Yeganli, Faezeh
Loading...
Profile URL
Name Variants
Yeganlı, Faezeh
Yeganli, F.
Yeganli, F.
Job Title
Email Address
faezeh.yeganli@ieu.edu.tr
Main Affiliation
05.06. Electrical and Electronics Engineering
Status
Current Staff
Website
ORCID ID
Scopus Author ID
Turkish CoHE Profile ID
Google Scholar ID
WoS Researcher ID
Sustainable Development Goals
Documents
17
Citations
114
h-index
6
Documents
16
Citations
62
Scholarly Output
10
Articles
1
Views / Downloads
18/915
Supervised MSc Theses
1
Supervised PhD Theses
0
WoS Citation Count
19
Scopus Citation Count
49
WoS h-index
3
Scopus h-index
3
Patents
0
Projects
0
WoS Citations per Publication
1.90
Scopus Citations per Publication
4.90
Open Access Source
1
Supervised Theses
1
| Journal | Count |
|---|
Current Page: 1 / NaN
Scopus Quartile Distribution
Competency Cloud
10 results
Scholarly Output Search Results
Now showing 1 - 10 of 10
Conference Object Citation - WoS: 3Citation - Scopus: 3Fast and Interpretable Deep Learning Pipeline for Breast Cancer Recognition(IEEE, 2022) Bonyani, Mahdi; Yeganli, Faezeh; Yeganli, S. FaeghehBreast cancer is one of the main causes of death across the world in women. Early diagnosis of this type of cancer is critical for treatment and patient care. In this paper, we propose a fast and interpretable deep learning-based pipeline for automatic detection of the metastatic tissues in breast histopathological images. Firstly, the proposed pipeline uses multiple preprocessing and data augmentation techniques to reduce over-fitting. Then, the proposed pipeline employs one - cycle policy technique in the pre-trained convolutional neural networks model in shallow and deep fine-tuning phases to find the optimal values. Finally, gradient-weighted class activation mapping (Grad-CAM) technique is utilized to produce a coarse localization map of the important regions in the image. Experiments on the PatchCamelyon dataset demonstrate the superior classification performance of the proposed method over the state-of-the-art.Conference Object Citation - WoS: 9Citation - Scopus: 16The Lidar and Uwb Based Source Localization and Initialization Algorithms for Autonomous Robotic Systems(Institute of Electrical and Electronics Engineers Inc., 2019) Bostanci B.; Tekkok S.C.; Soyunmez E.; Oguz-Ekim P.; Yeganli F.; Bostanci, Bekir; Yeganli, Faezeh; Soyunmez, Emre; Oguz-Ekim, Pinar; Tekkok, SercanThis paper covers the source localization algorithm based on the least squares techniques and the squared range measurements obtained from ultra-wide band (UWB) sensors to locate the robot in an indoor environment. Additionally, the initialization algorithm which is based on light detection and Ranging (LiDAR) scans is proposed. It takes the advantage of the estimated location to find the initial orientation of the robot with respect to the previously obtained map. Thus, the crucial problem of the autonomous initialization and localization in robotics is solved. To enable wide-spread adoption, we provide an open source implementation of our algorithms and the modules for the robot operating system (ROS) for real environment. Furthermore, an open source simulation environment is created for applications which employ UWB/LiDAR data. © 2019 Chamber of Turkish Electrical Engineers.Article Unified Deep Learning Method for Accurate Brain Tumor Segmentation Using Vertical Voxel Grouping and Wavelet Features(Springer London Ltd, 2025) Sahin, M. Faruk; Yeganli, S. Faegheh; Uludag, Goenuel; Yeganli, Faezeh; Anka, FerzatBrain tumor segmentation plays a vital role in medical imaging, enabling accurate diagnosis and guiding treatment decisions. Despite notable progress driven by deep neural networks (DNNs) and multi-parametric magnetic resonance imaging (mpMRI), the complexity and heterogeneity of tumor tissues make precise segmentation a persistent challenge. In this paper, we propose a novel method that integrates Vertically grouped Voxel Feature Extraction (VFE), wavelet-based multi-resolution detail enhancement, and a modified UNet-VGG16+ architecture. The VFE component enhances tumor region contrast and suppresses irrelevant background areas by grouping column-wise voxel intensities within each slice. As a result, the average image contrast is increased by 23.78%, thereby improving the ability of Deep Neural Networks (DNNs) to focus on tumor regions. The wavelet-based enhancement captures