the class of NP-hard problems, motivating the application of quantum computers. Focusing on the VPP optimal dispatch problem under wind and solar uncertainty, we employ the Model Predictive Control (MPC) framework to conduct the VPP intraday rolling dispatch. The classical model and the Quadratic Unconstrained Binary Optimization (QUBO) model for the MPC-based intraday rolling dispatch problem are formulated, respectively. The QUBO formulation of the VPP dispatch problem renders it directly solvable by a specialized quantum computer based on dissipative optical systems: the Coherent Ising Machine (CIM). Compared with the benchmark classical solvers, the experimental results demonstrate the significant computational time reduction capability of CIM. Specifically, compared to Gurobi, Simulated Annealing and Tabu Search, the CIM achieves relative computational time reductions of 75.25%, 99.95% and 99.96%, respectively, while maintaining competitive solution quality. Our work demonstrates the applicability of CIM and its acceleration potential in VPP intraday rolling dispatch, paving the way for the practical application of specialized photonic quantum computers in smart grids.

In light of the rapidly growing large-scale data in federated ecosystems, the traditional principal component analysis (PCA) is often not applicable due to privacy protection considerations and large computational burden. Algorithms were proposed to lower the computational cost, but few can handle both high dimensionality and massive sample size under distributed settings. In this paper, we propose the FAst DIstributed (FADI) PCA method for federated data when both the dimension d and the sample size n are ultra-large, by simultaneously performing parallel computing along d and distributed computing along n . Specifically, we utilize L parallel copies of p -dimensional fast sketches to divide the computing burden along d and aggregate the results distributively along the split samples. We present a general framework applicable to multiple statistical problems, and establish comprehensive theoretical results under the general framework. We show that FADI accelerates the computation while enjoying the same non-asymptotic error rate as the traditional PCA when L p ≥ d . We also derive inferential results that characterize the asymptotic distribution of FADI, and show a phase-transition phenomenon as L p increases. We perform extensive simulations to empirically validate our theoretical findings, and apply FADI to the 1000 Genomes data to study the population structure.

Engineering microbial consortia to perform robust population-level computations requires principled ways to achieve majority consensus in the presence of multiscale stochasticity. We develop a hybrid framework that couples time-varying intracellular gene-expression dynamics, including burst-like jump noise, to population-level birth, death, and collision reactions through the random time-change representation. Birth and death hazards are allowed to depend on evolving intracellular states through bounded hazard maps, while interspecific collisions are modeled by biologically interpretable SD or NSD encounter rules. The construction uses instantaneous hazard integrals, rather than frozen-rate SSA propensities, so that reaction times remain consistent with continuously evolving intracellular states. Under explicit bounded-per-capita hazard assumptions, we prove order-wise sufficient majority thresholds for two biologically motivated collision rules. With self-destructive collisions, where one cell of each species is removed per encounter, a new potential-supermartingale argument controls temporary excursions of the minority population above a given level and shows that only O(log n) non-collision events occur before consensus with high probability. Consequently, an Ω(log n) initial gap suffices for correctness with high probability. With non-self-destructive collisions, where only one side dies per encounter, the species-count difference after subtracting its predictable drift is a bounded-increment martingale at embedded gap-changing reaction times. Thus, if the number of such steps is O(nlog n) with high probability and the predictable drift is neutral or favorable to the initial majority, an Ω(nlogn) initial gap suffices. Heavy-tailed intracellular fluctuations alter event timing and consensus-time distributions, but not these order-wise sufficient gap scales, provided hazard maps are bounded. We also establish non-explosion of the hybrid process, clarify the distinction between the exact random-time-change model and numerical simulation, and treat representative-pool hazard approximations as simulation accelerations rather than assumptions used in the proofs. Expanded numerical diagnostics test the SD non-collision bound, the NSD step-count assumption, multi-size NSD normalized transitions, heavy-tail versus light-tail ablations, representative-pool accuracy, and thinning acceptance rates. The result is an assumption-transparent mathematical and computational framework for majority-like decision making in noisy synthetic microbial consortia.

Optical computing is an emerging field with the potential to enhance the performance and energy efficiency of artificial intelligence-related computing. One prospective application of optics in computation is random projection. Random projection, characterized by the use of a Gaussian independently and identically distributed (i.i.d.) matrix, is notable for its simplicity and ability to preserve distances within high-dimensional space. This property is leveraged by several machine learning algorithms to improve computational efficiency, including randomized singular value decomposition (RSVD), no-prior-knowledge exponentially weighted moving average (NEWMA) for change-point detection, and reservoir computing for time-series forecasting. Moreover, the application of random projection extends to neural network training methodologies such as direct feedback alignment (DFA). Optical implementations of random projection have also been explored, offering potential benefits in terms of time efficiency and energy savings, with initiatives like LightOn leading such efforts. This article provides a comprehensive analysis of these algorithms and identifies a trend in which random projection accounts for a small fraction of the total computational time in most algorithms examined. Additionally, we highlight the limitations of optical random projection, demonstrating that it does not significantly enhance performance. This finding raises questions about the relevance of random projection-based optical processing units.

