Abstract: Current approaches to cancer therapy often relate therapeutic input directly to outcomes, leaving the intermediate dynamics largely unrepresented. Recent work formulated cancer therapy as a continuous process in which an initial state-space organization and therapeutic input give rise to therapeutic trajectory evolution, attractor formation, and an observable outcome. The present work defines mesoscopic dynamics between therapeutic input and outcome. It then applies the probability chain rule to factorize the joint conditional distribution of possible therapeutic trajectories, attractors, and outcomes. The resulting hierarchical probabilistic formulation provides a probabilistic mathematical representation of mesoscopic therapeutic dynamics by making explicit the conditional organization of possible trajectories, attractors, and outcomes. Together, mesoscopic dynamics and its hierarchical probabilistic formulation provide a conceptual framework for connecting therapeutic input to clinical outcome.

Abstract: Background/Objectives: Older Adults experiencing low-trauma hip fractures (LTHF) may not regain pre-fracture functioning. This study examined whether different LTHF types (femoral neck (FN-F); trochanteric (Tr-F); acetabulum/periprosthetic (Ac/Pp-F) fractures) follow different healthcare pathways and achieve different functional outcomes. Methods: Retrospective cohort study 2020–2023 of (Bern Rehab Registry, 2020-2023; Swiss Ethics Project-ID 2024-00480) included patients aged ≥65 years with low-trauma FN-F, Tr-F, or Ac/Pp-F undergoing inpatient rehabilitation. Pathways comprised chronological stays in acute care, inpatient rehabilitation, and temporary nursing homes, bounded by pre-fracture situation and discharge destination. Pairwise standardized mean differences (SMDs) were calculated for between-group comparisons. Results: Of 294 individuals (mean age 81.0 ± 7.4 years; 191 (65%) women; 152 Tr-F, 118 FN-F, and 24 Ac/Pp-F) four main pathways were identified: (1) Acute care to inpa-tient rehabilitation in 208 (70.7%) individuals, 2) temporary nursing home stays before rehabilitation in 25 (8.5%), 3) temporary nursing home stays after rehabilitation in 17 (5.8%), and 4) re-hospitalization in 44 (15.0%). Median pathway length was 33.0 days (IQR 26.0–40.0) for FN-F, 35.0 (IQR 29.0–46.0) for Tr-F, and 60.5 (IQR 37.0–86.5) for Ac/Pp-F. Between-group differences in pathway type were moderate to large (SMDs: FN-F vs. TR-F = 0.47; FN-F vs. Ac/Pp-F = 1.25; TR-F vs. Ac/Pp-F = 0.84), and for length small to large (SMDs: FN-F vs. TR-F = 0.36; FN-F vs. Ac/Pp-F = 1.2; TR-F vs. Ac/Pp-F = 0.69). Overall, 246 (83.7%) returned to a private living setting. 88 (29.9 %) required new or increased home support or care. At discharge from rehabilitation, Functional Independence Measure (FIM) showed small to moderate and Timed Up and Go test (TUG) showed small between-group differences. Conclusions: Low-trauma hip fractures are not a uniform clinical entity. When pathways respect fracture-specific needs, functional outcomes and return to independent living can be comparable. Future rehabilitation should be tailored to these pathways and extended beyond inpatient rehabilitation to optimize care.

Abstract: Glycosylation is a fundamental molecular mechanism that integrates immune regulation, oncogenesis, inflammation, and pathogen interaction. Despite significant progress in understanding glycan biology, the clinical implementation of glycan-targeted strategies faces significant challenges due to the multifunctionality of glycans, their context-dependent nature, and methodological limitations. Overcoming these challenges requires the development of integrative platforms that combine glycomics with other omics disciplines, improved analysis methods, and the creation of more selective therapeutics. Glycobiology is one of the most promising areas of modern biomedicine, and further development of this discipline has the potential to revolutionize the diagnosis, prevention, and treatment of many human diseases.

