Search Results (1,205)

Machine learning (ML) integrated with metabolomics has emerged as a promising strategy to advance precision pharmacotherapy, enabling data-driven prediction of drug response. This review provides an overview of commonly applied ML methodologies in metabolomics-based pharmacological studies, including supervised models (Random Forest, Extreme Gradient Boosting, Support Vector Machine, Logistic Regression, K-Nearest Neighbors), unsupervised models (K-Means Clustering, Principal Component Analysis), and deep learning approaches. We summarize recent progress in the application of metabolomics-driven ML to personalized medication, with a focus on drug dosage optimization, therapeutic efficacy prediction, and adverse drug reaction assessment. Despite these advances, significant challenges remain, including limited explainability, insufficient prospective clinical validation, lack of standardization and reproducibility, and data dimensionality and quality issues. Addressing these issues will be essential for the clinical translation of ML-metabolomics integration. Looking ahead, continued methodological innovation, large-scale multi-center prospective validation, and integration with other omics platforms will be key to unlocking the full potential of metabolomics combined with ML in precision healthcare. Full article

The resistome, defined as the complete set of antibiotic resistance genes (ARGs) present in the microbiota of a given environment, is a critical component for understanding the evolutionary dynamics of antimicrobial resistance (AMR) and its impact on human, animal, and environmental health. This review summarizes current methods and technological advances and offers a forward-looking perspective on resistome research. A systematic literature search was conducted. References on short-read and long-read sequencing, amplicon sequencing, shotgun metagenomics, and multi-omics integration were included, as were bioinformatics tools for the detection, quantification, and annotation of ARGs. The results indicate that next-generation sequencing (NGS) technologies have significantly improved the characterization of ARGs across ecosystems, enabling high-resolution microbial profiling and the discovery of new variants. Furthermore, integrating multi-omics approaches with computational tools improves data accuracy, reduces analysis and reporting times, and facilitates the development of predictive models. However, significant challenges remain, which will be key to strengthening epidemiological surveillance under the One Health approach. Full article

Remote sensing building change detection is important for monitoring urban expansion, rural settlement dynamics, post-disaster reconstruction, and human-induced land transformation. However, most existing change detection models are optimized under in-domain protocols, while practical deployment often requires direct transfer from one labeled source dataset to unseen target domains without target images, labels, validation data, adaptation, or threshold calibration. This source-only cross-dataset setting is challenging because changes in sensor characteristics, spatial resolution, viewing geometry, scene composition, and background appearance can cause missed detections and pseudo-change false alarms. To address this problem, we propose DLV-CD, a frozen-DINOv3-based framework that trains only task-specific adapters, a multi-level difference fusion decoder, and a hierarchical evidence fusion module for source-only cross-dataset building change detection. Transfer evaluation on four datasets, namely LEVIR-CD, WHU-CD, S2Looking, and DSIFN-CD, shows that DLV-CD achieves the best F1-score compared with seven reproduced baselines, including classic supervised CD models, recent supervised CD models, and the SAM-based foundation-model baseline SAM-CD. Specifically, DLV-CD improves the average F1-score from 32.99% to 65.38%, outperforming the strongest reproduced baseline by 32.39 percentage points. Precision–recall analysis and qualitative comparisons further show that DLV-CD reduces both recall collapse and pseudo-change over-detection. These results demonstrate that frozen visual foundation representations provide a strong basis for target-free cross-dataset building change detection. Full article

