Abstract: Background/Objectives: While COVID-19 is typically more severe in males, there is limited data on sex-specific differences in the predictive value of common inflammatory biomarkers. To address this gap, their predictive capacity was evaluated in a single-center study of male and female patients during the Alpha variant wave. Methods: Univariate, multivariable, and receiver operating characteristic (ROC) analyses were used to evaluate the association of acute-phase proteins, cytokines, and white blood cell counts (measured on admission and day seven) with COVID-19 severity and mortality in severely/critically ill COVID-19 subjects. Results: On admission, the combination of ferritin and D-dimer effectively predicted disease severity in both sexes, though cut-off values and diagnostic accuracy (specificity and sensitivity) varied by sex. In males, neutrophil and lymphocyte counts provided additional clinically relevant predictive value. Seven days post-admission the combination of ferritin, D-dimer and fibrinogen in males and ferritin (as independent predictor comprising model with lactate dehydrogenase) in females emerged as efficient predictors of severe/critical COVID-19. On this evaluation-point, lymphocytes in males and neutrophil-to-lymphocyte ratio in females were also identified as independent predictors of severe/critical COVID-19. Notably, on this evaluation-point C-reactive protein and neutrophil count independently predicted mortality in males with severe/critical disease. Conclusion: Different acute-phase proteins (or the same proteins with distinct cut-off values and predictive characteristics) and white blood cell indices should be considered as independent predictors of severe/critical COVID-19 in males and females and ii) the prognostic capacity of many of them evolves during disease progression, indicating their sex-specific, time-dependent pathogenetic role in COVID-19.

Abstract: In this paper, the different emerging metaverse technologies are identified and a comprehensive understanding of the various technologies that can empower supply chains in various parts of the world is provided. It also presents a structure that shows how each of these classified technologies would work towards a robust and sustainable system of supply chain. Moreover, the study uses the fuzzy TOPSIS method to determine the most significant metaverse technology that can significantly enhance the resilience and sustainability of the supply chain networks across the world. The basic aim of this research is to arm the organizations with the latest technology in the metaverse, which enables them to develop a future-proof supply chain network capable of surviving in this ever-evolving world.

Abstract: In precision mold manufacturing, the machining of HRC52 hardened steel causes se-vere tool wear and high noise in multi source sensor signals, making accurate remain-ing useful life (RUL) prediction challenging. To address this, we propose a hybrid mod-el that integrates one dimensional deep convolution (DCNN), low resolution self attention (LRSA) with 1D 2D spatiotemporal reconstruction, and a multi channel bidirectional long short term memory network (McBiLSTM). A Gaussian smoothing filter is first applied to denoise the 50 kHz signals, followed by physical period sliding windows for feature extraction. A multi strategy fusion pooling layer (mean, max, and last quarter features) further improves prediction accuracy. Using the PHM 2010 milling cutter dataset under leave one out cross validation, the proposed model achieves a mean absolute percentage error (MAPE) of 1.45% and a root mean square error (RMSE) of 2.76 mm, reducing prediction error by up to 75.6% compared to Transformer, LSTM, and GRU baselines. These results demonstrate that the model ef-fectively extracts degradation features even during the accelerated wear stage, offer-ing a reliable solution for tool health monitoring and predictive maintenance under complex cutting conditions.

Abstract: Background: Live BCG vaccination is absolutely contraindicated in Severe Combined Immunodeficiency (SCID) due to the invariably fatal risk of disseminated BCGosis [5,6,7]. This creates a critical unmet need: SCID neonates are deprived of BCG's potent heterologous protection precisely when they are most vulnerable to opportunistic infections. Hypothesis: Heat-Killed BCG (HK-BCG), being entirely non-viable, cannot cause BCGosis or systemic mycobacterial disease. We propose that HK-BCG retains sufficient structural pattern-associated molecular patterns (PAMPs) to engage functional innate immune cells — specifically monocytes, macrophages, and NK cells — that remain present in the majority of SCID subtypes, thereby inducing trained immunity and heterologous protection without adaptive immune cell dependency [2,3,4]. Methods/Evidence: This paper synthesizes molecular mechanism data on TLR2/TLR4/NOD2-driven trained immunity [23,24], epigenetic reprogramming pathways [2,3], SCID immunophenotype data [32,33], and available non-viable mycobacterial clinical trial evidence [34] to construct a mechanistic and clinical rationale for HK-BCG use in SCID.Conclusions: HK-BCG represents a paradigm-shifting, potentially life-saving immunomodulatory platform for SCID patients. The complete absence of viable bacilli eliminates all infectious risk, while preserved mycobacterial PAMPs can train the residual innate immune compartment. This approach is particularly compelling in the post-HSCT reconstitution window. A dedicated Phase I/IIa clinical trial in SCID is urgently warranted.

