Seral Özşen
Research Subjects
Seral Özşen has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Enes Efe
Department of Electrical and Electronics Engineering, Hitit University, Corum, Turkey
3 shared publications - Hülya Vatansev
Department of Pulmonary Medicine, Necmettin Erbakan University, Selçuklu, Türkiye
3 shared publications Gülay Tezel
2 shared publications- Gülay Tezel
Konya Teknik Üniversitesi
2 shared publications - Güzin Özmen
Selcuk University
2 shared publications Serkan Küçüktürk
Karamanoglu Mehmetbey University
2 shared publications
Latest Publications
- 14 October 2026
- Published by Journal of the Faculty of Engineering and Architecture of Gazi University in Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi
- No. Advanced O
Spektral analiz görsellerinde ölçekleme etkisi: Derin öğrenmede gizli bir parametre
Derin öğrenme günümüzde birçok alanda kullanılmakta ve gelecek için önemli bir vizyon oluşturmaktadır. Bundan kaynaklı olarak modelin güvenilir ve yüksek doğrulukta sınıflama yapabilmesi oldukça önemlidir. Yapılan çalışmalar, model ve analiz yöntemlerinin parametrelerine odaklanmıştır. Fakat spektral analiz ile elde edilen verilerde skala aralığı da modelin performansında önemli bir etkiye sahiptir. Çalışmamızda MAEEG ve LSEEG veri setinde dinamik ve sabit skala yöntemi kullanılarak görsel veriler oluşturulmuştur. MAEEG veri setinde analizler özne-bağımlı, LSEEG veri setinde ise özne-bağımsız gerçekleştirilmiştir. Dinamik skala yönteminde MAEEG ve LSEEG veri setlerinde sırasıyla %70,35 ve %59,61 sınıflama doğruluğu elde edilmiştir. Aynı veri setlerinde önerilen sabit skala yöntemi kullanıldığında sırasıyla doğruluk %87,31 ve %89,41 oranına çıkmıştır. Bu durum skalanın da derin öğrenme modelleri için oldukça önemli bir parametre olduğunu göstermektedir. Kısa Zamanlı Fourier Analizi için skala belirleme yöntemi önerilmiştir.
- 14 October 2026
- Published by Journal of the Faculty of Engineering and Architecture of Gazi University in Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi
- No. Advanced O
Spektral analiz görsellerinde ölçekleme etkisi: Derin öğrenmede gizli bir parametre
Derin öğrenme günümüzde birçok alanda kullanılmakta ve gelecek için önemli bir vizyon oluşturmaktadır. Bundan kaynaklı olarak modelin güvenilir ve yüksek doğrulukta sınıflama yapabilmesi oldukça önemlidir. Yapılan çalışmalar, model ve analiz yöntemlerinin parametrelerine odaklanmıştır. Fakat spektral analiz ile elde edilen verilerde skala aralığı da modelin performansında önemli bir etkiye sahiptir. Çalışmamızda MAEEG ve LSEEG veri setinde dinamik ve sabit skala yöntemi kullanılarak görsel veriler oluşturulmuştur. MAEEG veri setinde analizler özne-bağımlı, LSEEG veri setinde ise özne-bağımsız gerçekleştirilmiştir. Dinamik skala yönteminde MAEEG ve LSEEG veri setlerinde sırasıyla %70,35 ve %59,61 sınıflama doğruluğu elde edilmiştir. Aynı veri setlerinde önerilen sabit skala yöntemi kullanıldığında sırasıyla doğruluk %87,31 ve %89,41 oranına çıkmıştır. Bu durum skalanın da derin öğrenme modelleri için oldukça önemli bir parametre olduğunu göstermektedir. Kısa Zamanlı Fourier Analizi için skala belirleme yöntemi önerilmiştir.