Data routing protocols play a vital role in Wireless Sensor Networks (WSNs). However, large network sizes and constrained resources demand more energy-efficient routing strategies. In this context, conventional routing protocols often show weak load balancing and inefficient energy use. Low-Energy Adaptive Clustering Hierarchy (LEACH) and Low-Energy Adaptive Clustering Hierarchy Centralized (LEACH-C) remain the two most widely adopted hierarchical routing protocols in WSNs. LEACH operates as a non-geographic distributed routing protocol, whereas LEACH-C is a geographic-based centralized routing protocol. Compared with flat routing protocols, both can prolong network lifetime, but they still suffer from limited energy efficiency. To address this limitation, we in this research proposed an enhanced LEACH protocol based on cluster configuration and Quantum Beluga Whale Optimization (QBWO-LEACH). During the setup phase, the central base station (BS) employs the proposed QBWO approach, which integrates Beluga Whale Optimization (BWO) with the strengths of quantum computing, to centrally organize the clusters. This process includes determining the cluster centroids, assigning cluster members, and evaluating cluster energy, cluster priority, and cluster lifetime. In the cluster heads (CHs) rotation phase, local clusters use the position and energy information of all cluster members to perform distributed CHs switching, distributing cluster energy approximately evenly among all members. In the steady-state phase, the relay forwarding of monitored data flows is implemented. Compared with traditional LEACH and other improved variants of the LEACH protocols, the comprehensive performance of the protocol proposed in the present research is found to be superior. We compare our proposed QBWO-LEACH with the existing LEACH protocols in terms of node survival, network residual energy, half node dies (HND), last node dies (LND), and first node dies (FND), in all four cases using both simulation and statistical analysis. QBWO-LEACH demonstrates an average improvement of 51.87% over LEACH, 17.69% over Particle Filter LEACH (PF-LEACH) and 4.31% over a 2-stage Genetic Algorithm-based LEACH (GA2-LEACH) in node survival and network residual energy in all four cases.

Physical neural networks (PNNs) are neural-like computational frameworks that exploit the intrinsic dynamics of physical media to achieve ultrafast and energy-efficient information processing. However, the complex and strongly-coupled physical nature of PNNs in disordered environments makes them resistant to accurate differentiable modeling. Here, we propose a concept of computational space that empowers the chaotic environment itself with computational capabilities. This space constitutes a large-scale, model-agnostic PNN through distributed intelligent metasurfaces. To enable effective training, we develop a fully-forward learning framework that estimates zeroth-order gradients from in-situ measurable electromagnetic feedback, thereby circumventing the rigorous modeling requirements of conventional backpropagation. In experiments, we construct such computational space that achieves recognition accuracies of 97% for alphabetic characters and 99% for numeric patterns. Furthermore, the space exhibits the functionalities of enhanced focusing under disordered scattering conditions and reliable human position localization. This emerging paradigm of self-evolving physical intelligence holds potential for advancing embodied intelligence, autonomous cyber-physical systems, and next-generation human-machine interaction, marking a fundamental transition from computing the physics to computing with physics.

Genome assembly from long-read sequencing data has become a standard approach for resolving complex genomic regions and producing high-contiguity assemblies. However, the diversity of available assemblers, their varying performance across species, and the need for reproducible workflows present ongoing challenges. We developed LORA, an easy-to-use and reproducible application for assembling genomes from long-read data. LORA integrates several well-established assemblers, including Canu, HiFiasm, Flye, and Unicycler, as well as more recent tools such as Necat and Pecat. It is implemented as a Snakemake pipeline to parallelize tasks and support seamless execution on both local machines and computing clusters. LORA includes multiple quality assessment steps, interactive HTML reports for interpretation, BLAST-based taxonomic identification, and completeness evaluation. Together, these features provide users with a comprehensive view of assembly quality and potential problems. We illustrate the capabilities of LORA using datasets from bacterial genomes and unicellular eukaryotes, sequenced with both PacBio and Oxford Nanopore technologies, highlighting typical outcomes and common pitfalls encountered during long-read assemblies. LORA is distributed as part of the Sequana project, an open-source framework designed for reproducibility, maintainability, and straightforward deployment across computing environments.