Abstract: Due to randomness factors in the machine learning model construction process, reproducibility is compromised. This study investigates the impact of randomness on model stability and evaluates techniques for reducing this impact, using the widely adopted shallow neural network (NN) model as a testbed. Randomness in this NN model arises from three events: randomly initializing model parameters, randomly selecting a validation subset, and randomly sampling batches for parameter updates. Among these, batch randomness exerts a significantly weaker impact than the other two factors. In this study, the model training is stopped when the validation performance fails to improve or when a preset threshold for loss or epoch-number is met. The final model stability is significantly better using threshold criteria than using validation criterion, as the former avoids the randomness associated with validation subset. Sensitivity experiments show that scaling the model's initial parameters to 0.1 times their original values can mitigate the impact of initialization randomness, thereby significantly improving model stability while also markedly enhancing predictive skill. Furthermore, weight decay and multi-model ensembles, which are two commonly used techniques, can also significantly enhance model stability. Moreover, the inherent instability of individual sub-models may actually benefit the overall predictive skill of a multi-model ensemble.

Abstract: Unmanned aerial vehicles (UAVs) have wide application prospects in disaster search and rescue, ecological monitoring and environmental inspection tasks, where target search is a key link to realize autonomous task execution. UAVs often face challenges related to limited onboard computational resources and inefficient environmental coverage when used for target search. To address these issues, this paper proposes an autonomous search method for UAVs based on combined lightweight target detection and coverage path planning. In this method, the target search task was decomposed into two core parts: target recognition and path planning. Firstly, in terms of target recognition, the YOLOv8n model was subjected to channel pruning and INT8 quantization to reduce its computational complexity, while HSV space data augmentation was incorporated to enhance recognition robustness in complex environments. Secondly, path planning was formulated as a dual-layer task comprising "spatial coverage + target confirmation." A grid-based search environment model was constructed, and a coverage path planning strategy was put forward that integrated Breadth-First Search (BFS) with local greedy optimization to achieve efficient traversal of unknown areas. Simultaneously, the A* algorithm was employed for path backtracking to cover omitted regions. Finally, a simulation platform for UAV target search was built to validate the recognition performance and search efficiency of the proposed method. The experimental results demonstrated that the proposed method significantly improved the UAV target search efficiency and reduced the path redundancy while ensuring the recognition accuracy, thereby offering an effective solution for autonomous UAV search on resource-constrained embedded platforms.

Abstract: The proliferation of unmanned aerial vehicles (UAVs) in civil, commercial, and defence domains has exposed a critical architectural gap: existing platforms optimise either communication or perception independently, leaving safety coverage incomplete under simultaneous stress in Beyond Visual Line of Sight (BVLOS) operations. This paper proposes the Risk-Aware UAV Safety Architecture (RASA), a three-layer conceptual framework integrating multi-modal sensor fusion, satellite communication (SATCOM), and AI-driven risk modelling aligned with functional safety principles such as ISO 26262. The RASA framework quantifies operational risk as R(t) = α·U_sensor(t) + β·L_c_norm(t) + γ·U_sensor(t)·L_c_norm(t) — a function of normalised sensor uncertainty and normalised communication latency, with an interaction term capturing compound degradation effects — enabling onboard risk estimation without ground-in-the-loop dependency. Building on prior validated work in multi-modal sensor fusion for safety-critical human detection [10] and SATCOM communication architectures for UAV connectivity [15], this paper extends those contributions to the BVLOS domain. Monte Carlo simulations across three representative operational scenarios validate the risk model’s behaviour and demonstrate that the interaction term produces steeper risk escalation under compound failure conditions compared to the linear baseline. This paper addresses the critical gap in BVLOS UAV safety architectures by integrating perception and communication reliability within a single, auditable, risk-aware framework.