- 24 June 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (4) , 11805-11829
Machine learning based detection of depression from task-based fMRI using weighted-3D-DWT denoising method
- 24 June 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (4) , 11805-11829
Machine learning based detection of depression from task-based fMRI using weighted-3D-DWT denoising method
- Open Access
- 21 June 2023
- Published by Konya Muhendislik Bilimleri Dergisi in Konya Journal of Engineering Sciences
- Vol. 11 (3) , 783-800
DETERMINING THE MOST POWERFUL FEATURES IN THE DESIGN OF AN AUTOMATIC SLEEP STAGING SYSTEM
Spending too much time on manual sleep staging is tiring and challenging for sleep specialists. In addition, experience in sleep staging also creates different decisions for sleep experts. The search for finding an effective automatic sleep staging system has been accelerated in the last few years. There are many studies dealing with this problem but very few of them were conducted with real sleep data. Studies have been carried out on mostly processed and cleaned-ready data sets. In addition, there are few studies in which the data distribution in sleep stages is balanced (equal numbers of epochs from each stage are used), and it is seen that the performance of these studies is quite low compared to other studies. When the literature studies are examined, there is a wide range of studies in which many features are extracted, many feature selection methods are used, many classifiers are applied and various combinations of these are available. For this reason, to determine the best-performing features and the most powerful features, 168 features were extracted from the real EEG, EOG, and EMG signals of 124 patients. These features were selected with 7 different feature selection methods, and classification was carried out with 4 classifiers. In general, the ReliefF feature selection method has performed best, and the Bagged Tree classifier has reached the highest classification accuracy of 67.92% with the use of nonlinear features.
- 21 June 2023
- Published by Konya Muhendislik Bilimleri Dergisi in Konya Journal of Engineering Sciences
- Vol. 11 (3) , 783-800
Open AccessDETERMINING THE MOST POWERFUL FEATURES IN THE DESIGN OF AN AUTOMATIC SLEEP STAGING SYSTEM
Spending too much time on manual sleep staging is tiring and challenging for sleep specialists. In addition, experience in sleep staging also creates different decisions for sleep experts. The search for finding an effective automatic sleep staging system has been accelerated in the last few years. There are many studies dealing with this problem but very few of them were conducted with real sleep data. Studies have been carried out on mostly processed and cleaned-ready data sets. In addition, there are few studies in which the data distribution in sleep stages is balanced (equal numbers of epochs from each stage are used), and it is seen that the performance of these studies is quite low compared to other studies. When the literature studies are examined, there is a wide range of studies in which many features are extracted, many feature selection methods are used, many classifiers are applied and various combinations of these are available. For this reason, to determine the best-performing features and the most powerful features, 168 features were extracted from the real EEG, EOG, and EMG signals of 124 patients. These features were selected with 7 different feature selection methods, and classification was carried out with 4 classifiers. In general, the ReliefF feature selection method has performed best, and the Bagged Tree classifier has reached the highest classification accuracy of 67.92% with the use of nonlinear features.
- 31 May 2023
- Published by Trans Tech Publications, Ltd. in Journal of Biomimetics, Biomaterials and Biomedical Engineering
- Vol. 60, 119-133
Automatic Sleep Stage Classification for the Obstructive Sleep Apnea Patients with Feature Mining
Automatic sleep scoring systems have being much more attention in last decades. Whereas a wide variety of studies have been used in this subject area, the accuracies are still under acceptable limits to apply these methods in real life data. One can find many high accuracy studies in literature using standard database but when it comes to the using real data reaching such a high performances is not straightforward. In this study, five distinct datasets were prepared using 124 persons including 93 unhealthy and 31 healthy persons. These datasets consist of time-, nonlinear-, welch-, discrete wavelet transform-and Hilbert-Huang transform-features. By applying k-NN, Decision Trees, ANN, SVM and Bagged Tree classifiers to these feature sets in various manners by using feature-selection highest classification accuracy was searched. The maximum classification accuracy was detected in case of Bagged Tree classifier as 95.06% with the use of 14 features among a total of 136 features. This accuracy is relatively high compared with literature for a real-data application.
- 31 May 2023
- Published by Trans Tech Publications, Ltd. in Journal of Biomimetics, Biomaterials and Biomedical Engineering
- Vol. 60, 119-133
Automatic Sleep Stage Classification for the Obstructive Sleep Apnea Patients with Feature Mining
Automatic sleep scoring systems have being much more attention in last decades. Whereas a wide variety of studies have been used in this subject area, the accuracies are still under acceptable limits to apply these methods in real life data. One can find many high accuracy studies in literature using standard database but when it comes to the using real data reaching such a high performances is not straightforward. In this study, five distinct datasets were prepared using 124 persons including 93 unhealthy and 31 healthy persons. These datasets consist of time-, nonlinear-, welch-, discrete wavelet transform-and Hilbert-Huang transform-features. By applying k-NN, Decision Trees, ANN, SVM and Bagged Tree classifiers to these feature sets in various manners by using feature-selection highest classification accuracy was searched. The maximum classification accuracy was detected in case of Bagged Tree classifier as 95.06% with the use of 14 features among a total of 136 features. This accuracy is relatively high compared with literature for a real-data application.