Electronic health records (EHRs) increasingly anchor clinical decision support and population-scale analytics, yet their concentration of sensitive information amplifies disclosure risk, widens the attack surface, and faces emerging threats from quantum computing. Existing frameworks fail to simultaneously address privacy preservation, quantum-resistant security, and cross-institutional federated learning. We introduce CITADEL (Cryptographically Integrated Temporal Architecture for Distributed EHR Ledger), integrating five co-designed components: NIST-standardized CRYSTALS-Kyber (ML-KEM-768) and CRYSTALS-Dilithium (ML-DSA-65) post-quantum cryptography via the validated pqcrypto library; a genomic-aware privacy engine with beacon query protection and calibrated randomized response; temporally-partitioned federated learning with hospital-specific weighted aggregation; multi-modal health data tokenization; and an adaptive regulatory compliance engine for HIPAA and GDPR. Evaluation used a synthetic EHR dataset comprising 5,000 patients across 10 healthcare institutions, with 30-day hospital readmission as the primary prediction task. CITADEL achieves 84.5% accuracy and 0.866 AUC-ROC, exceeding nine baselines including centralized neural networks and differentially-private federated learning. Privacy metrics include k-anonymity of 13, l-diversity of 2.0, 99.0% linkage attack resistance, 42.2% attribute inference resistance, and 100% correlation preservation. The ledger sustains 285.3 transactions per second with ML-DSA-65 signing in 2.16 ms and verification in 0.46 ms. Multi-seed evaluation confirms robustness (accuracy 0.854 ± 0.012, AUC-ROC 0.880 ± 0.014). CITADEL demonstrates that privacy preservation, quantum-resistant security, and usable federated analytics can be reconciled within one cohesive architecture. Results suggest a practical route to healthcare data management that remains credible in a post-quantum computing era and compatible with decentralized governance.

Latency sensitive, computation intensive and mobility aware applications in Edge Fog Cloud environments have increased the demand to develop intelligent offloading task mechanism that can dynamically scale to changing network conditions whilst remaining scalable, energy efficient, and preserving data privacy. Traditional, heuristic and centralized based learning offloading methods frequently have problems in accommodating heterogeneous workloads, non- stationary environments and privacy limitations associated with the next generation distributed computer system. In order to overcome these drawbacks, the present paper will suggest a Federated Deep Q-Learning (FDQL)-based task offloading framework, which incorporates deep reinforcement learning and federated learning to support adaptive, decentralized and privacy-conscious decision-making across hierarchical Edge Fog Cloud architectures. The framework proposed solves task offloading as a Markov Decision Process, with the execution decisions being trained based on the joint consideration of the latency, bandwidth availability, queue length, computational load, and energy state, as well as user mobility, without sharing raw data during federated model aggregation. In comparison to the current CNN-, LSTM-, SVM-, and rule-based methods, which use fixed threshold values or rely on centralized training, the FDQL architecture allows collaborative learning between distributed edge nodes, enhancing generalization as well as resilience as network conditions evolve. Large-scale experimental analysis is performed using a trace-driven simulation based on a publicly available task offloading dataset of tasks and the performance is evaluated based on the latency, energy consumption, task success rate, robustness analysis, and computational efficiency. Experimental findings indicate that the proposed FDQL framework demonstrates improved performance under distributed and resource-constrained environments compared to baseline approaches since shorter latency, increased energy efficiency, and more predictable execution-layer selection are achieved. The significance of federated learning, mobility awareness, and bandwidth-aware optimization in the stability of the performance is also confirmed by ablation studies. In order to achieve a better level of transparency and trustworthiness, SLA-based confusion matrix analysis and ROC analysis are performed as well as SHAP-based explainability analysis, which proves that the decisions made by FDQL are based on physically interesting, as well as SLA-relevant features, like latency, bandwidth, and resource use. All in all, the designed FDQL framework is a successful, interpretable, and scalable approach to intelligent task offloading, so it would fit perfectly into the implementation of the 6G-enabled application, such as smart cities, industrial internet of things, and autonomous systems in the future.