Abstract: Background: The dendritic cell initiates and directs antigen-specific immunity. Three Nobel Prizes frame the system it controls: Steinman (2011) recognized the dendritic cell as the conductor of adaptive immunity; Allison and Honjo (2018) recognized CTLA-4 and PD-1 as checkpoints restraining effector responses; Sakaguchi, Brunkow, and Ramsdell (2025) recognized that Foxp3-positive regulatory T cells maintain peripheral tolerance. Hypothesis: This article proposes that these discoveries describe a single bidirectional circuit with the dendritic cell as its fulcrum, and that the tolerogenic default observed in aging, cancer, chronic infection, and senescence represents a correctable failure of dendritic cell instruction driven by a specific molecular chain: SASP–STAT3–DNMT–IRF8 silencing. The proposed mechanism is self-reinforcing: senescence-associated secretory phenotype cytokines activate STAT3 in hematopoietic progenitors, STAT3 recruits DNMT1 and DNMT3B to methylate the IRF8 promoter, and IRF8 silencing—locked by a BATF3-dependent bistable switch with no stable intermediate—simultaneously eliminates IL-12 transcription, disarms target cells against apoptosis, collapses genome surveillance, and installs bilateral disarmament across the immune synapse. STAT3 simultaneously drives PD-L1 transcription on tolerogenic dendritic cells and tumor cells, and a fifth biosynthetic lock operates at the metabolic level: mitochondrial depolarization collapses citrate export and secretory pathway capacity, eliminating IL-12 production independently of transcriptional silencing. Four metabolic-epigenetic feedback loops—alpha-ketoglutarate depletion blocking TET demethylases, NAD-plus decline disabling SIRT1 deacetylation of STAT3, SOCS3 promoter methylation removing the STAT3 negative feedback brake, and lactate-derived lysine lactylation driving immunosuppressive gene programs—ensure the tolerogenic default is metabolically as well as epigenetically locked. Proposed correction: The alpha-type-1 polarized dendritic cell, manufactured ex vivo with IFN-gamma and multi-TLR engagement, escapes this architecture because its maturation commitment is made outside the STAT3 field. It initiates a self-amplifying four-phase cascade that progressively restores IRF8 expression across the immune surveillance network and removes the upstream cause through senescent cell clearance. Significance: If correct, this framework implies that many current therapies fail not because targets are wrong, but because they operate downstream of a corrupted instructional system. It redefines the therapeutic target from the effector compartment to the instructor.

Abstract: The long‑term evolution of the Earth–Moon system is traditionally attributed to tidal friction, which transfers angular momentum from Earth’s rotation to the Moon’s orbit. Present‑day measurements show that Earth’s rotational angular‑momentum loss closely matches the Moon’s orbital gain, consistent with this framework. However, deep‑time constraints from fossil growth increments and tidal rhythmites reveal a persistent and significant mismatch between these two quantities over the past 3.2 billion years. At 900 million years ago, Earth’s rotational angular‑momentum loss exceeded the Moon’s orbital gain by ~40 %, and at 3.2 billion years ago, by nearly a factor of three. These discrepancies cannot be reconciled by classical tidal friction, even when accounting for solar tides, ocean‑basin evolution, atmospheric tides, or core–mantle coupling. The Earth exhibited significantly greater flattening in the past than it does today and is projected to approach a near-spherical configuration in approximately 3 billion years. Using empirically fitted histories of the length of day (LOD), number of days per year (DOY), and Earth–Moon distance (DOM), I show that the angular‑momentum imbalance is robust and increases exponentially backward in time. The Dark Matter Field Fluid (DMFF) model provides a natural explanation: Earth loses rotational angular momentum to a pervasive dark‑matter‑like medium, while the Moon’s orbital evolution is driven by DMFF drag and anti‑gravitational effects. The DMFF‑derived equations for LOD, DOY, and DOM match both modern astronomical measurements and deep‑time geological records, including the critical LOD and DOM constraints at 3.2 billion years ago. The angular‑momentum discrepancy is therefore not a flaw in the data but a signature of DMFF physics, revealing a deeper dynamical structure of the Earth–Moon system.

