Browsing by Author "Turk, Omer"

Now showing 1 - 8 of 8

Automatic Detection of Brain Tumors With the Aid of Ensemble Deep Learning Architectures and Class Activation Map Indicators by Employing Magnetic Resonance Images
(Elsevier, 2024) Turk, Omer; Ozhan, Davut; Acar, Emrullah; Akinci, Tahir Cetin; Yilmaz, Musa; Türk, Ömer
Today, as in every life-threatening disease, early diagnosis of brain tumors plays a life-saving role. The brain tumor is formed by the transformation of brain cells from their normal structures into abnormal cell structures. These formed abnormal cells begin to form in masses in the brain regions. Nowadays, many different techniques are employed to detect these tumor masses, and the most common of these techniques is Magnetic Resonance Imaging (MRI). In this study, it is aimed to automatically detect brain tumors with the help of ensemble deep learning architectures (ResNet50, VGG19, InceptionV3 and MobileNet) and Class Activation Maps (CAMs) indicators by employing MRI images. The proposed system was implemented in three stages. In the first stage, it was determined whether there was a tumor in the MR images Tumor) were detected from MR images (Multi-class Approach). In the last stage, CAMs of each tumor group were created as an alternative tool to facilitate the work of specialists in tumor detection. The results showed that the overall accuracy of the binary approach was calculated as 100% on the ResNet50, InceptionV3 and MobileNet architectures, and 99.71% on the VGG19 architecture. Moreover, the accuracy values of 96.45% with ResNet50, 93.40% with VGG19, 85.03% with InceptionV3 and 89.34% with MobileNet architectures were obtained in the multi-class approach.
Classification of EEG Records for the Cursor Movement with the Convolutional Neural Network
(IEEE, 2018) Turk, Omer; Ozerdem, Mehmet Sirac; Türk, Ömer; Ozyildirim, BM; Yildirim, T
Nowadays, very successful results are obtained with deep learning architectures which can be applied to many fields. Because of the high performances it provides in many areas, deep learning has come to a central position in machine learning and pattern recognition. In this study, electroencephalogram (EEG) signals related to up and down cursor movements were represented as image pattern by using obtained approximation coefficients after wavelet transform. The Obtained image patterns were classified by applying Convolutional Neural Network. In this study, EEG records related to cursor movements were classified and classification accuracy was obtained as 88.13%.
Classification of Mental Task EEG Records Using Hjorth Parameters
(IEEE, 2017) Turk, Omer; Seker, Mesut; Akpolat, Veysi; Ozerdem, Mchmet Sirac; Türk, Ömer
The effects of mental activities on brain dynamics is the main field that studied for a long time, but the results of studies have not reached the desired level. The aim of present study was to classify the mental task EEG records by using Hjorth parameters. hi this study, EEG signals that recorded from 9 subjects were used. EEG signals were recorded by applying a experimental paradigm which contains five stimuli related to different mental task. These stimuli are defined as condition word mental subtraction spatial navigation right hand motor imagery and feet motor imagery Wavelet packet transform was used to obtain sub bands of EEC signals. Statistical parameters that consist of mobility, complexity and Mahalanobis distance were applied to sub-bands. Feature vectors were classified by using artificial neural network. When classification performances related to mental activities were examined, the best classification accuracy was obtained as nearly 80% for 'condition word - mental subtraction', ('spatial navigation feet motor imagery;' and 'spatial navigation - condition word'. The lowest classification accuracy was obtained for 'mental subtraction - right hand motor imagery,', 'condition word - right hand motor imagery' and 'spatial navigation right hand motor imagery'. The classification accuracies related to all stimuli that classifed among themselves were obtained as 77,61%.
The Deep Learning Method Differentiates Patients With Bipolar Disorder From Controls With High Accuracy Using Eeg Data
(Sage Publications inc, 2024) Metin, Baris; Uyulan, Caglar; Erguzel, Turker Tekin; Farhad, Shams; Cifci, Elvan; Turk, Omer; Tarhan, Nevzat; Türk, Ömer
Background: Bipolar disorder (BD) is a mental disorder characterized by depressive and manic or hypomanic episodes. The complexity in the diagnosis of Bipolar disorder (BD) due to its overlapping symptoms with other mood disorders prompted researchers and clinicians to seek new and advanced techniques for the precise detection of Bipolar disorder (BD). One of these methods is the use of advanced machine learning algorithms such as deep learning (DL). However, no study of BD has previously adopted DL techniques using EEG signals. Method: EEG signals of 169 BD patients and 45 controls were cleaned from the artifacts and processed using two different DL methods: a one-dimensional convolutional neural network (1D-CNN) combined with the long-short term memory (LSTM) and a two-dimensional convolutional neural network (2D-CNN). Additionally, Class Activation Maps (CAMs) acquired from the bipolar and control groups were used to obtain distinctive regions to specify a particular class in an image. Results: Group identifications were confirmed with 95.91% overall accuracy through the 2D-CNN method, demonstrating very high sensitivity and lower specificity. Also, the overall accuracy obtained from the 1D-CNN + LSTM method was 93%. We also found that F4, C3, F7, and F8 electrode activities produce predominant features to detect the bipolar group. Conclusion: To our knowledge, this study used EEG-based DL analysis for the first time in BD. Our results suggest that the raw EEG-based DL algorithm can successfully differentiate individuals with BD from controls. Class Activation Map (CAM) analysis suggests that prefrontal changes are predominant in EEG data of patients with BD.
