Sürdürülebilir Kalkınma Amaçları
Yayınlar
- MakaleAçık Erişim2025Scopus SJRQ1Deep learning and optimization-based feature selection for fetal health classification using CTG data
Kaya, Turgay; Kaya, Duygu; Atar, Fatmanur
This study introduces a DL and metaheuristic optimization-based framework for fetal health assessment using cardiotocography (CTG) signals to mitigate maternal and neonatal mortality. One-dimensional CTG signals were transformed into 2D representations, and deep feature extraction was performed using AlexNet. Feature vectors FC6, FC7, and their combination were subjected to optimization via Whale Optimization Algorithm (WOA + DL) and War Strategy Optimization (WSO + DL), utilizing updated fitness functions tailored for feature selection. Experimental results with SVM classifiers demonstrated superior performance with FC6 (89.98 %) and WSO + DL (90.17 %). FC6 exhibited strong discriminative capacity, while FC7 contained semantically richer features. The concatenated FC6 + FC7 vector increased feature diversity. WSO + DL achieved optimal balance across classification accuracy, feature subset size, convergence rate, and overall performance metrics. The integration of DL and metaheuristic algorithms effectively isolated informative feature subsets, improving training efficiency, minimizing redundant/noisy data, reducing overfitting risk, and enhancing classification accuracy. Optimization method selection proved critical to overall model performance.
AtıfScopus6WoS1OpenAlex6
İncele - MakaleAçık Erişim2025Scopus SJRQ1New comparative approach to multi-level thresholding: chaotically initialized adaptive meta-heuristic optimization methods
Serbet, Fatmanur; Kaya, Turgay
One method aimed at enhancing the performance of meta-heuristic optimization techniques is the incorporation of chaotic systems. Instead of irregular distributions in the search space, chaotic distributions are employed in the initial population of optimization algorithms to improve the efficiency of the search process. This approach enables search agents distributed in a chaotic manner to effectively explore the search space. The initial populations of both the well-established PSO algorithm and the enhanced WSO algorithm, which incorporates advanced search techniques, are distributed in the search space according to the characteristics of Logistic, Chebyshev, Circle, Sine, and Piecewise chaotic maps in this study. The original PSO and WSO algorithms, as well as the resulting chaotically initialized PSO and chaotically initialized WSO algorithms, were tested using 23 benchmark functions. Subsequently, the Otsu method was integrated into the tested optimization algorithms to obtain multi-level thresholding values. These algorithms were applied to five different test images with a manually determined number of thresholds. The results obtained were presented in the study and evaluated using statistical tests.
AtıfScopus13OpenAlex12İncele - MakaleYayıncıda açık erişim2023Scopus SJRQ3Optimization Approach in Window Function Design for Real-Time Filter Applications
Serbet, Fatmanur; Kaya, Turgay
Eliminating the Gibbs oscillations that occur during the Finite Impulse Response (FIR) digital filter design with the Fourier Series method will ensure correct filtering. For this reason, the development of the window improves the performance of the filter and, therefore, the system. In this study, the cosh window function is designed using Particle Swarm Optimization, which is a preferred optimization method in many areas. Thus, alternatives to the standard results obtained from the existing traditional calculations will be produced, and different windows that perform the same function will be obtained. In addition, exponential and cosh window functions were designed in LabVIEW environment, which is a graphical programming language-based program, and the designed windows were analyzed at different parameter values. LabVIEW provides a fast and easy programming environment, and it provides the opportunity to realize real-Time applications with its external hardware. Utilizing this feature, the amplitude spectrum of cosh window designed in LabVIEW is displayed in real time for different window parameter values. As a result, FIR digital filters were designed using cosh window based on optimization and the cosh window designed in LabVIEW, and the distorted EEG signal was filtered using these filters and displayed in real time.
AtıfScopus8WoS5OpenAlex5İncele - MakaleAçık Erişim2023Scopus SJRQ3Statistical Analysis and EEG Signal Filtering Using Design of Window Function Based on Optimization Methods
Serbet, Fatmanur; Kaya, Turgay
The window functions, which are widely used in Finite Impulse Response (FIR) digital filter design with the Fourier Series Methods, aim to eliminate the oscillations adversely affecting the filter performance that occur during the filter design. Window functions, which were developed with different methods in the literature, were designed with three different optimization techniques in this study. By using Particle Swarm Optimization (PSO), Grey Wolf Colony Optimization (GWCO), Cuckoo Search Optimization (CSO) algorithms, which are among current metaheuristic optimization methods, new window functions are designed for FIR digital filter design such that the designed new window functions have different window coefficients and design parameters. The difference of the designed window functions was analyzed with the Friedman and Wilcoxon tests, which are statistical data analysis methods, and the originality of the designed window functions was proven. FIR digital filters that can perform the same operation as the designed window functions have been produced and the results obtained by performing the filtering application of the EEG signal with the produced filters are presented in the study.
AtıfScopus9WoS6OpenAlex7İncele