Abstract: Hydrazine (N₂H₄) and its derivatives are components of rocket fuel that contaminate soils around aerospace facilities and military training grounds. The toxicity of hydrazine hydrate, 2,4-dinitrophenylhydrazine (DNPH) and 2,4-dinitroaniline to terrestrial invertebrates remains unexplored. This study examined the effects of these substances on the mortality, body mass, pupation, and intensity of Gregarina spp. infestation in larvae of Tenebrio molitor Linnaeus, 1758 (Coleoptera, Tenebrionidae) at temperatures of 21–23 and 26–28 °C. Six hundred larvae were distributed into 10 treatment groups at each of two temperatures, with three containers per group, over a 5-day exposure. A dose-dependent effect on the physiological parameters of T. molitor was recorded: mortality rose with dose for all three compounds and at 90 mg/kg reached 26.7% (DNPH, 26–28 °C). Pupation proved the earliest indicator of exposure: none was recorded in the control (0.0%), whereas 59.6% of the exposed larvae pupated, already at the lowest concentration of 10 mg/kg, where no deaths occurred. The proportion pupating was similar across the three compounds (55.0–62.2%) but declined with increasing concentration, from 67.2% at 10 mg/kg to 52.2% at 90 mg/kg. Temperature had no effect on it (57.8% at 21–23 °C versus 61.5% at 26–28 °C). All pupae survived (100.0%), compared with 85.3% of those that remained larvae. Initial body mass did not differ between these groups, so no size threshold for initiating metamorphosis was detected. Within the 21–28 °C range, no main effect of temperature on mortality, pupation, or body mass was found, although a statistically significant temperature × concentration interaction was recorded for the last of these. All three compounds reduced the intensity of Gregarina spp. infestation with increasing concentration, and raising the temperature to 26–28 °C lowered the mean parasite count from 18.0 to 13.0 specimens/larva. Under acute 5-day exposure, therefore, hydrazine compounds cause little larval mortality but alter pupation, body mass and the intensity of parasitic infestation at concentrations that kill no larvae. This makes the T. molitor–Gregarina spp. system suitable for the early detection of substrate contamination by hydrazine compounds.

Abstract: Background. Most pharmacy regulators now mandate continuing professional development, replacing or adding to traditional ongoing education. However, the broad evidence supporting this change has not been well characterized. Objective. This mapping aims to explore the scope, design, geography, and impact of research on continuing education and professional development for practicing pharmacists and pharmacy technicians. It also seeks to identify which links in the implicit chain, from requirement to patient outcome, have been examined. We distinguish three often-overlapping dimensions: the mechanism (what practitioners are asked to do, such as accrual or documented cycles), the requirement (whether participation is mandatory), and the warrant (the record's claimed proof). Five research questions are posed to address the determinants of map, mechanism, mandate, warrant, and engagement. Methods. An evidence map of PubMed was created using eight structured queries that focused on fundamentals, effectiveness, reviews, revalidation, barriers, settings in low- and middle-income countries, digital delivery, and the technician workforce. These queries retrieved a total of 1,145 records over two rounds. Of these, 743 were unique entries, and 724 were eligible for screening. The screening process involved reviewing titles and abstracts, and each selected study was categorized by topic, design, country, sample size, and the highest Kirkpatrick level of evidence for outcomes. Both screening and data extraction were completed in a single step using a language model for classification, while design and outcome coding were verified by a single author. When multiple publications reported the same study, they were linked, ensuring that counts reflect individual studies rather than reports. Results. Following an extensive search and a detailed audit, 359 studies met the inclusion criteria, of which 201 (56.0%) were published after 2020. The majority were surveys conducted at a single time point (129, 35.9%) or evaluations of a single group before and after an intervention (63, 17.5%). Only 12 studies (4.6%) included a comparison group, of which six were randomized trials. Outcome measurements mainly focused on knowledge, skills, and attitudes (161, 44.8%). Fifty-two studies (14.5%) examined behavior change, and just 3 (0.8%, 95% CI 0.3 to 2.4) assessed patient or system outcomes. All three were uncontrolled, and no controlled study measured outcomes beyond behavior. Only 4 of the 12 comparative studies considered the cycle itself; the remaining 8 compared different continuing education formats. No study evaluated the impact of introducing, removing, or varying a requirement against any outcome. One study independently measured engagement and competence over a single register and found that every pharmacist who failed the external competence review had previously passed the engagement review. Conclusions. Among the four links in the field’s implicit chain from requirement to patient outcome, one has not been tested in pharmacists at all, while another has been examined by four studies. The warrant claim based on this chain is challenged where it has been tested. The strongest positive claim, supported by evidence, is that the cycle influences pharmacists' actions more than accrual-based education does. This conclusion relies on self-reported behavior from a single, unblinded trial involving 100 pharmacists, reported across three papers, with effects diminishing over three years. No pharmacy trial has directly compared these approaches against patient outcomes, and the closest relevant research in medicine has been conducted twice without showing a signal. Therefore, the most valuable questions in the field are narrower than replication. They are whether the cycle offers additional benefits beyond any educational intervention, whether that effect can be sustained, and whether employer-led levers, such as protected time, are effective.

