Abstract: Achieving integrated water resources management at all levels, as called for by Sustainable Development Goal (SDG) target 6.5, requires assessment tools that operate at the local administrative scale. However, watershed sustainability assessments are mostly conducted at the whole-basin or provincial scale, which masks the spatial disparities that matter for local water management. This study develops a sub-district-scale Watershed Sustainability Index (WSI) for the Nong Han Basin, Thailand, by integrating the HELP framework (Hydrology, Environment, Life, and Policy) with the Pressure–State–Response structure, a calibrated QSWAT hydrological model, and spatial analysis in a geographic information system, covering 25 sub-districts. The results show that the basin has a moderate-to-high level of sustainability, with a mean WSI of 0.620: 18 sub-districts are classified as high and 7 as moderate, and none fall into the low category. The Life and Hydrology dimensions are the strongest, whereas the Policy dimension is the limiting factor in most sub-districts. This limitation arises from a low Response component (0.19) rather than from a lack of institutional capacity, as confirmed by the finding that sub-districts with low and high policy scores differ only in the Policy dimension. The apparently uniform aggregate index, combined with the high disparity among dimensional scores, confirms the value of diagnosis at the sub-district scale. The proposed framework translates the assessment results into spatial prioritization, an agency-linked decision matrix, and an intervention typology, thereby supporting evidence-based water management by local administrative organizations.

Abstract: We propose an ‘Invisible Bridge’ framework that links in situ measurements to remote sensing observations of atmospheric pollutants via atmospheric chemical transformations. An exploratory multivariate workflow, non-negative matrix factorization (NMF, used here as a PMF-type decomposition), principal component analysis (PCA) and k-means clustering identified three robust atmospheric regimes. One factor is a primary circulation signal (high levels of NOx and VOCs linked to traffic), another is a magnification/resuspension regime, and a third is a secondary ‘regenerated’ regime dominated by O3, SO2 and secondary PM. Compositions based on remote sensing data regimes show that the secondary/regenerated regime represents the crucial chemical and vertical link between surface emissions and satellite measurements integrated in the column, thus improving the interpretation of satellite observations in terms of surface exposure. The results show that chemical regeneration and regime context strongly modulate surface-column consistency, which has implications for the interpretation of satellite air quality indicators considering regime.

Abstract: Point to the manufacturing process of variable-section rectangular tube, a four-die radial extrusion process was designed in this article. The general deformation law was researched in two-dimensions, the influence of deformation parameters on the deformation results was explored, and the method by using the mold anti-deformation to eliminate the concave defects was proposed. For the tube parts with common thickness to diameter ratio, the corresponding optimal die arc degree is obtained through simulation and optimization. After that, the two-dimensional deformation law is applied to three dimensions, and the methods to eliminate the defect are carefully studied and successfully realized. Taking a variable-section rectangular tube as an example, the experiment was carried out. The experimental results were compared with the simulation results, and the accuracy of the simulation results and the feasibility of the process were verified.

Abstract: A series of molecularly imprinted polymers (MIPs) based on hydroxypropyl-β-cyclodextrin (HPCD) crosslinked with 1,6-hexamethylene diisocyanate (HMD) or toluene diisocyanate (TDI), as well as hybrid chitosan–HPCD polymers crosslinked with genipin, were synthesized using levofloxacin and fluorescein as template molecules. The structure and spatial organization of the obtained materials were characterized by FTIR spectroscopy, FTIR microscopy mapping, and ζ-potential measurements. The influence of pH, crosslinker content, and template structure on sorption performance was investigated. All MIPs exhibited maximum sorption at pH 3.0. The highest sorption capacity toward levofloxacin was achieved for the Chit–HPCD–G polymer (74.8 mg/g), whereas the fluorescein-imprinted HPCD–TDI (1:1) MIP demonstrated the highest sorption capacity (135.5 mg/g) and selectivity coefficient (84.1). Dynamic column experiments con-firmed efficient analyte extraction, reducing the analyte concentration by more than 90% after ten loading cycles. All synthesized MIPs exhibited excellent regenerability, with less than 3% loss of sorption efficiency after ten consecutive sorption–desorption cycles. The applicability of the developed sorbents to real matrices was demonstrated using milk and blood plasma samples after minimal sample preparation. Fluorescein extraction efficiencies reached 97.6% and 93.6% for milk and plasma, respectively. The obtained results demonstrate that HPCD-based MIPs combine high sorption capacity, exceptional selectivity, operational stability, and applicability to complex biological matrices, making them promising materials for selective sample preparation, analyte pre-concentration, and controlled drug delivery systems.

