Intraoperative intracranial electrophysiological recordings provide unique access to human cortical dynamics but remain difficult to translate across patients due to inconsistent localization of transient surface electrodes. Unlike chronic implantations, intraoperative electrodes are placed transiently, rarely visible on imaging, and often inconsistently documented. We present an open-source imaging pipeline, ALIGNER (Advanced Localization and Imaging Guidance for Neurosurgical Electrode Recording), designed to reconstruct intraoperative surface electrode array placements and quantitatively map neural activity to individualized anatomical and pathological substrates. By enabling anatomical localization of these electrodes, this framework supports systematic analysis of spatial gradients in neural activity relative to pathological tissue. We developed a multimodal reconstruction framework integrating pre- and postoperative MRI and CT, cortical surface modeling, semi-automated pathology segmentation, intraoperative photographs or videos when available, and physics-based electrode modeling. To improve robustness in cases with distorted anatomy, artificial intelligence tools such as SynthSR were used to enable reliable cortical surface reconstruction prior to FreeSurfer processing. A monocular depth-estimation network was incorporated to constrain electrode placement in conjunction with Blender cloth-physics simulation when photographic images were available, while atlas- and note-guided inference supported reconstruction otherwise. The pipeline was applied to 38 neurosurgical patients across drug-resistant epilepsy resection (n = 24), malformation (n = 1), brain tumor (n = 11), and deep brain stimulation (n = 2) cases, achieving some type of reconstruction and electrode localization in all participants. By exporting electrode coordinates for quantitative spatial analyses, including distance-based mapping relative to lesions and resection cavities, ALIGNER enables anatomically grounded and reproducible analysis of intraoperative electrophysiology. This open-source framework provides foundational infrastructure for cancer neuroscience studies of tumor-neuron interactions and establishes a scalable platform for future neurostimulation, implantable neurodevice, and brain-computer interface applications requiring precise anatomical localization.
Metagenomics analysis is a critical tool in identifying and typing viral samples to aid surveillance, clinical, epidemiological, and other workflows. Despite advances in sequencing technology and analysis pipelines, there are still limitations that lead to reduced taxonomic resolution or false positives from highly recombinant or challenging samples. Here we describe MGtree, a novel metagenomics pipeline that utilizes a combination of full-length read alignments and phylogenetic analysis to classify samples of interest. We demonstrate that MGtree accurately genotypes viral samples from challenging norovirus and HPV datasets. MGtree outperforms the popular metagenomics programs Kraken2 and Centrifuge, and it succeeds with low-input samples where de novo assembly fails. MGtree's correct assignments across highly mutant and coinfected samples highlights its ability to resolve viral genotypes and its potential to improve classification precision in complex samples.
Recent Department of Education actions implementing graduate loan limits under the One Big Beautiful Bill Act excluded nursing from "professional program" designation, creating a policy misalignment that threatens the advanced nursing workforce. Using data from the 2022 National Sample Survey of Registered Nurses, we demonstrate that educational debt is a structural feature of nursing careers and a barrier to progression into advanced practice and faculty roles. Debt accumulates across nursing careers and disproportionately affects nurses from historically marginalized racial and ethnic groups, nurses with dependents, and those advancing through associate-degree pathways. We argue that excluding nursing from professional program loan classification misaligns federal borrowing policy with the cumulative structure of nursing education and threatens equitable access to advanced training and the pipelines essential to health system capacity.