multi-resolution details to more clearly delineate tumor boundaries while also reducing noise. The UNet-VGG16+ architecture leverages transfer learning to efficiently process these enhanced features for accurate segmentation. Extensive experiments on the BraTS21 dataset demonstrate that the proposed method achieves a mean Dice score of 94.69%, with segmentation accuracies of 93.3%, 93.1%, and 94.4% for Enhancing Tumor (ET), Whole Tumor (WT), and Tumor Core (TC), respectively. Comparative evaluations show consistent and statistically significant improvements over state-of-the-art models (p< 0.001). Further validation on the BraTS18 dataset confirms the model's generalizability. These results highlight the effectiveness of combining spatially structured voxel aggregation with frequency-domain analysis for robust and high-precision brain tumor segmentation.Conference Object Citation - Scopus: 1Finger-Print Image Super-Resolution Via Gradient Operator Based Clustered Coupled Sparse Dictionaries(Institute of Electrical and Electronics Engineers Inc., 2019) Yeganli F.; Singh K.In this paper, a novel approach is employed for fingerprint image super-resolution based on sparse representation over a set of coupled low and high-resolution dictionary pairs. The primary step of fingerprint super-resolution involves learning a pair of coupled low-and high-resolution sub-dictionaries for each cluster of patches sampled from training set of fingerprint images. The clusters are formulated based on patch sharpness and the dominant phase angle via the magnitude and phase of the gradient operator for each image patch. In the reconstruction stage, for the low-resolution patch the most appropriate dictionary pair is selected, and the sparse coding coefficients are calculated with respect to the low-resolution dictionary. The equality assumption of the sparse representation of the low and high-resolution patches is the link between the low and high-resolution features space. For the reconstruction of high resolution patch, the sparse coefficients calculated for low-resolution patch are directly multiplied with corresponding high-resolution dictionary. The conducted experiments over fingerprint images show that the algorithm is competitive with the state-of-art super-resolution algorithms. © 2019 IEEE.Conference Object Citation - WoS: 1Citation - Scopus: 4Decoding Olfactory EEG Signals Using Multi-Domain Features and Machine Learning(IEEE, 2024) Sude Pehlivan, Akbugday; Akbuğday, Burak; Yeganli, Faezeh; Akan, Aydin; Rıza Sadıkzade; Akbugday, Sude Pehlivan; Sadikzade, RizaAccurate detection of human emotion is an important topic for affective computing. Especially with the rise of artificial intelligence in the marketing industry, the tools available are subjective and often heavily dependent on sample sizes and demographics. This study explores the neural responses to olfactory stimuli by analyzing EEG data collected from 57 participants exposed to a perfume scent in correlation with self-reported survey results. The electroencephalogram (EEG) signals were processed to extract time-domain, spectral-domain, and nonlinear features, which were subsequently classified using various machine learning algorithms. The classification outcomes were mapped onto a two-dimensional pleasure-arousal plane, with the Medium Gaussian support vector machine (SVM) achieving the highest performance, including 99.8 % validation accuracy and 100 % test accuracy. These results highlight the significant potential of EEG-based approaches in decoding the neural underpinnings of sensory experiences, with implications for applications in neuromarketing and therapeutic contexts.Master Thesis Automatic Sleep Stage Scoring Based Eeg Evoked Response /(İzmir Ekonomi Üniversitesi, 2021) Dıaı, Wassım; Yeganlı, FaezehUyku uzmanları sıklıkla uyku laboratuvarlarında ki hastalardan elde ettiği nörofizyolojik sinyalleri analiz ederek uyku değerlendirmeleri yaparlar. Bu iş temelinde çok zor, sıkıcı ve zamana mal olan bir iştir. Manuel uyku aşaması kaydının kısıtlanması, Otomatik Uyku Aşaması Sınıflandırma sistemlerinin (ASSC) üretimine olan ihtiyacını artırdı. Uyku evrelerinin belirlenmesi, farklı uyku dönemlerinin tanınması için gereklidir ve doktorların ilişkili uyku anormalliklerini tanımasına ayrıca tedavi etmesine izin veren çok önemli bir adımdır. Çalışma boşluklarını tanımlamak ve potansiyel