Altınbaş University Institutional Repository
Not a member yet
5805 research outputs found
Sort by
Recording student attendance and recognizing their faces using deep learning
Get PDFThere are several methods available to monitor student attendance in classes, such as
biometric, radiofrequency, face recognition, and paper-based systems. However, the face
recognition-based approach has been found to be both efficient and secure. In this study, a
threshold to confidence has been implemented through Euclidean distance values to
enhance the identification process. The Local Binary Pattern Histogram (LBPH) algorithm
has been utilized for this purpose, as it has been demonstrated to be more effective than
other methods such as Eigenfaces and Fisher faces. The Haar cascade method has been
used for facial detection due to its robustness. The system's performance has been assessed
in various scenarios, including recognition rates, false-positive rates, and detecting
unknown individuals with or without a threshold. The system has demonstrated an
impressive 79% recognition rate for students, with a 24% false-positive rate, and can
identify students wearing glasses or a beard. The LBPH algorithm and Haar Cascade
method contribute to the system's exceptional performance. The recognition rate for
unregistered individuals in facial recognition technology is noteworthy even without the
use of a threshold value, sitting at a commendable 64%. Moreover, the rate of false
positives is impressively low, remaining at approximately 15% and 31%
Contribution to the implementation of the HACCP system within the orange nectar workshop
Get PDFThis thesis focuses on implementing the Hazard Analysis and Critical Control Points
(HACCP) system in orange nectar production, emphasizing the integration of prerequisite
programs and adherence to the 12-step HACCP methodology. The goal is to design a
tailored HACCP system for orange nectar production, incorporating prerequisite programs
for general food safety and following the Codex Alimentarius Commission's 12 steps.
Expected outcomes include enhanced control over critical points, improved product safety,
reduced contamination risks, compliance with regulations, and increased consumer
confidence. This research contributes practical insights for applying HACCP in the fruit
beverage industry, specifically in orange nectar production. While limited to a specific
timeframe, geographic location, and industry context, the study aims to establish a
comprehensive framework for enhancing orange nectar safety and quality
Deep fake image detection based on deep learning using a hybrid CNN-LSTM with machine learning architectures as classifier
No full textOne of the most important and difficult subjects in social communication is detecting deepfake images and videos. Deepfake techniques have developed widely, making this technology quite available and proficient enough so that there is worry about its bad application. Considering this issue, discovering fake faces is very important for ensuring security and preventing sociopolitical issues on a private and general level. Deep learning provides higher performance than typical image processing approaches when it comes to deepfake detection. This work presents construction of an artificial intelligence system, which is capable of detecting deepfake from more than one dataset. This study proposes neural network models based on deep learning using random forest (RF) and support vector machines (SVM) as classifier for deepfake detection. The use of two classifiers (RF) and (SVM) and their combination with a convolutional neural network is the first study of its kind in the field of deepfake detection in images from three open-source datasets (FaceForensics++, FaceAntiSpoofing, and iFakeFaceDB). This proposed method shows an accuracy of 96%, 87% and 52% in iFakeFaceDB, CelebA-Spoof, FaceForensics++ and respectively
Vitamin D Receptor Polymorphisms in Overweight/Obese Chronic Kidney Disease Patients on Dialysis