Supervised fine-tuning is often discussed in terms of data volume and target quality, but the format of the supervision itself can change the reasoning strategy a model expresses in its outputs. We study this effect on a six-category reasoning benchmark spanning symbol transformation, text decryption, bit manipulation, gravitational constant estimation, numeral conversion, and unit conversion (drawn from the NVIDIA Nemotron Model Reasoning Challenge). Using NVIDIA Nemotron-3-Nano-30B-A3B with matched LoRA training settings, we compare three symbol-supervision formats: verbose English rule descriptions, compact family tags, and compact formula notation. We hypothesize that supervision renderings bias token-level reasoning priors, and that these priors transfer across task boundaries in multi-task SFT. In the canonical strict-rescore inventory, the best compact tag and formula checkpoints are statistically equivalent in aggregate within a pre-specified ±4-point margin: K8A-800 reaches 72.3% strict-scored overall accuracy and K8B-700 reaches 71.2% (TOST p = 0.003). Compact tags nevertheless provide a cleaner behavioral profile: an earlier K8A-400 checkpoint reaches 66.4% overall, 98.7% gravity accuracy, and 36.9% bit accuracy without the same contamination signatures. In contrast, verbose English rule descriptions are associated with heuristic parroting, with up to 57% of symbol failures at audited verbose checkpoints collapsing to a single remove-operator template, while formula notation is associated with cross-category contamination: numeric-looking predictions appear more often in text decryption (higher at five of six matched training steps under the canonical seed; matched-step means 15.8 vs. 11.7 numeric predictions per 157 text rows), and gravity failures at a representative K8B formula checkpoint shift toward shortcut stubs and explicit g = 9.8/9.81 fallbacks. We further show that checkpoint selection and strict evaluation auditing materially change branch decisions. Across three training seeds, neither compact format shows a consistent aggregate advantage, while the contamination signatures are partly seed-specific: the gravity-shortcut severity difference persists but is not exclusive to the formula branch, and the numeric–text signature does not reproduce under reseeding. These results support treating supervision format as a first-class hyperparameter for multi-task reasoning SFT in large language models—at least in this benchmark-and-model setting—rather than a mere rendering detail. Because such symbolic and procedural reasoning tasks recur in domains including cybersecurity, mathematics, and code generation, the same formatting choices plausibly shape the policy that any later reinforcement-learning stage would inherit, which we flag as future work. Full article

Accurate multi-object tracking and metric localization support traffic monitoring and cooperative intelligent transportation at complex intersections. This study presents a fixed-camera vision-map fusion framework that addresses two practical difficulties: axis-aligned boxes poorly represent turning vehicles, and unconstrained image-plane tracking can produce physically implausible trajectories. A map-aided frontend first generates candidate detections using improved You Only Look Once version 8 nano (YOLOv8n) horizontal bounding box (HBB) branch and an improved YOLOv8 oriented bounding box (OBB) branch. A high-definition (HD) map selector then retains the candidate geometry consistent with the straight-driving or turning region and converts it into a unified detection record. The selected reference point is projected to the ground plane through an offline-estimated homography, whereas the appearance feature bypasses the homography and is passed directly to the association stage. The tracking backend uses a 12-dimensional joint image/metric state, symmetric central-difference evaluations of the process and measurement functions, appearance-motion association, and a feasible-road projection derived from HD-map lane polygons. On the evaluated public sequences, the complete configuration achieved a multiple object tracking accuracy (MOTA) of 74.5%, an identification F1 score (IDF1) of 82.6%, 614 identity switches, and a throughput of 26.8 frames per second (FPS) on an RTX 4090 workstation. In a descriptive Vehicle-in-the-Loop case study involving one instrumented vehicle at one intersection, the overall localization mean absolute error (MAE) was 0.180 m, compared with 0.208 m for the baseline end-to-end configuration. These results indicate the feasibility of combining branch-specific vehicle geometry with map-constrained tracking; controlled same-detector comparisons, repeated multi-vehicle trials, and embedded-device latency and power profiling remain necessary for broader claims. Full article

