Sivas Cumhuriyet Üniversitesi Kurumsal Akademik Arşivi - Sivas Cumhuriyet University Open Access System
Not a member yet
34175 research outputs found
Sort by
Reliability and Validity Study of the Turkish Version of the Motivation and Attitude to Change Health (MATCH) Scale
No full textObjectiveThis study aims to culturally adapt the 'Motivation and Attitudes Towards Changing Health (MATCH)' scale into Turkish and assess its reliability and validity within the Turkish population.MethodsThe Turkish version of the MATCH questionnaire, designed to gauge patient motivation for initiating or sustaining behaviour changes, was created through translation and adaptation. A total of 305 patients diagnosed with chronic illnesses completed the questionnaire and scale to evaluate its validity and reliability.ResultsThe Turkish version of MATCH demonstrated strong internal consistency and reliability. The Cronbach's alpha coefficient for the entire scale was 0.737, with subdimensions ranging between 0.69 and 0.75. Notably, no item displayed a total correlation value below 0.40. Furthermore, test-retest analysis indicated high intraclass correlation coefficients. Consequently, no items were removed due to their consistently high reliability values among all nine items.ConclusionThe Turkish-adapted MATCH scale proves to be a valid and reliable instrument for assessing motivational status within the Turkish population
Mortality in patients with hypoxic ischemic encephalopathy treated with therapeutic hypothermia
No full textPurpose: Hypoxic-ischemic encephalopathy is a heterogeneous clinical syndrome that occurs in the perinatal period and is characterized by altered consciousness or seizures, respiratory depression, and hypotension. The aim of this study was to evaluate mortality in hypoxic-ischemic encephalopathy patients receiving therapeutic hypothermia. Materials and Methods: The study included 97 hypoxicUnit. The cases were evaluated for mortality and were divided into two groups: group 1 (n: 9, non-survivors) and complications of hypoxic-ischemic encephalopathy, APGAR scores, blood support, and laboratory parameters were evaluated for mortality. the risk of death. Conclusion: Mortality rates were significantly higher cases that developed Meconium aspiration syndrome associated hypoxic-ischemic encephalopathy than hypoxic-ischemic encephalopathy cases without meconium aspiration syndrome. A low APGAR score, increased number of intubation days, acute kidney injury, thrombocytopenia, and need for fresh frozen plasma were associated with a high risk of mortality in infants receiving therapeutic hypothermia for hypoxic-ischemic encephalopathy, and the presence of meconium aspiration syndrome significantly increased this risk
Predicting optical properties of NiO films fabricated by RF magnetron sputtering: A machine learning approach
No full textNiO films with different thicknesses (100, 150, 200, 250, 300 and 400 nm) were grown on glass substrates using the RF Magnetron sputtering method and their optical transmittance properties were analysed with a spectrophotometer. An innovative aspect of this work was the application of machine learning techniques used to derive new insights from experimental data. Four different machine learning algorithms -ANFIS, Artificial Neural Networks (ANN), Support Vector Machines (SVM) and Gaussian Process Regression (GPR)- were tested. While the models were trained using films of different thicknesses, a randomly selected 75 % of the whole dataset was used for model testing and the remaining 25 % of the films were used for testing the models. Among these, ANN and GPR models were found to be the most successful models. Using these models, the energy band gaps were estimated at 1 nm intervals and the values ranged from approximately 3.50 eV to 3.76 eV. © 2024 Elsevier GmbHSivas Cumhuriyet University Nanophotonic Application and Research Cente
Development of machine learning models for project effort prediction and explanation using SHAP method