Knee motion is a key biomarker in chronic musculoskeletal diseases, yet conventional in-lab optical motion capture falls short of identifying how knee motion continuously impacts joint health outside the lab. Inertial measurement unit (IMU) provides a clinically attractive approach for continuous real-world motion tracking. Our goal was to establish a clinically practical, minimum-setup pipeline for leg-worn IMUs to estimate knee flexion and determine its concurrent validity to optical motion capture during various knee movements. We recorded thigh and shank-worn IMU data with concurrent marker-based and markerless optical motion capture on 10 healthy adults, who performed 10 common movements including walking, running, and stair navigation. We combined IMU functional alignment with data fusion to estimate knee flexion during each movement and compared IMU-based estimate against both motion capture systems using Pearson correlation (Rxy) and root-mean-square difference (RMSD). IMU-estimated knee flexion strongly correlated with motion capture (Rxy ≥ 0.9). RMSDs were smaller for slower movements like walking (RMSD = 4.4-6.0°) while larger during faster movements like running (RMSD = 5.4-9.4°). Wearable IMUs track knee flexion with comparable results to motion capture during daily activities typical to older adults, highlighting their potential for continuous patient monitoring. Our simple pipeline makes IMU-based knee motion tracking more practical and compatible with clinical research. Future research should seek IMU-wearing best practices to secure clinically meaningful data on real-world knee mobility.
Student behavior recognition in classroom environments is important for teaching quality assessment and intelligent education, yet it remains challenging due to dense student distributions, frequent occlusion, substantial scale variation, and the subtle nature of common classroom activities. To address these issues, this paper proposes RepYOLOv5-SF3D, a cascaded visual perception framework for fine-grained student behavior recognition in complex classroom scenes. The framework integrates a lightweight RepYOLOv5m detector with a dual-stream SlowFast-3D recognition branch, enabling automated inference from raw video input to behavior labels. To improve robustness in dense and occluded scenes, the front-end detector serves as a spatial-prior module, while a decoupled training strategy reduces the impact of localization instability on back-end spatiotemporal learning. In addition, two task-oriented modules are introduced in the recognition branch: the Spatiotemporal Depthwise-Separable 3D module (SDS3D) and the Normalization-Based Temporal Attention Mechanism (NTAM). Experimental results on a real classroom dataset show that RepYOLOv5-SF3D achieves a mean average precision (mAP) of 88.83%, outperforming the baseline SlowFast model by 3.36% and surpassing the existing LSTC method by 2.05%, while maintaining a front-end inference latency of 12.5 ms per frame and a total model size of 151.46 MB. These results demonstrate a favorable balance between fine-grained recognition accuracy and edge-deployment efficiency in practical classroom visual sensing.
Can oocyte maturation be adapted to a phenotypic assay within a drug screening pipeline to identify compounds that block meiotic progression for female non-hormonal contraceptive drug discovery? A complex phenotypic assay of mouse oocyte maturation identifies potent compounds that reversibly block meiotic progression at specific stages. During oocyte maturation, prophase I-arrested oocytes resume meiosis, undergo germinal vesicle (GV) breakdown, and complete meiosis I with extrusion of the first polar body. Oocyte maturation is critical for generating a fertilizable gamete, and thus, blocking meiotic progression is a promising target for non-hormonal contraception. Moreover, this process can be recapitulated in vitro, providing a powerful phenotypic assay for drug screening. Oocytes were collected from CD-1 mice following hyperstimulation and in vitro matured for 14-16 h in the presence of a single dose (10 µM) of compounds from a bioactive compound library. Incubation in dimethyl sulfoxide (DMSO) only or 10 µM milrinone, a known phosphodiesterase 3A (PDE3A) inhibitor, was used as vehicle and positive control, respectively. Primary screening and subsequent concentration-response testing were performed in duplicate. Oocytes were incubated with compounds during IVM to identify compounds with inhibitory effects on meiotic progression. We were specifically interested in compounds that maintained arrest at prophase of meiosis I (GV-intact) or prevented extrusion of the first polar body. Brightfield images were taken before and after IVM to assess maturation status based on morphological criteria. Using this platform, we screened a subset of 818 compounds from the compound library based on annotated target and structural diversity. A hit was defined as a compound that inhibited meiotic progression by ≥80%. Hits were validated through an independent source of compound and concentration responsiveness. Hit compounds were also tested in a prolonged culture treatment to confirm maintenance of meiotic arrest. A counterscreen assay was performed on compounds that maintained arrest at prophase of meiosis I to