Abstract: A current interest in lunar exploration led by Artemis program pushes scientists to search for lunar water deposits directly on the surface of the Moon with small robotic rovers. The Institute of Experimental and Applied Physics, Czech Technical University in Prague (IEAP CTU) is developing a miniature Timepix3-based detector Neutron HardPix capable of mapping the water deposits using non-invasive detection of neutrons created underground by cosmic rays and thermalized by hydrogen. This neutron spectrometer measures count rate variations of thermal, epithermal and fast neutrons attributed to the hydrogen abundance in lunar subsurface, while monitoring cosmic radiation as a natural source of neutrons. Neutron HardPix is based on the miniature (< 0.1 U, 130 g) radiation monitor HardPix with multiple space heritage.

Abstract: Sunday Schools provided access to books that would otherwise have not been available to rural children and teenagers in Australia. These books were largely written in the late nineteenth century by lower middle-class women, the daughters of clerks and grocers in England. The books produced by these writers were hardly sedate or regimented but laced with techniques deriving from the sensation novel, with unreliable narrators and unreliable figures of authority. An examination of one young woman’s reading choices reveals a world characterized by insecurity in identity, danger and uncertainty.

Abstract: Pancreatic ductal adenocarcinoma (PDAC) remains one of the deadliest malignancies worldwide because of delayed diagnosis, rapid metastatic progression, profound immune suppression, and limited therapeutic responsiveness. The pancreatic tumor microenvironment is characterized by severe hypoxia, stromal desmoplasia, metabolic stress, and oxidative imbalance, all of which contribute to tumor progression and resistance to therapy. Among immune cells involved in antitumor defense, natural killer (NK) cells play an important role through direct cytotoxicity and cytokine production. However, NK-cell activity is frequently impaired in PDAC due to oxidative stress, altered nutrient availability, mitochondrial dysfunction, and dysregulated signaling pathways such as AMP-activated protein kinase (AMPK) and mammalian target of rapamycin (mTOR). This review examines current evidence regarding the interactions among redox biology, nutrient sensing, NK-cell metabolism, and pancreatic cancer progression. Importantly, much of the available evidence derives from in vitro studies, animal models, or early-phase clinical investigations, and several findings remain controversial or inconsistent. Further well-designed clinical trials are needed to determine whether nutritional interventions, vitamin supplementation, and strategies targeting metabolic and redox pathways can safely and effectively enhance NK-cell function and improve clinical outcomes in patients with PDAC.

Abstract: Remote sensing change detection technology is widely used in land-use monitoring, urban planning, and disaster assessment. However, during imaging and transmission, bi-temporal remote sensing images are vulnerable to Gaussian noise, which makes it difficult for change detection algorithms to distinguish truly changed areas from noise-affected regions. To address this issue, this study proposes an unsupervised Gaussian-noise-robust change detection algorithm, termed FRIH-SEEDSAM. The proposed method first applies the Fast and Robust Fuzzy C-Means (FRFCM) algorithm to perform noise-resistant fuzzy clustering on bi-temporal remote sensing images. To establish reliable correspondences between the clustering results, the Integrated Region Matching (IRM) algorithm is introduced to construct weighted matching relationships while reducing the influence of abnormal memberships. The change intensity of spatially corresponding pixels is then calculated to generate a more stable change intensity map. Subsequently, the change intensity map is input into the Hybrid Conditional Random Field (HCRF) to infer pixel-level change labels, where the object potential function is constructed from the segmentation results of the Energy-Driven Sampling (SEEDS)-guided Segment Anything Model (SEEDSAM), which uses the centroids of the SEEDS superpixel regions as point prompts for SAM, thereby enhancing change-label consistency within the same changed object region. The experimental results show that the FRIH-SEEDSAM algorithm maintains stable change detection performance across different datasets and under varying Gaussian noise levels. It outperforms the comparison algorithms in terms of several accuracy evaluation indicators, including Kappa and F1. Furthermore, even when the Gaussian noise variance increases to 0.05, Kappa remains at 0.8 or above on multiple dataset images.