- 20 October 2022
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 80
CoSleepNet: Automated sleep staging using a hybrid CNN-LSTM network on imbalanced EEG-EOG datasets
- 20 October 2022
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 80
CoSleepNet: Automated sleep staging using a hybrid CNN-LSTM network on imbalanced EEG-EOG datasets
- Open Access
- 7 June 2022
- Published by Orclever Science and Research Group in The European Journal of Research and Development
- Vol. 2 (2) , 13-22
EYE: A New Method for Detection of Electrode Disconnection in Sleep Signals
Biological signals that occur during sleep are recorded and classified by specialists. This process is called sleep staging. However, this is a very long and laborious process. Therefore, automatic sleep staging systems are needed. Nevertheless, automatic sleep staging studies to date have not provided satisfactory performance. The main reasons for this are inter-channel interference, electrode disconnection, and noise. In this paper, a new method (eye method) based on the Euclidean distance measurement method has been developed to solve the electrode disconnection or non-contact problem. This method was applied to three different datasets and detected all electrode disconnections with 100% accuracy. Thanks to this advanced method are aimed to increase the success of automatic sleep staging systems to be designed in the future.
- 7 June 2022
- Published by Orclever Science and Research Group in The European Journal of Research and Development
- Vol. 2 (2) , 13-22
Open AccessEYE: A New Method for Detection of Electrode Disconnection in Sleep Signals
Biological signals that occur during sleep are recorded and classified by specialists. This process is called sleep staging. However, this is a very long and laborious process. Therefore, automatic sleep staging systems are needed. Nevertheless, automatic sleep staging studies to date have not provided satisfactory performance. The main reasons for this are inter-channel interference, electrode disconnection, and noise. In this paper, a new method (eye method) based on the Euclidean distance measurement method has been developed to solve the electrode disconnection or non-contact problem. This method was applied to three different datasets and detected all electrode disconnections with 100% accuracy. Thanks to this advanced method are aimed to increase the success of automatic sleep staging systems to be designed in the future.
- 28 March 2022
- Published by Trans Tech Publications, Ltd. in Journal of Biomimetics, Biomaterials and Biomedical Engineering
- Vol. 55, 109-130
Comparison of Time-Frequency Analyzes for a Sleep Staging Application with CNN
Sleep staging is the process of acquiring biological signals during sleep and marking them according to the stages of sleep. The procedure is performed by an experienced physician and takes more time. When this process is automated, the processing load will be reduced and the time required to identify disease will also be reduced. In this paper, 8 different transform methods for automatic sleep-staging based on convolutional neural networks (CNNs) were compared to classify sleep stages using single-channel electroencephalogram (EEG) signals. Five different labels were used to stage the sleep. These are Wake (W), Non Rapid Eye Movement (NonREM)-1 (N1), NonREM-2 (N2), NonREM-3 (N3), and REM (R). The classifications were done end-to-end without any hand-crafted features, ie without requiring any feature engineering. Time-Frequency components obtained by Short Time Fourier Transform, Discrete Wavelet Transform, Discrete Cosine Transform, Hilbert-Huang Transform, Discrete Gabor Transform, Fast Walsh-Hadamard Transform, Choi-Williams Distribution, and Wigner-Willie Distribution were classified with a supervised deep convolutional neural network to perform sleep staging. The discrete Cosine Transform-CNN method (DCT-CNN) showed the highest performance among the methods suggested in this paper with an F1 score of 89% and a value of 0.86 kappa. The findings of this study revealed that the transformation techniques utilized for the most accurate representation of input data are far superior to traditional approaches based on manual feature extraction, which acquires time, frequency, or nonlinear characteristics. The results of this article are expected to be useful to researchers in the development of low-cost, and easily portable devices.