rule out phenotypes due to inhibition of PDE3A activity. From our primary screen, 29 hits were identified that blocked meiotic maturation. Following hit validation via independent sources and concentration-response, 18 compounds were confirmed. Five compounds resulted in arrest at prophase of meiosis I, but only one was not based on PDE3A inhibition. Thirteen compounds enabled resumption of meiosis but not polar body extrusion, and 12 of these maintained arrest during prolonged culture. Thus, using this screening pipeline, we identified a total of 13 compounds for an overall confirmed hit rate of 1.6%. Using this phenotypic assay, we also compared the activity of structural analogs, enabling the establishment of preliminary structure-activity relationships. Additionally, washout of compound-treated oocytes allowed for normal meiotic resumption in 33% and 50% of oocytes arrested at GV and GV breakdown stages, respectively, demonstrating the potential for reversibility. N/A. Although this screening platform can identify potent compounds, their protein targets and potential mechanisms of action are unknown. As such, further studies are required to deconvolute targets and generate more specific compounds. Given that our phenotypic oocyte screening assay is based on the mouse model, validation using human oocytes will also be necessary for translational consideration. Lastly, all experiments were conducted on denuded oocytes and do not account for activity within an intact cumulus-oocyte complex. Through our screening pipeline, we identified compounds that potently inhibit meiotic progression. Beyond furthering our understanding of oocyte maturation, we highlight potent compounds as promising starting points toward novel drug candidates or in novel target identification for female non-hormonal contraception. This work was supported by the Gates Foundation [INV-003385]. Under the grant conditions of the Foundation, a Creative Commons Attribution 4.0 Generic License has already been assigned to the Author Accepted Manuscript version that might arise from this submission. The authors have no conflict of interest to disclose.
Accurate vehicle localization must be maintained even in tunnel sections where GNSS reliability is degraded. However, conventional GNSS/INS-based localization rapidly accumulates errors in such environments, affecting lane-level decision-making and path-following stability. To address this problem, this study proposes a dedicated localization support sign for stable LiDAR observation and a point-cloud-registration-based correction algorithm. The proposed method detects a dedicated sign using a PointPillars-based detector, and the corresponding point cloud is registered to a pre-built reference map to estimate a rigid correction transform online. The sign was installed in a tunnel section of a proving ground that reproduces real-road conditions. For evaluation, the driving sequence was analyzed by separating the pre-entry section, the tunnel section before dedicated-sign recognition, and the section after dedicated-sign recognition. The proposed pipeline substantially reduced localization error after dedicated-sign recognition, compared with the GNSS/INS-only baseline. The dedicated sign also provided more stable correction than ordinary tunnel structures within the same registration pipeline. These results indicate that the proposed LiDAR-based pipeline can suppress localization drift in GNSS-degraded sections.
WiFi channel state information (CSI) has become a compelling sensing modality for contactless human activity recognition. However, differences in datasets, preprocessing protocols and model configurations make consistent comparison and reproducibility challenging. This study presents a unified baseline evaluation of four widely adopted deep learning architectures: multilayer perceptron (MLP), convolutional neural network (CNN), gated recurrent unit (GRU) and a hybrid CNN-GRU model across multiple publicly available CSI datasets encompassing a range of sensing tasks. We harmonize the datasets, implement a standardized preprocessing and training pipeline to reduce experimental inconsistencies and support controlled within-dataset comparisons of model behavior. Evaluations include single-person activity recognition, fall-risk estimation, multiperson occupancy classification and localization-aware activity recognition, representing progressively higher temporal and spatial complexity. Our results show dataset-dependent trends: CNNs provide an efficient accuracy-complexity trade-off in several structured activity scenarios, whereas GRUs are advantageous when temporal dynamics are more prominent, although with greater training and inference costs. In contrast, MLPs generally underperform due to limited capacity to capture spatial and temporal dependencies. Confusion matrix analysis reveals that dynamic behaviors and low-motion states remain challenging to distinguish, underscoring the importance of temporal modeling. By releasing the complete experimental pipeline and benchmarking results, this work establishes a reproducible reference framework for the research community and highlights directions for future investigation, including cross-dataset generalization, hybrid model design and lightweight deployment strategies.