Abstract: Background/Objectives: Celiac disease (CD) is a T-cell mediated autoimmune condition, triggered by gluten ingestion. Duodenal biopsy is the gold-standard diagnosis for CD, which is often limited by interobserver variability between pathologists. Immunohistochemistry (IHC) is a powerful technique for detecting biomarkers with potential diagnostic significance. This study aims to investigate five candidate biomarkers BTNL8, NKp46, TdT, THEMIS, and TCRδ that might improve the reproducibility of the diagnosis of CD. Methods: Formalin fixed paraffin-embedded material, surplus to diagnostic requirements was obtained from 46 subjects (untreated CD: n=21, CD treated with gluten-free diet: n=5; controls: n=20) and immunostained for BTNL8, NKp46, TdT, THEMIS and TCRδ. BTNL8 staining was scored on a 0-3 semi-quantitative scale. NKp46, TdT, THEMIS, and TCR delta-positive intra-epithelial lymphocytes (IELs) were quantified as mean counts per 100 epithelial cells (ECs). Results: TCRδ-positive IELs were markedly elevated in CD biopsies (median 9.4 IELs/100 ECs) compared to healthy controls (median 0.5 IELs/100 ECs; p<0.001), with a threshold of >2.1 TCRδ-positive IELs per 100 ECs yielding an AUC of 94% and interobserver agreement of 0.82. NKp46 expression was also increased in CD (median 13.8 IELs/100 ECs) versus controls (median 9.6; p<0.001), with >12.8 NKp46-positive IELs per 100 ECs achieving an AUC of 86% and interobserver agreement of 0.82. Immunostaining for the other biomarkers demonstrated less clear differences between CD and healthy controls. Conclusions: Corroborating several recent publications, TCRδ immunostaining provides high diagnostic accuracy and good interobserver agreement in the diagnosis of CD on duodenal biopsy, even for patients on a gluten-free diet.

Abstract: Miniaturization of microfluidic measurement systems offers several advantages, including reduced sample and reagent volumes, improved control over experimental conditions, and the ability to multiplex complementary measurement modalities, to enable new experimental approaches in microbial electrochemistry. We present a scalable glass-based microfluidic bioelectrochemical cell (µ-BEC) platform for multiplexed investigations of microbial extracellular electron uptake (EEU). The platform integrates eight independently addressable three-electrode cells in a 2×4 array, with transparent indium tin oxide working electrodes that support simultaneous electrochemical analysis and optical imaging. Systematic electrochemical characterization using the ferri/ferrocyanide redox couple demonstrated diffusion-controlled behavior and stable reference electrode performance, with well-to-well coefficients of variation in peak potentials of 0.6–1.5% and 0.6–1.1% for anodic and cathodic processes and device-to-device coefficients of variation of approximately 1.8% and 1.6%, respectively. Differential pulse voltammetry measurements demonstrated concentration-dependent electrochemical sensing over a three-order-of-magnitude range from 1 µM to 1 mM ferri/ferrocyanide, with peak currents exhibiting linear dependence on concentration for both anodic and cathodic processes across all tested wells. Biological compatibility was validated using the phototrophic bacteria Rhodopseudomonas palustris TIE-1, where reproducible light-dependent EEU was observed following 96 hours of incubation, and a reduction in current response after microfluidic removal of planktonic cells confirmed the contribution of surface attached cells to EEU. Together, these results establish the µ-BEC as a robust and reproducible microfluidic electrochemical platform suitable for parallelized, multimodal studies of microbial and abiotic electrochemical processes.

