Aykat, Şükrü
Loading...
Name Variants
Job Title
Doktor Öğretim Üyesi
Email Address
sukruaykat@artuklu.edu.tr
Main Affiliation
Department of Computer Engineering / Bilgisayar Mühendisliği Bölümü
Status
Website
ORCID ID
Scopus Author ID
Turkish CoHE Profile ID
Google Scholar ID
WoS Researcher ID
Scholarly Output
8
Articles
4
Citation Count
0
Supervised Theses
0
8 results
Scholarly Output Search Results
Now showing 1 - 8 of 8
Conference Object Brain Tumor Detection From Brain Mri Images With Deep Learning Methods;(Institute of Electrical and Electronics Engineers Inc., 2024) Aykat, S.; Aykat, ŞükrüIn this research, a deep learning model is proposed for brain tumor detection using brain MRI image collection. Three pre-trained convolutional neural networks are used as feature extractors. The obtained features are classified as brain tumors, normal, and tumorous using four different classifiers. Our proposed model has achieved a remarkable accuracy of 99.58% in its analysis, which is better than standard techniques. In addition, the proposed method has shown better performance than the convolutional neural network models used in the analysis. © 2024 IEEE.Article Advanced Detection of Retinal Diseases Via Novel Hybrid Deep Learning Approach(int information & Engineering Technology Assoc, 2023) Aykat, Sukru; Senan, Sibel; Aykat, ŞükrüDiabetic drusen, choroidal neovascularization (CNV), and macular edema (DME) are some retinal diseases that can cause severe blindness and vision loss. Early diagnosis of retinal diseases is vital to prevent this irreversible damage to the eye. The problem statement of this study can be given as presenting new deep learning based results for detecting these retinal diseases. For this purpose, OCT dataset was used to detect CNV, DME and Drusen patients. This data set, which is frequently used in the literature, consists of CNV, DME, Drusen and Normal retina images. RestNet50, InceptionV3, InceptionResnetV2, MobileNet, DenseNet-201, Xception, EfficentNetB4, EfficentNetB7 and EfficentNetV2S models of the CNN architecture were applied to the data set and the performance results of these models were obtained. Then, in order to increase the classification performance of each of these models, hyperparameter tuning was performed by reducing the learning rate by half in each epoch. Later, a hybrid version of the EfficientNetV2S and Xception convolutional neural network models, the most successful of these hyperparameter-tuned models, was developed. The performance analysis results of our proposed hybrid deep learning model are given by comparing them with traditional deep learning models in the literature. These comparison results show that the classification success of the proposed model is higher than the success of traditional deep learning models in the literature. Thus, the proposed hybrid model can shorten the clinical diagnosis time. In addition, the costs of healthcare services can be reduced by intervening in treatable diseases earlier, instead of more costly interventions in the advanced stages of the disease.Article Derin Öğrenme Kullanılarak Fundus Görüntülerinden Katarakt ve Diyabetik Retinopati Tespiti(BANDIRMA ONYEDİ EYLÜL ÜNİVERSİTESİ, 2023) Aykat, Şükrü; Senan, Sibel; Aykat, ŞükrüDiyabetik retinopati ve katarakt ciddi körlüğe ve görme kaybına neden olabilen bazı retina hastalıklarıdır. Gözde meydana gelen bu geri dönüşü olmayan hasarı önlemek için retina hastalıklarının erken teşhisi hayati önem taşımaktadır. Bu çalışmanın problem cümlesi, bu retina hastalıklarının tespiti için derin öğrenme tabanlı sonuçların sunulması olarak verilebilir. Bu amaçla ilk önce ham bir veri seti üzerinde histogram eşitleme yöntemi kullanılarak yeni bir seti oluşturulmuştur. Ardından beş geleneksel derin öğrenme modeline hiperparametre ayarı yapılarak veri setleri üzerinde eğitimler gerçekleştirilmiştir. En son olarak veri setleri üzerinde en yüksek başarıya sahip MobileNet tabanlı bir hibrit model geliştirilmiştir. Önerilen hibrit model, ön işlenmiş veri seti üzerinde %99 doğruluk oranı elde etmiştir. Hibrit modelin sınıflandırma başarısının literatürdeki derin öğrenme modellerinin başarısından daha yüksek olduğu görülmüştür. Bu çalışma diyabetik retinopati ve katarakt hastalarının teşhis sürecine katkı sağlayacaktır.Conference Object Classification of Epileptic and Healthy Individual Eeg Signals Using Neural Networks(Ieee, 2020) Aykat, Sukru; Senan, Sibel; Ensari, Tolga; Aykat, ŞükrüElectroencephalogram (EEG) are signals used for the analysis of the electrical and functional activity of the brain. These signals are commonly used to detect epileptic seizures. The aim of this study is to classify healthy and epileptic individual EEG signals using artificial neural networks (ANN). For this purpose, the open data source of the University of Bonn was used. The success rates of the classification results obtained with the designed ANN model show the effectiveness of this ANN structure in the application under consideration.Article Using Machine Learning To Detect Different Eye Diseases From Oct Images(Prof.Dr. İskender AKKURT, 2023) Aykat, Ş.; Senan, S.; Aykat, ŞükrüDiseases or damage to the retina that cause adverse effects are one of the most common reasons people lose their sight at an early age. Today, machine learning techniques, which give high accuracy results in a short time, have been used for disease detection in the biomedical field. Optical coherence tomography is an advanced tool for the analysis, detection and treatment of retinal diseases by imaging the retinal layers. The aim of this study is to detect eight retinal diseases that can occur in the eye and cause permanent damage as a result, using machine learning from eye tomography