Abstract: (1) Background: To analyze the prevalence, temporal trends, and epidemiological pro-file of VDRL seroreactivity among blood donors deemed ineligible due to syphilis in the 1st Social Protection Region of the State of Pará, Brazil, between 2017 and 2023. (2) Study design and methods: This was a cross-sectional, retrospective, census-based epidemiological study conducted using secondary data from the SBS.Web system of the Pará State Hemotherapy and Hematology Center Foundation. All donations with reactive VDRL results in laboratory screening were included. Descriptive and inferential analyses were performed, including temporal trend analysis, prevalence estimates, correlation between municipal rates, and hierarchical clustering analysis, adopting a significance level of 5%. (3) Results: Among 362,705 recorded donations, 3730 (1.02%) showed VDRL seroreactivity. The mean age was 37.6 ± 11.5 years, with a progressive decrease over the study period and a higher concentration among young adults. There was a predominance of males (51%), single individuals (71.5%), those with secondary education (61.7%), and self-reported mixed-race individuals (82.2%). Belém accounted for 67.9% of cases and was the only municipality with a significant increasing trend. The highest mean prevalence rates were observed in Marituba and Barcarena. A strong positive correlation was identified between prevalence rates in Belém, Ananindeua, and Marituba, indicating high epidemiological similarity among these municipalities. (4) Conclusion: VDRL seroreactivity showed a heterogeneous sociodemographic, temporal, and territorial distribution, with greater concentration in urbanized and socially vulnerable areas. These findings reinforce its usefulness as an indirect indicator of population exposure and as a tool to support planning within the blood network and regional epidemiological surveillance.

Abstract: We give a certified finite verification of the literal vertex formulation currently displayed as Erd\H{o}s Problem \#580. Namely, for every $1\le n\le 19$, an $n$-vertex graph having at least $\lceil n/2\rceil$ vertices of degree at least $\lceil n/2\rceil$ contains every tree on at most $\lfloor n/2\rfloor$ vertices. The only order requiring new computer-assisted analysis is $n=18$. An edge-minimal counterexample is reduced to a host partition $V(G)=L\sqcup S$ with $|L|=|S|=9$, degree exactly $9$ on $L$, and $S$ independent. Exact embedding theorems cover $42$ of the $47$ non-isomorphic trees on nine vertices. The five remaining trees reduce, by deleting their leaves, to four rooted cores. For each core we construct a Boolean formula whose models are precisely the reduced hosts avoiding that rooted core. All four formulas are unsatisfiable. The deposited data include complete CNF instances and DRUP refutations. The archived traces were validated by reverse unit propagation, all four formulas are independently solved as unsatisfiable by a second SAT solver, a standalone semantic audit reconstructs the complete formulas---including all $3735$ base clauses and every avoidance clause---without importing the production generator, the complete $47$-tree witness table is generated from the classifier, and regeneration from the published encoder reproduces all four CNFs byte for byte. This is a finite partial result concerning trees on at most $n/2$ vertices; it does not settle the stronger classical formulation asking for trees with at most $n/2$ edges.

Abstract: Medial thighplasty after massive weight loss carries high rates of wound dehiscence, seroma and prolonged edema, and there is evidence that thigh lift can alter lower limb lymphatic drainage. Conventional refinements such as liposuction assistance, limited undermining and fascial anchoring still rely on the surgeon’s implicit understanding of distorted lymphatic anatomy in post bariatric tissues. Direct intraoperative visualization of lymphatic pathways is not part of routine practice. We describe a simple adjunct based on preoperative intradermal methylene blue injection to map superficial lymphatic collectors before medial thigh lift. After standard markings in the standing position, a dilute methylene blue solution is injected intradermally along the planned medial thigh resection pattern once anesthesia is induced and before skin preparation. At flap elevation, lymphatic collectors appear as fine blue channels within the dermis and immediate subdermis, running parallel to the great saphenous axis and defining a practical limit for safe depth of dissection. The surgeon maintains dissection superficial to the stained collectors and avoids transfixing them with suspension sutures or deep liposuction passes. The technique is particularly useful in massive weight loss patients with thin, inelastic skin and deep folds in whom depth perception is unreliable, and for less experienced surgeons who are still developing three dimensional familiarity with the medial thigh. Methylene blue is inexpensive and easily integrated into the operative workflow without specialized equipment or relevant time cost. This lymphatic sparing mapping strategy merits prospective evaluation regarding its effect on seroma, lymphocele and postoperative edema after medial thighplasty.