Low back pain (LBP) is a major global health problem and can result in a variety of movement impairments. Advances in smart technology have enabled the collection of novel streams of movement data, and machine learning (ML) methods have been increasingly used for data analysis. However, many existing technologies remain expensive and unsuitable for widespread clinical use, and ML approaches have largely focused on distinguishing people with LBP from healthy controls rather than identifying meaningful subgroups within the LBP population. Motion Tape (MT) is a recently developed wearable strain sensor that translates skin deformation from underlying movement and muscle engagement into electrical signals. In this exploratory study involving 10 participants with LBP, we demonstrate that MT data from six sensors applied on the lower back capture rich movement information capable of characterizing movement patterns among participants with LBP. We propose a feature engineering approach based on biomechanical features as well as time-series causal discovery applied to multivariate sensor time-series data to extract directed inter-segment coordination patterns. We further develop an exploratory subgroup discovery pipeline by aggregating clustering coassociation information across diverse movement tasks. Our causal coordination features show promising discriminative information across several movement types, capturing aspects of motor control not reflected in amplitude-based or embedding-based features alone, such as asymmetries and movement restrictions. Preliminary ensemble clustering analysis indicates three potential LBP subgroups distinguished by biomechanical and inter-segment coordination patterns, which may reflect varied strategies under different movement demands. We investigate the differences in clinical characteristics among these LBP subgroups. We show that time-series foundation models are not well suited for LBP subgrouping due to their uninterpretability, which is improved in our feature engineering pipeline. This framework could reveal additional subgroups with larger cohorts and may generalize to other sensor modalities.
In many university and healthcare projects, models are built for very different data types such as tables, institutional time series, and medical images, but they are deployed as separate applications. In this work, that separation made testing and maintenance difficult because each module had its own pipeline and runtime requirements. This paper presents an integrated AI lakehouse-style implementation that runs three model pipelines inside one containerized backend. For medical imaging, we used MRI datasets from IEEE DataPort: a four-class classification set with 7012 images (5708 train/1304 test) and a segmentation set with 3063 image-mask pairs. The classification model (ResNet50 transfer learning) is evaluated using a proper train-validation-test protocol across multiple splits (80/10/10, 70/10/20, 60/10/30, and 10/30/60), achieving a test accuracy of 99.00% under the standard 80/10/10 split. Additionally, a patient-level evaluation is conducted using an external glioma dataset to provide a more realistic assessment without data leakage. The segmentation model (DeepLabV3-ResNet50) achieved 83.09% validation mIoU and 88.79% Dice score. For university KPI forecasting, we used annual IPEDS and NSF HERD data from 2010 to 2023 for three universities (BSU, EOU, and UAB). To examine the effect of preprocessing on forecasting performance, two case studies are conducted. In the first case, linear interpolation is applied to generate semester-level data. In the second case, the original annual data is used directly without interpolation. Random Forest regression and ARIMA models are evaluated using MAE, RMSE, MAPE, and R2. The results showed that interpolation improved apparent forecasting performance due to smoothing, while evaluation on the original annual data provided a more realistic assessment of model behavior. To further validate the framework on a larger dataset, an additional case study is conducted using a student dropout dataset. For water potability, we trained and compared multiple tabular classifiers on a large dataset (1,048,575 samples). A Random Forest model (100 trees, max depth 10) achieved 85.86% test accuracy and high recall for unsafe samples (0.8447). All modules are served via FastAPI and deployed together using Docker, with workflow automation routing requests to the correct endpoint. System-level benchmarking indicates that the backend maintains stable throughput and latency under concurrent requests.