The global energy sector is undergoing rapid and, in many respects, irreversible transformation driven by the convergence of digital disruption, sustainability mandates, and shifting investor expectations. Technologies such as artificial intelligence (AI), blockchain, and digital twin systems are fundamentally reshaping energy operations and strategic decision-making, while ESG governance quality and renewable energy adoption have emerged as two of the most consequential determinants of corporate financial competitiveness and equity valuation. Despite growing practitioner and regulatory interest in these dynamics, limited empirical evidence exists on how ESG governance, renewable adoption, and digital disruption jointly influence financial performance and environmental outcomes across multiple sectors simultaneously. This study addresses that gap using panel data from 26 large-cap US-listed firms across five sectors over 2015–2022 (N = 208 firm-year observations for Revenue/Market Cap/ROA models; N = 91 for the CO₂ model). A multi-method econometric framework is employed, comprising Fixed Effects and Random Effects panel regression with Hausman specification testing, Difference in Differences quasi-experimental analysis, and sequential OLS path analysis with HC3 robust standard errors. Three of four hypotheses are supported. ESG governance quality generates a significant market capitalisation premium of approximately 10–14% per unit Bloomberg ESG Score improvement, after controlling for firm size and R&D intensity; no significant revenue channel effect is found once firm size is properly accounted for. Renewable energy adoption shows a marginal association with market capitalisation at the 10% significance level (FE β = 0.019, p = 0.086; RE β = 0.016, p = 0.077), suggesting capital markets may price clean energy adoption as a forward-looking signal. ESG governance quality drives within-firm CO₂ emission reduction substantially more powerfully than renewable energy quantity alone, with the Fixed Effects estimator identifying a governance-led eco-efficiency mechanism. Firm profitability functions as a cross-model financial capacity moderator, enabling simultaneous ESG investment and environmental improvement. The findings carry direct implications for corporate managers, institutional investors, and policymakers aligned with SDG 7, SDG 9, and SDG 13. Full article

Smart meters are the core terminals for distribution network data acquisition in smart cities, yet their collected data commonly suffer from quality issues caused by harsh operating environments, communication failures, hardware degradation, and human factors. This paper presents a systematic review of anomaly detection and data repair methods for smart meter data based on a critical analysis of many publications. First, we characterize five typical anomalies—sudden jumps, reading stagnation, reverse readings, pulse spikes, and gradual drifts—from physical root causes to data manifestations and provide unified mathematical definitions with explicit traceability to the existing literature. Additional anomaly types including meter replacement jumps, data duplication from retransmission, complete missing segments, and timestamp errors are also discussed to present a more complete picture of operational data quality challenges. Second, existing anomaly detection methods are systematically reviewed and classified into four categories—statistical, machine learning, deep learning, and dedicated time-series methods—with representative studies, quantitative performance metrics, and scenario-specific applicability examined for each. Third, data repair approaches are reviewed across four categories—traditional interpolation, matrix completion, generative models, and time-series prediction—with systematic comparison of their accuracy and limitations across different anomaly types and durations. Based on the synthesized evidence, we identify three cross-cutting structural limitations that persist across method categories: the performance ceiling of data-only detection without physical constraint embedding, the open-loop architecture that separates detection from repair and allows error propagation, and the exclusive reliance on statistical error metrics that fails to distinguish physically plausible repairs from those violating conservation laws. To address these gaps, we discuss a physics-guided integrated framework incorporating physical constraint embedding, joint anomaly diagnosis, scenario-adaptive repair, and posterior verification as a promising forward-looking direction. Finally, open challenges and future research directions are outlined, including parameter adaptation in unlabeled scenarios, multi-source data fusion for physical disambiguation, new power system extensions, explainable AI integration, edge-computing deployment, and standardized benchmark development. This review provides a comprehensive theoretical reference and technical roadmap for smart meter data quality research in the context of smart city energy systems. Full article

Maintaining durable road infrastructure is crucial for reducing resource consumption, minimizing repair costs, and supporting sustainable urban mobility. However, accurately detecting small and morphologically diverse pavement cracks remains challenging due to variations in lighting, road textures, and crack shapes across different geographic regions. YOLO (You Only Look Once) is one of the most widely adopted deep learning (DL) frameworks for object detection. Traditional inspection methods are labor-intensive and often inconsistent, while existing DL models can be computationally heavy or limited to single crack types, restricting real-time deployment and scalability. To address these challenges, this study presents YOLO26s, a lightweight DL model for multi-class pavement crack detection across diverse environmental and geographic conditions. Using a curated subset of 6972 annotated images from the Road Damage Dataset 2022, YOLO26s identifies four crack types: longitudinal, transverse, pothole, and alligator cracks. Compared to baseline models (YOLOv8s, YOLOv8n, YOLO26n), YOLO26s achieves higher detection accuracy (mAP@0.5 = 89.0%) while reducing computational complexity by 14.3% in parameters and 7.7% in FLOPs, enabling real-time deployment on edge devices. By facilitating early and accurate crack detection, the proposed approach supports proactive maintenance, extends pavement lifespan, and reduces material and energy usage, contributing to more sustainable road network management. These findings highlight the potential of efficient AI-driven inspection systems to enhance environmental and economic sustainability in civil infrastructure. Full article