Abstract: Lung cancer is one of the leading causes of cancer-related mortality. The effectiveness of treatment is limited by several factors, including significant tumoral heterogeneity, rapid development of drug resistance, high metastatic potential, and a complex tumor microenvironment. In recent years, natural products have emerged as valuable resources for the discovery of novel anti-tumor strategies against lung cancer. These compounds possess diverse chemical properties, target multiple pathways, are abundant in nature, and exhibit relatively low toxicity. The utilization of existing medications with established pharmacokinetic profiles and safety records, combined with shorter development timelines, shows promise for advancing lung cancer treatment research. A growing body of evidence indicates that both naturally occurring compounds and commercially available drugs exert effects that extend beyond traditional cytotoxic mechanisms. These agents influence processes such as ferroptosis, oxidative stress, metabolic reprogramming, autophagy, apoptosis, epithelial-mesenchymal transition (EMT), tumor immune microenvironments, and epigenetic networks, suggesting that their activities can be leveraged for a robust multi-target anti-tumor strategy. Accordingly, this review systematically reviews the research advancements regarding natural products and the repurposing of marketed drugs in lung cancer, summarizes their potential for combination with chemotherapy, targeted therapy, radiotherapy, and immunotherapy, and discusses future directions for clinical translation.

Abstract: This article is a review of developed in 2007-2025 years Planck scale informational physical model that is based on 3 main points. First of all on the 2007 “The Information as Absolute” conception, where the fundamental phenomena/notions “Matter”, “Consciousness”, “Space”, “Time”, “Energy”, “Information”, which are fundamentally transcendent in conventional philosophy and sciences, are rigorously scientifically defined. The conception completely rigorously scientifically legitimates the outstanding C. F. von Weizsäcker “Ur hypothesis”, and E. Fredkin “Digital Philosophy/Physics”, which posit that Matter is constructed from some binary reversible logical elements; and on all reliable experimental data. Correspondingly in the model it is postulated that Matter’s ultimate base is the [4+4+1]4D dense lattice of primary [4+4+1]4D binary reversible fundamental logical elements [FLE], which is placed in the corresponding Matter’s fundamentally absolute “Cartesian” [4+4+1]4D spacetime with metrics (cτ,X,Y,Z, g,w,e,s,ct), while everything in Matter is/are some specific disturbances in the lattice. Basing on the above in the model a number of physical problems are either solved or essentially clarified, e.g. of what is real Matter’s spacetime above, what are the physical sense of Lorentz transformations; uncertainty and wave-particle duality in QM; particles and antiparticles, etc. Besides initial Planck scale models of fundamental Gravity, Electric, and Nuclear/Strong Forces are developed, where it is shown that these Forces strengths ratio is in accordance with experimental values only if the FLE size and FLE flip time are Planck length and Planck time; the model rather probably really scientifically clarifies some cosmological problems, including the “matter-antimatter asymmetry: one. Etc. more of the problems see in the article.

Abstract: (1) Background: Social media marketing, the use of social platforms to build brand awareness, engage customers, gather market insights, distribute content, and drive sales has become essential to business strategy across firm sizes. (2) Methods: A systematic review of literature from Google Scholar, Semantic Scholar, and Scispace, published between 2020 and 2026, was conducted using PRISMA-based screening, yielding 14 studies for full-text analysis, a supplementary VOSviewer keyword co-occurrence analysis of study titles and abstracts was also performed. (3) Results: Five core domains were identified, expressed as each domain’s share of the 43 total domain-study codings across the 14 studies: brand awareness and promotion (26%), customer engagement and interaction (30%), market research and insights (9%), content creation and distribution (14%), and sales and lead generation (21%). Customer engagement and brand awareness were the most consistently supported outcomes, while market research remained comparatively under-investigated; the keyword analysis independently corroborated this gap, surfacing no cluster corresponding to market research or content creation. (4) Conclusions: Social media effectively drives brand visibility, engagement, and sales through quality content and consistent interaction, though its use as a market-intelligence tools remains underutilized. Businesses should prioritize content quality, platform-appropriate strategy, and internal social media capability, while future research should address this documented gap across broader contexts.

