Community detection is a critical challenge in analysing real graphs, including social, transportation, citation, cybersecurity, and many other networks. This article proposes three new, general, hierarchical frameworks to deal with this task. The introduced approach supports various linkage-based clustering algorithms, vertex proximity matrices, and graph representation learning models. We compare over a hundred module combinations on the Stochastic Block Model graphs and real-life datasets. We observe that our best pipelines (Wasserman-Faust and the mutual information-based PPMI proximity, as well as the deep learning-based DNGR representations) perform competitively to the state-of-the-art Leiden and Louvain algorithms. At the same time, unlike the latter, they remain hierarchical. Thus, they output a series of nested partitions of all possible cardinalities which are compatible with each other. This feature is crucial when the number of correct partitions is unknown in advance.
Purpose This study examines how SMEs operating in the Polish transport, shipping and logistics (TSL) sector perceive the business value of digital freight exchanges, and how they associate platform use with operational efficiency, empty vehicle movements, transport resource utilization and competitive positioning. The study investigates whether the functionalities offered by digital freight platforms constitute distinct value dimensions or a single integrated platform value dimension. Methodology Empirical data were collected in 2025 through a questionnaire survey of 65 purposively selected Polish SMEs actively using digital freight exchanges. The sample is therefore not representative of the broader Polish SME road-freight population. The analysis included descriptive and reliability statistics, Exploratory Factor Analysis (EFA) (Principal Axis Factoring), Spearman's correlation, multiple linear regression, and Kruskal–Wallis tests with Dunn–Bonferroni post hoc comparisons. Factor stability was assessed using bootstrap analysis. Findings Respondents primarily used digital freight exchanges to obtain transport orders, improve fleet utilization, monitor shipments and maintain operational continuity during periods of lower demand. The most valued functionalities were contractor credibility assessment, freight search, fleet utilization, transaction security and responsiveness to demand fluctuations. Exploratory Factor Analysis identified a provisional one-factor solution explaining 50.67% of the total variance, interpreted as a Perceived Integrated Platform Value dimension, and suggesting that these SMEs evaluate platform functionalities as complementary elements of an integrated business-support system, because the items measure perceived importance. This dimension is interpreted as a proxy for perceived platform value rather than as a measure of realized benefits rather than as separate capabilities. Perceived platform value was positively associated with competitive advantage (ρ = 0.411, p < 0.001). While overall platform value did not differ significantly by enterprise size, micro-enterprises reported greater competitive benefits and more frequent platform use than medium-sized enterprises.
Accurate, precise, and non-invasive determination of cell density is a critical challenge in cell culture research, directly influencing experimental reproducibility and reliability of biological interpretations. Conventional methods, such as cell counting and colorimetric assays, are invasive and often unsuitable for continuous monitoring. This study evaluates excitation-emission matrix (EEM) fluorescence spectroscopy (multispectral fluorescence) combined with chemometric analysis as an innovative, data-driven approach for quantifying cell density in adherent cultures. The experiments were performed using two adherent cell lines, i.e., HaCaT and A375, simultaneously, to verify if the observed trends might extend beyond a single cell line and thus not be cell-line specific, which could further support the general applicability of the proposed cell number determination strategy. The number of cells in culture was determined using two reference methods, i.e., classical cell counting or the MTT assay providing information on the number of metabolically active cells. Unfolded partial least squares (PLS) regression was used to develop models for predicting cell numbers based on information encoded in excitation-emission spectra of cell cultures. Statistical analysis shows that multispectral fluorescence surpasses cell counting in sensitivity, while the MTT assay exhibits the highest precision. By enabling label-free, non-destructive, and repeatable measurements with applicability across different cell lines, multispectral fluorescence can be proposed as an interesting alternative, especially for automated and dynamic monitoring of cell cultures.
Point-of-Care (PoC) diagnostic systems require simplified workflows, high reliability, and long-term stability of biochemical reagents to meet the REASSURED criteria. A major challenge in achieving fully self-contained PoC devices is integrating reagent storage, particularly for sensitive biomolecules such as antibodies and enzymebased labels. While drying-based approaches are widely used, they often introduce structural damage, rehydration variability, and increased formulation complexity. In this work, a liquid-phase storage strategy based on hybrid polyethylene/aluminum/nylon blister systems is presented to preserve nanozyme-antibody conjugates for use in immunoassays. Bimetallic Au:Pt nanoparticles exhibiting high peroxidase-like activity were synthesized and conjugated with anti-CRP antibodies to form detection probes. The conjugates were encapsulated in thermally sealed blisters, and the impact of fabrication conditions, including short-term thermal exposure, on their functionality was evaluated. The results demonstrated that, despite a partial loss of activity during thermal sealing, the conjugates retained sufficient functionality to enable reliable CRP detection. The multilayer blister structure effectively limited evaporation, resulting in minimal mass loss over 12 weeks. Functional stability was maintained across various storage conditions, with over 60% activity retained at 4 degrees C and over 40% at room temperature after 12 weeks. Even at elevated temperature (37 degrees C), analytical performance remained sufficient for target detection. These findings demonstrate that blister-based liquid storage is a viable approach for integrating functional biochemical reagents into self-contained PoC systems, combining ease of use with preserved biomolecular activity.
We demonstrate a high-sensitivity fiber-optic sensor formed by cascading two Fabry-Perot (FP) interferometers (FP1 and FP2) for precise refractive index (RI) prediction. Both FP1 and FP2 are fabricated directly on the fiber tip via two-photon polymerization (2PP) based 3D printing, with an open-ended structure designed for operation in liquid environments. Meanwhile, the cascading FP2 with FP1 exploits the vernier effect, which amplifies the sensitivity of FP1, and experimental results show high sensitivity. The sensitivity of FP1 is 1143.78 nm/RIU (0.162 nm/(mg/mL)), while the vernier–effect of the sensor achieves a higher sensitivity of 3478.23 nm/RIU (0.494 nm/(mg/mL)), representing a 3.041–fold enhancement. A physics-guided data augmentation strategy is introduced to train SpecCMamba, and a hybrid deep learning model that integrates a convolutional neural network for local feature extraction with a Mamba encoder for global sequence modeling. The SpecCMamba enables precise RI prediction, effectively handling spectral overlap and non-linearity in the data. The model achieves a determination coefficient (R2) of 0.9996 and a Root Mean Square Error of 2.855 × 10−5. With its simple structure and high sensitivity, the proposed sensor assisted by the SpecCMamba offers significant potential for applications in highly precise refractive index prediction and related fields.
