Browsing by Author "Cura, Ozlem Karabiber"

Now showing 1 - 12 of 12

Citation - WoS: 1
Citation - Scopus: 2
Alzheimer's Dementia Detection: an Optimized Approach Using Itd of Eeg Signals
(IEEE, 2024) Sen, Sena Yagmur; Akan, Aydin; Cura, Ozlem Karabiber
This paper presents a novel early-stage Alzheimer's dementia (AD) disease detection based on convolutional neural networks (CNNs). As it is widely used in detection and classification of AD disease, a time-frequency (TF) method has been proposed for AD detection. It has been described to address the problem of detecting early-stage AD by combining TF and CNN methods. The method is developed by utilizing the well-known structural similarity index measure (SSIM) to obtain discriminative features in each TF image. Experimental results demonstrate that the proposed method outperforms the early-stage AD detection using advanced signal decomposition algorithm that is intrinsic time-scale decomposition (ITD), and it achieves a notable improvement in terms of the detection success rates compared to AD detection from TF images of raw EEG signals.
Citation - WoS: 14
Citation - Scopus: 19
Analysis of Epileptic Eeg Signals by Using Dynamic Mode Decomposition and Spectrum
(Elsevier, 2021) Cura, Ozlem Karabiber; Akan, Aydin
Dynamic mode decomposition (DMD) is a new matrix decomposition method proposed as an iterative solution to problems in fluid flow analysis. Recently, DMD algorithm has successfully been applied to the analysis of non-stationary signals such as neural recordings. In this study, we propose single-channel, and multi-channel EEG based DMD approaches for the analysis of epileptic EEG signals. We investigate the possibility of utilizing the DMD Spectrum for the classification of pre-seizure and seizure EEG segments. We introduce higher-order DMD spectral moments and DMD sub-band powers, and extract them as features for the classification of epileptic EEG signals. Experiments are conducted on multi-channel EEG signals collected from 16 epilepsy patients. Single-channel, and multichannel EEG based DMD approaches have been tested on epileptic EEG data recorded from only right, only left, and both brain hemisphere channels. Performance of various classifiers using the proposed DMD-Spectral based features are compared with that of traditional spectral features. Experimental results reveal that the higher order DMD spectral moments and DMD sub-band power features introduced in this study, outperform the analogous spectral features calculated from traditional power spectrum. (c) 2020 Nalecz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences. Published by Elsevier B.V. All rights reserved.
Citation - WoS: 12
Citation - Scopus: 13
Attention Deficit Hyperactivity Disorder Recognition Based on Intrinsic Time-Scale Decomposition of Eeg Signals
(Elsevier Sci Ltd, 2023) Cura, Ozlem Karabiber; Atli, Sibel Kocaaslan; Akan, Aydin
Attention deficit hyperactivity disorder (ADHD), a neuro-developmental condition, is characterized by various degrees of impulsivity, hyperactivity, and inattention. Treatment of this condition and minimizing its negative impact on learning, working, forming relationships, and quality of life depends heavily on the early identifi-cation. The Electroencephalography (EEG) is a useful neuroimaging technique for understanding ADHD. This study examines the brain activity of children with ADHD by analyzing the EEG signals using the intrinsic time-scale decomposition (ITD). Different combinations of the modes, known as Proper Rotation Components (PRCs), produced by ITD, are used to extract a variety of connectivity-based features (magnitude square coherence, cross power spectral density, correlation coefficient, covariance, cohentropy coefficient, correntropy coefficient). EEG signals of 15 ADHD children and 18 age-matched health children are recorded while resting with the eyes closed. Mentioned features are calculated using different channel pairs chosen from longitudinal and transversal planes. Through various machine learning approaches and a 10-fold cross-validation method, the proposed approach is evaluated to distinguish between ADHD patients and healthy controls. Classification accuracies are obtained for the longitudinal and transverse planes, between 92.90% to 99.90% and 91.70% to 100.00%, respectively. Our results support the remarkable performance of the proposed approach, and represent a substantial advance over similar studies in terms of recognizing and classifying ADHD.