No full textGünümüzde işletmeler, dijital dönüşüm sürecine uyum sağlayabilmek ve rekabet avantajı elde edebilmek adına etkin bir proje yönetim sürecine ihtiyaç duymaktadır. Özellikle yazılım projelerinin giderek artan bir ivmeyle yaygınlaşması, yazılım projelerinde doğru ve güvenilir efor tahmini yapmayı kritik bir gereklilik hâline getirmiştir. Efor tahmini, bir projenin tamamlanması için ihtiyaç duyulan iş gücü ve zamanın öngörülmesini sağlayarak kaynak planlamasının optimize edilmesine ve maliyetlerin etkin bir şekilde yönetilmesine katkıda bulunmaktadır. Ancak, yazılım projelerinin dinamik ve karmaşık yapısı, geleneksel yöntemlerle gerçekleştirilen efor tahminlerinin doğruluğunu sınırlayabilmektedir. Bu nedenle, son yıllarda veri odaklı yaklaşımlar ve makine öğrenmesi teknikleri, efor tahmini sürecinde önemli bir araştırma alanı hâline gelmiştir. Bu çalışmada, yazılım projelerinde efor tahmininin doğruluğunu artırmaya yönelik olarak rastgele orman, karar ağacı, doğrusal regresyon, yapay sinir ağı, GradientBoost ve AdaBoost gibi yaygın kullanılan makine öğrenmesi yöntemleri uygulanmıştır. Çalışmanın deneysel aşamasında, china_original, cocomonasa_v1, humans2, nasa93, usp05 ve usp05-ft olmak üzere altı farklı veri seti üzerinde model performansları değerlendirilmiştir. Model sonuçlarının güvenilirliğini artırmak amacıyla 50 tekrarlayan sınama yaklaşımı kullanılmış ve modellerin performansları ortalama mutlak hata, ortalama logaritmik kare hatası, belirleme katsayısı ve ortalama göreli büyüklük hatası gibi yaygın metrikler aracılığıyla karşılaştırılmıştır. Deneysel analizler sonucunda, yapay sinir ağı, rastgele orman, karar ağaçları ve GradientBoost modellerinin belirli veri setlerinde yüksek başarı gösterdiği tespit edilmiştir. Bununla birlikte, genel değerlendirmeler doğrultusunda karar ağacı modelinin proje efor tahmini açısından en başarılı yöntem olduğu sonucuna varılmıştır. Çalışmanın bir diğer önemli aşaması olarak, geliştirilen modellerin karar verme süreçlerini anlamlandırmak ve model şeffaflığını artırmak amacıyla SHAP yöntemi kullanılmıştır. SHAP analizi sonuçları, her bir veri setinde bazı özniteliklerin model karar alma süreçlerinde diğer özniteliklere kıyasla daha etkili olduğunu ortaya koymuştur. Bu bulgular, proje yöneticilerinin ve yazılım geliştiricilerinin karar süreçlerini daha bilinçli bir şekilde yönlendirmelerine olanak tanıyacak önemli içgörüler sunmaktadır. Elde edilen sonuçlar, makine öğrenmesi tabanlı modellerin yazılım projelerinde efor tahmini süreçlerinde başarılı bir şekilde uygulanabileceğini ve açıklamalı yapay zekâ teknikleri ile bu modellerin yorumlanabilirliğinin artırılabileceğini göstermektedir. Bu bağlamda, çalışmanın yazılım mühendisliği ve proje yönetimi alanlarında ileriye dönük araştırmalar için değerli bir kaynak teşkil etmesi beklenmektedir.In today's digital era, businesses require an effective project management process to successfully adapt to digital transformation and gain a competitive advantage. The increasing prevalence of software projects has made accurate and reliable effort estimation a critical requirement. Effort estimation plays a crucial role in optimizing resource planning and managing costs effectively by predicting the workforce and time needed to complete a project. However, the dynamic and complex nature of software projects poses significant challenges, limiting the accuracy of effort estimation using traditional methods. Consequently, data-driven approaches and machine learning techniques have emerged as a prominent research area in effort estimation. This study explores the application of common machine learning methods, including random forest, decision tree, linear regression, artificial neural networks, GradientBoost, and AdaBoost, to enhance the accuracy of effort estimation in software projects. The experimental phase involved evaluating the performance of these models on six distinct datasets: china_original, cocomonasa_v1, humans2, nasa93, usp05, and usp05-ft. To ensure the reliability of the model results, a 50 repeated holdout approach was implemented, and the models were compared using widely used evaluation metrics, including mean absolute error, mean squared logarithmic error, coefficient