Background/Objectives: First-generation CDK4/6 inhibitors (palbociclib, ribociclib, abemaciclib) target the conserved ATP-binding pocket of CDK4 and, despite clinical success, are limited by acquired resistance and insufficient exploration of alternative regulatory sites. This study aimed to identify a putative allosteric small-molecule candidate at the CDK4 αE-helix-Cyclin D1 α1-helix protein-protein interaction (PPI) interface within the CDK4/Cyclin D1/p21 ternary complex using RapidFunnel-AI, a decision-interpretable virtual-screening pipeline. Methods: Starting from 50,000 ChEMBL 33 molecules, the pipeline sequentially applied a Q-Fold/RapidFunnel topological Tanimoto scan based on clinical CDK4/6 inhibitor motifs, fragment-level electronic-property enrichment, ADMET/PAINS filtering, dry Vina-GPU docking, hydration-mediated AutoDock-GPU (Version 1.6) docking, explicit-solvent molecular dynamics, contact-retention analysis, and MM-GBSA energy decomposition. The Q-Fold Thermo-Core surrogate model provided fragment-level enrichment, predicting the HOMO-LUMO gap (R2 = 0.93) and isotropic polarizability (R2 = 0.98) on QM9. Candidate selection did not rely on the lowest docking or MM-GBSA score alone, but on pose persistence, contact continuity, and energy-component consistency. Results: The workflow reduced the initial library to 43 topologically prioritized candidates, 25 ADMET/PAINS-filtered ligands, and 9 docking-derived complexes for MD validation. Ligand_020 emerged as the only candidate that preserved a persistent binding mode at Site 2 during a 500 ns simulation-an interface engagement reproduced across three independent 500 ns replicates with no full dissociation in any replicate-with a protein Cα RMSD of 2.88 ± 0.32 Å, a ligand heavy-atom RMSD of 3.56 ± 0.28 Å, and a van der Waals-dominated MM-GBSA profile (ΔGbind = -28.23 ± 3.57 kcal/mol). In contrast, palbociclib and ribociclib, forcibly placed at Site 2 as negative controls, lost most initial contacts within 5 ns and tended to detach despite more favorable MM-GBSA values. Conclusions: These results suggest that single-score docking or MM-GBSA ranking can generate false positives at shallow PPI interfaces. By integrating AI-assisted prioritization, multipocket docking, explicit-solvent MD, contact-retention analysis, and energy-component consistency, RapidFunnel-AI nominated Ligand_020 as an experimentally testable putative allosteric hit targeting the CDK4/Cyclin D1 interface, offering a reusable platform for PPI-focused oncological drug discovery.
The process of drug discovery is one of the most expensive, time-consuming, and high-risk endeavors in modern science. Translating initial scientific insights into safe and effective therapies, supported by genomics, structural biology, and computational chemistry, typically requires more than a decade and substantial financial investment. Machine learning (ML) has emerged as a powerful tool for improving efficiency across the drug discovery pipeline. By enabling the analysis of large and complex datasets, ML supports target identification, lead discovery, optimization, and prediction of preclinical and clinical outcomes. Its integration with experimental validation and automation is illustrated by recent advances such as protein structure prediction, AI-driven antifibrotic compound discovery, and antibiotic identification. Despite these advances, significant challenges remain. Model generalizability is limited by data scarcity, heterogeneity, and hidden biases. In addition, the translation of in silico predictions into clinically validated outcomes remains a major bottleneck, and regulatory acceptance is constrained by limited model interpretability. Ethical considerations, including data privacy, equitable representation, and the potential misuse of generative models, further complicate adoption. This review examines the applications of ML across the drug discovery pipeline, with a focus on translational and regulatory considerations. It also discusses emerging directions, including hybrid physics-AI approaches, multimodal foundation models, federated learning, and explainable AI. The effective integration of ML will depend on rigorous validation, interdisciplinary collaboration, responsible data governance, and alignment with regulatory frameworks.