Abstract: Large language models, multimodal foundation models, and agent systems are increasingly being integrated into geographic research by linking natural-language interaction with remote sensing, geospatial data, and specialized analytical tools. This review combines bibliometric analysis with qualitative synthesis to examine this convergence through a Perception--Reasoning--Action--Decision framework. A Web of Science search covering 2022--2026 year-to-date yielded a bibliometric corpus of 1,147 records. From this corpus, 151 representative studies were purposively selected for detailed narrative synthesis across four analytical stages: multimodal perception, geospatial reasoning, agentic action, and operational decision support. The review compares developments in cross-modal alignment, geographic cognition, spatial code generation, tool use, multi-agent collaboration, and applications in urban governance, transportation, disaster response, environmental monitoring, agriculture, energy, and satellite scheduling. Across these areas, recurring limitations concern precise spatial grounding, cross-sensor and cross-region generalization, hallucination, workflow verification, computational efficiency, real-time deployment, and responsible decision-making. The literature therefore points toward three priorities for future geospatial intelligence: explicit spatiotemporal grounding, verifiable tool-augmented workflows, and reliable integration of multimodal observations with professional geographic models and human expertise.

Abstract: Water allocation in capacity-constrained canal trees with conveyance losses requires local limits and guaranteed service to be considered simultaneously. We develop a deterministic three-stage lexicographic model for a rooted tree with fixed routes. Stage 1 maximizes the minimum service ratio, Stage 2 maximizes weighted seasonal satisfaction subject to that guarantee, and Stage 3 minimizes temporal service-ratio variation while preserving the preceding optima. The nonnegative packing structure yields the Stage-1 optimum in closed form from capacity-to-full-load ratios; the loss-aware path operator is equivalent to node balance and determines unique gross flows. In five exact benchmarks, the maximum discrepancy between the closed-form and linear-programming solutions was 1.11×10⁻¹⁶, with zero operator–balance residual. Scaling tests with up to 500 users and 1022 edges yielded a maximum discrepancy of 2.22×10⁻¹⁶. In the three-period test, the temporal variation ranged from 0.40 to 1.05 over the Stage-2 optimal face, and Stage 3 attained 0.40. The price of fairness was 3.29% in the controlled Gone Abat Jap scenario. The results provide a model-specific analytical characterization and a reproducible computational framework for loss-aware allocation on fixed-route irrigation trees.

Abstract: The technological foundations for creating new composite materials based on solid emulsions of UV photopolymers are considered. Samples of photopolymer composites with artificial porosity, representing solid emulsions based on photopolymer resin and various liquids and colloidal solutions, have been obtained. The work develops methods for synthesis and formation of artificial porous structures, methods for controlling pore morphology, and approaches to filling pores with active liquid phases. Mechanisms for controlling the electrodynamic parameters of such materials are examined, as well as methods for their diagnostics in a wide frequency range. The electrophysical characteristics of the composites in the low-frequency and microwave ranges are investigated. It is shown that the features of the frequency dispersion of the dielectric parameters of the samples are influenced by the presence of a liquid phase in the pores of the composite. The obtained type of composite materials is promising for the development of next-generation acoustic and electrodynamic media with controllable parameters (acoustic impedance, dielectric susceptibility, polarizability, magnetization, attenuation for different types of acoustic and electromagnetic waves). The developed technologies may find practical application in the creation of new functional materials that can be effectively used in the fabrication of metamaterials with unique properties, tunable over a wide frequency range for radio engineering modules, as well as in adaptive electrodynamic and acoustic systems.