Current clinical care following ischemic stroke provides intermittent assessments, potentially missing early detection of neurological deterioration. Optical imaging offers the potential for a continuous, portable alternative to the current combination of physical exams and radiological imaging. However, optical imaging technology has lacked the crucial combination of portability, resolution, and coverage required to assess spatially distributed brain function. Here, we present a proof-of-principle study that applies recent advancements in portable high-density diffuse optical tomography (HD-DOT) instrumentation and a functional connectivity analysis strategy to assess spatially distributed brain connectivity. Point-of-care brain health assessments with these tools may inform clinical care and our understanding of brain injury throughout acute stages of stroke recovery. We measured spatially distributed cortical oxygenation with HD-DOT within the intensive care unit in N = 13 patients with ischemic stroke within the first 72-h since last known normal. To assess brain function integrity, we developed a Similarity metric of functional connectivity (FC) through comparisons to FC in a healthy young adult population. To assess the sensitivity of this FC Similarity metric to disruptions caused by stroke, we compared these data with FC Similarity assessed in an older healthy cohort. We administered the NIH Stroke Scale to stroke patients to evaluate the potential of the FC Similarity metric to inform the severity of functional disruptions. The FC Similarity metric applied to HD-DOT data in both acute stroke patients and healthy older controls exhibited a significant sensitivity to the presence of a stroke (Cohen's d = 1.5 , p = 1.2 × 10 - 3 ) and a significant relationship with the degree of neurological disruption as measured by the NIH Stroke Scale ( R 2 = 0.69 , p = 4.7 × 10 - 4 ). We present a proof-of-principle for HD-DOT and the FC Similarity metric for assessing brain function during the acute stage of ischemic stroke recovery at the point-of-care.

Industrial Internet of Things (IIoT) ecosystems are expanding rapidly. Scalable and reliable intrusion detection systems (IDS) are needed to protect critical infrastructures from evolving cyber threats. This study proposes a hybrid IDS framework that combines Graph Attention Networks (GAT) and Bidirectional Gated Recurrent Units (BiGRU) for privacy‑preserving distributed detection. The model is optimized with the Grey Wolf Optimizer (GWO) and enhanced through Federated Learning (FL). In IIoT traffic, GAT captures complex structural links, while BiGRU analyzes bidirectional temporal patterns, enabling accurate anomaly detection. GWO automates hyperparameter tuning and offers faster convergence than traditional methods such as Ant Colony Optimization. FL trains models locally on distributed IIoT devices, preserving data privacy and supporting decentralized deployment. The framework demonstrates improved scalability potential through decentralized training and reduced communication overhead (20% lower in a 10-node simulation), achieving detection accuracies of up to 95% across diverse attack scenarios, including Distributed Denial of Service (DDoS), Advanced Persistent Threats (APTs), and Zero‑Day exploits. It has been evaluated on the Edge‑Industrial Internet of Things dataset (Edge‑IIoTset), Canadian Institute for Cybersecurity - Internet of Things 2023 Dataset (CICIoT2023), and Real‑Time Internet of Things 2022 (RT‑IoT2022) datasets. An Explainable AI (XAI) module further improves interpretability by leveraging GAT's attention mechanism. Overall, this technology demonstrates competitive offline performance on the EDGE-IIoTset, CICIoT2023, and RT-IoT2022 benchmark datasets, achieving F1-scores of up to 0.94, and shows promising scalability potential through decentralized Federated Learning with 20% lower communication overhead in a 10-node simulation. However, inference latency on resource-constrained edge hardware remains a challenge (e.g., 120-180 ms per sample on Raspberry Pi 4), which limits its strict real-time feasibility in mission-critical environments. Therefore, further model compression, adversarial robustness testing, and real-world deployment validation are required before practical edge-level applicability can be confirmed.

Artificial intelligence (AI) is reshaping paediatric healthcare, offering new capabilities across diagnosis, monitoring and treatment personalisation. Modern AI systems integrate multimodal data, including imaging, genomics, electronic health records, wearable sensors, environmental exposures and patient-reported outcomes, to generate insights tailored to children's developmental, physiological and psychosocial needs. Advances in machine learning, deep learning, natural language processing, computer vision and generative models are enabling earlier detection of rare diseases, dynamic risk stratification and personalised care pathways. Emerging technologies such as digital twins simulate individual disease trajectories and treatment responses, reducing reliance on traditional trial designs and supporting anticipatory, precision care.The next frontier of AI in healthcare includes adaptive decision support powered by reinforcement learning and advanced time-series modelling, allowing systems to respond to real-time physiological changes and support complex sequential decisions in areas such as ventilation, insulin dosing, deterioration prediction and medication titration. Parallel progress in remote monitoring and smart sensors is shifting care from hospitals to homes, supporting long-term condition management and reducing avoidable admissions. Causal AI offers further potential by uncovering true cause-and-effect relationships, enabling clinicians to understand why interventions work and explore counterfactual 'what-if' scenarios. Looking further ahead, quantum AI, neuromorphic computing and privacy-preserving federated learning may unlock new computational capabilities, enabling ultra-fast analysis, on-device learning and the secure use of distributed paediatric datasets. Realising this future requires rigorous governance, paediatric-specific validation, safeguards for privacy and autonomy and equitable digital access. When developed responsibly, AI has the potential to augment clinical expertise, reduce health inequalities and transform child health outcomes.