Supramolecular polymer blends (SPBs) offer tunable morphologies that dictate their macroscopic properties, yet their rational design is limited by the absence of predictive structure-morphology models. Here, we introduce a data-driven high-throughput workflow that integrates modular polymer synthesis, robotic formulation, automated morphology characterization, and machine learning (ML) for accelerated SPB discovery. Using a plug-and-play synthetic strategy, 33 hydrogen-bonding end-functional homopolymers were prepared and orthogonally combined to generate 260 SPBs in 1 day. A fully automated atomic force microscopy (AFM) pipeline enabled systematic imaging, producing 2340 morphology data sets with minimal human intervention. Domain spacings were extracted through complementary image-processing methods and used to train ML models. A support vector regression (SVR) model accurately predicted target phase-separation sizes (50, 100, and 150 nm), which were experimentally validated. This work demonstrates the power of coupling high-throughput experimentation with ML to accelerate morphology discovery and provides one of the first large-scale experimental data sets for supramolecular polymer systems.
Fiber-degrading microorganisms are widely recognized for their potential to convert renewable lignocellulosic biomass into animal feed. However, translating this potential into practical application faces five critical yet underappreciated challenges. First, current screening methods, primarily including plate dilution and Congo red staining, are low-throughput, poorly reproducible and fail to capture the synergistic actions of natural microbial consortia. Second, the lack of standardized assays for quantifying cellulolytic activity compromises the reliability of cross-study comparisons. Third, safety assessments for fiber-degrading microorganisms remain superficial, with most studies neglecting mycotoxin production, antibiotic resistance gene transfer and long-term colonization risks. Fourth, fundamental differences between fungal and bacterial degradative systems, such as enzyme multiplicity, oxygen requirements and cellulosome assembly, are rarely considered in strain selection, leading to suboptimal application outcomes. Finally, the vast majority of positive in vitro degradation results fail to translate into improved animal performance in vivo, owing to poor microbial survival in the gastrointestinal tract, mismatched enzyme activity with gut pH and temperature, coupled with the absence of dose-response validation. This review critically evaluates these five bottlenecks across fiber-degrading microorganism types, screening platforms and practical livestock production applications. Overall, future progress should depend less on discovering "novel" strains and more on establishing standardized screening pipelines, rigorous safety frameworks and mechanistic understanding of in vivo efficacy, including direct head-to-head comparisons between fungal enzymes and bacterial probiotics under identical conditions.
Air quality forecasting and environmental health research at urban and regional scales depend on the combination of measurements from heterogeneous sensor networks, yet the construction of integrated multi-source datasets is rarely described or released as a self-contained deliverable. This paper presents an open dataset that combines four sensor-derived sources covering the whole of Spain over the period from 2022 to 2024: hourly air quality observations from the 588 stations of the national network operated by the Ministerio para la Transición Ecológica y el Reto Demográfico (MITECO), daily meteorological records from the Agencia Estatal de Meteorología (AEMET), daily mobility indicators derived from anonymised mobile telephony events published by the Ministerio de Transportes y Movilidad Sostenible (MITMA) at the municipality level, and a calendar of national and Autonomous Community public holidays. The processing pipeline harmonises sources that differ in temporal resolution, spatial codification and quality regime into a tidy hourly table indexed by station and timestamp, with a fixed feature schema of 56 variables per record. Air quality stations are paired with their nearest AEMET station through a three-tier distance rule, and the daily exogenous features are aligned to the air quality time axis through a two-variant temporal-alignment scheme (lag-and-expand to the hourly grid for the hourly release, same-calendar-day join for the daily release). A complementary daily resolution variant of the dataset is also released, with 72 columns and the same feature schema except for the air quality block, which is aggregated to daily mean, minimum and maximum. The integrated dataset contains approximately 15 million hourly records across the 588 stations and is released on Zenodo (DOI 10.5281/zenodo.20196221) under a Creative Commons Attribution 4.0 International (CC BY 4.0) licence. It is intended as a substrate for research on air quality forecasting, environmental epidemiology and multi-source data fusion at the nationwide scale.