Abstract: Ferromagnetic-ferroelectric composites are of interest for studies of magneto-electric (ME) coupling between the two phases, which is mediated by mechanical strain arising from magnetostriction and the piezoelectric effect. A strong ME response in such composites typically requires an external DC bias magnetic field to compensate for weak zero-bias piezomagnetic deformation in the ferromagnetic phase. This reliance on an external field remains a key barrier to device miniaturization. Here we report strong, self-biased ME coupling achieved in a bilayer structure consisting of lead zirconate titanate (PZT) and a dual-phase ferromagnetic composite. The composite pairs hard barium hexaferrite BaFe12O19 (BaM), with its high magneto-crystalline anisotropy field, with soft nickel-zinc spinel ferrite Ni0.8 Zn0.2 O4 (NZFO), with its high magnetostriction. Samples consisting of (100-x) wt.% BaM- x wt.% NZFO (BNZx) were prepared via conventional solid-state sintering, and x-ray diffraction confirmed phase purity. Magnetic characterization revealed an in-plane anisotropy field (HA) ranging from 4.8 kOe to 8.3 kOe, which decreased systematically with increasing spinel content, alongside a high saturation magnetostriction of -25 ppm. The ME coupling strength of the BNZx-PZT bilayers was evaluated via the ME voltage coefficient (MEVC) across both low-frequency and electromechanical resonance regimes. The MEVC increased markedly with higher x-values, peaking at x=90. Under zero external DC bias, the bilayer showed a low-frequency (100 Hz) MEVC of 114 mV/cm Oe and a remarkably high MEVC of 3.6 V/cm Oe at resonance (10.6 kHz). The corresponding MEVC values under an external bias were 161 mV/cm Oe and 6 V/cm Oe. This pronounced self-biased behavior, driven by the internal field of the hard BaM phase, eliminates the need for external biasing elements and provides a practical framework for compact magnetic sensors, imaging arrays, and energy harvesters.

Abstract: Active learning is commonly formulated as a problem of selecting the next observation that is maximally informative for the current predictive model. Uncertainty sampling, information gain, and representative subset selection all instantiate variants of this idea. This paper investigates a different hypothesis- a useful query may be valuable not because it most reduces present uncertainty, but because its incorporation changes the organization of what the learner can subsequently construct, infer, or learn. We call this quantity generative reach. We report a sequence of progressively harder computational experiments designed both to test and to falsify this proposal. We test it on a hard construction in which local structural importance, immediate utility, and long-horizon generativity are deliberately separated. Deep modules have low immediate branching but support delayed compositional reuse; decoy modules have high local branching but terminate. Generative reach yields a mean final reach of 42.71 versus 33.45 for uncertainty sampling, 38.35 for immediate reach, and 32.01 for a local structural heuristic. Crucially, these gains occur with nearly the same number of primitive products discovered, indicating that the advantage lies in the organization and reuse potential of acquired knowledge rather than in raw acquisition volume. Adding uncertainty as an independent objective does not reliably improve over pure expected future reach. Active learning under generative, compositional, and non-stationary conditions should optimize expected future learnability, not merely present uncertainty.

Abstract: 3D scene reconstruction and novel view synthesis have undergone a paradigm shift with the advent of 3D Gaussian Splatting (3DGS). Unlike computationally intensive volumetric rendering, 3DGS leverages a point-based representation with a differentiable tile-based rasterizer. This elegant synthesis achieves state-of-the-art visual fidelity at real-time rendering frame rates. As the 3DGS literature grows rapidly, existing surveys have primarily organized the field around downstream applications. In contrast, this survey provides a rasterization-centric analysis, treating the differentiable rasterization pipeline as the core computational engine of 3DGS. We begin by delineating the mathematical foundations of Gaussian splatting and tracing its lineage from classical volume rendering. Subsequently, we propose a fine-grained taxonomy that categorizes the literature across three hierarchical dimensions: (1) representation and optimization, (2) rasterization pipeline innovations, and (3) scenario-driven rasterization extensions. By deconstructing these algorithmic advances in rasterization, we offer quantitative insights into hardware-algorithm co-design and outline critical trajectories for future research. To support the community, we maintain a continually updated repository of relevant literature and open-source implementations at https://github.com/3DAgentWorld/Advanced3DGS.