images. For this purpose, hyperparameter settings were applied to six deep learning models, training was performed on the OCT-C8 dataset and performance analyzes were made. The performance of these hyperparameter-tuned models was also compared with previous eye disease detection studies in the literature, and it was seen that the classification success of the hyperparameter-tuned DenseNet121 model presented in this study was higher than the success of the other models discussed. The fine-tuned DenseNet121 classifier achieved 97.79% accuracy, 97.69% sensitivity, and 97.79% precision for the OCT-C8 dataset. © IJCESEN.thesis.listelement.badge Mesleki lise öğretmenlerinin, öğrencilerinin ve idarecilerinin hizmetiçi eğitim öncesi ile sonrası etkileşimli tahtaya ilişkin görüşlerinin incelenmesi(2017) AYKAT, Şükrü; Aykat, ŞükrüBu çalışmanın amacı; FATİH Projesi uygulanan mesleki liselerdeki öğretmenlerin, ETEKUK eğitimi öncesi ve sonrası etkileşimli tahtayı kullanma düzeylerinde, öz yeterliliklerinde ve görüşlerinde farklılık olup olmadığını belirlemektir. Örneklem 2015-2016 eğitim öğretim yılında Mardin ili Midyat ilçesinde FATİH Projesi çerçevesinde etkileşimli tahta kurulumu yapılmış Telkari Mesleki ve Teknik Anadolu Lisesi’nde görev yapan ETEKUK eğitimine katılan öğretmenler ve okul idarecileri ile okulda eğitim gören öğrencilerden oluşmaktadır. Bu çalışmada karma yöntem modeli kullanılmıştır. Veriler ölçek, anket ve yarı yapılandırılmış görüşme formları ile toplanmıştır. Nicel veriler ortalama ve t-Testi, nitel veriler ise içerik analizi ile çözümlenmiştir. Araştırma sonunda öğretmenlerin ETEKUK eğitimi sonrasında etkileşimli tahta kullanma öz yeterliliklerinde ve etkileşimli tahta kullanma düzeylerinde anlamlı bir değişimin olmadığı görülmüştür. Ayrıca öğretmenlerin ETEKUK eğitiminden önce etkileşimli tahta için materyal geliştiremedikleri ETEKUK eğitiminden sonra ise kısmen materyal hazırlayabildikleri görülmüştür. Bu da ETEKUK eğitim içeriğinin etkileşimli tahta için materyal geliştirmede yeterli olmadığını göstermektedir. Öğrenciler ise; etkileşimli tahta kullanılan dersleri tercih ettiklerini, öğretmenlerin etkileşimli tahtayı kullanmaya devam etmelerini istediklerini belirtmişlerdir. Etkileşimli tahta arızaları dersi olumsuz yönde etkilediğini de belirtmişlerdir. Öğretmenler, öğrenciler ve okul idarecileri derste etkileşimli tahta kullanılmasıyla; öğrenci başarısının arttığını, derse ilgi ve katılımı olumlu yönde etkilediğini belirtmişlerdir.Article Review of the Opinions of Vocational High School Teachers, Students, and Administrators on the Interactive Whiteboard1(Turkish Online Journal of Qualitative Inquiry, 2020) Aykat, Şükrü; Günüç, Selim; Aykat, ŞükrüThe primary aim of this study is to determine whether there was a difference in teachers' levels of interactive whiteboard usage, their self-efficacy, and opinions before and after the Use of Technology in Education Course (UTEC) in vocational high schools where FATIH project was implemented. Research participants constituted of teachers and school administrators, and students attending Telkari Vocational and Technical Anatolian High School in which interactive whiteboard installation was made within the framework of FATIH project in Midyat district of Mardin province and who attended UTEC training in the 2015-2016 academic year. In this study, a mixed research method model was used. Data were collected through scale, survey and semi-structured interview forms. Quantitative data were analyzed by mean and ttest, and qualitative data were analyzed by content analysis. In conclusion of the study, it was observed that there was no significant change in the self-efficacy of teachers using interactive whiteboard and their level of using interactive whiteboard after UTEC training. Furthermore, it was observed that teachers were unable to develop materials for the interactive whiteboard before UTEC training, and after the UTEC training, they were able to prepare materials in part. This fact has revealed that the UTEC training content was not sufficient in material development for the interactive whiteboard. The students, on the other hand, indicated that they preferred the courses used on the interactive whiteboard and asked the teachers to continue using the interactive whiteboard. Participant students indicated that interactive whiteboard failures also adversely affected the lesson. Teachers, students, and school administrators reported that use an interactive whiteboard in the course increased student success and positively affected interest and participation in the course.Conference Object A Review Of Cuda Based Face Detection And Recognition Applications;(Institute of Electrical and Electronics Engineers Inc., 2019) Aykat, S.; Sertbas, A.; Aykat, ŞükrüFace recognition is an important biometry used in many areas such as building security systems, biometric passports and identification, surveillance systems. Systems used in these areas need to be fast. In recent years, many applications have been developed by taking advantage of the GPU's parallel processing. In this study, face detection and face recognition studies with CUDA supported by GPGPU, which is a parallel computing platform and programming module developed by NVIDIA, are examined. To date, the literature on face detection and recognition studies with CUDA has been conducted. As a result of the study, it was observed that the face detection and face recognition operations performed by using the parallel processing power of CUDA can be performed much faster. Furthermore, it was concluded that if deep learning is used in CUDA based face recognition applications, face recognition operations will be performed in much shorter periods. © 2019 IEEE.