The increasing deployment of IoT-enabled electric-vehicle charging networks has created a rapidly evolving cyber-physical environment in which security mechanisms must operate amid ever-changing data patterns and resource constraints. In these environments, static Machine Learning (ML) pipelines are often insufficient because they struggle to adapt to concept drift issues, emerging attacks, and real-time operational requirements. We analyzed cybersecurity vulnerabilities, challenges of conventional ML approaches, and the possibilities of AI-powered, adaptive security measures. This paper examines Online AutoML and its advantages, including automated adaptation to streaming data, reduced human intervention, and privacy-preserving, resource-aware learning. Furthermore, this paper discusses adversarial attacks and defences in Online AutoML systems, highlighting the need for frameworks that jointly address concept drift, scalability, privacy, and adversarial threats. Finally, this study emphasizes the importance of establishing comprehensive public benchmarks for Online AutoML research.
Parry-Romberg syndrome (PRS) is a rare facial asymmetric deformity. The pathogenesis of PRS is poorly understood. The role of predisposing genetic factors in the development of PRS is still unclear. This study aims to identify pathogenic variants associated with PRS by performing whole-genome sequencing (WGS) and genetic variation analysis. Nineteen peripheral whole blood samples were collected and sent for WGS from sporadic PRS patients in our department from September 2020 to January 2023. Single-nucleotide variation (SNV), insertion or deletion (InDel), and copy number variation (CNV) were called, filtered, and interpreted for pathogenicity using the PUMP pipeline. An enrichment analysis was performed. A common pathogenic gene related to PRS was not found through the analysis of SNV/InDel and CNV annotation files. A young female patient diagnosed with right Parry-Romberg syndrome carried a missense mutation in the PORCN gene on the X chromosome. Verifying the relationship between this mutation and the pathogenesis of PRS is needed. Based on the genetic variation analysis of the WGS data of 19 patients, we believe that the combined effect of genetic and environmental factors should be considered in the pathogenesis of PRS, and the role of genetic factors in the development of PRS needs to be further explored.
Precise respiration assessment is crucial for heart rate variability (HRV) interpretation as respiratory components-particularly respiratory sinus arrhythmia (RSA)-provide essential information on vagally mediated regulation. Conventional single-lead electrocardiogram-derived respiration (EDR) methods measure the amplitude modulation of the QRS-waveform caused by respiratory chest movements. This causes a displacement of the electrical heart axis in relation to the ECG lead axis, typically within the 2D frontal plane of the Einthoven electrode montage. Another approach is based on heartbeat acceleration and deceleration during respective inspiration and expiration causing RR interval modulation. However, interval-based methods depend on the complexity of sympathovagal factors that affect RSA. The present feasibility study accounts for the 3D rotational movement of the electrical heart axis during the respiratory cycle and avoids non-respiratory neuromodulatory confounds. The beat-to-beat cardiac rotation was extracted from Frank-XYZ coordinates reconstructed via a four-electrode EASI device. In a pilot study with data from 19 healthy adults performing acoustically paced breathing (6-18 bpm), three surrogates (RR-IntervalEDR, R-AmplitudeEDR, HeartmovementEDR) were compared using a unified Python 3.11.13 pipeline (3D VCG R-peak detection, multivariate Mahalanobis artifact correction, wavelet-based analysis) against a synthetic reference derived from the instructed breathing schedule. The results demonstrated a consistently lower estimation error and higher reference-based signal-to-noise ratio (refSNR), measuring spectral alignment with the paced-breathing trajectory for HeartmovementEDR and achieving a mean refSNR of 6.01 dB (vs. 4.62 dB for RR-IntervalEDR and 3.20 dB for R-AmplitudeEDR) and a mean absolute estimation error of 0.016 Hz (vs. 0.050 Hz and 0.032 Hz, respectively). Notably, HeartmovementEDR and R-AmplitudeEDR performance slightly improved at higher heart rates, consistent with the interpretation that higher cardiac sampling density benefits spectral resolution for chest movement-based methods, whereas RR-IntervalEDR showed no significant heart rate dependence. Furthermore, HeartmovementEDR was compared with the EDR results obtained by applying the Kubios-HRV Premium software (version 3.5.0). Kubios-EDR yielded higher precision at elevated breathing frequencies, whereas HeartmovementEDR outperformed Kubios-EDR at breathing rates below 10 bpm-a range that is particularly relevant for vagally activating slow breathing protocols or treatments. Future work should validate this method using a direct respiration measurement under spontaneous natural breathing conditions.
