Mahmut Ozturk
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mahmut Ozturk has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Aydin Akan
Department of Electrical and Electronics Engineering, Izmir University of Economics, 35330 Izmir, Türkiye
5 shared publications - Gökçe Koç
Biomedical Engineering Department, Faculty of Engineering and Architecture, Istanbul Yeni Yuzyil University, Istanbul 34010, Türkiye
5 shared publications - Mosab A. A. Yousif
Department of Biomedical Engineering, Institute of Graduate Studies, İstanbul University-Cerrahpasa, İstanbul 34320, Türkiye
5 shared publications - Zaied Alhaj
Department of Biomedical Engineering, Institute of Graduate Studies, Istanbul University - Cerrahpasa, Istanbul, Türkiye
5 shared publications - Nadide Gulsah Gulenc
Department of Biomedical Engineering, Çorlu Engineering Faculty, Tekirdag Namık Kemal University, Tekirdag, Turkey
4 shared publications - Yalcin Cekic
Department of Mechatronics Engineering, Bahcesehir University, 34349 Istanbul, Türkiye
4 shared publications
Latest Publications
- 12 November 2026
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 20 (17) , 795
Comprehensive sub-frequency analysis of EEG bands for major depressive disorder diagnosis using empirical wavelet transform
- 12 November 2026
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 20 (17) , 795
Comprehensive sub-frequency analysis of EEG bands for major depressive disorder diagnosis using empirical wavelet transform
- Open Access
- 7 October 2026
- Published by Springer Nature in Discover Computing
- Vol. 29 (1) , 707
Heart sound classification using machine learning approaches based on mel-frequency cepstrum coefficients (MFCCs) and statistical features
- 7 October 2026
- Published by Springer Nature in Discover Computing
- Vol. 29 (1) , 707
Open AccessHeart sound classification using machine learning approaches based on mel-frequency cepstrum coefficients (MFCCs) and statistical features
- Open Access
- 7 August 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Patient-level topology-aware graph transformer for breast histopathology classification
- 7 August 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessPatient-level topology-aware graph transformer for breast histopathology classification
- Open Access
- 5 July 2026
- Published by MDPI AG in Diagnostics
- Vol. 16 (13) , 2107
A Comparative Analysis of EWT and EMD Techniques in the Diagnosis of Major Depressive Disorder Using EEG: Asymmetry Features and Explainability via SHAP
Background/Objectives: Major Depressive Disorder is a serious mental disorder that negatively affects an individual’s health and quality of life. The diagnosis of this disease is based on clinical interviews, questionnaires, and the patient’s self-reports. The objective of this study is to develop a biological diagnostic system based on the analysis of EEG signals and brain regions, rather than relying on self-reports. Methods: In this study, the EEG signals in the Multimodal Open Mental Disorder Analysis (MODMA) dataset were divided into six anatomical regions: prefrontal, frontal, central, parietal, temporal, and occipital. Empirical Wavelet Transform and Empirical Mode Decomposition methods were applied separately to the channels in each region, resulting in three IMF components. A total of 23 features, including statistical, nonlinear, spectral, and model-based (AR) features, were extracted from each IMF component. In addition to these features, asymmetry features between the left and right hemispheres were also included. Feature dimensions ranging from 10 to 40 were selected via the mRMR method, and the extracted feature sets were classified using SVM, k-NN, RUSBoost, Random Forest, and Meta-Ensemble machine learning models with Leave-One-Subject-Out (LOSO) validation. Results: According to the analysis results, the highest accuracy rate in Major Depressive Disorder (MDD) diagnosis was achieved by classifying features extracted from the frontal and prefrontal regions. The EMD signal processing method demonstrated superior performance compared to the EWT method. An accuracy rate of 98.11% was achieved using Random Forest and Meta-Ensemble models. Conclusions: In the proposed method, Explainable Artificial Intelligence (XAI) based SHAP analysis was applied to provide reliable and interpretable features for MDD diagnosis based on brain regional analysis.
