Abstract: The integration of artificial intelligence (AI) in digital classrooms has introduced both opportunities and challenges for academic honesty. This narrative review study explored how AI tools influence students’ learning behaviors, assessment practices, and ethical decision-making in academic tasks. Data were collected from students and educators through surveys, interviews, and document analysis, focusing on AI-assisted writing, digital platforms, and institutional policies. Findings reveal that while AI can enhance learning efficiency and engagement, it also blurs the boundary between legitimate academic support and misconduct. Many students perceive AI use as similar to peer assistance, resulting in uncertainty regarding ethical practices. Lower proficiency students were particularly prone to reliance on AI-generated outputs, highlighting the need for targeted instructional support. Traditional assessment formats, such as essays and take-home assignments, were identified as vulnerable to AI misuse, prompting calls for process-oriented evaluations, reflective tasks, and in-class assessments. The study also emphasizes the importance of clear institutional policies and AI literacy programs in promoting responsible use. Moreover, emerging technological risks, including deepfake content, underscore the necessity of proactive guidance and monitoring. Overall, the research suggests that fostering academic integrity in AI-mediated classrooms requires a balanced approach, combining ethical education, innovative pedagogy, and policy development. By cultivating transparency, critical thinking, and responsible AI engagement, institutions can maximize AI’s educational benefits while safeguarding authenticity and integrity in student work.

Abstract: Grid-forming (GFM) renewable energy sources are increasingly integrated into power grids to enhance the stability of high-penetration renewable energy systems, while the fault current characteristics of GFM-based outgoing lines lead to the inapplicability of conventional longitudinal differential protection, which suffers from reduced sensitivity or even refusal to operate under weak grid conditions. To address this issue, this paper proposes a novel active detection-based protection strategy for GFM photovoltaic power station outgoing lines based on the amplitude ratio of characteristic harmonic signals. First, the sequence equivalent circuits of the GFM system during grid faults are established to analyze the fault current characteristics, and the inapplicability mechanism of conventional pilot differential protection is revealed. Considering the filter cutoff frequency, harmonic interference avoidance and power quality constraints, the 8th harmonic is selected as the characteristic signal, and a proportional-resonant (PR) controller is adopted to realize the independent and flexible injection of the characteristic signal and power frequency signal. Based on the distribution difference of characteristic signals under internal and external faults, a protection criterion is constructed using the amplitude ratio of the harmonic component of the differential current to the characteristic signal injected on the station side. The simulation results on the MATLAB/Simulink platform show that the proposed strategy can quickly and accurately distinguish various internal and external faults of the transmission line, and operate reliably under different fault types, fault locations and high transition resistance.

Abstract: Banana ripening is a complex biological process that determines fruit quality and shelf life, yet the integrated behavior of physicochemical, nutritional, and enzymatic attributes throughout ripening remains insufficiently understood. This study applied a multivariate approach to characterize and classify Ecuadorian bananas across nine ripening stages (T1–T9). Twenty-seven samples (three replicates per stage) were analyzed considering 29 variables, including carbohydrates, proximate composition, minerals, vitamins, color, texture, and enzymatic activity (PPO and POD). Data were evaluated using PERMANOVA, Principal Component Analysis (PCA), Spearman’s correlation, and k-means clustering. PERMANOVA confirmed that ripening stage explains nearly all multivariate variation (pseudo-F = 2758.3; R² = 0.999; p = 0.001). PCA revealed a dominant gradient (Dim1 = 86.2%) describing a coordinated transition from green stages characterized by high starch content, firmness, and mineral concentration to overripe stages associated with sugar accumulation, increased PPO/POD activity, and tissue softening. Vitamin C reached its maximum value at the intermediate stage (T5). These findings indicate that banana ripening follows a synchronized physiological gradient, allowing the identification of functional ripening stages based on multivariate signatures, which may support improved postharvest management and the development of non-destructive monitoring strategies aligned with sustainable food systems.

Abstract: The microscopic pairing mechanism in unconventional superconductors remains elusive, largely because the extreme flatness of the superconducting band often obscures key energy-momentum dispersion features observed in angle-resolved photoemission spectroscopy. In this work, we re-examine high-resolution dispersion data from cuprates (Bi2212 and Bi2201) and iron-based superconductors (monolayer FeSe) to test the predictions of a newly proposed chiral electron-hole (CEH) pairing mechanism. Unlike Cooper pairs in BCS-like theories that form a single quasiparticle band with a smooth back-bending dispersion, CEH pairs exhibit a distinct two-band structure in quasiparticle dispersion with sharp cusps at the back-bending points. Our analysis identifies clear empirical signatures of these CEH-predicted features, concluding that quasiparticle dispersions in these strongly correlated materials deviate significantly from BCS-like behavior. Further comprehensive and targeted experimental strategies are proposed to definitively resolve the subtle dispersion features and rigorously test the CEH model for unconventional superconductivity.