Citation - Scopus: 5
Classification of Alzheimers' Dementia by Using Various Signal Decomposition Methods
(IEEE, 2021) Cura, Ozlem Karabiber; Yilmaz, Gulce Cosku; Ture, Hatice Sabiha; Akan, Aydin
Neurological disorders may spring from any disorder in the brain or the central and autonomic nervous systems. Among the neurological disorders, while Alzheimer's disease and other dementias are the fourth-largest contributors of disabilityadjusted life years, they are the second largest contributor of deaths. In the proposed study, various signal decomposition methods such as EMD, EEMD, and DWT are presented to classify EEG segments of control subjects and Alzheimer' dementia patients. Time-domain features are calculated using selected 7 IMFs and 5 detail and approximation coefficients of DWT. Various classification techniques namely Decision Tree (DT), Support Vector Machine (SVM), k- Nearest Neighbor (kNN), and Random Forest (RF) are utilized to distinguish two groups. Simulation results demonstrate that the proposed approaches achieve outstanding validation accuracy rates.
Citation - WoS: 28
Citation - Scopus: 29
Classification of Epileptic Eeg Signals Using Synchrosqueezing Transform and Machine Learning
(World Scientific Publ Co Pte Ltd, 2021) Cura, Ozlem Karabiber; Akan, Aydin
Epilepsy is a neurological disease that is very common worldwide. Patient's electroencephalography (EEG) signals are frequently used for the detection of epileptic seizure segments. In this paper, a high-resolution time-frequency (TF) representation called Synchrosqueezing Transform (SST) is used to detect epileptic seizures. Two different EEG data sets, the IKCU data set we collected, and the publicly available CHB-MIT data set are analyzed to test the performance of the proposed model in seizure detection. The SST representations of seizure and nonseizure (pre-seizure or inter-seizure) EEG segments of epilepsy patients are calculated. Various features like higher-order joint TF (HOJ-TF) moments and gray-level co-occurrence matrix (GLCM)-based features are calculated using the SST representation. By using single and ensemble machine learning methods such as k-Nearest Neighbor (kNN), Logistic Regression (LR), Naive Bayes (NB), Support Vector Machine (SVM), Boosted Trees (BT), and Subspace kNN (S-kNN), EEG features are classified. The proposed SST-based approach achieved 95.1% ACC, 96.87% PRE, 95.54% REC values for the IKCU data set, and 95.13% ACC, 93.37% PRE, 90.30% REC values for the CHB-MIT data set in seizure detection. Results show that the proposed SST-based method utilizing novel TF features outperforms the short-time Fourier transform (STFT)-based approach, providing over 95% accuracy for most cases, and compares well with the existing methods.
Citation - WoS: 2
Citation - Scopus: 6
Deep Time-Frequency Feature Extraction for Alzheimer's Dementia Eeg Classification
(IEEE, 2022) Cura, Ozlem Karabiber; Yilmaz, Gulce C.; Ture, H. Sabiha; Akan, Aydin
Alzheimer's Dementia (AD), one of the age-related neurological disorders, causes loss of cognitive functions and seriously affects the daily life of patients. Electroencephalogram (EEG) is one of the most frequently used clinical tools to investigate the effects of AD on the brain. In the proposed study, a time-frequency representation and deep feature extraction based model is introduced to distinguish EEG segments of control subjects and AD patients. TF representations of EEG segments are obtained using high-resolution SynchroSqueezing Transform (SST), and conventional short-time Fourier transform (STFT) methods. The magnitudes of SST and STFT are used for deep feature extraction. Various classifiers are used to classify the extracted features to distinguish the EEG segments of control subjects and AD patients. STFT based deep feature extraction approach yielded better classification results than that of the SST method.
Detection of Alzheimer's Dementia by Using Eeg Feature Maps and Deep Learning
(IEEE, 2024) Sude Pehlivan, Akbugday; Cura, Ozlem Karabiber; Akbugday, Burak; Akan, Aydin
One of the most frequent neurological conditions that impair cognitive abilities and have a major negative impact on quality of life is dementia. In this work, a novel approach for identifying Alzheimer's disease (AD) by utilizing electroencephalogram (EEG) signals via signal processing techniques is proposed. Five spectral domain characteristics are computed for one-minute EEG segment duration using EEG data. Each feature is mapped onto a 9 x 9 matrix called topographic EEG feature maps (EEG-FM) to represent spectral as well as spatial information on the same image. Images were then classified using a 2-layer convolutional neural network (CNN) to classify healthy and AD cases. Results indicate that the constructed CNN generalizes well, and the proposed method can accurately classify AD from EEG-FMs with up to %99 accuracy, precision, and recall with loss values as low as 0.01.