of determination, and mean magnitude of relative error. The results of the experimental analyses indicate that artificial neural networks, random forests, decision trees, and GradientBoost models demonstrated high performance on specific datasets. However, the decision tree model was identified as the most successful approach for project effort estimation based on overall evaluations. Additionally, to enhance the interpretability of the developed models and improve model transparency, this study employed the SHAP method. The SHAP analysis results revealed that certain attributes had a greater influence on the decision-making process of the models across different datasets. These findings provide valuable insights that can help project managers and software developers make more informed decisions. The results demonstrate that machine learning-based models can be effectively applied to effort estimation in software projects, and their interpretability can be significantly enhanced using explainable artificial intelligence techniques. In this context, the study is expected to serve as a valuable resource for future research in software engineering and project management
Statistical and seismotectonic analyses of the Marmara region under existing stress regime in the west of the NAFZ
No full textThe Marmara Region is an active tectonic region in northwestern Türkiye, which comprises some important strike-slip active fault mechanisms and important tectonic units, located near the western part of the North Anatolian Fault Zone. In the historical and instrumental period, the Marmara Region experienced large/devastating earthquakes. Considering this continuous activity, in this study, we investigate the tectonic structure and performed future seismic hazard estimation of the region based on some seismotectonic parameters. For this evaluation, we plot the Coulomb stress change maps of 1912 Mürefte-Şarköy, 1953 Yenice-Gönen and 1999 İzmit mainshocks with the earthquakes (MW ≥ 4.5) that occurred in the study region after 2003. For the estimation of b-value, occurrence probabilities and return periods of earthquakes, we used a homogenous local seismicity catalogue consisting of 119.029 events for the period between 1912 and 2023. In the findings of this study, the lower b-values and increasing Coulomb stress changes which are trigger stress failure compatible are observed in the west and northwest of the Marmara Sea. In contrast, the higher/moderate b-values and decreasing Coulomb stress values are observed in the east and southeast of the Marmara Sea. The results of probability assessments show that an earthquake with Mw = 6.5 may occur with a probability of 98% in the west of the Marmara Sea after 2025. As a remarkable fact, a comprehensive assessment of these types of variables will supply important findings for earthquake hazard and potential in the study region. © The Author(s) under exclusive licence to Institute of Geophysics, Polish Academy of Sciences 2024.Bogazici UniversityKandilli Observatory and Research InstituteGeneric Mapping Tools, (2019
mir-188-5p emerges as an oncomir to promote chronic myeloid leukemia via upregulation of BUB3 and SUMO2
No full textBackgroundChronic myeloid leukemia (CML) is an aggressive malignancy originating from hematopoietic stem cells. miRNAs play a role in physiological and developmental processes, including cellular proliferation, apoptosis, angiogenesis, and differentiation, and in CML's prognosis, diagnosis, and treatment. This study aimed to investigate the function and possible mechanisms of action of miR-188-5p in the development and progression of chronic myeloid leukemia.Methods and resultsmiRNA expression profiles were obtained from the GSE90773 dataset in the Gene Expression Omnibus (GEO). GEO2R was used to identify differentially expressed miRNAs. miRNET, miRDB, CancerSEA, GeneMANIA, and BioGRID databases were applied to assess the biological function of miRNA and target molecules in CML. RT-PCR performed validation analyses of miRNA and target molecules in CML. To determine the power of miR-188-5p expression