Abstract: Standard accuracy benchmarks evaluate whether a language model produces correct outputs but not whether it produces them consistently. We demonstrate that accuracy and output predictability are independent dimensions (Pearson r = -0.24, p = 0.56, N = 8 medical LLMs) when evaluated at a critical clinical summarization position. This independence yields a four-class behavioral taxonomy: IDEAL (convergent and accurate), EMPTY (convergent but inaccurate), DIVERGENT (high variance with incomplete outputs), and RICH (moderate variance with high accuracy).The DIVERGENT class exhibits stochastic incompleteness—summaries that are factually accurate but randomly incomplete across trials, with zero hallucinations. LAD occlusion, a critical clinical finding in STEMI cases, appears in only 22% of Llama 4 Scout summaries despite the model correctly identifying it when directly queried. This failure mode is invisible to standard benchmarks that average across outputs rather than measuring trial-to-trial variance.We propose a two-dimensional framework (Predictability × Accuracy) as a minimum requirement for clinical AI assessment, identify specific models unsuitable for deployment (Llama 4 Scout with Variance Ratio = 7.46; Llama 4 Maverick with Variance Ratio = 2.64), and flag one model requiring safety filter reconfiguration (Gemini Flash, 16% accuracy due to over-refusal). These findings demonstrate that current single-metric evaluation approaches systematically miss critical safety failures in clinical AI systems.

Abstract: This study evaluated the predictive performance of Random Forest, Bagged Trees, Support Vector Machines (SVM), and Least Squares Boosting (LSBoost) for estimating Tunnel Boring Machine (TBM) penetration rate (ROP). While all models achieved acceptable accuracy, LSBoost outperformed the others, showing the highest correlation (R = 0.965) and coefficient of determination (R² = 0.909), along with the lowest RMSE and MAE. Its performance remained robust after Z-score normalization, highlighting its ability to capture nonlinear parameter interactions and generalize well on limited geotechnical datasets. Random Forest and Bagged Trees showed similar performance, with Bagged Trees only slightly improved by normalization. SVM performed less effectively, indicating limited capacity to model complex TBM penetration behavior. Feature importance and SHAP analyses identified discontinuity spacing (DPW) and uniaxial compressive strength (UCS) as the primary controlling factors, while brittleness index (BI) was more influential within the SVM model. Agreement between Jacobian-based derivative analyses and SHAP results confirmed both mathematical sensitivity and engineering interpretability. Overall, TBM penetration prediction is a multivariate and inherently nonlinear problem. LSBoost provides reliable and high-accuracy predictions even under data-constrained conditions. The combination of SHAP- and PDP-based feature importance analyses enhances interpretability, supporting engineering decision-making in TBM design and operation. These findings emphasize the applicability of machine learning approaches for accurate, interpretable, and robust TBM performance prediction.

Abstract: The effective treatment of neurodegenerative diseases (NDDs), such as Alzheimer's disease, Parkinson's disease, and amyotrophic lateral sclerosis, remains a critical challenge in modern medicine. Given the limitations of current therapies, alternative strategies to slow neurodegeneration are urgently needed. This study presents a critical review of the current evidence regarding low-dose ionizing radiation (LDIR) as a promising modality for modulating neurodegenerative processes. This study examines current experimental data on the effects of LDIR on cellular protective and compensatory mechanisms, including evidence from in vivo models of NDDs. Our analysis demonstrates that LDIR enhances antioxidant activity and DNA repair, stimulates autophagy and neuroplasticity, and modulates neuroinflammatory signaling. Collectively, these findings support the hypothesis of the neuroprotective potential of LDIR, underscoring its translational viability provided that strict dosimetric guidelines are followed and individual biological responses are rigorously monitored.