- 5 July 2026
- Published by MDPI AG in Diagnostics
- Vol. 16 (13) , 2107
Open AccessA Comparative Analysis of EWT and EMD Techniques in the Diagnosis of Major Depressive Disorder Using EEG: Asymmetry Features and Explainability via SHAP
Background/Objectives: Major Depressive Disorder is a serious mental disorder that negatively affects an individual’s health and quality of life. The diagnosis of this disease is based on clinical interviews, questionnaires, and the patient’s self-reports. The objective of this study is to develop a biological diagnostic system based on the analysis of EEG signals and brain regions, rather than relying on self-reports. Methods: In this study, the EEG signals in the Multimodal Open Mental Disorder Analysis (MODMA) dataset were divided into six anatomical regions: prefrontal, frontal, central, parietal, temporal, and occipital. Empirical Wavelet Transform and Empirical Mode Decomposition methods were applied separately to the channels in each region, resulting in three IMF components. A total of 23 features, including statistical, nonlinear, spectral, and model-based (AR) features, were extracted from each IMF component. In addition to these features, asymmetry features between the left and right hemispheres were also included. Feature dimensions ranging from 10 to 40 were selected via the mRMR method, and the extracted feature sets were classified using SVM, k-NN, RUSBoost, Random Forest, and Meta-Ensemble machine learning models with Leave-One-Subject-Out (LOSO) validation. Results: According to the analysis results, the highest accuracy rate in Major Depressive Disorder (MDD) diagnosis was achieved by classifying features extracted from the frontal and prefrontal regions. The EMD signal processing method demonstrated superior performance compared to the EWT method. An accuracy rate of 98.11% was achieved using Random Forest and Meta-Ensemble models. Conclusions: In the proposed method, Explainable Artificial Intelligence (XAI) based SHAP analysis was applied to provide reliable and interpretable features for MDD diagnosis based on brain regional analysis.
- Open Access
- 29 June 2026
- Published by Springer Nature in Journal of King Saud University Computer and Information Sciences
- Vol. 38 (6)
CrossTaskFormer: Bidirectional cross-task attention for joint brain tumor segmentation and classification
- 29 June 2026
- Published by Springer Nature in Journal of King Saud University Computer and Information Sciences
- Vol. 38 (6)
Open AccessCrossTaskFormer: Bidirectional cross-task attention for joint brain tumor segmentation and classification
- Open Access
- 19 June 2026
- Published by Springer Nature in Journal of King Saud University Computer and Information Sciences
- Vol. 38 (6)
A unified biologically informed framework for bone marrow cell classification under ambiguity and imbalance
- 19 June 2026
- Published by Springer Nature in Journal of King Saud University Computer and Information Sciences
- Vol. 38 (6)
Open AccessA unified biologically informed framework for bone marrow cell classification under ambiguity and imbalance
- 10 March 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Neural Systems
- Vol. 36 (07) , 2650022
Motor and Cognitive Imagery Detection from MEG Signals Using Wavelet-Based Common Spatial Pattern Analysis
Brain–computer interface (BCI) technology supports the interactions of individuals with severe neuromuscular limitations with their environment. This work presents a classification approach for distinguishing motor imagery (MI) from speech-related cognitive imagery (CI) such as word generation and arithmetic subtraction, using magnetoencephalography (MEG) signals. Differentiating MI and CI/SI processes is relevant for expanding command diversity in hybrid BCI systems and for clarifying the distinct neural mechanisms underlying motor versus verbal–semantic processing. Although a large proportion of noninvasive BCI studies focus on MI, this distinction has received relatively limited attention, particularly in MEG-based approaches. Making this distinction is important for increasing command diversity in hybrid BCI systems and for improving the understanding of neural mechanisms associated with motor and verbal–semantic processing. Tasks from an open-access MEG dataset were analyzed across six binary pairs (H–F, H–W, H–S, F–W, F–S, W–S). MEG signals were processed using two frequency-separation strategies: a broad-band configuration (FSB-1: 8–14[ ]Hz and 14–30[ ]Hz) and a narrow-band configuration (FSB-2: six sub-bands between 8 and 32[ ]Hz). Time–frequency features were extracted using continuous wavelet transform (CWT), and spatial features via the common spatial pattern (CSP) method. Feature selection followed a two-stage procedure: (i) t-test ranking to obtain a shared feature set for all task pairs; and (ii) subject- and task-specific optimization of feature number. The initial evaluation based on the shared feature set showed that the FSB-2/CWT approach yielded better classification accuracies compared to FSB-1/CWT (H-F: 56%, H-W: 71%, H-S: 66% versus 54%, 68%, 64%). With subject- and task-adaptive optimization, additional improvements were observed. Accuracies increased to 60%, 72%, and 69% for FSB-1, and to 63%, 75%, and 71% for FSB-2, for H–F, H–W, and H–S, respectively. Overall, the findings indicate that the proposed CWT[ ]CSP framework, particularly when combined with adaptive feature optimization, offers an interpretable analysis approach that can contribute to MI–CI discrimination in MEG-based BCI systems under limited data conditions.