Citation - WoS: 14
Citation - Scopus: 21
Detection of Alzheimer's Dementia by Using Signal Decomposition and Machine Learning Methods
(World Scientific Publ Co Pte Ltd, 2022) Cura, Ozlem Karabiber; Akan, Aydin; Yilmaz, Gulce Cosku; Ture, Hatice Sabiha
Dementia is one of the most common neurological disorders causing defection of cognitive functions, and seriously affects the quality of life. In this study, various methods have been proposed for the detection and follow-up of Alzheimer's dementia (AD) with advanced signal processing methods by using electroencephalography (EEG) signals. Signal decomposition-based approaches such as empirical mode decomposition (EMD), ensemble EMD (EEMD), and discrete wavelet transform (DWT) are presented to classify EEG segments of control subjects (CSs) and AD patients. Intrinsic mode functions (IMFs) are obtained from the signals using the EMD and EEMD methods, and the IMFs showing the most significant differences between the two groups are selected by applying previously suggested selection procedures. Five-time-domain and 5-spectral-domain features are calculated using selected IMFs, and five detail and approximation coefficients of DWT. Signal decomposition processes are conducted for both 1 min and 5 s EEG segment durations. For the 1 min segment duration, all the proposed approaches yield prominent classification performances. While the highest classification accuracies are obtained using EMD (91.8%) and EEMD (94.1%) approaches from the temporal/right brain cluster, the highest classification accuracy for the DWT (95.2%) approach is obtained from the temporal/left brain cluster for 1 min segment duration.
Detection of Attention Deficit Hyperactivity Disorder Using Eeg Signals and Douglas-Peucker Algorithm
(IEEE, 2022) Cura, Ozlem Karabiber; Aydin, Gamze N.; Celen, Sibel; Atli, Sibel Kocaaslan; Akan, Aydin
Attention Deficit Hyperactivity Disorder (ADHD) is a neurological disease that typically appears in childhood. The disease has three main symptoms in children: inattention, hyperactivity, and impulsivity. Treatment of the disease is based on behavioral studies; however, there is no definitive diagnosis method. Hence, the electroencephalography (EEG) signals of ADHD subjects are often investigated to understand changes in the brain. In the proposed study, it is aimed to process and reduce the EEG data of ADHD and control subjects (CS) by using the Douglas-Peucker algorithm and to investigate the effects of the algorithm on EEG signal analysis. EEG data obtained from 18 control subjects (4 boys, 14 girls, mean age 13) and 15 ADHD patients (7 boys, 8 girls, mean age 12) are collected. By using reduced EEG data; time features such as energy, skewness, kurtosis, mean absolute deviation (MAD), root mean square (RMS), peak to peak (PTP) value, Hjorth parameters, and non-linear features such as largest Lyapunov Exponent (LLE), correlation dimension (CD), Hurst exponent (HE), Katz fractal dimension (KFD), Higuchi fractal dimension (HFD), are calculated to examine different signal characteristics. Extracted features are used to distinguish the EEG data of ADHD and CS by using various machine learning algorithms.