Digital health tools are increasingly used in mental health care to passively collect patient data and analyze health status outside of clinical settings. While technologies such as digital phenotyping, affective computing, and computational behavioral analysis offer new insights into symptom manifestation in daily life, they generate large volumes of potentially sensitive data that raise significant data privacy concerns, requiring high levels of patient awareness and consent. Empirical research is lacking on stakeholder understandings toward the sensitivity of these data and expectations for data stewardship, perspectives that are critical for developing robust informed consent and data protection policies for digital health data use. This study aimed to explore key stakeholder perspectives on the sensitivity of computer perception (CP) data, trust in existing data protections, willingness to share CP data externally, and desire for transparency of CP data transactions outside of the clinical space. As part of a larger, multisite study, we conducted qualitative interviews (n=40) via Zoom (Zoom Communications, Inc) with 20 adolescents (aged 12-17 years) familiar with CP tools and their caregivers (n=20). Interviews consisted of a series of open-ended questions regarding stakeholders' perspectives on privacy, data security, and the use and exchange of CP data. We developed a qualitative codebook to identify and label thematic patterns in responses to questions addressing the topics above, using thematic content analysis to identify themes inductively. Each interview was coded by merging work from at least two separate coders, and several team members contributed to qualitative analysis. Most adolescents and caregivers viewed CP data as highly sensitive and expressed a reluctance to share these data beyond their clinical teams. While many participants expressed trust in existing data protections to protect CP data, they often misunderstood or overestimated the extent of protections to safeguard CP data. Our findings underscore the critical need for clear and effective patient communication and education about the risks, benefits, and protections associated with CP data through informed consent protocols. To promote greater transparency, understanding, and trust, we recommend 5 strategies: educating patients about data protection; studying secondary data exchange and reidentification risks; strengthening transparency regulations; improving data traceability mechanisms, such as distributed ledger technologies, to enhance data traceability and auditability; and adopting dynamic consent models.

Lung cancer remains the most significant cause of cancer-related death worldwide due to the critical challenges in diagnosis. Despite the promising efforts, the existing models faced challenges in capturing the complex patterns in medical imaging data while minimizing the computational complexity. In this research, the lung cancer detection using Computed Tomography (CT) images is performed using the Reverse Task attention-enabled Distributed Elman convolutional neural Network (RTsDEN) model that helps in mitigating the challenges in existing methods and improving the detection performance for real-time applications. The proposed model, combining the Reverse Task attention-(RTsAt) module and the distributed Elman concept, significantly contributes to capturing the intricate disease patterns from the complex backgrounds and varying environmental conditions. In addition, the proposed method exploits the adaptive lobe and multigranular nodule segmentation stage to facilitate better understanding and interpretation for accurate diagnosis. Experimental results reveal that the proposed RTsDEN outperforms other existing models by attaining 97.12% accuracy, 98.03% precision 96.22% recall using LUNA 16 dataset and 97.72% accuracy, 98.31% precision, 97.14% recall using the LIDC-IDRI dataset. The research introduces an efficient DL model with an ensemble approach, which significantly influences effective lung cancer detection.

Wearable and textile-based technologies are transforming health monitoring by enabling continuous, non-invasive, and context-aware assessment of physiological and biochemical signals in daily life. Advances in flexible electronics, conductive fibers, smart materials, and artificial intelligence have driven a shift from rigid, device-centric wearables toward textile-integrated and garment-based systems capable of distributed sensing, actuation, energy harvesting, storage, and communication. This roadmap provides a textile-centric overview of the current state and future trajectory of wearable technologies for healthcare, with electronic textiles positioned as a distinct and strategically important class within the broader wearable ecosystem. We synthesize progress across textile-integrated sensing, therapeutic and protective garments, textile body-area networks, energy-autonomous systems, and embedded computing, while critically examining challenges related to signal reliability, manufacturability, scalability, data governance, regulation, and equity. Market trends and adoption patterns are discussed to contextualize translational pathways from laboratory prototypes to clinically deployable and scalable textile systems. By identifying key scientific, technological, and societal priorities, this roadmap outlines actionable directions to accelerate the integration of textile-based wearable technologies into preventive, personalized, and decentralized healthcare.