Metal ions are crucial for viral processes like replication, structural integrity and immune modulation, despite that the metalloproteome of Monkeypox virus (MPXV) remains largely unexplored. Monkeypox virus is a re-emerging zoonotic Orthopoxvirus with a 197 kb genome encoding over 183 proteins. Here, in this report we aimed to identify and explore the metal-associated proteome of the MPXV. Derived from employing a structure driven pipeline, followed by the functional annotation, the subcellular localization, evolutionary aspect, and structural validation with the established experimental evidences. Yielded a set of approximately 21 % high-confidence putative metal-associated proteins with a potential as metal-binding proteins. Functional annotation, Gene Ontology (GO) enrichment, and KEGG Orthology (KO) term assignment also revealed significant enrichment of pathways primarily related to functions such as, viral replication, genome maintenance, transcription, virion assembly, and host immune modulation. These metal-associated proteins are hypothesized to perform critical biological roles throughout the viral life cycle and pathogenesis. This includes nucleotide metabolism, transcriptional regulation, redox balance and immune evasion, and virion morphogenesis. These findings were further strengthened by subcellular localization analysis, which predicts the presence of metalloproteins within MPXV and its viral factories. Further, suggesting the spatial distribution of various metals and its utilization in the host organism. Facilitating the activities from viral attachment, entry, replication, transcription, viral assembly, and release as either mature virion (MV) or intracellular mature virion (IMV). Comparison and mapping these identified metal-associated MPXV proteins to Vaccinia virus followed by the virus-host network analysis highlighted the proposed role of metal-associated proteins within conserved Orthopoxvirus interaction pathways.
Massive Internet of Things (IoT) and sensor-network services in 5G/6G systems increasingly rely on network slicing to support large-scale sensing, monitoring, and mission-critical applications. In such sliced infrastructures, Proportional Fair (PF) allocation assigns resources according to slice-reported demands. This reliance on trusted demand reporting makes coexisting slices, including mMTC-based IoT sensor slices, vulnerable to resource exhaustion attacks, where a malicious slice inflates its demand to monopolize shared resources and induce Service Level Agreement (SLA) violations. Existing unsupervised defenses mainly focus on anomaly detection, while the translation of detection results into resource-level mitigation remains insufficiently addressed. To bridge this gap, this paper proposes AutoGuard-Hybrid, an unsupervised detection-to-mitigation framework that combines complementary anomaly detectors with allocation-aware mitigation policies to preserve slice-level service availability. Unlike prior detection-only approaches, AutoGuard-Hybrid converts unsupervised anomaly evidence into allocation-aware demand purification before PF scheduling. Its key design is a closed-loop integration of Isolation Forest (IF) and Long Short-Term Memory Autoencoder (LSTM-AE) as spatial and temporal front-end detectors with Adaptive Clipping and a Safety Cap, which translate anomaly scores into demand purification actions. Experiments show that AutoGuard-Hybrid remains comparable to Isolation Forest under Continuous attacks and improves the mean system-wide SLA violation rate by 27.6% under Adaptive Probing attacks. Stage activation analysis further shows that LSTM-AE activations increase from 9.3 under Continuous attacks to 29.4 under Adaptive Probing attacks. Ablation results show that Adaptive Clipping alone reduces the system-wide SLA violation rate by 75.0%, while the full mitigation pipeline achieves an 84.6% total reduction. AutoGuard-Hybrid operates within the 1 ms Transmission Time Interval (TTI) constraint and provides a practical defense framework for next-generation network slicing-enabled IoT and sensor-network services.