Scopus İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/8651
Browse
719 results
Search Results
Article Beam Shaping on a Fluorescent On-Chip Imaging System(IOP Publishing Ltd, 2026) Arpali, Caglar; Arpali, Serap AltayThe fluorescent on-chip imaging system differs from a conventional fluorescent microscope in terms of the imaging method because the sample is directly placed on the imaging sensor (i.e., charge-coupled device (CCD)). While this imaging modality presents several advantages, including a wide field of view and rapid scanning speed, it can be difficult to detect certain particles in dense and scattering environments, such as whole blood and tissue. These difficulties lead to a decreased signal-to-noise ratio (SNR) in the captured images, influenced by both the medium's light-transmitting capability and the excitation techniques used. In this paper, we quantitatively examine the effect of beam shaping techniques on a fluorescent on-chip imaging system from the SNR perspective. An experimental comparison is conducted between a Gaussian beam and plane-wave illumination generated by a novel phase modulation schema using our developed imaging platform. The results indicate that the Gaussian beam produces higher SNR images than plane waves when detecting fluorescent particles in a microchannel. Gaussian beam's higher energy confinement ability enhances the image quality of on-chip fluorescent imaging systems, particularly involving scattering-like medium limitations.Article An Investigation of Discontinuities in Time-Dependent 2D and 3D Parabolic Partial Differential Equations Utilizing Collocation Methods: A Comparative Analysis of Complex Interface Problems(Springer Heidelberg, 2025) Faheem, Muhammad; Asif, Muhammad; Amin, Rohul; Haider, Nadeem; Jarad, FahdParabolic double interface problems have many applications in the fields such as materials science, fluid dynamics, and heat transfer. This paper presents a comparison of the Haar wavelet-based collocation method and two variants of radial basis function (RBF) method for solving 2D and 3D, linear as well as nonlinear, parabolic double interface problems. The two variants of RBF methods are the multiquadric RBF method and the integrated RBF method. For linear problems, the system of equations obtained from the integrated RBF method is solved using Moore-Penrose pseudoinverse. Error analysis is performed using L infinity\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_\infty $$\end{document} norm error and root mean square error, and the findings are discussed in detail. The methods are compared based on their accuracy and efficiency in solving different benchmark problems. The results show that both the Haar wavelet collocation method and the integrated RBF method perform better than the conventional RBF method in terms of accuracy.Article Numerical and Experimental Investigation of Effects of Porous Layer on Cooling of Electronic Components(American Society of Mechanical Engineers (ASME), 2026) Kocak, E.; Türkoǧlu, H.In this study, heat transfer and temperature distribution characteristics of an electronic component covered with a porous medium were investigated both experimentally and numerically. An experimental setup was designed and constructed to conduct the experiments. For the numerical analysis, a computational fluid dynamics (CFD) software was developed on the OPENFOAM platform. The experimental results were used to validate the mathematical model and the computer program developed. The validated computer program was used to investigate the effects of Reynolds number, porosity, Darcy number, porous layer sizes, and the channel height on the heat transfer rate from the heat dissipating elements (electronic component) to the flow in wider ranges of the parameters. Using the Nusselt number values obtained both experimentally and numerically, a correlation equation was developed, and an artificial neural network architecture was trained for the Nusselt number. Results show that the Nusselt number increases with increasing Reynolds number, porosity, and the ratio of the height of the porous layer to the channel height. It was observed that the width of the porous medium has no noticeable effect on the Nusselt number. The correlation equation developed with four independent parameters predicts the Nusselt number with an average error of 7.59%. The artificial neural network architecture developed prevails as a more accurate tool, with a maximum error of 1%, for the prediction of the Nusselt number in the range of the parameters considered. © © 2026 by ASME.Article Propagation Characteristics of Higher-Order Annular Gaussian Beams in Oceanic Turbulence(Institute of Physics, 2025) Arpali, S.A.; Arpali, Ç.; Baykal, Y.This study aims to explore