- 10 March 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Neural Systems
- Vol. 36 (07) , 2650022
Motor and Cognitive Imagery Detection from MEG Signals Using Wavelet-Based Common Spatial Pattern Analysis
Brain–computer interface (BCI) technology supports the interactions of individuals with severe neuromuscular limitations with their environment. This work presents a classification approach for distinguishing motor imagery (MI) from speech-related cognitive imagery (CI) such as word generation and arithmetic subtraction, using magnetoencephalography (MEG) signals. Differentiating MI and CI/SI processes is relevant for expanding command diversity in hybrid BCI systems and for clarifying the distinct neural mechanisms underlying motor versus verbal–semantic processing. Although a large proportion of noninvasive BCI studies focus on MI, this distinction has received relatively limited attention, particularly in MEG-based approaches. Making this distinction is important for increasing command diversity in hybrid BCI systems and for improving the understanding of neural mechanisms associated with motor and verbal–semantic processing. Tasks from an open-access MEG dataset were analyzed across six binary pairs (H–F, H–W, H–S, F–W, F–S, W–S). MEG signals were processed using two frequency-separation strategies: a broad-band configuration (FSB-1: 8–14[ ]Hz and 14–30[ ]Hz) and a narrow-band configuration (FSB-2: six sub-bands between 8 and 32[ ]Hz). Time–frequency features were extracted using continuous wavelet transform (CWT), and spatial features via the common spatial pattern (CSP) method. Feature selection followed a two-stage procedure: (i) t-test ranking to obtain a shared feature set for all task pairs; and (ii) subject- and task-specific optimization of feature number. The initial evaluation based on the shared feature set showed that the FSB-2/CWT approach yielded better classification accuracies compared to FSB-1/CWT (H-F: 56%, H-W: 71%, H-S: 66% versus 54%, 68%, 64%). With subject- and task-adaptive optimization, additional improvements were observed. Accuracies increased to 60%, 72%, and 69% for FSB-1, and to 63%, 75%, and 71% for FSB-2, for H–F, H–W, and H–S, respectively. Overall, the findings indicate that the proposed CWT[ ]CSP framework, particularly when combined with adaptive feature optimization, offers an interpretable analysis approach that can contribute to MI–CI discrimination in MEG-based BCI systems under limited data conditions.
- Open Access
- 15 January 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (1) , 99
Lean-NET-Based Local Brain Connectome Analysis for Autism Spectrum Disorder Classification
Autism spectrum disorder (ASD) is a neurodevelopmental condition characterized by impairments in social interaction and communication, along with atypical behavioral patterns. Affected individuals often seem isolated in their inner world and exhibit particular sensory reactions. The World Health Organization has indicated a persistent increase in the global prevalence of autism, with approximately 1 in 127 persons affected worldwide. This study contributes to the growing research effort by presenting a comprehensive analysis of functional connectivity patterns for ASD prediction using rs-fMRI datasets. A novel approach was used for ASD identification using the ABIDE II dataset, based on functional networks derived from BOLD signals. The sparse functional brain connectome (Lean-NET) model is employed to construct subject-specific connectomes, from which local graph metrics are extracted to quantify regional network properties. Statistically significant features are selected using Welch’s t-test, then subjected to False Discovery Rate (FDR) correction and classified using a Support Vector Machine (SVM). Our experimental results demonstrate that locally derived graph metrics effectively discriminate ASD from typically developing (TD) subjects and achieve accuracy ranging from 70% up to 91%, highlighting the potential of graph learning approaches for functional connectivity analysis and ASD characterization.
- 15 January 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (1) , 99
Open AccessLean-NET-Based Local Brain Connectome Analysis for Autism Spectrum Disorder Classification
Autism spectrum disorder (ASD) is a neurodevelopmental condition characterized by impairments in social interaction and communication, along with atypical behavioral patterns. Affected individuals often seem isolated in their inner world and exhibit particular sensory reactions. The World Health Organization has indicated a persistent increase in the global prevalence of autism, with approximately 1 in 127 persons affected worldwide. This study contributes to the growing research effort by presenting a comprehensive analysis of functional connectivity patterns for ASD prediction using rs-fMRI datasets. A novel approach was used for ASD identification using the ABIDE II dataset, based on functional networks derived from BOLD signals. The sparse functional brain connectome (Lean-NET) model is employed to construct subject-specific connectomes, from which local graph metrics are extracted to quantify regional network properties. Statistically significant features are selected using Welch’s t-test, then subjected to False Discovery Rate (FDR) correction and classified using a Support Vector Machine (SVM). Our experimental results demonstrate that locally derived graph metrics effectively discriminate ASD from typically developing (TD) subjects and achieve accuracy ranging from 70% up to 91%, highlighting the potential of graph learning approaches for functional connectivity analysis and ASD characterization.