Hacettepe University Reserach Information System
Not a member yet
118745 research outputs found
Sort by
Slippery Liquid-Infused Porous Surfaces Using Eucalyptus Oil-Loaded Bead-on-String Nanofibers for Antibacterial, Antifungal, and Antibiofilm Applications
No full textSlippery liquid-infused porous surfaces (SLIPS) are emerging as promising antifouling potential surfaces. However, conventional fabrication requires multistep chemical processes and synthetic lubricants with risks of accumulation. Herein, a simplified, nature-inspired route to multifunctional SLIPS by electrospinning bead-on-string polysulfone nanofiber-coated surfaces (bPSU_NfCS) directly onto glass, followed by infusion with Eucalyptus globulus oil (EGO) as a natural lubricant, resulting in EGO-infused bPSU_NfCS-SLIPS (bPSU_NfCS-EGO/SLIPS), is presented. Bead-on-string morphology and nanofibrous architecture provide roughness and internal reservoirs that trap EGO, thereby eliminating the need for etching or salinization. With infusion, bPSU_NfCS-EGO/SLIPS undergoes a hydrophobic-to-slippery transition, with contact angle decreasing from 121 degrees to 61 degrees and sliding angle to 5 degrees, enabling repellency toward diverse liquids. The surfaces exhibited remarkable antibacterial efficacy, achieving 100% inhibition against Escherichia coli (ATCC 25922) and Staphylococcus aureus (ATCC 6338), along with effective suppression of Candida albicans (ATCC 30028). Biofilm formation by Pseudomonas aeruginosa (ATCC 27853) decreased by >80%, while mineral fouling from CaCO3 was strongly inhibited. Cytocompatibility with HEK293-T cells confirmed nontoxicity, with >85% viability and adhesion. These findings validate the dual functionality of SLIPS, with passive repellency via slipperiness and active bioactivity. This strategy offers a scalable, biocompatible, sustainable pathway to multifunctional SLIPS with broad-spectrum antifouling performance, applicable to marine, environmental, and healthcare technologies
Non-targeted metabolomic profiling of Cistus species and association with anticholinesterase efficacy for Alzheimer's disease: In vitro and in silico evaluation
No full textEthnopharmacological relevance: Alzheimer's disease (AD) is characterized by decreased glucose utilization, and insulin therapy has been associated with improved memory. Therefore, AD is suggested to be classified as "Type 3 diabetes". Cistus L. species are traditionally used to treat diabetes, which is highly associated with AD, and the potential of this genus for treating AD has not been sufficiently investigated. Aim: This study focused on the untargeted metabolomic profiling of methanolic extracts from five Cistus species to investigate the correlation between the metabolites and bioactivity. Materials and methods: Gas chromatography-mass spectrometry and liquid chromatography quadrupole time-offlight mass spectrometry were employed for metabolomics analysis. The inhibitory activity of the extracts on acetylcholinesterase (AChE) and butyrylcholinesterase (BChE), as well as their antioxidant capacity, were assessed. Additionally, molecular modeling techniques were utilized to corroborate the correlation between the metabolites and their cholinesterase inhibitory potency. Results: The plant extracts demonstrated inhibitory effects on BChE with IC50 values ranging from 1.80 to 9.83 mu g/mL, which were notably lower than those observed for AChE. Correlation analysis revealed that chlorogenic acid demonstrated strong correlations with AChE inhibitory activity, while sinapyl alcohol was closely associated with BChE inhibitory activity. Additionally, molecular modeling studies supported the inhibitory potential of these metabolites. Conclusion: This study highlights the substantial cholinesterase inhibitory capabilities of Cistus species, with C. creticus demonstrating particularly strong activity against both AChE and BChE. The results indicate that extracts from these species could be valuable natural sources of active metabolites with potent cholinesterase inhibitory effects, presenting promising new options for AD therapy
İleri Glikasyon Son Ürünlerinden Zengin Bir Diyetle Beslenen Ratlarda Diyete İlave Edilen Zeytin Yaprağı Ekstraktının Glikasyon, İnflamasyon ve Gen Ekspresyonları Üzerine Etkisi
No full textLeveraging large language models to mimic domain expert labeling in unstructured text-based electronic healthcare records in non-english languages