levels to distinguish patients with CML from control, the ROC analysis was performed. miR-188-5p is significantly increased in K-562 cells, and overexpression of miR-188-5p was associated with clinicopathological features. miR-188-5p showed significantly higher AUC values (AUC = 1.0, p = 0.0001). The cut-off of miR-188-5p was 6.74. miRDB and mirNET predicted BUB3 and SUMO2 as a potential target gene of miR-188-5p. Additionally, increased expression of BUB3 and SUMO2 was observed in the K-562 cell. Bub3 is implicated in apoptosis and the cell cycle, whereas Sumo2 protein sumoylation and DNA binding are believed to contribute to catabolic processes.ConclusionsOur results suggest that miR-188-5p acts as an oncomiRNA in CML pathogenesis and may be a promising therapeutic target for CML
A critical and comprehensive review of the removal of thorium ions from wastewater: Advances and future perspectives
No full textThis review article critically evaluates recent developments in removing thorium (Th) ions from wastewater and discusses future perspectives. Moreover, it is imperative to develop effective and sustainable techniques for removing Th due to the significant environmental and health risks its contamination produces. This review is done to understand the recent technologies and strategies developed for the removal of Th ions from wastewater and to evaluate their efficiency, feasibility, and environmental friendliness. Removal of Th ions from wastewater is an urgent issue due to the radiotoxicity and chemical toxicity of Th. This review may facilitate a broad understanding of the advances in various methods used for the removal of thorium ions through adsorption, ion exchange, membrane technologies, and bioremediation. Again, the article tries to spotlight the problems or gray areas associated with the different techniques developed so far, thereby suggesting scopes for future research and improvement. It is hoped that this review can successfully guide the development of new and more powerful approaches to mitigating Th contamination in wastewater by synthesizing leading-edge research findings and advances for better environmental protection and public health safety. It is of critical importance to protect the environment and human health from the toxic effects of radioactive thorium mixed into life-threatening wastewater. Because a clean environment means a more livable world left to future generations
Methane Production and Struvite Recovery from Disintegrated Biosludge at Different Microwave Powers
No full textThis study examined the impact of microwave (MW) irradiation on the disintegration of waste biological sludge (WBS) and CH4 production during anaerobic digestion (AD), alongside the recovery of NH4-N and PO4-P via struvite precipitation in AD tailings. MW treatment was conducted under controlled conditions (120 degrees C and 2 degrees C/min) at the power levels of 900 W and 1,800 W. The soluble chemical oxygen demand (sCOD) increased by factors of 465 and 432 at 900 W and 1,800 W, respectively, while sugar and protein concentrations rose by factors of 840-110 and 1.93-2.15 compared to the untreated WBS. However, MW irradiation was less effective for releasing NH4-N and PO4-P. The disintegration of WBS improved CH4 production in the biochemical methane potential (BMP) test by 26% and 35% at 900 W and 1,800 W, respectively, relative to the control. Despite these enhancements, the process was deemed uneconomical due to the high energy demand of MW irradiation compared to the energy gained from the increased CH4 yield. Additionally, the sludge dewatering properties, measured as sludge filter resistance (SFR), deteriorated significantly, increasing from 2.87 x 10(14) for the untreated WBS to 6.50 x 10(14) and 6.70 x 10(14) at 900 W and 1,800 W, respectively. Kinetic modeling of the BMP tests revealed that the Transference Function provided the best fit to the experimental data. In the struvite precipitation, the optimal recovery of NH4-N and PO4-P (96%) was achieved at a molar ratio of 1.25/1/1 for Mg/N/P. [GRAPHICS] .The Research Fund of Cumhuriyet University (CUBAP) [M-2022-838]This study was supported by The Research Fund of Cumhuriyet University (CUBAP) under Grant No. M-2022-838, Sivas, Turkiye