Abstract:

Food and agricultural consumption are responsible for nearly a quarter of the world's greenhouse gas (GHG) emissions, so eating is a critical element in slowing climate change. Accurate estimation of meals carbon price is necessary to promote sustainable food intakes, but current methods rely heavily on self-administered dietary questionnaires, nutrition databases, or manual input, which are time-consuming, subject to errors, and difficult to scale up. In response to these challenges, we provide a novel machine learning model that predicts the carbon footprint of meals from images of food. Our approach marries deep-learning-based food identification and carbon intensity data from established life cycle assessment (LCA) studies. With ubiquitous food image datasets such as Food-101 and UECFood256, we train convolutional neural networks (CNNs) and newer models such as Efficient Net to classify meal ingredients. Each of the recognised food items is then cross-mapped to a carbon footprint database, where emission factors (in g CO₂-eq/100 g) are summed up to create a composite meal-level estimate. In addition to prediction, our system suggests alternative, lower-emission foods, providing actionable evidence for environmentally conscious dietary changes. Experimental results demonstrate high accuracy of food classification and footprint estimation, with prediction errors in an acceptable range compared to ground-truth values for emissions. Survey bias is eliminated by the suggested system, real-time estimation is achieved, and the system can be incorporated as part of mobile or web-based diet-tracking tools without any difficulty. This research is one of the first to combine computer vision and sustainability strategies, and it offers a scalable and automated platform to guide people and organizations toward sustainable food consumption patterns.

Abstract: Based on the complexification of the modular flow parameter in the Tomita-Takesaki theorem and the thermal time hypothesis, we propose a complex-time picture: as a system approaches absolute zero, real time freezes while imaginary time emerges. Mathematically, this is equivalent to a Wick rotation. In this picture, applying the heat diffusion equation at absolute zero forces this rotation, transforming the diffusion equation into the Schrödinger equation and ensuring entropy invariance as required by the third law of thermodynamics. This complex-time picture thus offers a unified, temperature-based origin for two fundamental facts: why microscopic particles obey the Schrödinger equation, and why an arrow of time emerges in macroscopic systems.

Abstract: Fiber-reinforced polymer composites are increasingly used in lightweight aerospace structures due to their high strength-to-weight ratio, excellent corrosion resistance, and superior mechanical performance compared with conventional metallic materials. Among these materials, glass fiber-reinforced polymer (GFRP) and carbon fiber-reinforced polymer (CFRP) composites have gained widespread attention for use in unmanned aerial vehicle (UAV) structures, where structural efficiency, durability, and cost-effectiveness are critical design considerations. Understanding the compressive behaviour and failure mechanisms of composite laminates is therefore essential for ensuring structural reliability and safe operation in aerospace applications. This study presents an experimental investigation of the compressive behaviour of woven E-glass fiber-reinforced epoxy and carbon fiber-reinforced epoxy composite laminates. Rectangular specimens were prepared from commercially manufactured composite laminate plates with approximate dimensions of 100 mm × 95 mm and a laminate thickness of approximately 1.5 mm. Compression tests were performed using a universal testing machine under displacement-controlled loading conditions until structural failure occurred. The results revealed significant differences in the mechanical response of the two composite systems. Carbon fiber-reinforced laminates exhibited considerably higher stiffness and compressive load capacity due to the higher modulus of carbon fibers. However, carbon fiber specimens exhibited brittle failure, characterized by sudden fiber fracture and a rapid loss of load-carrying capacity. In contrast, E-glass laminates exhibited lower stiffness but showed more progressive damage, including matrix cracking and fiber buckling, prior to final failure. These findings highlight the trade-off between stiffness and damage tolerance in fiber-reinforced composites and provide useful experimental insight into the compressive performance of commonly used aerospace composite materials. The results contribute to the development and optimization of lightweight composite structures for UAV structural applications.

