Aynur Sevinc

Dr. Aynur Sevinc

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 April 2025 in Diagnostics
Verified
Journal Article
Open access
PDF file
ASAynur Sevinc
Aynur Sevinc
Murat UCANMurat UCAN
Murat UCAN
BKBuket Kaya
Buket Kaya
Background/Objectives: Although transformer-based deep learning architectures are preferred in many hybrid architectures due to their flexibility, they generally perform poorly on image classification tasks with small datasets. An important improvement in performance when transformer architectures work with limited data is the use of distillation techniques. The impact of distillation techniques on classification accuracy in transformer-based models has not yet been extensively investigated. Methods: This study investigates the impact of distillation techniques on the classification performance of transformer-based deep learning architectures trained on limited data. We use transformer-based models ViTx32 and ViTx16 without distillation and DeiT and BeiT with distillation. A four-class dataset of brain MRI images is used for training and testing. Results: Our experiments show that the DeiT and BeiT architectures with distillation achieve performance gains of 2.2% and 1%, respectively, compared to ViTx16. A more detailed analysis shows that the distillation techniques improve the detection of non-patient individuals by about 4%. Our study also includes a detailed analysis of the training times for each architecture. Conclusions: The results of the experiments show that using distillation techniques in transformer-based deep learning models can significantly improve classification accuracy when working with limited data. Based on these findings, we recommend the use of transformer-based models with distillation, especially in medical applications and other areas where flexible models are developed with limited data.

30 June 2025 in Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi
Verified
Journal Article
Open access
PDF file
ASAynur Sevinc
Aynur Sevinc
BKBuket Kaya
Buket Kaya
MGMehmet Gül
Mehmet Gül
In conventional methods, the detection of tumour disease from brain images using magnetic resonance imaging is a difficult and human error-prone field of study that requires an expert medical doctor. Incomplete or inaccurate detection of brain tumours can have significant undesirable consequences such as shortening of human life. In order to overcome these difficulties, many researchers are working on autonomous disease detection supported by artificial intelligence. The aim of this study is to utilise brain magnetic resonance images with deep learning architectures for fast and reliable autonomous cancer detection. In this study, brain images are classified using two different datasets and a convolutional neural network infrastructure, which are widely used in the publicly available literature. The results obtained as a result of experiments with similar parameters in training, validation and testing processes are compared in detail with other studies in the literature and the differences between them are presented. The new convolutional neural network-based model proposed in the study achieved 99.76% classification result in the accuracy evaluation metric. The results obtained showed that the model proposed in the study can be used with high accuracy in brain tumour detection and can shed light on other fields of study.

31 March 2026 in Yassawi Journal of Engineering Science
Verified
Journal Article
Open access
PDF file
ЖКЖансая Калимбетова
Жансая Калимбетова
ASAynur Sevinc
Aynur Sevinc
ЧГЧетин Генчер
Четин Генчер
АСАйнур Севинч
Айнур Севинч
Ветровая энергия представляет собой чистое и эффективное решение для производства электроэнергии. Разработка приложений в области ветроэнергетики требует тщательного анализа ветровых характеристик и точного прогнозирования энергии ветра на исследуемой площадке. В данном исследовании была оценена генерация ветровой энергии как вида возобновляемой энергии. Для оценки данных о производстве энергии учитывались геологические и метеорологические данные ветропарка в Адыямане. Для проведения этой оценки использовались прямые сети с обратным распространением ошибки (feedforward backpropagation) и адаптивная нечеткая система на основе сети (ANFIS), благодаря их успешному прогнозированию линейных и нелинейных моделей, как одного из применений искусственного интеллекта. В исследовании было установлено, что оцененное значение производства энергии (МВт·ч) довольно близко к фактическим данным о выработке. В будущих исследованиях по прогнозированию приложения искусственного интеллекта могут успешно использоваться в качестве замены традиционным методам.