Abstract:

Background. Perivascular adipose tissue (PVAT) contains adipocytes and a stromal-vascular fraction with immune cells that modulate the adjacent vasculature. The presence of immune cells in PVAT of vascular beds is poorly understood—are they resident or recruited? We propose a novel resident microbiome is present in PVAT, given the immune-rich stromal environment. Hypothesis. We hypothesized the existence of distinct bacterial and viral communities in healthy PVAT compared to non-PVAT adipose tissues. Methods. PVAT samples from thoracic and abdominal aorta, mesenteric resistance arteries, non-PVAT tissues (subscapular brown adipose tissue, retroperitoneal white adipose tissue), and fecal samples were collected one year apart from male Dahl SS rats, split into two cohorts (2023 and 2024, n=3 each). Whole-genome shotgun sequencing (CosmosID) and 16S rRNA gene analysis assessed microbial relative abundance. Results. PVAT harbored bacterial and viral sequences, and species composition varied significantly between cohorts. Bacterial and viral fecal samples showed lower variability. Conclusions. This confirms a microbiome in PVAT that differed dramatically from the fecal microbiome, with temporal influences on bacterial and viral diversity, marking the first such report. These findings establish the potential of PVAT microbiota in vascular biology and immune modulation, paving the development of microbiome-targeted drugs to address vascular dysfunctions.

Abstract:

In a field experiment with papaya (Carica papaya L.), the effects of three soil cover types (bare soil, living cover of Canavalia ensiformis, and senescent cover) and three microbial biofertilisers (Bacillus subtilis, Pseudomonas putida, and Trichoderma viride) on crop growth and yield were evaluated. Vegetative and reproductive variables were monitored over 187 days, and the data were analysed using generalised linear mixed models (GLMMs) and generalised linear models (GLMs). The results indicated that soil cover was the dominant factor, explaining the largest proportion of variation in plant growth and final yield (p < 0.001), whereas biofertilisers did not exhibit significant main effects when applied independently. However, significant interactions between soil cover and biofertiliser were detected (p < 0.05), demonstrating that inoculant efficacy was strongly context-dependent. Trichoderma viride increased stem diameter by approximately 7% under living cover only, while Pseudomonas putida showed a comparative advantage under bare soil conditions, increasing final fruit weight by approximately 32%. Principal component analysis (PCA) further confirmed that treatment groupings were primarily driven by soil cover type. These findings provide field-based evidence that the efficiency of microbial biofertilisers in promoting papaya growth depends on edaphic conditions shaped by soil cover management. A hierarchical management strategy is therefore proposed, in which establishing a favourable soil habitat through plant cover is a prerequisite for maximising the benefits of microbial inoculants in tropical fruit production systems.

Abstract:

Hyaluronic acid (HA)–based nanocapsules containing plant-derived bioactives are promising formulations for dermatological applications. In this study, nanocapsules containing extracts of Arnica montana, Calendula officinalis and Aesculus hippocastanum were synthesized and their structural and functional properties were characterized. Scanning electron microscopy confirmed the formation of spherical nanostructures with uniform morphology, while rheological analyses demonstrated stable viscoelastic behavior suitable for topical application. Their antimicrobial potential was assessed on microorganisms isolated from multiple regions of healthy human skin and opportunistic pathogens. A diverse panel of approx.. 100 bacterial and fungal isolates was identified using MALDI-TOF MS. Antimicrobial activity of formulations was compared with commonly used disinfectants: H₂O₂, octenidine, isopropanol and topical ophthalmic antiseptic. Arnica-based formulations showed the strongest inhibitory effect against both Gram-positive and Gram-negative bacteria, whereas chestnut extract demonstrated selective activity against Candida spp. Calendula-based formulation exhibited limited antimicrobial activity. These findings demonstrate that plant-extract-loaded HA nanocapsules exhibit selective antimicrobial properties dependent on extract type and microbial group, supporting their potential as multifunctional components of future dermatological formulations.

Abstract: This study investigated the effects of different application rates of spent mushroom substrate (SMS) from Morchella sextelata on soil properties and microbial communities in a moso bamboo (Phyllostachys edulis) plantation. Three SMS rates (2.4, 4.7, and 9.4 kg·m⁻² ) were applied, and soil samples were collected at 6 and 12 months from two depths (0–20 cm and 20–40 cm). One year after application, topsoil total phosphorus (TP) increased 12–20 fold, while available phosphorus (AP) and potassium (AK) were significantly elevated. Soil pH initially decreased but partially recovered, whereas electrical conductivity (EC) continued to rise, indicating salt accumulation. Urease (UA) and sucrase (SA) activities increased 10–17 fold and 3–5 fold, respectively, while catalase (CAT) and acid phosphatase (ACP) were temporarily suppressed. SMS application significantly altered microbial community composition, with Acidobacteriota and Basidiomycota becoming more abundant. Correlation analysis identified pH, organic matter, AP, and UA as key factors linked to microbial changes. The medium application rate (4.7 kg·m⁻² ) provided the best balance between soil improvement and environmental risk. These findings demonstrate that M. sextelata SMS can effectively enhance soil fertility while modulating microbial communities, but salt accumulation and short-term acidification warrant attention.

