5916 research outputs found
Sort by
SERS on analyte-enriched blood for rapid, culture-free sepsis recognition and causative pathogen identification with super operational neural networks
No full textSepsis remains a leading cause of morbidity and mortality, yet routine diagnostics are slow, culture-dependent, and often lack the sensitivity or specificity required for early intervention. Prior studies rarely demonstrate clinical-grade performance on blood culture samples or in independent external cohorts. We address these gaps with a surface-enhanced Raman spectroscopy and deep learning workflow (SERS-DL) that performs sepsis instance recognition and causative pathogen identification directly from target-analyte enriched blood. We assembled a primary dataset of SERS spectra acquired from 653 analyte-enriched blood samples collected at a tertiary hospital in Qatar and an external blind cohort of 70 independent samples. After rigorous preprocessing and class-weighted augmentation of SERS spectra, we trained SuperRamanNet, a lightweight one-dimensional classifier based on super operational neural networks. In five-fold, sample-contained cross-validation, the system achieved 99.67 % accuracy for binary sepsis recognition and 98.84 % accuracy for six-class pathogen identification. On the external cohort, performance remained high at 98.28 % for pathogen typing, indicating robust generalizability. Comparative benchmarks and ablation studies confirmed consistent gains over convolutional and operational baselines and quantified the impact of augmentation and architectural choices. Residual confusions were concentrated between control and Escherichia coli and among certain Gram-negative classes, underscoring the need for improved raw class balance during blood sample collection. Overall, this rapid, culture-free, and portable SERS-DL pipeline delivers near clinical-grade accuracy for sepsis detection and pathogen identification directly from blood. The compact model and streamlined workflow support point-of-care translation, with potential to accelerate triage, guide early therapy, and reduce the global sepsis burden. © 2025 The Authors.Qatar Research, Development and Innovation Council (QRDI) ; Qatar National Research Fund (QNRF)Publisher versio
Micro-laser-induced breakdown spectroscopy using GHz repetition rate pulses at nJ pulse energy
No full textUsing GHz repetition rate pulses in burst mode has shown significant attention, particularly in laser micromachining, due to its ability to enable highly efficient, low-energy material removal. In this work, we introduce the first Laser-Induced Breakdown Spectroscopy (LIBS) in the ablation-cooled regime. By employing GHz repetition rate pulses, we present a high-speed (100 kHz) micro-LIBS system that operates with ultra-low pulse energies in the range of 10 - 200 nJ. To achieve this, we employ our home-built 2.8 GHz burst-mode Yb-doped fiber laser, which delivers ∼40 ps pulses to the sample with a beam diameter of around 18 μ m. A systematic LIBS study was conducted on stainless steel (SS) under varying burst durations and burst energies to investigate their effects on the optical emission spectrum. Finally, the electron temperature and electron density were determined using the Boltzmann plot method and Stark-broadened line profile analysis, respectively. © 2025 The Author(s). Published by IOP Publishing Ltd.TÜBİTAKPublisher versio
Optimizing collection processes using conservative Q-learning
No full textThis study proposes a reinforcement learning framework based on Conservative Q-Learning (CQL) to optimize debt collection strategies while mitigating customer churn. Traditional rule-based approaches often fail to adapt to individual customer profiles or evolving behaviors. To address this limitation, our framework dynamically recommends actions tailored to each customer's characteristics. Using customer datasets, we evaluate the performance of the proposed model across various reward coefficient settings (representing potential future profit loss in case of churn). The results show that while standard Q-Learning generally underperforms the rule-based strategy, CQL achieves overall performance comparable to rule-based approaches. Notably, product-level analysis shows statistically significant improvements for general-purpose installment loan (GPL) customers, while outcomes for credit card (CC) and overdraft (OD) customers are weaker. This likely reflects reinforcement learning's tendency to prioritize higher-value cases, suggesting that product-specific models may further enhance performance across loan types
Wear performance of additive-manufactured and heat-treated CoCrMo alloy
No full textThis study examines the wear behavior of additive-manufactured CoCrMo alloy in as-built and heat-treated conditions. Suitable thermal processing modified fine cellular grains and grain-boundary carbides into equiaxed FCC grains and minor HCP phase. With increased load, coefficient of friction declined due to larger contact area and tribo-oxide formation during reciprocating wear experiments. At 10 N, wear is mainly abrasive in the as-built condition, whereas after heat treatment at 1200 degrees C it becomes dominated by stick-slip mode. At 40 N, there is a transition to delamination, ploughing, and subsurface fatigue cracking. Heat-treated samples, despite lower hardness, outperformed at 40 N while exhibiting lower wear rate.Turkiye Cumhuriyeti Kalkinma Bakanlig
Metaverse acceptance in younger and older cohorts: Testing technology acceptance model
No full textThe monthly users of Metaverse, a multi-user virtual reality platform, are over 400 million around the world (Nikolovska, 2023). However, research about individual differences in acceptance of the metaverse is still limited. In the present study, we examined individuals’ cognitive responses, attitudes toward using, and willingness to engage in metaverse based on the Technology Acceptance Model (Davis, 1987). Hence, we developed the Metaverse Acceptance Scale (MAS) and explored how young adults and older adults differ in subscales of attitude, intention to use, perceived usefulness, perceived ease of use, and eagerness to know more about metaverse. The participants (N = 721) filled out a demographic questionnaire and MAS online. MAS demonstrated a 4-factor structure with adequate validity and reliability: Attitude, behavioral intention, perceived usefulness, and perceived ease of use. Not only did the subscales and items show variations between younger and older individuals, but also the associations between components of MAS. The associations between perceived ease of use and attitude, perceived usefulness and intention to obtain further information, and attitude and intention to acquire further information showed variations in the two cohorts. The acceptance, engagement, and intention to adopt metaverse can vary based on age. Thus, different age groups may be active in different domains of the metaverse. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2026
