Ayça Kurnaz Türkben
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ayça Kurnaz Türkben has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Sefer Kurnaz
Altinbas University
2 shared publications - Awatef Bshir M Alzririg
Electrical and Computer Engineering, Altınbaş University, Istanbul, Turkey
1 shared publication - Awatef Bshir M. Alzririg
Altinbas University
1 shared publication - Davron Aslonqulovich Juraev
Scientific Research Center, Baku Engineering University, Baku, Azerbaijan
1 shared publication - Ebrahim Eldesoky Elsayed
Department of Electronics and Communications Engineering, Faculty of Engineering, Mansoura University, Mansoura 35516, Egypt
1 shared publication - Maan Muataz Abdulwahid
Institute of Graduate Studies, Altinbas University, Istanbul, Türkiye
1 shared publication
Latest Publications
- Open Access
- 9 February 2026
- Published by Springer Nature in Journal of King Saud University Computer and Information Sciences
- Vol. 38 (3)
Optimized intrusion detection using bees algorithm enhanced deep neural networks with perfect ROC separability
- 9 February 2026
- Published by Springer Nature in Journal of King Saud University Computer and Information Sciences
- Vol. 38 (3)
Open AccessOptimized intrusion detection using bees algorithm enhanced deep neural networks with perfect ROC separability
- 22 January 2026
- Published by Springer Nature
- p. 239-255
Intrusion Detection Using TabNet with Recursive Feature Elimination and Deep Supervision
- 22 January 2026
- Published by Springer Nature
- p. 239-255
Intrusion Detection Using TabNet with Recursive Feature Elimination and Deep Supervision
- Open Access
- 1 January 2026
- Published by Springer Nature in Research Square
Bio-Inspired Leafy Seadragon Optimization for Automated Segmentation of Skin Cancer in Dermoscopy
Accurate segmentation of skin lesions in dermoscopic images is essential for early melanoma detection and reliable computer-aided diagnosis. This paper proposes a novel bio-inspired segmentation algorithm, the Leafy Seadragon Optimization (LSDO) method, which models the drifting locomotion and appendage-based micro-refinement behavior of the leafy seadragon to achieve robust boundary localization. LSDO incorporates a dual-phase search strategy combining global drifting with fine local adjustments, guided by a boundary-aware fitness function that enhances lesion–background separation. The algorithm was evaluated on the ISIC 2018 and HAM10000 datasets and demonstrated superior performance compared with six strong baselines: HFOA, GJO, FA, Pyramid U-Net, Attention-based CNN, and YOLOSAMIC. On ISIC 2018, LSDO achieved a Dice score of 0.94, IoU of 0.88, Sensitivity of 0.95, Specificity of 0.98, Accuracy of 0.96, and BF-score of 0.92. On the more challenging HAM10000 dataset, the method maintained high performance with a Dice of 0.90, IoU of 0.82, Sensitivity of 0.93, and Accuracy of 0.94. Qualitative comparisons further revealed that LSDO produces cleaner, more complete lesion masks and superior boundary adherence. These results confirm that the proposed LSDO algorithm is a powerful, stable, and generalizable segmentation tool for dermoscopic image analysis.
- 1 January 2026
- Published by Springer Nature in Research Square
Open AccessBio-Inspired Leafy Seadragon Optimization for Automated Segmentation of Skin Cancer in Dermoscopy
Accurate segmentation of skin lesions in dermoscopic images is essential for early melanoma detection and reliable computer-aided diagnosis. This paper proposes a novel bio-inspired segmentation algorithm, the Leafy Seadragon Optimization (LSDO) method, which models the drifting locomotion and appendage-based micro-refinement behavior of the leafy seadragon to achieve robust boundary localization. LSDO incorporates a dual-phase search strategy combining global drifting with fine local adjustments, guided by a boundary-aware fitness function that enhances lesion–background separation. The algorithm was evaluated on the ISIC 2018 and HAM10000 datasets and demonstrated superior performance compared with six strong baselines: HFOA, GJO, FA, Pyramid U-Net, Attention-based CNN, and YOLOSAMIC. On ISIC 2018, LSDO achieved a Dice score of 0.94, IoU of 0.88, Sensitivity of 0.95, Specificity of 0.98, Accuracy of 0.96, and BF-score of 0.92. On the more challenging HAM10000 dataset, the method maintained high performance with a Dice of 0.90, IoU of 0.82, Sensitivity of 0.93, and Accuracy of 0.94. Qualitative comparisons further revealed that LSDO produces cleaner, more complete lesion masks and superior boundary adherence. These results confirm that the proposed LSDO algorithm is a powerful, stable, and generalizable segmentation tool for dermoscopic image analysis.
- Open Access
- 28 October 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (21) , 11500
Application of Graph Neural Networks to Model Stem Cell Donor–Recipient Compatibility in the Detection and Classification of Leukemia
Stem cell transplants are a common treatment for leukemia, and close donor–recipient matching improves their success. Machine learning models like support vector machine (SVM), convolutional neural networks (CNNs), and recurrent neural networks (RNNs) can have difficulty handling the complexity of genomic and immune data, which then lowers the accuracy of clinical predictions. This study looks at using graph neural networks (GNNs) in a different way. This method combines data such as single-nucleotide polymorphisms (SNPs), human leukocyte antigen (HLA) typing, and clinical details to create a graph that shows the relationship between donor and recipient pairs. The framework uses graph attention networks (GATs) to focus on key compatibility traits and Dynamic GNNs (DGNNs) to monitor changes in the immune system and the disease’s progression. With data from the 1000 Genomes Project, the model correctly identified matches with 97.68% to 99.74% accuracy and classified them with 98.76% to 99.4% accuracy, outperforming standard machine learning models. The model uses SNP similarity and HLA mismatches to assess compatibility, which enhances its match prediction and compatibility explanation capabilities. The results suggest that GNNs offer a helpful and understandable way to model donor–recipient matching, potentially assisting in early leukemia detection and personalized stem cell transplant plans.