Abstract: Cyber Ranges (CRs) are complex socio-technical ecosystems, combining infrastructure resources, software services, learning mechanisms, and human-in-the-loop processes for cybersecurity training, education, and experimentation. However, their design and representation are conventionally described by diverse architectural representations and a lack of standardization, making them difficult to compare, integrate, and reason in an automated manner. This paper proposes a novel framework that uniquely integrates the structural, functional, informational, and decisional aspects of CR platforms, formalizing them into a common semantic framework. It models the architectural and learning characteristics of CRs, allowing the representation of design choices, operational processes, information resources, and capability development. The ontology is implemented using OWL 2 DL, which includes logical constraints and enables consistency checking and automated reasoning. Validation through instantiation and competency question assessment shows that the model allows for structured querying, traceability across abstraction levels, and capability-level reasoning. The findings indicate that ontology-based modeling can serve as a basis for more formalized CR configuration analysis and capability-focused evaluation of diverse CR platforms.

Abstract: Social sustainability, encompassing equitable development, livable urban construction, and inclusive social governance, has become a core dimension of sustainable development goals (SDGs) and urban planning practice in the context of rapid urbanization. Taking the Yangtze River Delta (YRD) urban agglomeration—the most economically developed and urbanized region in China—as the study area, this research constructed a comprehensive evaluation index system of social sustainability from three dimensions: equitable resource allocation, livable urban environment, and inclusive social development. Based on multi-source data from 2010 to 2020 (including socioeconomic statistics, remote sensing imagery, and open geospatial data), the entropy weight-TOPSIS model was used to measure the spatiotemporal evolution characteristics of social sustainability in the YRD. Additionally, the GeoDetector model was employed to identify the key driving factors and their interactive effects on the spatial differentiation of social sustainability, and the future development trends of social sustainability under three scenarios (urban expansion, ecological priority, and coordinated development) were predicted using the PLUS model. The results showed that the overall social sustainability of the YRD presented a steady upward trend from 2010 to 2020, with a spatial pattern of "high in the core, low in the periphery" and significant inter-city disparities. Equitable resource allocation was the primary constraint on social sustainability in peripheral cities, while livable urban environment was the main advantage of core cities such as Shanghai, Nanjing, and Hangzhou. The driving factor detection indicated that per capita GDP, urban green space rate, and the number of medical and educational institutions per 10,000 people were the top three key factors affecting social sustainability, with the interactive effect of any two factors showing a dual-factor enhancement pattern. Under the coordinated development scenario, the social sustainability of the YRD will achieve the most balanced and high-quality growth by 2030, with the peripheral cities narrowing the development gap with the core cities significantly. These findings imply that future urban development in the YRD should adhere to the concept of coordinated and inclusive development, optimize the spatial allocation of public resources, and promote the integrated construction of livable cities, so as to realize the high-quality social sustainability of the urban agglomeration. This study provides a quantitative method and empirical reference for the evaluation and optimization of social sustainability in China’s urban agglomerations and even in other rapidly urbanizing regions worldwide.

Abstract: Soluble solid content (SSC) is a critical indicator of ‘Red Fuji’ apple quality, directly governing fruit grading and maturity assessment processes. Conventional SSC measurement by refractometry is destructive and time-consuming, rendering near-infrared diffuse reflectance spectroscopy (NIR-DRS) a promising nondestructive alternative. In this study, a low-cost and compact embedded spectrometer named as DLP NIR-scan Nano EVM was used to acquire NIR-DRS spectra of ‘Red Fuji’ apples for SSC prediction. To improve prediction accuracy, we combined spectral preprocessing with machine learning methods. The dataset was cleaned using Monte Carlo outlier detection, and samples were divided into calibration and validation sets via Kennard–Stone (KS) and joint X-Y distance (SPXY) algorithms. Among preprocessing methods tested, a 12-point second derivative performed best when paired with KS partitioning. For feature-wavelength selection on the preprocessed KS data, competitive adaptive reweighted sampling, Monte Carlo uninformative variable elimination, and Random Frog were applied to the second-derivative spectra. Partial least squares regression (PLSR) models were then built using both full-spectrum data and four sets of selected wavelengths. The best preprocessed PLSR model achieved R2c = 0.916, RMSEC = 0.4093%, R2p = 0.8632, and RMSEP = 0.537%. These results demonstrate that NIR-DRS, combined with appropriate preprocessing and modeling strategies, offers a reliable, rapid, and nondestructive method for apple SSC quantification, paving the way for portable, cost-effective instruments for commercial fruit quality monitoring.