No full textBackground: Little is known about the possible association of vitamin D receptor (VDR) gene polymorphisms in obese patients with chronic kidney disease on dialysis (CKD-G5D). Therefore, we aimed to investigate VDR gene TaqI, ApaI, and FokI single-nucleotide polymorphisms (SNPs) in overweight/obese CKD-G5D patients. Methods: Seventy-one normal-weight and 68 overweight/obese CKD-G5D patients were included in the study. The polymerase chain reaction-restriction fragment length polymorphism method was used for genotyping. Demographic and laboratory data were obtained from the medical records of patients. Results: For all 3 SNPs, no significant association was found between normal-weight and overweight/obese patients ( P > .05). High-density lipoprotein (HDL) concentrations were lower, but triglyceride (TG) and glucose levels were higher in overweight/obese patients compared to normal-weight patients ( P < .001 for HDL and TG and P = .023 for glucose). In overweight/obese patients, individuals with the TaqI CC genotype had higher (PTH) levels than those with TC and TT genotypes (CC = 717.1 +/- 616.4, TC = 342.7 +/- 360.8, and TT = 310.2 +/- 323.4 pg/mL; P = .028). Similarly, patients with the ApaI genotype (627.3 +/- 653.0 mg/dL) had higher TG levels than those with the AA and AC genotypes (CC = 627.3 +/- 653.0, AA = 223.3 +/- 156.6, AC = 193.1 +/- 85.4; P < .001). Overweight/obese patients with the FokI TT genotype had higher glucose concentrations than those with the CC and CT genotypes (CC = 183.4 +/- 128.4 mg/dL, TT = 151.9 +/- 66.1 mg/dL, and CT = 107.6 +/- 41.9 mg/dL; P = .008). Conclusion: Our study suggests that VDR TaqI, ApaI, and FokI polymorphisms are not associated with obesity in CKD-G5D patients. However, they might increase the risk of secondary hyperparathyroidism, dyslipidemia, and hyperglycemia
Investigation of the degree of monomer conversion in dental composites through various methods: an in vitro study
No full textThe degree of monomer conversion (DC) values of three different dental composites were examined using three different methods: surface microhardness (ratio of bottom/top), Fourier-transform infrared (FT-IR), and differential scanning calorimetry (DSC). Two of the dental composites included in the study were nanohybrid (Dentsply Neo Spectra ST HV and Omnichroma), and one was a microhybrid-labeled newly marketed composite containing nanoparticles (Dentac Myra). Composite discs were prepared according to the methodology for all methods and analyzed (2 mm thickness x 5 mm diameter). Surface microhardness values were measured in Vickers Hardness Number (VHN), while FT-IR and DSC values were obtained in percentage (%). Significant differences were observed in both bottom/top surface microhardness values and DC values obtained from FT-IR. However, there was no statistical difference in the ratio of bottom/top microhardness values. Neo Spectra ST HV exhibited superior performance in both microhardness and monomer conversion compared to the other two composites. Newly marketed Myra showed values close to Omnichroma. It was found that the values obtained by the DSC method were parallel to those obtained by FT-IR. In conclusion, the structure of dental composites leads to different mechanical properties. Additionally, DSC measurements and FTIR spectra were found to be complementary techniques for characterizing monomer conversion values
Interleukin-8 (IL-8) levels in gingival crevicular fluid during root canal treatment of molar teeth with symptomatic irreversible pulpitis: impact of varying sodium hypochlorite concentrations
Get PDFBackground: The aim of this study was to evaluate the effect of the use of different NaOCl concentrations (1%, 2.5%, and 5.25%) during root canal treatment of molar teeth with symptomatic irreversible pulpitis on the change of the IL-8 level in gingival crevicular fluid (GFC).
Methods: GCF sampling was performed on experimental tooth with irreversible pulpitis before and after treatment and also contralateral healthy tooth of 54 patients. The patients were divided into three groups according to concentration of NaOCl solution (n = 18); 1%, 2.5%, and 5.25% NaOCl solution. GCF sampling from experimental teeth was repeated one week after root canal treatment. Statistical analysis was performed using Mann-Whitney U, Wilcoxon test, one-way ANOVA and Pearson correlation analysis.
Results: There was a significant correlation between IL-8 levels in GCF samples taken from teeth with pulpitis before treatment and from healthy contralateral teeth (p = .000). Furthermore, the pretreatment IL-8 level was significantly higher than the posttreatment IL-8 level(p .05).
Conclusions: The use of NaOCl during root canal treatment can effectively reduce the levels of IL-8 in GCF and improve clinical outcomes.