The relationship between radial artery stenosis and whole blood viscosity after transradial coronary angiography
No full textBACKGROUND: The transradial approach (TRA) is preferred for coronary procedures due to improved outcomes and lower complication rates. However, complications such as radial artery stenosis (RAS) and occlusion (RAO) post TRA require investigation. This study aimed to explore the link between whole blood viscosity (WBV) and RAS/RAO after TRA coronary angiography. METHODS: A retrospective analysis of 215 TRA coronary angiography patients was conducted. Doppler ultrasonography assessed RAS/RAO one month post procedure. WBV, calculated from haematocrit and total plasma protein (TP), was evaluated at low (LSR) and high shear rates (HSR). RESULTS: RAS/RAO incidence was 15.3%, with 7.4% of patients experiencing RAO. Patients with RAS/RAO showed significantly elevated HSR, LSR and TP levels, with lower blood urea nitrogen levels. Multivariable analysis identified body mass index, HSR and LSR as independent RAS/RAO predictors. CONCLUSIONS: This study established WBV association with RAS/RAO after TRA, suggesting WBV as a potential predictor and aiding pre-TRA risk assessment for alternative angiography routes
Evaluation of COVID-19 Awareness and Concerns of Patients Admitted to Dentist
No full textAmaç: Bu çalışmada, Cumhuriyet Üniversitesi Diş Hekimliği Fakültesi Ağız Diş ve Çene Cerrahisi Anabilim Dalı'na başvuran erişkin hastaların Coronavirüs Hastalığı-2019 (COVID-19) pandemisi hakkındaki bilgi düzeylerinin ve diş hekimi ziyaretleri sırasında COVID-19'un bulaşmasına ilişkin endişelerinin incelenmesi amaçlanmıştır. Gereç ve Yöntem: Veriler yüz yüze görüşmeler yoluyla elde edilmiştir. Evreni belli olmayan örneklem hacmi formülüne göre (n=t 2 PQ/d 2 ), 385 kişi örnekleme alınmıştır. Araştırma verilerinin tanımlanmasında frekans ve yüzde (%) dağılımı kullanılmıştır. Açık uçlu sorulardan elde edilen veriler kategorize edilerek değerlendirilmiştir. Değişkenler pearson ki-kare analizine göre analiz edildi. Bulgular: Çalışmaya katılanların büyük bir çoğunluğu COVID-19 ile ilgili endişe duymaktadır. Bu endişe nedeniyle insanlar acil olmayan diş işlemleri için diş hekimine başvurmaktan çekinmektedirler. Sonuç: Diş hekimine başvuran hastaların kaygılarının anlaşılması ve buna yönelik önlemlerin alınması ile bireysel ve toplum diş sağlığının devamlılığı sağlanabilir. Pandemiyi bitirmek ve normal iş akışımıza dönmek için bağışıklık kazanmanın en güvenli yolu aşı olmaktır. Yaş, cinsiyet, eğitim, kronik hastalıkların varlığı ve COVID-19 endişesi de COVID-19'a karşı aşılamayı etkileyebilir.Aim: Study aimed to examine the level of knowledge among adult patients admitted to Cumhuriyet University, Faculty of Dentistry, Department of Oral and Maxillofacial Surgery, regarding the Coronavirus Disease-2019 (COVID-19) pandemic and their concerns regarding the transmission of COVID-19 during dental visits. Material and Method: The data were obtained through face to face interviews. According to the sample size formula with unknown population (n=t 2 PQ/d 2 ), 385 people were taken into the sample. Frequency and percentage (%) distribution were used to define the research data. The data obtained from the open-ended questions were categorized and evaluated. The variables were analyzed according to pearson chi-square analysis. Results: A large majority of the participants in the study are concerned about COVID-19. Because of this concern, people are hesitant to contact the dentist for non-emergency dental procedures. Conclusion: Continuity of individual and community dental health can be ensured by understanding the concerns of the patients who apply to the dentist and taking precautions accordingly. The safest way to gain immunity to end the pandemic and return to our normal workflow is by getting vaccinated. Age, gender, education, the presence of chronic diseases and COVID-19 concern can also affect vaccination against COVID-19