Abstract: Introduction: With age and injury, the infiltration of fat in the paraspinal muscles can cause degeneration, disorganizing the structural integrity of the connective tissue, causing lower back pain (LBP). Human umbilical cord tissue allografts (UCTa) have a collagen-rich matrix with various ECM components that can replace damaged connective tissue. The objective of this study is to evaluate preliminary findings on the safety and efficacy of UCTa for the supplementation of degenerated tissue in thoracic and lumbar paraspinal muscles refractory to standard conservative methods. Materials and Methods: A total of 141 patients from an observational repository were identified with paraspinal muscle degeneration. Patients received one to three applications of UCTa, outcomes were tracked using the Numeric Pain Rating Scale, the Western Ontario and McMaster University Arthritis Index, and the Quality-of-Life Scale. Results: All groups showed positive improvement in the NPRS and WOMAC scales. Multi-application groups revealed statistically significant differences in the analyses. No adverse events or complications were reported. Discussion: Limitations included a lack of a control group and the increase of recall and response bias due to using patient-reported measures. Conclusion: This pilot investigation highlights the need for continued research through randomized controlled trials to validate efficacy, establish optimal dosage protocols, and compare UCTa to other conservative interventions.

Abstract: This article seeks to explore and analyze the interrelationship between environmental factors, the structure of the energy sector, and stability/resilience within the financial sector by employing data from OECD countries between 2004 and 2021. The article utilizes new data sets provided by the World Bank Group's Global Financial Development Data and Sovereign ESG Data, with specific emphasis placed on the bank capitalization indicator, which is described as the bank capital asset ratio, and is considered an important factor in sectoral stability/resilience. Using fixed effect panel data econometrics, the article suggests that methane emissions, PM2.5 air pollution, and net energy imports have statistically significant impacts on the bank capitalization process, while renewable energy and bank capitalization have positive and statistically significant associations. The positive association between fossil fuel consumption and bank capitalization suggests that there is an inherent contradiction between current sectoral stability/resilience and the challenges associated with the energy transition process. The Hausman test suggests that omitted variables may exist and that fixed effect econometrics is an appropriate model. Clustering analysis suggests that each country has an underlying regime driven by environmental factors, the structure of the energy sector, and sectoral stability/resilience. Moreover, machine learning regression analysis employing K-Nearest Neighbors (KNN) and Random Forest models indicate that significant predictive potential is possible and that energy dependence, renewable energy, and air pollution are important factors in bank capitalization processes. The article suggests that robust evidence is provided regarding environmental quality and its interrelationship with sectoral stability/resilience and has significant implications for developing macroprudential frameworks that incorporate elements of the energy transition process.

Abstract: Water-In-Salt (WIS) electrolytes are expected to replace the expensive and environmentally harmful organic electrolytes while delivering high voltages and improved system safety. In this study, we conducted a failure modes, mechanisms, and effects analysis of a highly concentrated potassium acetate (KAc) electrolyte, evaluated and degraded at 2V in a conventional EDLC carbon-based symmetric configuration. The adopted method provides a simplified yet effective approach for assessing the complexity and interconnectivity of degradation mechanisms in a WIS supercapacitor. The effects analysis included electrochemical stability studies, post-mortem characterizations (SEM-EDS and XPS), low-frequency impedance fitting, and cell reassembly using end-of-life electrodes. Among the failure modes analyzed, electrolyte decomposition and pore blocking exhibit strong physicochemical correlations and high failure rates. Therefore, they should be prioritized in the design of new WIS electrolyte compositions for next-generation energy storage systems.