Abstract: Elastic scaling in cloud native environments is essential for maintaining service quality and resource efficiency. In practice, frequent traffic bursts and shifts in workload distributions make rule-based methods or approaches with a single optimization objective insufficient. They struggle to ensure system stability and decision reliability at the same time. To address this challenge, this study formulates elastic scaling as a risk-constrained reinforcement learning problem from a sequential decision perspective. A unified framework is used to model resource adjustment actions, system state evolution, and potential instability costs. By explicitly incorporating risk constraints into policy optimization, the proposed approach achieves a dynamic balance between performance optimization and safety control. It prevents service level objective violations and system oscillations caused by aggressive decisions. Resource utilization efficiency and service response behavior are jointly considered, which improves consistency and controllability in complex cloud environments. Comparative evaluation based on real cloud cluster traces shows advantages in service reliability, response performance, and resource usage over existing baselines. These results confirm the effectiveness of risk-aware decision-making under non-stationary workloads. This work provides a systematic modeling approach for the safe application of reinforcement learning in cloud resource management and lays a methodological foundation for stable and efficient intelligent elastic scaling.

Abstract:

Dental implant therapy demonstrates high long-term survival; however, biological, behavioral, and technical complications remain prevalent. The objective of this study was to introduce GF-Predictability for Dental Implants (GF-PreDImp), the first multidomain predictive tool in the literature, designed to quantify implant success predictability through a structured, evidence-based scoring system. The model integrates six domains: Biological, Behavioral, Hard tissue, Soft tissue, Implant, and Prosthetic, approaching systemic, behavioral, anatomical, surgical, and prosthetic variables into a 100-point composite index. The Biological/Systemic point (20 points) involves diabetes (HbA1c), bisphosphonates, head and neck radiation, cardiovascular disease, osteoporosis, and immunosuppression; the Behavioral/External topic (20 points) approaches post-implant smoking, oral hygiene, plaque/calculus index, brushing performance, alcohol usage, and patient’s compliance; the Hard Tissue (20 points) analyzed bone quality (densities: D1–D4), bone quantity, arch position, guided-bone regeneration (GBR) need, sinus lift, cone beam computed tomography (CBCT) height/width; the Soft Tissue evolution (15 points) observes keratinized mucosa width (KMW), periodontal history, gingival phenotype, bleeding on probing (BoP), and probing depth (PD); the Implant Parameters topic (15 points) assessed tooth position, loading timing, primary stability (ISQ), length/diameter, and surface treatment; and the last point analyzed, Prosthetic/Surgical (10 points), appraisal bruxism characteristic, occlusal contacts, crown-to-implant ratio, cantilever, surgeon experience, and antibiotic protocol. The final GF-PreDImp score could be excellent (≥ 85), good (70 – 84), moderate to guarded (55-69), guarded to high risk (40-54), and poor (<40). Results: Predictors were derived from literature on implant failure, peri-implant disease, and risk assessment. The tool generates dynamic visual outputs, including radar charts and domain-specific scores, enabling real-time clinical interpretation. Each domain can achieve up to 100%, and the average results predict the predictability of dental implant therapy. The final screen of the GF-PreDImp outcome presents a summary of the worst areas to clarify possible risks for clinicians and patients. The graphic and result can be printed for electronic filing and/or shown and given to the patient. The GF-PreDImp system can provide a comprehensive framework for individualized risk stratification and treatment optimization. Its implementation can improve clinical decision-making and enhance long-term implant outcomes. Further clinical assessments must be done to confirm the findings in future studies.

Abstract: Education for Sustainable Development (ESD) has emerged as a key priority in contemporary education systems, emphasizing the need to equip learners with the knowledge and competencies required to address complex environmental and societal challenges. Mathematics education can play an important role in achieving these goals by enabling students to analyse data, interpret real-world problems, and develop critical thinking skills related to sustainability issues. This study investigates the impact of sustainability-oriented mathematical modelling activities on pre-service primary teachers’ perceptions of integrating sustainability into mathematics education. The study employed a quasi-experimental design involving 68 pre-service primary teachers enrolled in a mathematics education course at a university. Participants engaged in a six-week intervention consisting of modelling activities based on real-world sustainability contexts, including water consumption, energy use, waste management, and sustainable transportation. Data were collected using a pre- and post-intervention questionnaire examining participants’ perceptions of sustainability integration, mathematical modelling, and teaching confidence. Statistical analyses, including reliability analysis, descriptive statistics, paired-sample t-tests, and correlation analysis, were conducted to examine the impact of the intervention. The results indicate significant improvements in participants’ perceptions of sustainability-oriented mathematics teaching and their confidence in designing modelling-based sustainability activities. The findings suggest that mathematical modelling can serve as an effective pedagogical approach for integrating sustainability topics into mathematics education and preparing future teachers to connect mathematical reasoning with real-world environmental challenges. The study contributes to the growing body of research at the intersection of mathematics education, teacher education, and sustainability education by providing empirical evidence on the potential of modelling-based learning for supporting sustainability-oriented teaching practices.