Deep Learning-Based Artificial Intelligence Can Differentiate Treatment-Resistant and Responsive Depression Cases With High Accuracy
(Sage Publications inc, 2025) Metin, Sinem Zeynep; Uyulan, Caglar; Farhad, Shams; Erguzel, Tuerker Tekin; Turk, Omer; Metin, Baris; Tarhan, Nevzat; Türk, Ömer
Background: Although there are many treatment options available for depression, a large portion of patients with depression are diagnosed with treatment-resistant depression (TRD), which is characterized by an inadequate response to antidepressant treatment. Identifying the TRD population is crucial in terms of saving time and resources in depression treatment. Recently several studies employed various methods on EEG datasets for automatic depression detection or treatment outcome prediction. However, no previous study has used the deep learning (DL) approach and EEG signals for detecting treatment resistance. Method: 77 patients with TRD, 43 patients with non-TRD, and 40 healthy controls were compared using GoogleNet convolutional neural network and DL on EEG data. Additionally, Class Activation Maps (CAMs) acquired from the TRD and non-TRD groups were used to obtain distinctive regions for classification. Results: GoogleNet classified the healthy controls and non-TRD group with 88.43%, the healthy controls and TRD subjects with 89.73%, and the TRD and non-TRD group with 90.05% accuracy. The external validation accuracy for the TRD-non-TRD classification was 73.33%. Finally, the CAM analysis revealed that the TRD group contained dominant features in class detection of deep learning architecture in almost all electrodes. Limitations: Our study is limited by the moderate sample size of clinical groups and the retrospective nature of the study. Conclusion: These findings suggest that EEG-based deep learning can be used to classify treatment resistance in depression and may in the future prove to be a useful tool in psychiatry practice to identify patients who need more vigorous intervention.
Determination Of Changes in Frequencies of EEG Signal in Eyes Open/Closed Duration
(IEEE, 2015) Turk, Omer; Ozerdem, Mehmet Sirac; Türk, Ömer
In this study, the changes of the power spectral density (PSD) in the EEG data during eyes-closed and eyes-open states were analyzed. In the analysis, the interval of dominant frequencies was roughly determined with different approaches. The EEG signal is separated into sub bands with wavelet transform (WT). The Welch method which is the one of the classical methods was used for PSD prediction and the Burg and Yule-Walker parametric methods were used also for PSD prediction of the EEG signal. It was observed that the alpha rhythm is dominant band in the eyes closed state compared to eyes open state.
A Hybrid 2d Gaussian Filter and Deep Learning Approach With Visualization of Class Activation for Automatic Lung and Colon Cancer Diagnosis
(Sage Publications inc, 2024) Turk, Omer; Acar, Emrullah; Irmak, Emrah; Yilmaz, Musa; Bakis, Enes; Türk, Ömer
Cancer is a significant public health issue due to its high prevalence and lethality, particularly lung and colon cancers, which account for over a quarter of all cancer cases. This study aims to enhance the detection rate of lung and colon cancer by designing an automated diagnosis system. The system focuses on early detection through image pre-processing with a 2D Gaussian filter, while maintaining simplicity to minimize computational requirements and runtime. The study employs three Convolutional Neural Network (CNN) models-MobileNet, VGG16, and ResNet50-to diagnose five types of cancer: Colon Adenocarcinoma, Benign Colonic Tissue, Lung Adenocarcinoma, Benign Lung Tissue, and Lung Squamous Cell Carcinoma. A large dataset comprising 25 000 histopathological images is utilized. Additionally, the research addresses the need for safety levels in the model by using Class Activation Mapping (CAM) for explanatory purposes. Experimental results indicate that the proposed system achieves a high diagnostic accuracy of 99.38% for lung and colon cancers. This high performance underscores the effectiveness of the automated system in detecting these types of cancer. The findings from this study support the potential for early diagnosis of lung and colon cancers, which can facilitate timely therapeutic interventions and improve patient outcomes.
MENTAL ACTIVITY DETECTION FROM EEG RECORDS USING LOCAL BINARY PATTERN METHOD
(IEEE, 2017) Turk, Omer; Ozerdem, Mehmet Sirac; Türk, Ömer
Electroencephalogram signals are widely used in the detection of different activities but not in the desired level. In this study with this motivation, it is aimed to obtain the attributes by using the Local Bilinear Pattern (LBP) method of EEG records for various mental activities and to classify these features by k-Nearest Neighbor (k-NN) method. The binary classification performance of these EEG records containing 5 mental tasks was evaluated. In addition, in order to evaluate classification performance, confusion matrix was used as model performance criterion. In the study, the average of the classification performance of all participants was found as 87.38%. As a model performance criterion from the participants' classification of mental activity, accuracy was 85.03%, precision was 85.40% and sensitivity was 85.47%. So, as a result the obtained results support the literature and the applicability of the LBP method for EEG markings has been confirmed.