Citation - WoS: 73
Citation - Scopus: 85
Epileptic Eeg Classification by Using Time-Frequency Images for Deep Learning
(World Scientific Publ Co Pte Ltd, 2021) Ozdemir, Mehmet Akif; Cura, Ozlem Karabiber; Akan, Aydin
Epilepsy is one of the most common brain disorders worldwide. The most frequently used clinical tool to detect epileptic events and monitor epilepsy patients is the EEG recordings. There have been proposed many computer-aided diagnosis systems using EEG signals for the detection and prediction of seizures. In this study, a novel method based on Fourier-based Synchrosqueezing Transform (SST), which is a high-resolution time-frequency (TF) representation, and Convolutional Neural Network (CNN) is proposed to detect and predict seizure segments. SST is based on the reassignment of signal components in the TF plane which provides highly localized TF energy distributions. Epileptic seizures cause sudden energy discharges which are well represented in the TF plane by using the SST method. The proposed SST-based CNN method is evaluated using the IKCU dataset we collected, and the publicly available CHB-MIT dataset. Experimental results demonstrate that the proposed approach yields high average segment-based seizure detection precision and accuracy rates for both datasets (IKCU: 98.99% PRE and 99.06% ACC; CHB-MIT: 99.81% PRE and 99.63% ACC). Additionally, SST-based CNN approach provides significantly higher segment-based seizure prediction performance with 98.54% PRE and 97.92% ACC than similar approaches presented in the literature using the CHB-MIT dataset.
Citation - WoS: 59
Citation - Scopus: 72
Epileptic Seizure Classifications Using Empirical Mode Decomposition and Its Derivative
(Bmc, 2020) Cura, Ozlem Karabiber; Atli, Sibel Kocaaslan; Ture, Hatice Sabiha; Akan, Aydin
Background Epilepsy is one of the most common neurological disorders associated with disruption of brain activity. In the classification and detection of epileptic seizures, electroencephalography (EEG) measurements, which record the electrical activities of the brain, are frequently used. Empirical mode decomposition (EMD) and its derivative, ensemble EMD (EEMD) are recently developed methods used to decompose non-stationary and nonlinear signals such as EEG into a finite number of oscillations called intrinsic mode functions (IMFs). Our main objective in this study is to present a hybrid IMF selection method combining four different approaches (energy, correlation, power spectral distance, and statistical significance measures), and investigate the effect of selected IMFs extracted by EMD and EEMD on the classification. We have applied the proposed IMF selection approach on the classification of EEG signals recorded from epilepsy patients who are under treatment at our collaborator hospital. Multichannel EEG signals collected from epilepsy patients are decomposed into IMFs, and then IMF selection was performed. Finally, time- and spectral-domain, and nonlinear features are extracted and feature sets are created for the classification. Results The maximum classification accuracies obtained using various combinations of IMFs were 94.56%, 95.63%, 96.8%, and 96.25% for SVM, KNN, naive Bayes, and logistic regression classifiers, respectively, by using EMD analysis; whereas, the EEMD approach has provided maximum classification accuracies of 96.06%, 97%, 97%, and 96.25% for SVM, KNN, naive Bayes, and logistic regression, respectively. Classification performance with the same features obtained using direct EEG signals instead of the decomposed IMFs was worse than the aforementioned 2 approaches for every combination. Conclusion Simulation results demonstrate that the proposed IMF selection approach affects the classification results. Also, EEMD provides a robust method for feature extraction from EEG signals in order to classify pre-seizure and seizure segments.
Citation - WoS: 69
Citation - Scopus: 79
Time-Frequency Signal Processing: Today and Future
(Academic Press Inc Elsevier Science, 2021) Akan, Aydin; Cura, Ozlem Karabiber
Most real-life signals exhibit non-stationary characteristics. Processing of such signals separately in the time-domain or in the frequency-domain does not provide sufficient information as their spectral properties change over time. Traditional methods such as the Fourier transform (FT) perform a transformation from time-domain to frequency-domain allowing a suitable spectral analysis but looses the spatial/temporal information of the signal components. Hence, it is not easy to observe a direct relationship between the time and frequency characteristics of the signal. This makes it difficult to extract useful information by using only time- or frequency-domain analysis for further processing purposes. To overcome this problem, joint time-frequency (TF) methods are developed and applied to the analysis and representation of non-stationary signals. In addition to revealing a time-dependent energy distribution information, TF methods have successfully been utilized in the estimation of some parameters related to the analyzed signals. In this paper, we briefly summarize the existing methods and present several state-of-the-art applications of TF methods in the classification of biomedical signals. We also point out some future perspectives for the processing of non-stationary signals in the joint TF domain. (C) 2021 Elsevier Inc. All rights reserved.