- 28 October 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (21) , 11500
Open AccessApplication of Graph Neural Networks to Model Stem Cell Donor–Recipient Compatibility in the Detection and Classification of Leukemia
Stem cell transplants are a common treatment for leukemia, and close donor–recipient matching improves their success. Machine learning models like support vector machine (SVM), convolutional neural networks (CNNs), and recurrent neural networks (RNNs) can have difficulty handling the complexity of genomic and immune data, which then lowers the accuracy of clinical predictions. This study looks at using graph neural networks (GNNs) in a different way. This method combines data such as single-nucleotide polymorphisms (SNPs), human leukocyte antigen (HLA) typing, and clinical details to create a graph that shows the relationship between donor and recipient pairs. The framework uses graph attention networks (GATs) to focus on key compatibility traits and Dynamic GNNs (DGNNs) to monitor changes in the immune system and the disease’s progression. With data from the 1000 Genomes Project, the model correctly identified matches with 97.68% to 99.74% accuracy and classified them with 98.76% to 99.4% accuracy, outperforming standard machine learning models. The model uses SNP similarity and HLA mismatches to assess compatibility, which enhances its match prediction and compatibility explanation capabilities. The results suggest that GNNs offer a helpful and understandable way to model donor–recipient matching, potentially assisting in early leukemia detection and personalized stem cell transplant plans.
- Open Access
A Hybrid Tree Convolutional Neural Network with Leader-Guided Spiral Optimization for Detecting Symmetric Patterns in Network Anomalies
In the realm of cybersecurity, detecting Distributed Denial of Service (DDoS) attacks with high accuracy is a critical task. Traditional machine learning models often fall short in handling the complexity and high dimensionality of network traffic data. This study proposes a hybrid framework leveraging symmetry in feature distribution, network behavior, and model optimization for anomaly detection. A Tree Convolutional Neural Network (Tree-CNN) captures hierarchical symmetrical dependencies, while a deep autoencoder preserves latent symmetrical structures, reducing noise for better classification. A Leader-Guided Velocity-Based Spiral Optimization Algorithm is proposed to optimize the parameters of the system and achieve better performance. A Leader-Guided Velocity-Based Spiral Optimization Algorithm is introduced to maintain a symmetrical balance between exploration and exploitation, optimizing the autoencoder, Tree-CNN, and classification thresholds. Validation using three datasets—UNSW-NB15, CIC-IDS 2017, and CIC-IDS 2018—demonstrates the framework’s superiority. The model achieves 96.02% accuracy on UNSW-NB15, 99.99% on CIC-IDS 2017, and 99.96% on CIC-IDS 2018, with near-perfect precision and recall. Despite a slightly higher computational cost, the symmetrically optimized framework ensures high efficiency and superior detection, making it ideal for real-time complex networks. These findings emphasize the critical role of symmetrical network patterns and feature selection strategies for enhancing intrusion detection performance.
Open AccessA Hybrid Tree Convolutional Neural Network with Leader-Guided Spiral Optimization for Detecting Symmetric Patterns in Network Anomalies
In the realm of cybersecurity, detecting Distributed Denial of Service (DDoS) attacks with high accuracy is a critical task. Traditional machine learning models often fall short in handling the complexity and high dimensionality of network traffic data. This study proposes a hybrid framework leveraging symmetry in feature distribution, network behavior, and model optimization for anomaly detection. A Tree Convolutional Neural Network (Tree-CNN) captures hierarchical symmetrical dependencies, while a deep autoencoder preserves latent symmetrical structures, reducing noise for better classification. A Leader-Guided Velocity-Based Spiral Optimization Algorithm is proposed to optimize the parameters of the system and achieve better performance. A Leader-Guided Velocity-Based Spiral Optimization Algorithm is introduced to maintain a symmetrical balance between exploration and exploitation, optimizing the autoencoder, Tree-CNN, and classification thresholds. Validation using three datasets—UNSW-NB15, CIC-IDS 2017, and CIC-IDS 2018—demonstrates the framework’s superiority. The model achieves 96.02% accuracy on UNSW-NB15, 99.99% on CIC-IDS 2017, and 99.96% on CIC-IDS 2018, with near-perfect precision and recall. Despite a slightly higher computational cost, the symmetrically optimized framework ensures high efficiency and superior detection, making it ideal for real-time complex networks. These findings emphasize the critical role of symmetrical network patterns and feature selection strategies for enhancing intrusion detection performance.
- 31 January 2025
- Published by Springer Nature in Journal of Optics
- Vol. 55 (4) , 1-16
Performance analysis of input power variations in high data rate DWDM-FSO systems under various rain conditions
- 31 January 2025
- Published by Springer Nature in Journal of Optics
- Vol. 55 (4) , 1-16
Performance analysis of input power variations in high data rate DWDM-FSO systems under various rain conditions