Machine learning in nanoscience and beyond: Workflows, data processing, XAI and ITAP metrics, language-based models.

Adhesives for cultural heritage conservation: Functions, performance evaluation, and application development

In neurosurgery, accurate brain tissue characterization via probe-based Confocal Laser Endomicroscopy (pCLE) has become popular for guiding surgical decisions and ensuring safe tumour resections. In order to enable surgeons to trust a tissue classification model, interpretability of the result is required. However, state-of-the-art (SOTA) deep learning models for pCLE data classification exhibit limited interpretability. This paper introduces a novel image classification framework for interpretable brain tissue characterisation using pCLE data. Firstly, instead of the commonly employed cross-entropy based classification loss, we propose Label Contrastive Learning (LCL) loss to learn intra-category similarities and inter-category contrasts. We are then able to generate highly representative data embeddings, which not only improve classification performance but also distinguish characteristics from different tissue classes. Secondly, we design a Saliency Consistency (SC) module to enable the trained model to generate clinically relevant saliency maps of the input data. To further refine the saliency maps, a novel Top-K Maximum and Minimum Pooling (TK-MMP) layer is introduced to our SC module, to increase the contrast of saliency values between non-clinically relevant and clinically relevant areas. For the first time, the Exponential Moving Average (EMA) is used in a novel fashion to update global embeddings of the different tissue categories rather than the weights of the model. In addition, we propose a Global Embedding Inference (GEI) layer to replace learnable classification layers to achieve more robust classification by estimating the cosine similarity between the input data embeddings and global embeddings. Performance evaluation on ex-vivo and in-vivo pCLE brain data verifies that our proposed approach outperforms SOTA classification models in terms of accuracy, robustness and interpretability. Our source codes are released at: https://github.com/XC9292/LCL-SC.git.

Performance evaluation in healthcare: a narrative review of recent advances and challenges in Italy

Microwave-assisted regeneration of CuxO-modified activated carbon saturated with sulfamethazine: Mechanism and performance evaluation

TransformerCARE: A novel speech analysis pipeline using transformer-based models and audio augmentation techniques for cognitive impairment detection.

Bimetallic (Fe–Zn) MIL MOFs for enhanced textile dye adsorption: synthesis, characterization, and performance evaluation

This research aims to study the role of green human resource management in achieving sustainable development at the Libyan Iron and Steel Company in Misrata, focusing on measuring the availability of its key practices and their impact on the three dimensions of sustainable development (environmental, economic, and social). The study adopted a descriptive-analytical approach, and data were collected through a questionnaire consisting of four main axes (green recruitment, green training and development, green performance evaluation, and green rewards and incentives). The questionnaire was distributed to a stratified random sample of 100 employees at the Libyan Iron and Steel Company in Misrata. The results showed partial to moderate adoption of these practices, with the green training and development axis recording the highest mean score (3.84), followed by green recruitment (3.75), then green performance evaluation (3.72), and finally green rewards and incentives (3.62). These results indicate that industrial companies have begun investing in raising environmental awareness through training, but they face clear gaps in linking environmental performance to job evaluation and financial incentives, which limits the continuous promotion of green behaviors. The findings are consistent with previous studies that have confirmed the positive impact of green human resource management on sustainable development. However, they reveal local challenges specific to the industrial sector, including weak financial incentives and the absence of quantitative environmental performance indicators. The study recommends developing comprehensive green recruitment policies, linking training to career development plans, incorporating environmental indicators into performance evaluations, establishing an integrated incentive system that combines financial and non-financial aspects, and fostering a green organizational culture.

Problem definition: Much of the focus of queueing theory (QT) is on performance evaluation that supports comparative analytics—that is, comparing performance measures under different interventions. However, closed-form queueing models are very sensitive to assumptions. We develop a data-driven Structural Causal Queueing Model (SCQM)—a form of structural causal models that automatically adapts to the data-generating process of queueing systems, finds causal relations, and supports comparative analytics. Numerical experiments show that the accuracy of SCQM is competitive with QT, even for examples where analytical queueing solutions are available. Methodology: We employ structural causal modeling methodology that uses queueing-relevant features to develop a simulator that replicates the system’s data-generating process without requiring prior knowledge of its dynamics. We apply Machine Learning models for identifying the parent sets and causal relations. We then provide intervention analysis using Monte Carlo simulation. Managerial implications: We use queueing knowledge to develop an accurate self-adapting data-driven performance evaluator for congested systems that requires no prior knowledge of the system dynamics. Using this method, companies can perform comparative analytics of interventions for queueing systems that may not be analytically solvable. History: This paper was selected as part of the 1RR initiative between M&SOM and the MSOM Society. This paper was part of the 2024 MSOM Service Operations Service Management Special Interest Group Conference. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2024.1515 .