the propagation characteristics of higher-order annular Gaussian (HOAG) beams in oceanic turbulence. We provide an analytical derivation of the average intensity at the receiver plane based on excitation from a HOAG source field. Additionally, we conduct a detailed analyses of various beam intensity moments including kurtosis parameter, power-in-the-bucket (PIB) and the beam size variation. As oceanic turbulence strength increases, the HOAG beam gradually transforms into a pure Gaussian beam. As the strength of turbulence increases, PIB values for all modes of HOAG beams gradually decrease in an exponential manner until they stabilize, exhibiting behavior similar to that of Gaussian beams. It is also observed that modes of HOAG beams having larger mode numbers carry less energy to the receiver compared to lower-order modes as turbulence strength increases. Analyses of the kurtosis parameter for HOAG beams indicate that during propagation over intermediate distances, there is a tendency for more beam energy to be distributed toward the wings rather than to the center. In contrast, at longer distances, the beam redistributes its energy, resulting in a lower energy concentration in the wings compared to the center. This research can enhance our understanding of the effects of higher-order laser beams, thereby potentially facilitating longer communication distances in underwater wireless optical communication technologies. © 2025 IOP Publishing Ltd.Article Self-Supervised Learning With BYOL for Non-Alcoholic Fatty Liver Disease Diagnosis Using Ultrasound Imaging(Springer London Ltd, 2025) Buktash, Ali; Gorur, Abdul KadirPurpose:The study aims to evaluate the effectiveness of Bootstrap Your Own Latent (BYOL), a self-supervised learning method for diagnosing NAFLD from ultrasound images using limited labeled data, which represents a novel approach in this domain. Self-supervised learning provides an alternative approach to traditional supervised learning by learning useful representations from unlabeled data, thereby reducing the time and cost required by radiologists to annotate images.Methods:The pre-trained ResNet-50 and ResNet-101 on the labeled ImageNet dataset were used for BYOL pre-training on ultrasound images without relying on labels. The training was conducted using default and custom augmentation, as well as balanced and imbalanced class distribution protocols. The model was then evaluated using linear and fine-tuning protocols with varying percentages of labeled data. The model was trained using three shuffled subsets, each trained 10 times. The custom augmentation set was derived by testing various augmentation settings using 100% and 1% of the labels to enhance feature learning.Results:BYOL with ResNet-101 and using the proposed custom augmentation set achieved average accuracies of 93.44%, 92.29%, and 88.49% using 100%, 10%, and 1% of the training labels across three shuffled datasets. In addition, our proposed method attained an average accuracy of 96.9% using patient-specific leave-one-out cross-validation (LOOCV).Conclusion:BYOL, with the proposed custom augmentation set, can learn effective image representations without relying on a large amount of labeled data, thereby enhancing scalability since unlabeled images are easier to acquire. It surpasses BYOL with default augmentation and training under supervised learning, especially with a low-labeled data regime.Article Dielectric and Thermal Properties of the Ferroelectric NH4HSO4 Close to Phase Transitions(Elsevier, 2025) Kiraci, A.; Yurtseven, H.This work gives the analysis of the dielectric and thermal properties close to the second order and first order types of ferroelectric-paraelectric phase transitions in ferroelectrics, in particular, NH4HSO4. The power-law formula is used by adapting the Kouvel-Fisher (KF) method describing the magnetization (M) and magnetic susceptibility (chi M) in the case of the spontaneous polarization (PS) and the dielectric constant (& varepsilon;), respectively, in ferroelectric systems. Similar treatment is performed to describe the heat capacity (CP) and the thermal expansivity (proportional to P) close to the phase transitions in NH4HSO4. We show that the variations of PS and & varepsilon; with the temperature near the upper Curie point (TC1 = 270 K) exhibit linearity for the second order transition in NH4HSO4. A linear variation is also obtained between the CP and proportional to P with the temperature close to the lower Curie point (TC2 = 160 K) for the first order transition in this crystal. Experimental data are used from the literature for our analysis. Our approach given here to describe dielectric and thermal properties of NH4HSO4 close to the Curie points, can be applied to some other ferroelectric