Sivas Cumhuriyet University Research Information System
Not a member yet
61356 research outputs found
Sort by
Osteoarthritis Severity Classification in Knee X-Rays Using Optimized Deep Learning Approaches
No full textOsteoarthritis (OA) is a common, progressive joint disease that significantly reduces quality of life and limits mobility, especially in older adults. It is essential to accurately classify the disease in its early stages to develop effective treatments and slow its progression. This study introduces a deep learning-based system for classifying OA severity using knee joint X-ray images. EfficientNetB1, DenseNet169, and Xception architectures were employed for five-class OA classification (asymptomatic, early, mild, moderate, and severe), with hyperparameters in the fully connected layers optimized through the Gray Wolf Optimization (GWO) algorithm. By automatically selecting the most suitable parameters with GWO, the model learns more effectively and provides more accurate results in distinguishing OA levels. The dataset includes knee X-ray images from patients at University Training and Research Hospital, comprising 1000 images—200 per class. Model performance was assessed using accuracy, precision, recall, F1 score, and ROC curves. The experiments consisted of two stages: the first involved a five-class classification, and the second involved binary classification to distinguish between mild and severe OA. Correctly identifying moderate and severe stages, which are particularly serious, is vital for determining the need for surgical intervention. The best results were achieved with the DenseNet169 model: 74% in multi-class classification and 93.75% in binary classification. These findings show that the optimized models deliver high accuracy and effectiveness in OA diagnosis. This system helps specialists determine OA severity levels early, allowing for more informed treatment and surgical decisions
Mo/n-Sİ SCHOTTKY DİYOTLARIN ÜRETİMİ VE ELEKTRİKSEL ÖZELLİKLERİNİN I-V, C-V-F, G-V-F ÖLÇÜMLERİNDEN BELİRLENMESİ
No full textApplications of a novel deep neural network to the classification of liver steatosis and breast lesions in ultrasound images
No full textUltrasound is one of the most commonly used imaging modality in clinical practice. A computer-based approach is required for the non-invasive detection of chronic liver diseases or breast cancers. In particular, breast cancer is a substantial public health concern, and prompt detection and classification are crucial for the effectiveness of treatment. In this study, a divergence-based feature extractor (DivFE), a new deep learning model focusing on the identification of features, is used. Image features are extracted with convolutional neural networks (CNN) trained with Walsh vectors and the classification process is carried out with minimum distance network (MDN). In the literature, it is seen that ultrasound images of breast and liver diseases are successfully classified using deep neural networks (DNN). In this study, the same images were classified with high classification accuracy using a smaller number of nodes compared to the DNNs in the literature. To demonstrate the advantages of the DivFE, four widely used datasets were employed: three breast cancer datasets (Datasets I, II, and III) and one liver steatosis dataset (Dataset IV). Dataset I, Extended Dataset (I+II), Dataset III and Dataset IV, classification success rates of 100%, 97%, and 91%, and 100% were respectively achieved by using the DivFE with a small number of nodes. It is seen that classification accuracies obtained in the literature were achieved by using a new, small-sized DNN
A Plantago lanceolata L. extract-based cream enhances wound healing by modulating inflammatory mediators and growth factors in a full-thickness wound model: In vivo and In silico evidence
No full textThis study presents the wound healing supporting processes of cream formulations developed using Plantago lanceolata L. (PL) methanol extract in rats with full-thickness excisional skin wounds. Following the creation of full-thickness excisional skin wounds in rats, these formulations (25 mg/g, 50 mg/g, and 100 mg/g) and Fito cream (control drug) were applied topically to the wound areas for 14 days. Biochemical (TNF-α, IL-8, MMP-2), histopathological, and immunohistochemical (FGFR2/Bek, IGF-I, TGF-β1, VEGFR2, ERK 1/2, and iNOS) analyses were performed on wound tissues following sacrifice. Major components of the PL methanol extract were determined by LC-HRMS analysis. Interactions between major components and inflammation-associated human receptors were evaluated using molecular docking (Schrödinger Maestro version 13.9.138.) methods. Significantly greater closure of wound areas was observed on day 14 in the rats treated with the developed new cream formulations and Fito cream compared to the positive control group. Inflammation was significantly lower, and epithelialization was higher in the group treated topically with 100 mg/g PL methanol extract-containing cream formulation than in the other groups. Additionally, the major components of the PL methanol extract exhibited strong binding affinity with inflammation-related human receptors. The study findings show that the developed cream formulation supports wound healing by regulating antioxidant and anti-inflammatory pathways. The PL cream formulation (100 mg/g) was more effective in wound healing than Fito cream and possesses significant potential for clinical use
Synthesis and characterization of new Schiff base derivatives an imidazopyridine motif, followed by theoretical DFT and molecular docking studies