Accurate coal–gangue instance segmentation remains challenging because visually similar materials, irregular scale variation, and densely adjacent boundaries must be handled simultaneously. To address these coupled problems, this paper proposes STRATUM-Seg, a compact network based on the nano instance-segmentation variant of You Only Look Once version 11 (YOLO11n-seg), which coordinates parameter-sharing symmetry with heterogeneous and direction-selective feature processing. In the backbone, the Shared Dilated Pyramid Module applies one kernel across multiple dilation rates to extract multi-receptive-field texture features with limited parameter redundancy. In the neck, the Multi-Kernel Focus Fusion Neck aligns three pyramid levels, re-injects backbone features, and performs two-stage aggregation using heterogeneous depth-wise kernels. In the prediction head, the Difference-Enhanced Convolutional Head combines a shared trunk with direction-selective difference convolutions and inference-time re-parameterization to improve boundary representation. Experiments on the Wangjialing subset of the Dataset for Coal, Gangue, and Foreign Objects (DsCGF) show that STRATUM-Seg increases mask mean average precision (mAP) over intersection-over-union (IoU) thresholds from 0.50 to 0.95 from 0.615 to 0.668 over three independent runs while maintaining a compact deployment-form model and real-time 32-bit floating-point (FP32) end-to-end inference on the evaluated NVIDIA RTX 4090 platform. Full article

Accurate and real-time apple detection in complex orchard environments is essential for robotic harvesting but remains challenging because of illumination variation, foliage occlusion, and limited computational resources. This study proposes Fast-YOLO11n, a lightweight detector derived from the nano variant of You Only Look Once 11 (YOLO11n) and integrating three complementary components. A Fast-C3k2 module based on partial convolution (PConv) reduces redundant computation while preserving cross-layer feature transmission. A focal modulation (FM) mechanism enhances target-related responses and suppresses background interference under occlusion and uneven illumination. In addition, a parallel downsampling module, termed ADown, retains local geometric details and multi-scale semantic information during downsampling. Experiments were conducted on a field-collected orchard dataset comprising 2240 images and 22,673 annotated apple instances under diverse lighting, scale, and occlusion conditions. Fast-YOLO11n achieved mean average precision values of 75.76% across intersection-over-union (IoU) thresholds of 0.50–0.95 (mAP@50–95) and 91.29% at an IoU threshold of 0.50 (mAP@50), while operating at 366.19 frames per second (FPS) with 2.51 million parameters and 6.00 billion floating-point operations (FLOPs). Compared with the YOLO11n baseline, it improved mAP@50–95 and mAP@50 by 2.39 and 1.39 percentage points, respectively, while reducing the parameter count and FLOPs by 2.71% and 5.36%. Ablation experiments demonstrated the individual and combined effects of the three modules on detection performance and computational efficiency. The proposed model provides a favorable balance between detection accuracy and computational efficiency, indicating its potential for real-time orchard perception on resource-constrained platforms. Full article