Trial registration: This study was registred in the Institutional Review Board and the Ethics Committee of the University (No:11) on 15/12/2021
The relationship of erectile dysfunction severity with nocturnal blood pressure pattern and RDW
No full textAims: The study aimed to investigate the relationship between the severity of erectile dysfunction (ED), nocturnal blood pressure patterns, and red blood cell width distribution (RDW) in hypertensive patients. Methods: The study involved 106 hypertensive patients, categorized into non-dippers and dippers based on their nocturnal blood pressure patterns. Key parameters including smoking status, RDW, and International Index of Erectile Function (IIEF) scores, were compared between the groups. Results: The demographic data of the patients were similar. RDW was significantly higher in patients with non-dipper hypertension (HT) compared to the dipper group. Moderate and severe ED was seen more frequently in the non-dipper HT group (40.4% vs 20.4%; p=0.025). IIEF score was higher in the dipper HT group (17.6±6.9 vs 21.0±4.5; p=0.004). According to logistic regression analysis, age and smoking habit were significant predictors for moderate or severe ED. Conclusion: The study highlights the significant association between non-dipping blood pressure patterns, elevated RDW, and the severity of ED in hypertensive patients. The findings underscore the importance of monitoring nocturnal blood pressure patterns and RDW in understanding and managing ED in this population
FSI analysis and simulation of a fixed-wing UAV using composite materials
Get PDFThis thesis explores the design and simulation of fixed-wing drones, specifically the Boeing
MQ-25 Stingray, with a focus on aerodynamics, Computational Fluid Dynamics (CFD), and
Fluid-Structure Interaction (FSI) analysis. A unique model of the Boeing MQ-25 Stingray is
created using SOLIDWORKS software and is to be analysed using ANSYS Fluent,
showcasing the integration of different software tools in one project. A significant portion
of the study is dedicated to exploring the use of composite materials, especially in the wings
of the aircraft model. The comparison between traditional materials and composite materials
aims to show a clear improvement in both structural and aerodynamic properties when
composite materials are used. The findings are expected to highlight the importance of
material selection in drone design and the superior performance of composite materials. The
methods used in this thesis aim to address the challenges related to design, simulation, and
material selection. The results are anticipated to provide a strong foundation for further
exploration into choosing the right materials to achieve better aerodynamic and structural
performance in aerospace applications
Deep learning-based predictive model for cement strength in building construction
No full textThis research work presents the development and evaluation of machine learning (ML) and
deep learning (DL) models for predicting the CS in building construction. The data used in
this study were obtained from Kaggle website, and several features were considered as inputs
to the models. The ML models employed in this study include Support Vector Machine
(SVM), Random Forest, and XGBoost, while the DL models are Convolutional Neural
Network (CNN) and Artificial Neural Network (ANN). The achievement of the models was
assessed using various measures such as RMSE, R², MSE, MAE, EV, and MAPE. The
findings showed that the Random Forest and XGBoost models conducted better than the
SVM model in terms of accuracy (ACC), while the CNN and ANN models had inferior
performance compared to the other models
Serum vitamin D, hemoglobin A1c and vitamin B12 levels in patients with gingivitis and periodontitis stages
No full textAim: To compare the serum vitamin D, hemoglobin A1c (HbA1c) and vitamin B12 levels in patients with gingivitis and four different periodontitis stages diagnosed according to the 2017 Periodontal Disease Classification. Materials & methods: A total of 606 patients were included in the study who were diagnosed with gingivitis and stage I–IV periodontitis. Patients were divided into groups based on disease stage, and the HbA1c, vitamin D and B12 levels of the patients were compared and analyzed. Result: The highest HbA1c level and the lowest vitamin D level were seen in stage III–IV periodontitis. The highest vitamin D and B12 levels were seen in the gingivitis group. Conclusion: Serum HbA1c, vitamin D and B12 levels might vary depending on the presence or severity of periodontitis. Clinical Trial Registration: NCT05745779 (This study was registered and approved by www.clinicaltrials.gov)