Abstract: Understanding the hydration dynamics of montmorillonite clay minerals is critical for predicting their behavior in geotechnical and environmental applications. This study employs in situ environmental scanning electron microscopy (ESEM) combined with X-ray diffraction (XRD) to directly observe and quantify the wetting and drying processes of montmorillonite SWy-1 under controlled pressure and temperature conditions. To characterize the real-time wetting and drying morphologies of montmorillonite and determine the relationship between water-induced swelling and relative humidity, ESEM enabled direct visualization of water-clay interactions by precisely controlling chamber pressure (4–5.3 Torr) and temperature (~2°C) to manipulate relative humidity and induce water condensation on mineral surfaces, while quantitative analysis of particle areas before and after hydration determined swelling percentages, XRD measured basal spacing (d₀₀₁) changes across relative humidity gradients, and water-adsorption isotherms were constructed from ESEM thickness measurements. ESEM revealed distinct wetting stages with water preferentially condensing on unsaturated edge sites and external surfaces at low pressures (<4.6 Torr), followed by rapid interlayer filling at elevated pressures with characteristic structural rounding and aggregate formation, while anisotropic swelling ocurred predominantly perpendicular to clay layers, with single water-layer hydration (1W) producing ~19% swelling and two-layer hydration (2W) yielding ~32% swelling, water-adsorption isotherms exhibited exponential swelling behavior with pronounced type H3 hysteresis, logarithmic analysis revealed steeper pressure dependency during hydration (slope = 2.7249) versus dehydration (slope = 1.6702) indicating thermodynamically driven water uptake but kinetically limited desorption, and rapid dehydration kinetics occurred within 3 minutes with complete equilibration by 15 minutes. ESEM successfully bridges microscale observations and molecular-scale understanding of smectite hydration, establishing practical timescales for clay equilibration and providing critical insights for predicting clay behavior in geotechnical engineering, soil stabilization, contaminant transport, and engineered barrier design.

Abstract: Background: Care for knee osteoarthritis (KOA) is frequently fragmented, and pathway-level decisions within Physical Medicine and Rehabilitation (PM&R) are influenced by local organizations. We sought expert consensus on an ideal, function-oriented KOA care itinerary deliverable in PM&R services. Methods: A two-round Real-Time Delphi study was conducted using the SmartDelphi web platform. A steering committee of five PM&R physicians developed a 37-item questionnaire covering referral/access, functional and outcome assessment, conservative management, escalation/referral thresholds, and follow-up/discharge. Round 1 was online (SERMEF osteoarthritis working group; 46 invited, 40 completed; 87.0%) with responses collected until 30 April 2025. Round 2 was an in-person, facilitated validation round on 30 May 2025 at the SERMEF Congress (A Coruña; 85 invited, 70 completed; 82.4%). Items were rated on a 6-point Likert scale; consensus strength was defined by interquartile range (IQR): strong (0–1) vs weak (≥2). No patient-level data were collected; participant characteristics were comparable across rounds, suggesting consensus refinement reflected deliberation rather than panel shifts over time. Results: Consensus supported a longitudinal, function-first pathway structured into five phases: entry/referral to PM&R; comprehensive functional assessment using a minimum outcomes dataset (pain VAS/NRS, WOMAC function, quality-of-life scale); multimodal conservative rehabilitation combining exercise/physiotherapy, education/self-management support, and indicated oral/topical therapies; reassessment-guided escalation in non-responders, reserving interventional PM&R techniques, multidisciplinary musculoskeletal pain-unit management, or orthopaedic evaluation for persistent pain and/or functional limitation; and longitudinal monitoring with defined discharge criteria. Conclusions: SERMEF PM&R experts converged on an implementation-oriented, outcomes-driven KOA itinerary centred on functioning, conservative multimodal care, structured reassessment, and explicit discharge planning.

Abstract: Drought has become a major threat among extreme weather events impacting ecosystems, the economy, food production, and livelihoods. Since the beginning of this century, it has significantly affected Nigeria's economy by reducing agricultural productivity and internally generated revenue. In Northern Nigeria, the shift from meteorological to agricultural drought has not been adequately monitored, particularly concerning future predictions using Artificial Intelligence (AI) and Remote Sensing (RS) methods. Therefore, this study employs AI and EO techniques to analyse and forecast the spatiotemporal dynamics of agricultural drought propagation in North-Central Nigeria from 2000 to 2024. The Vegetation Condition Index (VCI), the Temperature Condition Index (TCI), the Temperature Vegetation Drought Index (TVDI), and the Standardized Precipitation Evapotranspiration Index (SPEI) were used to evaluate vegetation health, temperature variation, and drought severity during the study period. For the machine learning component, Gradient Boosting Regressor was used to predict drought events over five years using cross-validation methods. This study confirms persistent drought events in 2011, 2015, and 2022, with the propagation of meteorological to agricultural drought in 2015, as indicated by VCI, TCI, and TVDI. The integration of AI and EO approaches for drought propagation assessment could enhance climate resilience efforts (SDGs 2, 13 & 15) and provide a framework for drought mitigation strategies in regions prone to drought recurrence, including the study area.