Abstract: We present a four-dimensional spacetime-geometrodynamical description of electromagnetism as a \(U(1)\) holonomy, formulated in parallel with standard gravitational geometrodynamics. According to the Hamilton--Jacobi optomechanical analogy, interactions admit a dual reading: dynamically, as local changes of four-momentum; geometrically, as local modulations of spacetime recurrences. Spacetime curvature induces redshift and ruler deformation, reproducing gravitational interaction. Similarly, Local Lorentz Transformations associated to a congruence of observer-adapted Fermi--Walker accelerated frames induce local modulations of spacetime recurrences, analogous to the precession of a gyroscope. This yields an effective Abelian connection that reproduces the Lorentz force as a Coriolis-like inertial effect and leads to Maxwell equations for the induced \(U(1)\) curvature. At leading order, the observer-adapted local Lorentz transport generator \(\Omega^{EM}_{\mu\nu}\) determines the electromagnetic field strength through the geometrodynamical identity \(qF_{\mu\nu}\simeq - m \Omega^{EM}_{\mu\nu}\), up to integrable contributions generating \(U(1)\) gauge orbits. These results emerge within a field description implementing "Phase Harmony'', \ie covariant Periodic Boundary Conditions imposed consistently with the variational principle, thereby tracking the relativistic transformations of the natural Compton proper-time recurrence associated with the particle's mass, namely the "particle's internal Compton clock''. The resulting effective \(S^1\) fiber attached to each spacetime point allows a purely four-dimensional reinterpretation of the Kaluza--Klein mechanism and, in turn, a derivation of the Maxwell kinetic term directly from the Ricci tensor. By construction, the same local-isometry structure that governs electromagnetism also governs spin transport, yielding a unified kinematical and dynamical description of Thomas precession and the Bargmann--Michel--Telegdi equation. This suggests a possible geometrodynamical carrier for the anomalous magnetic-moment \(g-2\).

Abstract: Enterprise processes involve many interacting objects whose behavior depends on operational states. Object-centric process mining with OCEL 2.0 captures interactions between objects, and state-aware object-centric process mining adds the state evolution of selected objects. Identifying recurring local behavioral patterns that contribute to entering, maintaining, or recovering from undesired states is essential for process analysis and for designing improvement measures. However, detecting these patterns currently relies on manual inspection of state-aware directly-follows graphs, which is complex and does not scale. This paper presents an automated pattern detection approach for state-aware object-centric process mining. Given a leading object type, the method segments its state evolution, represents each segment as an object-centric graph, and aggregates structurally equivalent segments into ranked patterns. The method distinguishes patterns that occur inside a state from patterns that span state changes. In a real-life case study conducted with Europe’s leading pet retailer, the analysis reveals the behavioral patterns most strongly associated with understock and overstock states, providing a finer-grained diagnostic view of process behavior.

Abstract: Artificial intelligence (AI) has rapidly emerged as a transformative tool in virology, offering new opportunities for the detection, classification, and surveillance of viral pathogens. Recent advances in machine learning, deep neural networks, and multimodal data analysis now enable the identification of viral signatures from genomic sequences, medical images, environmental samples, and social-media-derived epidemiological signals. This review provides a comprehensive overview of state-of-the-art AI methodologies applied to viral pathogen research, with a particular focus on image-based diagnostics, automated quality assessment of virology-related digital content, and predictive modelling for outbreak monitoring. We will discuss how convolutional and transformer-based architectures are being used to classify infected tissues, detect viral particles, and support laboratory workflows. Furthermore, we will highlight the emerging role of AI in evaluating the reliability of user-generated images and short videos related to infectious diseases, an area increasingly relevant in the age of misinformation. Challenges such as dataset bias, limited annotated virological images, ethical concerns, and the need for standardized quality-assessment pipelines are critically examined. Finally, we will outline future research directions, including hybrid AI-biological models, IoT-supported viral surveillance in smart environments, and the integration of explainable AI to enhance clinical trust.