Abstract: Background: Pancreatic cancer is a leading cause of cancer-related mortality. Locally advanced pancreatic cancer (LAPC) remains associated with low resection rates and poor survival despite advances in systemic therapy. Intratumoral Phosphorus-32 (P-32) microparticle implantation is an emerging locoregional brachytherapy approach for unresectable LAPC. Methods: This prospective single-center observational cohort study included patients with unresectable LAPC treated with gemcitabine plus nab-paclitaxel and EUS-guided intratumoral P-32 microparticle implantation. Intratumoral distribution was assessed by SPECT/CT. Tumour response was evaluated according to RECIST 1.1, with local disease control rate (LDCR) assessed at approximately 16 weeks. Adverse events were graded according to CTCAE version 4.0. Results: Seventeen patients underwent P-32 implantation (median age, 70 years; ECOG PS 0–1). SPECT/CT confirmed appropriate intratumoral distribution in all patients, with no implantation-related procedural complications. At 16 weeks, 7/16 evaluable patients (43.8%) achieved partial response and 9/16 (56.2%) had stable disease, resulting in an LDCR of 100% among evaluable patients and 94.1% in the all-treated analysis. Mean target-lesion diameter decreased from 3.4 to 2.8 cm, with a median reduction of 21.4%. CA19-9 decreased by ≥50% in 13/16 evaluable patients (81.2%). Four patients (23.5%) underwent surgical resection, all achieving R0 resection. Adverse events were predominantly grade 1–2; one grade 3 event and no grade 4–5 events occurred. Conclusions: P-32 microparticle implantation combined with systemic chemotherapy was feasible and demonstrated encouraging early local activity with an acceptable safety profile. These findings support further prospective evaluation in larger cohorts.

Abstract: Feeding the growing global population without permanently overshooting local-to-planetary boundaries demands a transition from conventional resource extraction to holistic models. Current digital twin applications in (sea)food systems primarily focus on predictive monitoring but fail to holistically integrate socio-economic-ecological systems (SEES) and their emergent properties (e.g., sustainability, stability, dynamism, and circularity). However, it remains unknown whether a unified cybernetic architecture can simultaneously satisfy SEES-level properties by bridging asynchronous global supply-demand chains through embedded spatiotemporal feedback loops. This conceptual review argues that a cybernetic twin—defined as the continuous, bidirectional fusion of a SEES and its digital counterpart via recursively coupled algorithms—can bridge this gap by operating across seven interacting phases of the global food continuum at multiple spatiotemporal resolutions. By integrating ecosystem dynamics with real-time simulations, the twin would interact with decision-making processes by rendering transparent and dynamic interactions among SEES-level pathways across aquatic-terrestrial food systems. We propose how a cybernetic twin can be leveraged to achieve sustainable (sea)food quantity and quality in SEES.

Abstract: A broad reading of the literature finds that scientific and engineering discourse on AI safety has abundant theory, abundant measurement, and almost no identifiable coupling between them. The theory is asymptotic and non-binding; the measurement is atheoretical. Current results constraining deployment decision are not derived from any limit theorem arguments. The result appears as an ‘index hacking’ reminiscent of the ‘Integrated Information Theory’ catastrophe in consciousness studies. We provide some context for– and implications of– such lacunae using the asymptotic limit theorems of information and control theories to impose strict necessary condition constraints on the dynamics of interaction between cognitive agents that might represent markedly different intents, cultures, path-dependent developmental trajectories or modalities of ‘mind.’ We view each entity as intimately communicating with another in a Data Rate Theorem dyad under Rate Distortion Control Theory. A first outcome is a distribution-dependent ‘equipartition’ model of the dyad’s demand for a ‘materiel’ free energy-analog driven channel capacity that displays complex patterns of dysfunction. A second outcome is recovery of canonical delay distribution-dependent ‘eigenmode’ expressions for dyad dynamics characterizing a different class of pathologies. A third result explores the impact of a ‘logistics instability’ on ultimate conflict outcome. The model is then extended to many interacting composite entities. Models of cognitive phenomena not directly reflecting the constraints of the asymptotic limit theorems of information and control theories risk the same incoherence as perpetual motion ‘machines’ that do not respect the laws of thermodynamics. In sum, the probability models developed here can be reconstituted as statistical tools for data analysis and the crafting of more reliable patterns of interaction between closely-linked, but markedly-different, cognitive entities.