This tutorial takes the reader on a historical and technical journey from the simple Perceptron (1958) to modern Convolutional Neural Networks (CNNs) that dominate spatial data processing (images and video). We start with the Perceptron’s linear classifier, then expose its inability to learn non-linear patterns (e.g., XOR), which motivates the Multi-Layer Perceptron (MLP) and the backpropagation algorithm. Next, we discuss the limitations of MLP when faced with structured data like images—parameter explosion, loss of spatial information, and lack of translation invariance—and use these limitations as a natural springboard to the core ideas of CNNs: local connectivity, weight sharing, and hierarchical feature learning. Throughout, we provide intuitive explanations, mathematical formulations, and step-by-step numerical examples (e.g., a complete forward and backward pass for a small network, and a manual 2D convolution). Clear graphical representations and examples help readers understand each concept. The tutorial concludes with a detailed walkthrough of influential CNN architectures (LeNet-5, AlexNet, VGG, GoogLeNet, ResNet, DenseNet, and EfficientNet) and also discusses more recent attention-based models (e.g., Vision Transformers and ConvNeXt), explaining why each was necessary and how it advanced the field. Aimed at students and practitioners with a basic knowledge of calculus and linear algebra, this tutorial connects foundational ideas to state-of-the-art deep learning, focusing on spatial data. It is designed for readers who want to understand why each architectural choice was made, not just what the final model looks like. Full article

The Central Environmental Protection Inspection (CEPI) represents a landmark institutional arrangement for the centralization of environmental supervision authority from local governments to the central government. It has played a significant role in correcting the distortion of local government incentives for environmental governance under decentralized management. However, the persistence and spatial heterogeneity of its policy effects remain to be further examined. This paper treats the CEPI, which was progressively implemented from 2016, as a quasi-natural experiment of centralizing environmental supervision authority. Using daily city-level panel data from 333 Chinese cities over the period 2015–2022, we employ a multi-cutoff time regression discontinuity design in time to evaluate its policy effects. The findings are threefold. First, the CEPI significantly reduces urban concentrations of AQI, PM_2.5, PM₁₀, SO₂, NO₂, and CO, while O₃ concentrations increase significantly over the same period, indicating that the centralization of supervision authority is effective in mitigating conventional primary pollutants but faces limitations in addressing secondary pollutants. Second, the policy effects vary across inspection rounds, with the first round and the “look-back” generating larger short-run coefficients than the second round. These differences in short-run inspection-period coefficients across rounds indicate that marginal policy responsiveness diminishes as the inspection program becomes institutionalized. Third, the policy effects display significant spatial heterogeneity: cities in northern regions, resource-based cities, and cities with historically moderate air quality experience more pronounced improvements, while the grouping effect of administrative hierarchy is not statistically significant for most pollutants. This indicates that the governance effects of centralizing supervision authority vary across city characteristics but do not simply follow an administrative hierarchy gradient. This study provides empirical evidence for understanding the boundary of the effectiveness of centralizing environmental supervision authority and offers policy implications for optimizing the vertical allocation of environmental governance responsibilities. Full article

The human placenta is a temporary organ structured to optimize exchange between the maternal and fetal circulatory systems. Its fetal component consists of highly branched chorionic villi, which are anchored to the maternal uterine wall and project into the intervillous space. The outer surface of these villi is lined by a multinucleated, continuous layer called the syncytiotrophoblast, which is supported by an underlying layer of proliferative cytotrophoblast cells and the invasive extravillous trophoblast (EVT). This cellular bilayer forms a selective barrier that directly bathes in maternal blood, allowing for the efficient transfer of oxygen and nutrients while structurally preventing the direct mixing of maternal and fetal blood cells. Human placental studies have been stymied by ethical and accessibility constraints. Stem cell biology has now revolutionized the capacity to model human placental development, in particular with the derivation of human trophoblast stem cells (hTSCs) and organoids. Authentic, self-renewing human trophoblast stem cells (hTSCs) were first derived not from pluripotent stem cells but from primary tissue—first-trimester villous cytotrophoblasts and blastocysts. Derivation from human pluripotent stem cells (PSCs) followed only subsequently, along two principal routes: conversion of naive PSCs, which retain extraembryonic competence, and induction from primed PSCs, as well as by direct reprogramming of somatic cells to induced hTSCs. An important advance underlying these improvements is the mapping of a global reprogramming roadmap. Multi-omic and lineage-tracing experiments have mapped the stepwise transcriptional and epigenetic conversions of fibroblasts to hTSCs, including sequential chromatin reconfiguration, trophoblast gene network activation, and repression of somatic signatures. These results identify major regulatory bottlenecks and intermediate states, improving reprogramming fidelity. The derivation of stem-cell-based trophoblast organoids now enables complex modeling of placental architecture, function, and disease susceptibility in vitro. These organoids accurately recapitulate placental barrier functions and immunological features, allowing for examinations of maternal–fetal health, pregnancy disorders, and placental infection response to viruses like cytomegalovirus and SARS-CoV-2. Looking ahead, the integration of reprogramming and organoid technologies will propel patient-specific and tailor-made models for personalized diagnostics, drug screening, and mechanism studies. As we unravel the molecular ballet of trophoblast induction, such discoveries have the potential to bridge basic translational gaps in reproductive biology and maternal–fetal medicine. Full article