Abstract: Post harvest loss severely threatens smallholder livelihoods and food security across the Himalayan region of South Asian countries. Although improved solar dryers are technically viable, their adoption remains persistently low and unsustainable beyond donor-funded project lifecycles. This paper argues that the persistence of low adoption reflects shortcomings in business model design and institutional support rather than technological inadequacy alone. Employing a theoretically guided systematic literature review, this study integrates inclusive business model (IBM) theory, technology adoption theory (Rogers’ Diffusion of Innovations and UTAUT), and institutional theory (North; DiMaggio and Powell) to develop a three-level conceptual framework for solar dryer adoption in high altitude agroecosystems. The framework positions inclusive business model configurations as the primary driver of sustained solar dryer adoption, moderated by institutional context and mediated by six interdependent adoption barriers, namely, economic, technical, market, social-cultural, institutional, and geographic. Three archetypal IBM configurations are theorized and compared: the micro-enterprise model, the cooperative shared-services model, and the social enterprise service provider model. Nepal and Bhutan serve as contrasting institutional cases, representing market-driven and state-coordinated governance contexts, respectively, to demonstrate how institutional environments shape the effectiveness of each configuration. The framework generates three empirically testable propositions concerning IBM fit, institutional moderation, and gender inclusion. This study contributes to the literature by advancing a context-specific, multi-level framework for technology scaling at the Base of the Pyramid in mountain agroecosystems, and by establishing a research agenda for comparative qualitative investigation in Nepal and Bhutan. This perspective offer actionable guidance for entrepreneurs, development practitioners, and policymakers seeking to foster resilient and inclusive agri-food systems in high-altitude regions.

Abstract: Background/Objectives: Human adenoviruses (HAdVs) are widely used as vectors for vaccines and gene therapy; however, pre-existing immunity can reduce their efficacy. Therefore, rapid and accessible serological methods are required to assess antibody levels against adenoviruses. Lateral flow immunochromatographic assay (LFIA) sensitivity depends on the label and antigen. In this study, we aimed to develop and evaluate LFIA systems based on gold nanoparticles (GNPs) and quantum dots (QDs) using a recombinant HAdV hexon protein. Methods: A recombinant HAdV hexon protein fragment (rhHAdV, 35 kDa; amino acids A120–R316) was expressed in Escherichia coli and purified using Ni²⁺ affinity chromatography. Protein identity was confirmed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis, western blotting, and liquid chromatography–tandem mass spectrometry analyses. Two LFIA formats were developed using Protein G-conjugated GNPs (GNP-G) and QDs (QD-G). Analytical sensitivity was evaluated using serial dilutions of positive serum samples. Diagnostic performance was assessed using 90 human serum samples and compared with that of a commercial enzyme-linked immunosorbent assay (ELISA). Results: The rhHAdV antigen demonstrated high immunoreactivity in ELISA. Antibody detection was achieved at serum dilutions of up to 1:300 and 1:1000 for the GNP- and QD-based LFIA, respectively. Both LFIA formats showed high specificity (98.2%). Sensitivity was 96.9% and 100% for GNP-G and QD-G, respectively; ROC analysis demonstrated excellent diagnostic accuracy, with AUC values of 0.976 and 0.991, respectively. Conclusions: The findings of this study highlight the potential of QD-based LFIA as an advanced tool for rapid serodiagnostics and large-scale immunological monitoring.

Abstract: Radiation fields around collimated or shielded sources exhibit strong gradients whose accurate delineation is critical for worker protection and emergency response. Mobile robots can survey such fields, but they sample sparsely and irregularly along their trajectories, and it remains unclear which reconstruction method can be trusted, and where. Using a single dominant collimated source in a two-dimensional indoor setting, this study shows that the answer depends decisively on sampling geometry, and proposes a physics-guided Gaussian process (GP) that performs reliably under trajectory-constrained sampling. A tracked robot combining light detection and ranging (LiDAR)-based simultaneous localization and mapping (SLAM) with a γ dose-rate detector surveyed a collimated Cs-137 field in seven independent runs, and all methods were evaluated under both random hold-out (interpolation near visited locations) and spatial block cross-validation (extrapolation into unvisited regions), with Poisson-sampled simulations providing truth-referenced comparisons. Under uniform sampling, a multilayer perceptron (MLP) robustly outperformed GP variants (R²=0.95, stable across 18 seed combinations); under trajectory sampling, its advantage vanished at visited locations and reversed catastrophically in unvisited regions. The proposed physics-guided GP, which uses a fitted collimated-beam template as the GP mean with a Matérn 3/2 residual process, achieved the highest extrapolation R² (median 0.61; best baseline 0.31), reduced the extrapolation error by 32–69% relative to all eight baselines, recovered interpretable source parameters, and provided calibrated predictive uncertainty; a runtime fit-quality gate further renders the correctness of the embedded prior an observable quantity, so the method flags when its own assumptions fail. These results offer quantitative guidance for method selection in robotic radiation mapping under the as-low-as-reasonably-achievable (ALARA) principle.