- 28 March 2022
- Published by Trans Tech Publications, Ltd. in Journal of Biomimetics, Biomaterials and Biomedical Engineering
- Vol. 55, 109-130
Comparison of Time-Frequency Analyzes for a Sleep Staging Application with CNN
Sleep staging is the process of acquiring biological signals during sleep and marking them according to the stages of sleep. The procedure is performed by an experienced physician and takes more time. When this process is automated, the processing load will be reduced and the time required to identify disease will also be reduced. In this paper, 8 different transform methods for automatic sleep-staging based on convolutional neural networks (CNNs) were compared to classify sleep stages using single-channel electroencephalogram (EEG) signals. Five different labels were used to stage the sleep. These are Wake (W), Non Rapid Eye Movement (NonREM)-1 (N1), NonREM-2 (N2), NonREM-3 (N3), and REM (R). The classifications were done end-to-end without any hand-crafted features, ie without requiring any feature engineering. Time-Frequency components obtained by Short Time Fourier Transform, Discrete Wavelet Transform, Discrete Cosine Transform, Hilbert-Huang Transform, Discrete Gabor Transform, Fast Walsh-Hadamard Transform, Choi-Williams Distribution, and Wigner-Willie Distribution were classified with a supervised deep convolutional neural network to perform sleep staging. The discrete Cosine Transform-CNN method (DCT-CNN) showed the highest performance among the methods suggested in this paper with an F1 score of 89% and a value of 0.86 kappa. The findings of this study revealed that the transformation techniques utilized for the most accurate representation of input data are far superior to traditional approaches based on manual feature extraction, which acquires time, frequency, or nonlinear characteristics. The results of this article are expected to be useful to researchers in the development of low-cost, and easily portable devices.
- Open Access
- 1 March 2022
- Published by Politeknik Dergisi in Journal of Polytechnic
- Vol. 25 (1) , 47-58
3B T1 Ağırlıklı MR Görüntülerinde Atlas Tabanlı Hacim Ölçüm Yöntemini Kullanarak Alzheimer Hastalığının Teşhisi
Alzheimer Hastalığı yaşlılık ile beraber başlayan bir beyin hastalığıdır. Hastalığın teşhisi, takibi ve ilgili beyin bölgelerinin ölçümleri yüksek çözünürlüklü üç boyutlu yapısal manyetik rezonans görüntüleri ile yapılabilmektedir. Bu çalışmada, OASIS veri tabanından alınan 70 Alzheimer 70 Normal 3B T1 ağırlıklı MR görüntüleri üzerinde 116 subkortikal bölgenin hacimsel ölçümünü yapabilecek atlas tabanlı bir hacim ölçüm ve sınıflandırma modeli tasarlanmıştır. Ölçülen değerler her bir denekte gri madde, parankim, total beyin hacmi ile bölünerek normalizasyon işlemi yapılmıştır. Böylece ham ölçülen değerler dahil olmak üzere 140x116 matris boyutlu 4 farklı veri kümesi elde edilmiştir. Veri kümeleri entropi, t-test, roc, Bhattacharyya, Wilcoxon özellik derecelendirme yöntemleri ile en anlamlı özellikten en anlamsız özelliğe doğru derecelendirilmiştir. Derecelendirilen veriler her döngüde sırasıyla birleştirilmiş, lineer ve rbf kernel kullanan destek vektör makinelerine 10-kat çapraz geçerleme ile verilerek sınıflandırma işlemi yapılmıştır. Tüm senaryolar analiz edilerek, en az özellikle en iyi sonucu veren küme, özellik derecelendirme ve sınıflandırma metodu ortaya konulmuştur. Normalizasyon ve özellik derecelendirme yöntemlerinin sınıflandırma sonucuna etkisi incelenmiştir. Deneysel işlemler sonucunda roc özellik derecelendirme tabanlı lineer destek vektör makinesi, total beyin hacmi normalizasyonlu 107 özellik kullanarak %95.71 hassasiyet, %94.29 özgüllük, %95.00 doğruluk, 0.95 eğri altında kalan alan değerleri ile en yüksek oranları vermektedir.