Electricity is fundamental for all modern infrastructures, while disruptions in transmission networks could result in severe failures in healthcare, transportation, communication, industry, and all related to public safety. However, inspection of overhead lines and their components is costly and labor-intensive. The convergence of Unmanned Aerial Vehicles (UAVs), advanced imaging sensors, and deep-learning-based Computer Vision has reshaped this domain. To this end, this work presents a systematic review of Computer Vision applications in electric power transmission line inspection. From an initial 1493 Scopus records, 148 studies published between 2018 and 2026 were retained through a transparent, multi-stage screening and quality-scoring process based on PRISMA guidelines. The reviewed literature was synthesized across four axes: (1) monitoring platforms and sensor technologies, (2) Computer Vision approaches per vision task, (3) datasets, metrics, and evaluation practices, and (4) synthesis of results and industrial adoption. The analysis of the literature confirms the dominance of You Only Look Once (YOLO) family models for real-time edge deployment and the rising adoption of Transformer architectures, while exposing persistent gaps in dataset availability, domain generalization, and field validation. The review concludes with the open challenges and concrete future directions toward fully autonomous, robust, and sustainable inspection systems. Full article

Accurate multi-level grape counting in natural orchard images provides useful visual information for image-based crop-load characterization and vineyard management. However, cluster overlap, occlusion, boundary adhesion, and illumination variation complicate berry localization and cluster-wise counting. Existing detection- and density-regression-based methods remain susceptible to missed detections, false positives, and cross-cluster misassignment, whereas approaches relying on pixel- or instance-level supervision typically require costly annotations. To address these challenges, this paper proposes UniGC, a unified framework for multi-level grape counting. At the berry level, UniGC employs CSD-P2PNet, a P2PNet-based point-regression branch equipped with a Condition-Structure Decoupled Frequency Enhancement Block (CSD-Block), to improve the localization and counting of image-visible berries under complex illumination and occlusion using only berry point annotations. At the cluster level, a You Only Look Once (YOLO) detector predicts grape-cluster bounding boxes, which are used as prompts for the Segment Anything Model (SAM) to generate cluster instance masks. We further propose Boundary-Depth Guided Mask-Point Assignment (BDMPA), which assigns each predicted berry point to at most one grape-cluster instance and thereby produces cluster-wise berry counts. Selected key experiments on the mixed dataset organized from WGISD and GBISC were repeated using three random seeds. CSD-P2PNet achieved an MAE of 23.4407 ± 0.1577, an RMSE of 32.8898 ± 1.2682, and a point-localization F1 score of 0.8374 ± 0.0097. YOLO12-S achieved an F1 score of 0.8433 ± 0.0056 and an AP50 of 0.8789 ± 0.0080 for grape-cluster detection. For BpGC counting, UniGC obtained a matched-cluster MRD of 0.1288 ± 0.0043, a $1-\mathrm{FVU}$ of 0.8589 ± 0.0098, and an all-cluster penalized mean relative deviation (ACP-MRD) of 0.5062 ± 0.0124. These results demonstrate that UniGC achieves competitive multi-level grape counting performance while requiring only point and bounding-box annotations for training. Full article