olarak gerçekçi bir yaklaşımı dahil etmek için, bu çalışma Elektroensefalogram (EEG) konusunda yer alan uyku evreleme aşamalarının her birinde kullanılan uyandırılmış yanıt ve diğer yaklaşımlar dahil, veri işleme, özellik çıkarma ve sınıflandırma konularında ki ilerlemeleri ve zorlukları analiz etmeyi amaçlamaktadır. Bu tezde, sağlıklı deneklerin uyku-edf veri seti birkaç sınıflandırıcıyı ölçmektedir. % 85'in üzerinde bir test doğruluğu için optimize edilmiş model, kayda değer bir ilerleme olduğunu kanıtlıyor. Bulgular, farklı sınıflandırıcılar arasındaki eşitsizliği göstermektedir. Son olarak, sağlıklı kişiler tarafından birleştirilen 2 EEG kanalı kullanılarak ilgili sınıflandırma doğrulukları elde edilebilir. Aslında, algoritmaları daha fazla kişi tarafından kullanılabilmesi için daha fazla genellemek mümkündür.Conference Object Olfactory Emotion Recognition Using EEG Spectral Topographic Heatmaps and CNNs(Institute of Electrical and Electronics Engineers Inc., 2025) Yeganli, Faezeh; Sadikzade, Riza; Akan, AydinConference Object Citation - WoS: 6Citation - Scopus: 23Heat Leakage Detection and Surveiallance Using Aerial Thermography Drone(Institute of Electrical and Electronics Engineers Inc., 2018) Kayan, Hakan; Eslampanah, Raheleh; Yeganli F.; Askar M.In recent years, UAVs provide an excellent investigative tool used for detecting heat leakages and their surveillance using high-resolution thermal cameras. In this work a low cost optimal aerial drone for surveillance and heat leakage detection is developed. The developed hexacopter have been used to take thermal images. These images have been analyzed by a developed image processing toolkit to find a solution for important needs in civil applications like search and rescue, surveillance and heat loss mapping for buildings. Our toolkit illustrates the heat leakages in the picture and calculates the total waste of money due to heat leakages of buildings. © 2018 IEEE.Conference Object Citation - Scopus: 1DeepSurvLiver: Predicting Post-Operative Survival after Liver Transplantation(Institute of Electrical and Electronics Engineers Inc., 2023) Bonyani, M.; Yeganli, Faezeh; Yeganli, S.F.; Shahidi, N.Liver transplantation (LT) offers a vital solution for end-stage liver disease patients. Predicting post-LT survival, however, remains challenging. This paper introduces an artificial intelligence (AI)-based model to predict post-operative survival after LT. The proposed model employs a two-stream recurrent neural network (RNN) using deep long short-term memory (LSTM-RNN) and bidirectional long short-term memory (BiLSTM-RNN) to extract inherent features of donors and recipients, respectively. Additionally, a self-attention based module is developed to capture the influential features of donors' and patients' data. To eliminate errors in the prediction model caused by imbalanced distributions, implicit semantic data augmentation (ISDA) is employed. Tested with 5-fold cross-validation, the proposed model achieved 99.47% accuracy and 0.996 the area under the curve, outperforming existing models in prediction performance. © 2023 IEEE.Conference Object Citation - Scopus: 1Face Detection Via Hog and Ga Feature Selection With Support Vector Machines(Institute of Electrical and Electronics Engineers Inc., 2019) Binli M.K.; Can Demiryilmaz B.; Ekim P.O.; Yeganli F.; Can Demiryilmaz, Burak; Ekim, Pinar Oguz; Yeganli, Faezeh; Binli, Mustafa KemanThis work employs a new method of feature selection and classification of image objects by combining previous studies in literature of feature selection and classification of images. In the new algorithm, instead of sliding a window on the image and scaling the window by applying normalized cuts and image segmentation algorithms, information related to the position of objects is considered. Accordingly, the scaling of searching image window process is been exceeded. For this purpose, the obtained segments of image features extracted by employing Histogram Oriented Gradient (HOG) descriptor, which clears the images from the curse of dimensions. These extracted features optimized by applying Genetic Algorithm (GA) method, to detect faces and compared with HOG feature results. The main objective of this work is to improve the accuracy of Support Vector Machine (SVM) classifier in detail description applications. © 2019 Chamber of Turkish Electrical Engineers.
