Scopus İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/8651
Browse
6 results
Search Results
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 Low Signature UAVs: Radar Cross Section Analysis, Simulation, and Measurement in X-Band(Springer London Ltd, 2025) Unalir, Dizdar; Yalcinkaya, Bengisu; Aydin, ElifThe increasing prevalence of unmanned aerial vehicles (UAVs) is driving the development of radar systems capable of detecting them. This hampers the deployment of UAVs in military operations. While radar cross section reduction (RCSR) can be a valuable solution, the research on this subject is inadequate. This paper presents an RCSR approach adopting a shaping technique for UAVs, demonstrating the proposed approach's efficacy through simulations and actual experimental measurements performed in X-Band on a four-legged UAV model. Using electromagnetic computational instruments, the shaping is applied to the designed UAV model with parameter-based simulations, the simulated radar cross section (RCS) values are derived, and the comparative analysis of these instruments is conducted. Experimental measurements are performed in laboratory conditions using a vector network analyzer. Actual measurement results are validated by simulative findings with the examination of the influence of frequency, polarization, and aspect angle on RCS. The demonstrated measuring approach allows cost-effective and easily applicable research on RCS in X-Band, a commonly utilized frequency range in military. An average RCSR of 10 dBsm has been accomplished with the presented shaping approach.Article Citation - WoS: 11Citation - Scopus: 15Nimrad: Novel Technique for Respiratory Data Treatment(Springer London Ltd, 2014) Nigmatullin, R. R.; Ionescu, C.; Baleanu, D.This paper illustrates the efficiency and simplicity of a new technique which is determined in this paper as NIMRAD (the non-invasive methods of the reduced analysis of data) for describing information extracted from biological signals. As a specific example, we consider the respiratory data. The NIMRAD can be applied for quantitative description of data recorded for complex systems in cases where the adequate model is absent and the treatment procedure should not contain any uncontrollable error. The theoretical developments are applied to signals measured from the respiratory system by means of the forced oscillation technique based on non-invasive lung function test. In order to verify the feasibility of the proposed algorithm for developing new diagnosis tools, we apply NIMRAD on two different respiratory data sets, namely from a healthy subject and from a patient diagnosed with asthma. The results are promising and suggest that NIMRAD could be further tailored and used for specific clinical applications.Article Citation - WoS: 4Citation - Scopus: 6Analysis of Uv Spectral Bands Using Multidimensional Scaling(Springer London Ltd, 2015) Dinc, Erdal; Baleanu, Dumitru; Tenreiro Machado, J. A.; Machado, J. A. TenreiroThis study describes the change of the ultraviolet spectral bands starting from 0.1 to 5.0 nm slit width in the spectral range of 200-400 nm. The analysis of the spectral bands is carried out by using the multidimensional scaling (MDS) approach to reach the latent spectral background. This approach indicates that 0.1 nm slit width gives higher-order noise together with better spectral details. Thus, 5.0 nm slit width possesses the higher peak amplitude and lower-order noise together with poor spectral details. In the above-mentioned conditions, the main problem is to find the relationship between the spectral band properties and the slit width. For this aim, the MDS tool is to used recognize the hidden information of the ultraviolet spectra of sildenafil citrate by using a Shimadzu UV-VIS 2550, which is in the world the best double monochromator instrument. In this study, the proposed mathematical approach gives the rich findings for the efficient use of the spectrophotometer in the qualitative and quantitative studies.Article Citation - WoS: 3Citation - Scopus: 4Application of Continuous Wavelet Transform To the Analysis of the Modulus of the Fractional Fourier Transform Bands for Resolving Two Component Mixture(Springer London Ltd, 2015) Duarte, Fernando B.; Machado, J. A. Tenreiro; Baleanu, Dumitru; Dinc, ErdalIn this paper, the fractional Fourier transform (FrFT) is applied to the spectral bands of two component mixture containing oxfendazole and oxyclozanide to provide the multicomponent quantitative prediction of the related substances. With this aim in mind, the modulus of FrFT spectral bands are processed by the continuous Mexican Hat family of wavelets, being denoted by MEXH-CWT-MOFrFT. Four modulus sets are obtained for the parameter of the FrFT going from 0.6 up to 0.9 in order to compare their effects upon the spectral and quantitative resolutions. Four linear regression plots for each substance were obtained by measuring the MEXH-CWT-MOFrFT amplitudes in the application of the MEXH family to the modulus of the FrFT. This new combined powerful tool is validated by analyzing the artificial samples of the related drugs, and it is applied to the quality control of the commercial veterinary samples.Article Citation - WoS: 8Citation - Scopus: 10Sparse Coding of Hyperspectral Imagery Using Online Learning(Springer London Ltd, 2015) Toreyin, Behcet Ugur; Ulku, IremSparse coding ensures to express the data in terms of a few nonzero dictionary elements. Since the data size is large for hyperspectral imagery, it is reasonable to use sparse coding for compression of hyperspectral images. In this paper, a hyperspectral image compression method is proposed using a discriminative online learning-based sparse coding algorithm. Compression and anomaly detection tests are performed on hyperspectral images from the AVIRIS dataset. Comparative rate-distortion analyses indicate that the proposed method is superior to the state-of-the-art hyperspectral compression techniques.