Clean heating is essential for achieving carbon neutrality and accelerating the low-carbon transition of buildings. Electric radiant heating panels offer advantages such as flexible control, rapid thermal response, and simple installation. This study investigates the feasibility and performance of a Carbon Crystal Panel (CCP) floor heating system for buildings in China's hot summer and cold winter (HSCW) climate region. A numerical model of a CCP-heated room was developed, validated through experimental testing, and used to evaluate heating performance, temperature regulation behavior, thermostat placement, and the influence of wall thermal storage. The results demonstrate that while a higher heating intensity substantially speeds up the indoor temperature rise, it also triggers earlier and more frequent cycling due to faster attainment of the set point temperature. The optimal thermostat location is identified as approximately seven-eighths of the room width from the exterior wall. Lightweight walls lead to larger temperature fluctuations, whereas higher thermal storage reduces cycling frequency. Compared with continuous operation, a time-of-use intermittent strategy reduces daily heating cost by 23.44 %. This study provides a scientific basis for the design and practical application of CCP floor heating systems in HSCW regions.

Berry pomace-derived carbon quantum dots for antimicrobial active packaging: Hydrothermal synthesis, characterization, and performance evaluation

A cable-based polymer damping reinforcement system for confined masonry structures: Experimental investigation and performance evaluation

Raman spectroscopic characterization and thickness assessment of hydrolysis in polymer-encapsulated Lithium hydride.

Performance evaluation and methods comparison of transcriptomic-based approaches for the characterization of wastewater treatment effluent.

Evaluation of load reduction performance according to the modeling approach of the mooring load reduction device

ChatGPT (Chat Generative Pretrained Transformer), a large language model-based artificial intelligence, simulates doctor-patient communication. This study tested its applicability in injection-based cosmetic consultations using questions selected by 3 board-certified plastic surgeons with doctoral training and more than 10 years of experience at China's leading plastic surgery hospital. Professionalism and safety ratings were independently rated by 2 doctors for the test section and a third doctor for possible argument. Each question was asked 3 times independently to assess whether ChatGPT's performance on the same question was consistently satisfactory. Descriptive statistical analysis, multiple linear regression analysis, consistency analysis, 1-way analysis of variance, and repeated-measures analysis of variance were used to evaluate ChatGPT's safety, professionalism, empathy, and performance stability in this specific field. ChatGPT showed 58.3% professionalism, 94.54% safety, and only 11.48% empathy, with lowest ratings for questions related to injection effects and wrinkles and highest ratings for failed repairs. Reproducibility was 65.57%, indicating stability. Empathetic and professional answers were usually more comprehensive, whereas inconsistent ones were correct but insufficient. By analyzing the assessment results of ChatGPT in terms of safety, professionalism, empathy, and performance stability, it is suggested that ChatGPT could potentially serve as an auxiliary tool to help doctors and patients improve treatment outcomes; however, efficiency and patient satisfaction were not directly measured in this study and should be validated in future prospective studies involving real patients.

For nearly eight decades, the evaluation of disintegration performance has relied on a single disintegration time (DT) value, which serves as the sole pass or fail criterion for drug products. Although practical for regulatory compliance, this endpoint-based metric offers limited scientific insight into the underlying physical breakdown of the tablet matrix. In this study, the dynamic evolution of the gap height between the disk and mesh in the USP〈701〉 disintegration apparatus was monitored using a state-of-the-art non-contact distance sensor, enabling the construction of disintegration profiles that function as a mechanical fingerprint of the dosage form. Simvastatin 20mg tablets from six different brands were investigated, encompassing both originator and generic alternatives, and the resulting gap height profiles clearly discriminate among brands in terms of formulation-dependent disintegration behaviour. Within the measurable range, the erosion interface was found to propagate linearly towards the tablet core at a constant velocity, supporting the interpretation of the process as governed by linear erosion-controlled kinetics, in which the erosion rate and liquid ingress are synchronised. By linking the observed erosion velocity to the rate of surface area generation, this profiling approach provides the physical parameters needed to improve dissolution models within the Noyes-Whitney framework. The results demonstrate the potential of upgrading the conventional USP disintegration setup into an effective Process Analytical Technology tool for formulation development and quality assessment.

Evaluating the performance of Large language models in rheumatology for connective tissue Diseases: DeepSeek-R1, ChatGPT-4.0, Copilot, and Gemini-2.0.

Comparative evaluation of the thermal performance of a new sustainable energy system implemented with flat plates

Evaluation of the rock-cutting performance for TBM partial-wear cutters using vibration analysis and discrete element method