Abstract: Green bonds are intended to mobilise capital for environmental investment, yet evidence on their contribution to economy-wide circularity remains limited. This study examines whether national green-bond market development is associated with the circular material use rate and resource productivity in the EU-27. A balanced country-year panel covering 2014-2024 (297 observations) is analysed with country and year fixed effects, six controls, contemporaneous and one- and two-year-lagged green-bond shares, country-clustered standard errors, restricted wild cluster bootstrap-t inference with 9,999 repetitions, and Driscoll-Kraay sensitivity estimates. The complete-control samples contain 270, 243, and 216 observations for the current, L1, and L2 specifications. None of the six baseline models, 20 main robustness-model estimates, or 54 leave-one-country-out re-estimations yields a wild-bootstrap p-value below 0.10. Circular material use coefficients are consistently negative but imprecise, while resource-productivity coefficients change sign across timing and control choices. The findings do not support the two directional hypotheses and indicate no statistically robust conditional association over the observed period. Green-bond market size should therefore not be treated as sufficient evidence of circular-economy progress; allocation, additionality, and verifiable material outcomes remain decisive.

Abstract: Background: Hepatitis B (HBV) infection remains a major public health challenge in Latin America and the Caribbean (LAC), a region characterized by disparities in re-search capacity and limited molecular surveillance. A comprehensive assessment of research output and viral genetic diversity is crucial for informing region-specific control strategies. Methods: An integrated analysis combining bibliometrics, text-mining, and genomics was performed on HBV-related publications indexed in PubMed, Scopus, and Web of Science (1970–2024), along with 9,528 HBV nucleotide sequences from GenBank. Results: The study identified 4,083 publications, of which 72.1% (2,942) originated from Brazil, Mexico, and Argentina. Most were original articles (75.8%; 2,749/3,628) and involved human subjects (95.8%; 2,471/2,580), with a primary focus on serological markers, co-infections, vaccination, genotype characterization, liver pathology, and transfusion safety. Conversely, hepatocellular carcinoma, antiviral therapy, and met-abolic comorbidities were notably underrepresented. In vitro studies were limited, primarily examining viral replication, disease mechanisms, and gene expression, whereas animal studies focused on vaccine development, infection dynamics, and immune responses. Phylogenetic analysis identified eight circulating HBV genotypes (A–H), with genotypes A (43.4%; 4,138/9,528), F (29.3%; 2,787/9,528), and D (19.8%; 1,885/9,528) predominating. Genotype H was found mainly in Mexico (3.2%; 306/9,528). Notably, non-A genotypes accounted for most infections and showed dis-tinct geographic clustering (p < 0.001). Conclusion: HBV research in LAC is marked by pronounced regional disparities in scientific output, critical gaps in experimental and clinical research, and high viral di-versity. These findings underscore the need to strengthen regional research capacity and develop targeted strategies for HBV prevention, treatment, and vaccination to achieve hepatitis B elimination in the region.

Abstract: We prove that the optimal sphere packing density in \(\mathbb{R}^4\) is \(\Delta_4=\pi^2/16\), achieved uniquely by the \(D_4\) root lattice. The proof is local: every Voronoi cell of a unit-ball packing satisfies \(\mathrm{vol}(V_c)\ge 8\), with equality precisely when the active neighbours form the \(D_4\) kissing configuration at distance~$2$. The Voronoi domain decomposes into $176$ combinatorial chamber types under the Weyl group \(W(D_4)\) of order~$192$. For each type the free-volume surplus \(F_\Omega(\theta)\ge 0\) is established by two complementary certificates: \emph{Level~A}: an exact rational \(LDL^\top\) decomposition of a $6\times 6$ Hessian, with minimum pivot \(\tfrac{160}{13}\) at chamber C004, certifying the sum-of-squares identity \(F_\Omega^{(2)}(\tau)=\tau^\top H_\Omega\tau\ge 0\); and \emph{Level~B}: an explicit polynomial certificate combining a sum-of-squares identity \(H(c)=18(c-\tfrac{11}{18})^2+\tfrac{23}{18}\), a discriminant argument, and Bernstein expansions with positive rational coefficients (minimum \(\tfrac{1088}{625}\) on Piece~I; minimum \(\approx 0.765\) on Piece~II). All claims are verified independently by two computer algebra systems using exact rational arithmetic throughout. This resolves the four-dimensional sphere packing problem.