No full textBackgroundThe integration of big data and artificial intelligence (AI) in healthcare, particularly through the analysis of electronic health records (EHR), presents significant opportunities for improving diagnostic accuracy and patient outcomes. However, the challenge of processing and accurately labeling vast amounts of unstructured data remains a critical bottleneck, necessitating efficient and reliable solutions. This study investigates the ability of domain specific, fine-tuned large language models (LLMs) to classify unstructured EHR texts with typographical errors through named entity recognition tasks, aiming to improve the efficiency and reliability of supervised learning AI models in healthcare.MethodsTurkish clinical notes from pediatric emergency room admissions at Hacettepe University & Idot;hsan Do & gbreve;ramac & imath; Children's Hospital from 2018 to 2023 were analyzed. The data were preprocessed with open source Python libraries and categorized using a pretrained GPT-3 model, "text-davinci-003," before and after fine-tuning with domain-specific data on respiratory tract infections (RTI). The model's predictions were compared against ground truth labels established by pediatric specialists.ResultsOut of 24,229 patient records classified as poorly labeled, 18,879 were identified without typographical errors and confirmed for RTI through filtering methods. The fine-tuned model achieved a 99.88% accuracy, significantly outperforming the pretrained model's 78.54% accuracy in identifying RTI cases among the remaining records. The fine-tuned model demonstrated superior performance metrics across all evaluated aspects compared to the pretrained model.ConclusionsFine-tuned LLMs can categorize unstructured EHR data with high accuracy, closely approximating the performance of domain experts. This approach significantly reduces the time and costs associated with manual data labeling, demonstrating the potential to streamline the processing of large-scale healthcare data for AI applications
Approximation by Max-Min Neural Network Operators
No full textIn this paper, we introduce a max-min approach for approximation by neural network operators activated by sigmoidal functions. Our focus lies in addressing both pointwise and uniform convergence in the context of univariate functions. Then, we investigate the order of approximation. We also take into account the max-min quasi-interpolation operators. Finally, we present several practical applications of our approximation methods, including a comparative analysis between max-min neural network operators and their max-product and linear counterparts, as well as denoising 1D noisy signals
Multiple papillary fibroelastoma presenting with mitral, tricuspid, and pulmonary valve involvement and surgical treatment: case report
No full textApproximately 90% of primary paediatric cardiac tumours are benign lesions. Depending on their location and size, benign cardiac tumours may cause inflow and outflow obstructions, cyanosis, valvular insufficiencies, myocardial ischaemia, associated dysfunction, systemic embolisation, arrhythmias, and even sudden death. Decision-making and timing for surgery can be challenging in children. Herein, we present an asymptomatic 11-year-old girl with papillary fibroelastoma in the mitral, tricuspid, and pulmonary valves, discussing the decision-making process and successful surgical management
Copper(II) Oxide Spindle-like Nanomotors Decorated with Calcium Peroxide Nanoshell as a New Nanozyme with Photothermal and Chemodynamic Functions Providing ROS Self-Amplification, Glutathione Depletion, and Cu(I)/Cu(II) Recycling
No full textUniform, mesoporous copper(II) oxide nanospindles (CuO NSs) were synthesized via a method based on templated hydrothermal oxidation of copper in the presence of monodisperse poly(glycerol dimethacrylate-co-methacrylic acid) nanoparticles (poly(GDMA-co-MAA) NPs). Subsequent decoration of CuO NSs with a CaO2 nanoshell (CuO@CaO2 NSs) yielded a nanozyme capable of Cu(I)/Cu(II) redox cycling. Activation of the Cu(I)/Cu(II) cycle by exogenously generated H2O2 from the CaO2 nanoshell significantly enhanced glutathione (GSH) depletion. CuO@CaO2 NSs exhibited a 2-fold higher GSH depletion rate compared to pristine CuO NSs. The generation of oxygen due to the catalase (CAT)-like decomposition of H2O2 by CuO@CaO2 NSs resulted in a self-propelled diffusion behavior, characteristic of a H2O2 fueled nanomotor. These nanostructures exhibited both peroxidase (POD)-like and CAT-like activities and were capable of self-production of H2O2 in aqueous media via a chemical reaction between the CaO2 nanoshell and water. Usage of the self-supplied H2O2 by the POD-like activity of CuO@CaO2 NSs amplified the generation of toxic hydroxyl (center dot OH) radicals, enhancing the chemodynamic effect within the tumor microenvironment (TME). The CAT-like activity provided a source of self-supplied O2 via decomposition of H2O2 to alleviate hypoxic conditions in the TME. Under near-infrared laser irradiation, CuO@CaO2 NSs exhibited photothermal conversion properties, with a temperature elevation of 25 degrees C. The combined GSH depletion and H2O2 generation led to a more effective production of center dot OH radicals in the cell culture medium. The chemodynamic function was further enhanced by an elevated temperature. To assess the therapeutic potential, CuO@CaO2 NSs loaded with the photosensitizer, chlorine e6 (Ce6), were evaluated against T98G glioblastoma cells. The synergistic combination of photodynamic, photohermal, and chemodynamic modalities using CuO@CaO2@Ce6 NSs resulted in cell death higher than 90% under in vitro conditions