- 1 March 2022
- Published by Politeknik Dergisi in Journal of Polytechnic
- Vol. 25 (1) , 47-58
Open Access3B T1 Ağırlıklı MR Görüntülerinde Atlas Tabanlı Hacim Ölçüm Yöntemini Kullanarak Alzheimer Hastalığının Teşhisi
Alzheimer Hastalığı yaşlılık ile beraber başlayan bir beyin hastalığıdır. Hastalığın teşhisi, takibi ve ilgili beyin bölgelerinin ölçümleri yüksek çözünürlüklü üç boyutlu yapısal manyetik rezonans görüntüleri ile yapılabilmektedir. Bu çalışmada, OASIS veri tabanından alınan 70 Alzheimer 70 Normal 3B T1 ağırlıklı MR görüntüleri üzerinde 116 subkortikal bölgenin hacimsel ölçümünü yapabilecek atlas tabanlı bir hacim ölçüm ve sınıflandırma modeli tasarlanmıştır. Ölçülen değerler her bir denekte gri madde, parankim, total beyin hacmi ile bölünerek normalizasyon işlemi yapılmıştır. Böylece ham ölçülen değerler dahil olmak üzere 140x116 matris boyutlu 4 farklı veri kümesi elde edilmiştir. Veri kümeleri entropi, t-test, roc, Bhattacharyya, Wilcoxon özellik derecelendirme yöntemleri ile en anlamlı özellikten en anlamsız özelliğe doğru derecelendirilmiştir. Derecelendirilen veriler her döngüde sırasıyla birleştirilmiş, lineer ve rbf kernel kullanan destek vektör makinelerine 10-kat çapraz geçerleme ile verilerek sınıflandırma işlemi yapılmıştır. Tüm senaryolar analiz edilerek, en az özellikle en iyi sonucu veren küme, özellik derecelendirme ve sınıflandırma metodu ortaya konulmuştur. Normalizasyon ve özellik derecelendirme yöntemlerinin sınıflandırma sonucuna etkisi incelenmiştir. Deneysel işlemler sonucunda roc özellik derecelendirme tabanlı lineer destek vektör makinesi, total beyin hacmi normalizasyonlu 107 özellik kullanarak %95.71 hassasiyet, %94.29 özgüllük, %95.00 doğruluk, 0.95 eğri altında kalan alan değerleri ile en yüksek oranları vermektedir.
- Open Access
- 31 December 2021
- Published by Deomed Publishing in Anatomy
- Vol. 15 (3) , 2007-215
Quantitative evaluation of the cerebellum in patients with depression and healthy adults by VolBrain method
Objectives: Besides the well-known sensorimotor control function, the cerebellum is also associated with cognitive functions and mood via the cerebral-cerebellar circuit. This study aimed to investigate possible cerebellar morphometric changes in untreated patients with depression. Methods: Brain magnetic resonance (MR) images of 40 adults (age: 18–50 years), including 20 untreated depression patients and 20 healthy controls were analysed prospectively. Intracranial cavity and total cerebellar volumes were measured by using VolBrain. The cerebellum segmentation was performed with CERES to obtain the total gray matter volumes and cortical thickness of the lobules. Results: Total cerebellar volume was 141.27±13.12 cm3 in the depressed group and 142.63±8.01 cm3 in the control group (p>0.05). The difference between males and females in the depressed group was not statistically significant (p>0.05). Total cerebellar volume was approximately 11% of total intracranial volume in both groups. The cortical thickness of lobule V (right-total), lobule VIIIB (right), and lobule IX (right) was smaller in the depressed group, independent of sex (p<0.05). Lobule V, VIIIB and IX volume was smaller and Crus-I cortical thickness was increased in depressed females (p<0.05). Conclusion: The cerebellar volume and cortical thickness of cerebellar lobules in patients with depression show significant differences compared to healthy subjects.
- 31 December 2021
- Published by Deomed Publishing in Anatomy
- Vol. 15 (3) , 2007-215
Open AccessQuantitative evaluation of the cerebellum in patients with depression and healthy adults by VolBrain method
Objectives: Besides the well-known sensorimotor control function, the cerebellum is also associated with cognitive functions and mood via the cerebral-cerebellar circuit. This study aimed to investigate possible cerebellar morphometric changes in untreated patients with depression. Methods: Brain magnetic resonance (MR) images of 40 adults (age: 18–50 years), including 20 untreated depression patients and 20 healthy controls were analysed prospectively. Intracranial cavity and total cerebellar volumes were measured by using VolBrain. The cerebellum segmentation was performed with CERES to obtain the total gray matter volumes and cortical thickness of the lobules. Results: Total cerebellar volume was 141.27±13.12 cm3 in the depressed group and 142.63±8.01 cm3 in the control group (p>0.05). The difference between males and females in the depressed group was not statistically significant (p>0.05). Total cerebellar volume was approximately 11% of total intracranial volume in both groups. The cortical thickness of lobule V (right-total), lobule VIIIB (right), and lobule IX (right) was smaller in the depressed group, independent of sex (p<0.05). Lobule V, VIIIB and IX volume was smaller and Crus-I cortical thickness was increased in depressed females (p<0.05). Conclusion: The cerebellar volume and cortical thickness of cerebellar lobules in patients with depression show significant differences compared to healthy subjects.