Abstract: Asking questions is fundamental, but without a systematic framework, it remains a matter of intuition rather than design. The Generalized Coordinate System (GCS) was initially proposed for analyzing and generating rhetorical modes. In this paper, we apply the GCS to form an inquiry design framework—the GCS-based 10-dimensional inquiry generation framework: treating a question as a coordinate point across ten axes, so that we have potentially a billion ways to ask questions. The five low-dimensional axes (Thing, Feature, Quantitative Attribute, Qualitative Attribute, Formal Attribute) determine what and how the question expresses; the two mediating axes (Basic Element, Rhetorical Mode) transform a raw inquiry into a communicable question package; the three high-dimensional axes (Cognitive Function, Epistemic Purpose, Expression Staircase) determine what mental operation, why, and at what developmental level. This GCS-based 10-dimensional inquiry generation transforms questioning from an intuitive art into a designable, transferable, and evaluable cognitive methodology, and is potentially useful in applications such as education, research, communication, and language modeling.

Abstract: The proliferation of behavioral data in digital retail has not been matched by equally rigorous frameworks for converting that data into customer intelligence that practitioners can act on. This paper addresses that gap by introducing RFM-B, a behavioral segmentation framework that extends the classical recency–frequency–monetary (RFM) model with four additional indicators derivable from standard e-commerce event logs: conversion rate (CVR), category breadth, average order value (AOV), and brand diversity. Applied to 4,635,837 interaction events from 64,204 purchasing customers observed over 61 days on a large-scale multi-category platform, the framework produces five customer archetypes—Champions, Loyal Customers, Potential Loyalists, At-Risk, and Lost—whose behavioral profiles differ systematically in purchase efficiency, platform embeddedness, and commercial significance. A machine-learning recoverability analysis using a Random Forest classifier achieves 96.99% held-out accuracy (5-fold cross-validated: 96.84% ± 0.10%), confirming that the segments are operationally deployable in real-time marketing automation systems. The central empirical finding is the identification of a Potential Loyalist segment characterized by an average order value of USD 147.2—more than three times the platform-wide mean—combined with low purchase frequency, a profile that standard RFM frameworks would systematically misclassify as low-priority. The results show that enriching the behavioral feature space yields a customer typology that is both analytically coherent and directly actionable, and that interpretability is not a secondary concern but a functional prerequisite for organizational adoption.

Abstract: Regenerative medicine has emerged as a transformative paradigm in contemporary healthcare, shifting the therapeutic focus from symptomatic management toward the restoration of tissue structure and function through biologically active interventions. Within this framework, adipose-derived products have attracted substantial interest owing to their relative abundance, ease of harvesting, and rich cellular and paracrine composition, including mesenchymal stromal cells, pericytes, and bioactive mediators with immunomodulatory potential. Among these technologies, Lipogems® represents an innovative approach based on minimally manipulated microfragmented adipose tissue, because it preserves the native stromal vascular niche and extracellular matrix architecture while avoiding enzymatic processing. This characteristic not only maintains biological integrity but also facilitates regulatory compliance in multiple jurisdictions. This narrative review provides a comprehensive synthesis of the current evidence on Lipogems®, integrating biological rationale, mechanistic insights, and clinical applications across musculoskeletal disorders and chronic pain conditions. Particular attention is devoted to its capacity to modulate inflammatory pathways, promote angiogenesis, and support tissue regeneration within complex pathological environments. In addition, the review critically appraises the methodological limitations of existing clinical studies, including heterogeneity of design and limited high-quality randomized evidence. Finally, future perspectives are explored, emphasizing the integration of precision medicine approaches, biomarker-driven patient stratification, and combinatorial regenerative strategies aimed at optimizing therapeutic outcomes.