Abstract: Echocardiography is an important imaging modality in the assessment of cardiovascular diseases. Several echocardiographic signs have been described to aid in the diagnosis of specific clinical conditions and to help clinicians remember the key sonographic features. This review summarizes five classic signs in echocardiography, including the McConnell's Sign, the D-Sign, the 60:60 Sign, the SAM Sign, and the Swinging Heart Sign, as well as three novel signs, the Three Pipes Sign, the LTS Sign, and the LTI Sign. This review aims to provide an overview of the classic echocardiographic signs, focusing on their naming logics, definitions, and clinical implications to facilitate a better understanding of them, and to first propose three novel signs, the Three Pipes Sign, the LTS Sign, and the LTI Sign, which may aid in diagnosis and serve as teaching aids in echocardiography.

Abstract: Selecting an encoding configuration for Hyperdimensional Computing (HDC) determines whether a model fits the memory budget of an Internet of Things (IoT) device, yet the knowledge needed to make that choice is dispersed across thousands of papers in a form no optimizer can query. The configuration space is large and each evaluation is expensive, so exhaustive search is infeasible; sequential optimizers are the standard remedy, but they begin from an uninformed prior and spend early evaluations exploring blindly. We present an agentic Retrieval-Augmented Generation (RAG) recommendation system that converts the HDC literature into an informed starting point for this search. A corpus of 1,518 HDC papers is indexed multimodally. A vision-language model (VLM) describes 1,755 figure regions extracted from source PDFs, making plotted evidence such as accuracy against dimension curves retrievable alongside prose and tables. Given a task profile stating the dataset characteristics, target hardware, objectives, and memory budget, the agent retrieves evidence under an allocation that reserves context for figures, and returns an encoding configuration with supporting citations. That configuration seeds the first trial of a constrained multi-objective search over accuracy and inference memory, which then refines it, and feasibility is enforced by external measurement rather than by the model. On network intrusion detection (CICIDS2017) and bearing fault diagnosis (CWRU), the literature-grounded start raised the hypervolume of the recovered Pareto front by 17.4% and 19.1% over an identically budgeted search from a random draw, at no additional evaluation cost. A natural extension is to let the system perform the data preprocessing that the practitioner now supplies, so that a deployment can be described end to end from raw sensor streams.

Abstract: The recent emergence and rapid spread of Shamonda-like virus (SHAV/DE/2026) across several European countries raises questions about the mechanisms facilitating the circulation of this and related arboviruses. Although Culicoides biting midges are considered the most plausible vectors, the ecological pathways by which viruses may enter and persist within these insects remain incompletely understood. Here, we examine a potentially overlooked component of arbovirus transmission ecology: interactions among viruses, nematodes, and Culicoides. Published evidence demonstrates that mermithid nematodes can penetrate Culicoides larvae, develop within the insect, and in some systems persist through metamorphosis into adult midges. Importantly, experimental studies have also demonstrated nematode-associated enhancement of viral infection in Culicoides and virus carriage by nematodes in other arthropod systems. Based on these observations, we propose a testable environmental pathway in which virus-associated nematodes may facilitate viral entry into immature Culicoides, potentially linking environmental virus sources with adult vectors and susceptible vertebrate hosts. This hypothesis warrants targeted experimental investigation. If confirmed, such a pathway could expand current concepts of arbovirus transmission and should be considered in future biosurveillance, biorisk assessment, and preventive strategies against emerging arbovirus-associated biological threats.