materials.Article Unit Root Testing in the Presence of Mean Reverting Jumps: Evidence From US T-Bond Yields(Walter de Gruyter GmbH, 2019) Ilalan, Deniz; Ozel, OzgurMean reversion of financial data, especially interest rates is often tested by linear unit root tests. However, there are times where linear unit root test results can be misleading especially when mean reverting jump formations are at stage. Considering this framework, we provide a new unit root testing methodology and compute its asymptotic critical values via Monte Carlo simulation. Moreover, we numerically compare the power of this generalized mean reversion test with the pioneering linear unit root test in the literature namely the Augmented Dickey Fuller (ADF) test. We deduce that our test is a refinement of ADF test with a higher power. Weapply our findings to US 10-year Treasury bond yields. We aim to shed light to the discussion among researchers whether interest rates can sometimes revert to a long-term constant mean or not from an unorthodox point of view.Article Enhancing Content-Based Retrieval Through an End-to-End Approach Utilizing Deep Learning and Multidimensional Indexing(Springer London Ltd, 2025) Uzel, Omer; Arslan, SerdarRecent advancements in technology, coupled with reductions in hardware and software costs, have propelled visual search applications into the spotlight, making them both popular and indispensable. Consequently, the rapid and precise retrieval of images from vast databases through image queries has become a critical task. We introduce a novel end-to-end retrieval architecture that significantly enhances retrieval performance when compared to a baseline system that conducts database searches at the video frame level. Leveraging a pre-trained convolutional neural network model, we employ unsupervised image retrieval processes to extract and store low-level features for efficient indexing. To facilitate swift and effective access, we implement a tree-based indexing structure known as VP-Tree. This structure utilizes the extracted low-level features. To make these features compatible with our system, we employ dimension reduction techniques to represent them in a lower-dimensional space. Our experiments, conducted on three benchmark datasets, demonstrate that VP-Tree consistently outperforms k-nearest neighbor (KNN) search in terms of retrieval accuracy and efficiency. Specifically, for image data set, VP-Tree achieves a precision of 56.3903, an F1-score of 68.703, and an area under the curve (AUC) of 93.518719, all slightly surpassing KNN. Similarly, for news video data set, VP-Tree attains a precision of 38.704011, an F1-score of 55.029674, and an AUC of 64.6412, again outperforming KNN. For documentary data set, VP-Tree achieves a notable improvement with a precision of 73.511723, an F1-score of 84.734013, and an AUC of 80.981328, demonstrating superior performance over KNN. In addition to accuracy, we evaluated retrieval time across different dataset sizes. While KNN performs slightly faster on smaller datasets, VP-Tree scales significantly better as dataset size increases. For 100,000 images, VP-Tree reduces retrieval time from 79.77 to 54.34 ms, and for 200,000 images, it improves performance from 108.75 to 44.63 ms, confirming its efficiency in large-scale retrieval scenarios. These results highlight VP-Tree as a robust and scalable alternative to traditional KNN-based methods, ensuring both accuracy and efficiency in large-scale image retrieval tasks.Article Citation - WoS: 46Citation - Scopus: 48Solutions of the Fractional Davey-Stewartson Equations With Variational Iteration Method(Editura Acad Romane, 2012) Baleanu, Dumitru; Jafari, Hossain; Kadem, Abdelouahab; Yılmaz, Tuğba; Baleanu, Dumitru; Yilmaz, Tugba; Matematik; PsikolojiThis paper presents approximate analytical solutions for the fractional Davey-Stewartson equations using the Variational iteration method. The fractional derivatives are described in the Caputo sense. This method is based on the incorporation of the correction functional for the equation. The results obtained by this method have been compared with the exact solutions and show that the introduced approach is a promising tool for solving many linear and nonlinear fractional differential equations.Article Citation - WoS: 37Citation - Scopus: 38Newtonian Mechanics on Fractals Subset of Real-Line(Editura Acad Romane, 2013) Golmankhaneh, Alireza K.; Baleanu, Dumitru; Fazlollahi, Vahideh; Baleanu, Dumitru; MatematikIn this paper, we have studied the calculus on the fractals, meanwhile Newtonian mechanics on fractals subset of real-line has been suggested. Further, work and energy theorem on fractals with the examples has been explained. Finally Langevin F-alpha-Equation on fractals is derived.