No full textThe need for new bioactive compounds is becoming increasingly vital due to climate change and antibiotic resistance. The focus of this study was the synthesis and characterization of imidazopyridine derivatives with substitutions at the 3-position and characterization of structures was determined using spectroscopic methods NMR, LC-MS, and FT-IR). The structures were in accordance with the predicted above molecular structures. Quantum chemical calculations based on density functional theory (DFT) in gas and methanol phase were also performed to gain insight into the electronic behavioral patterns and additional structures of the mentioned molecules in order to acquire results on the stability and reactivity and being able to assess the energy of their optimized structure to avoid this type of consideration tool for the foregoing purposes. Among the synthesized compounds HA05 appeared to be the most stable. Molecular docking studies were also performed with hepatitis B virus protein data bank PDB with codes 1WZ4 and 3MS6 performed similar as with pharmacophore models provided indication of possible antiviral activity with HA05 having good binding affinities (–4.38 and –6.47). Moreover, ADME/T (absorption, distribution, metabolism, excretion, and toxicity) predictions provided some favorable pharmacokinetic/hazards profiles compared to approve drugs. While these results are promising, it will be important to perform additional experiment to confirm biological activity
Benzothiazole-thiadiazole hybrids as dual α-glucosidase and aldose reductase inhibitors: Synthesis, in vitro, and In Silico studies
No full textThe present study explores the synthesis, characterization, α-glucosidase and aldose reductase inhibitory activities of some novel benzothiazole-thiadiazole hybride compounds. All compounds exhibited a higher potential of aldose reductase (AR) inhibition (KI: 4.816 ± 0.342–19.47 ± 1.726 nM and IC50: 5.698–12.560 nM) compared to the reference inhibitor epalrestat (KI: 756.342 ± 52.874 nM, IC50: 787.142 nM), along with higher α-glucosidase (α-GLY) inhibitory activity (KI: 0.413 ± 0.032–20.971 ± 2.035 µM and IC50: 0.616–31.247 µM) relative to acarbose (KI: 119.43 ± 10.65 µM and IC50: 136.28 µM). Among the tested compounds, especially compounds 4a (KI: 4.816 ± 0.342 nM), 4b (KI: 6.244 ± 0.456 nM), and 4i (KI: 5.260 ± 0.386 nM) stand out as the most potent AR inhibitors while compounds 4b (KI: 1.166 ± 0.097 µM) and 4c (KI: 0.413 ± 0.032 µM) come forward as the strongest candidates for α-GLY inhibition. Also, in silico molecular docking studies of the most potent compounds 4a, 4b, 4c and 4i are performed to evaluate interactions between the active compound and binding site of the tested enzymes. Furthermore, the predicted ADMET characteristics of these compounds was calculated using QikProp to gain insights into their drug-like properties. Cytotoxicity study was performed with the L929 fibroblast cell line in vitro revealed that all the synthesized compounds were non-cytotoxic
Machine learning-driven predictive modeling of natural frequency and displacement in perforated diaphragms for enhanced structural analysis
No full textDisplacement and naturel frequency are the most important design parameters for diaphragms based microelectromechanical system (MEMS) pressure sensors. For nonconventional diaphragm design of MEMS devices, finite element method (FEM)-based analysis to obtain these two parameters requires quite long time and cost as compared to conventional diaphragm design including circular, square, and rectangular shape. Thus, one major disadvantage of FEM is the excessive time required for simulation. Machine learning (ML) algorithms might be an alternative approach to FEM analysis. ML algorithms, which is an easier, functional, and time and cost saving, might provide rapid prediction of essential information comprising displacement and naturel frequency of MEMS diaphragm design with accurate and reliable results. In this study, ML algorithms including XGBoost regressor, LightGBM regressor, CatBoost regressor, and TabNet regressor were used to estimate displacement (µm) and frequency (Hz) of perforated low temperature co-fired ceramic (LTCC) diaphragms using 200 FEM-based numerical results. Predicted results were compared by considering R2, MAE, RMSE, and MAPE metric. According to these results, best performance was obtained by CatBoost regressor with the values of R2 = 0.927 and R2 = 0.995 for the displacement and frequency prediction, respectively. It was realized that CatBoost strikes an exceptional balance between computational efficiency and predictive performance, while LightGBM emerges as a strong alternative for scenarios prioritizing speed and memory efficiency. As a result, it was concluded that ML algorithms might be a useful, cost, and time effective tools for rapid analysis of displacement and naturel frequency of perforated diaphragms without requiring FEM analysis
Cilazapril and benazepril mitigate neurodegeneration and α-synuclein accumulation in a cellular model of parkinson’s disease
No full textParkinson’s disease (PD) is characterised by progressive neuronal degeneration and oxidative stress, both of which significantly contribute to its pathology and clinical symptoms. This study investigated the antioxidant, anti-inflammatory and neuroprotective effects of cilazapril and benazepril in an in vitro PD model induced by 6-hydroxydopamine (6-OHDA) in SH-SY5Y neuroblastoma cells. We performed molecular docking of these compounds to dopamine-related receptors (AT1, AT2, D1A, D1B and D2), as well as dynamic simulations. Morphological analysis revealed structural preservation in treated cells, and key biomarkers (MTT, TAS-TOS, IL-10, IL-1β, TNF-α, LDH, MDA, SOD and TGF-β) were evaluated using ELISA. The gene expression of the D1 and D2 receptors was measured using real-time PCR, and the expression of α-synuclein and neuronal nitric oxide synthase was analysed using immunohistochemistry. Both drugs significantly reduced oxidative damage, α-synuclein aggregation and neuroinflammation (*p < 0.05). They modulated genetic pathways involved in dopaminergic signalling and showed high binding affinity for the AT1, AT2, D1A, D1B and D2 receptors. These findings suggest that cilazapril and benazepril exert multi-targeted protective effects that extend beyond simple antioxidant activity, and that they may be effective in treating or mitigating PD-related neurodegeneration
