Yazar "Acir N." seçeneğine göre listele
Listeleniyor 1 - 9 / 9
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe An adaptive noise canceller based on QLMS algorithm for removing EOG artifacts in EEG recordings(Institute of Electrical and Electronics Engineers Inc., 2017) Mengüc E.C.; Acir N.In this paper, a novel adaptive noise canceller (ANC) based on the quaternion valued least mean square algorithm (QLMS) is designed in order to remove electrooculography (EOG) artifacts from electroencephalography (EEG) recordings. The measurement real-valued EOG and EEG signals (FP1, FP2, AF3 and AF4) are first modeled as four-dimensional processes in the quaternion domain. The EOG artifacts are then removed from the EEG signals in the quaternion domain by using the ANC based on QLMS algorithm. The quaternion representation of these signals allows us to remove EOG artifacts from all channels at the same time instead of removing the EOG artifacts in each EEG recordings separately. The simulation results support the proposed approach. © 2017 IEEE.Öğe Automatic detection of epileptiform events in EEG by a three-stage procedure based on artificial neural networks(2005) Acir N.; Öztura I.; Kuntalp M.; Baklan B.; Güzeliş C.This paper introduces a three-stage procedure based on artificial neural networks for the automatic detection of epileptiform events (EVs) in a multichannel electroencephalogram (EEG) signal. In the first stage, two discrete perceptrons fed by six features are used to classify EEG peaks into three subgroups: 1) definite epileptiform transients (ETs); 2) definite non-ETs; and 3) possible ETs and possible non-ETs. The pre-classification done in the first stage not only reduces the computation time but also increases the overall detection performance off the procedure. In the second stage, the peaks falling into the third group are aimed to be separated from each other by a nonlinear artificial neural network that would function as a postclassifier whose input is a vector of 41 consecutive sample values obtained from each peak. Different networks, i.e., a backpropagation multilayer perceptron and two radial basis function networks trained by a hybrid method and a support vector method, respectively, are constructed as the postclassifier and then compared in terms of their classification performances. In the third stage, multichannel information is integrated into the system for contributing to the process of identifying an EV by the electroencephalographers (EEGers). After the integration of multichannel information, the overall performance of the system is determined with respect to EVs. Visual evaluation, by two EEGers, of 19 channel EEG records of 10 epileptic patients showed that the best performance is obtained with a radial basis support vector machine providing an average sensitivity of 89.1%, an average selectivity of 85.9%, and a false detection rate (per hour) of 7.5.Öğe Automatic removal of ocular artefacts in EEG signal by using independent component analysis and artificial neural network(Institute of Electrical and Electronics Engineers Inc., 2017) Cinar S.; Menguc E.C.; Acir N.Ocular artefacts caused by eye movements can distort Electroencephalogram (EEG) recordings. It is important to obtain clean EEG signals in diagnosing and interpreting diseases. Meaningful EEG signals should not be distorted during the removal of artefacts. In this study, Independent Component Analysis and Artificial Neural Network were used together to remove ocular artefacts. The method was tested by using the real dataset. The Relative Error (RE) and Correlation Coefficient (CC) was used to test the performance of the method. Relative error = 0.227±0.229 and correlation coefficient = 0.941 ±0.088 was calculated in the performance analysis. According to the results, the proposed method is successful in removing ocular artefacts in EEG signals. © 2017 IEEE.Öğe Automatic removal of ocular artefacts in EEG signal by using independent component analysis and Chauvenet criterion [Baglmslz bileşen analizi ve chauvenet kriteri kullanarak EEG sinyallerindeki oktiler artefaktlan otomatik yok etme](Institute of Electrical and Electronics Engineers Inc., 2017) Cinar S.; Acir N.Eye movements (saccade, blink and etc.) cause artefacts in Electroencephalogram recordings. The ocular artefact can distort the EEG signals. Removal of ocular artefact is important issue in EEG signal analysis. The main task of artefact removal algorithms is to obtain cleaned EEG without losing meaningful EEG signal. The main focus of this work is to remove ocular artefact automatically by using Independent Component Analysis and Chauvenet criterion. The method is tested on real dataset. Relative error and Correlation coefficient are used for the performance test. The performance of the proposed method was Relative error= 0.273±0.148, Correlation coefficients 0.943± 0.042 in the dataset. The results show that the porposed method effectively removes ocular artefacts in EEG. © 2016 IEEE.Öğe Complex-valued least mean Kurtosis adaptive filter algorithm [Kompleks-Degerli En Küçük Ortalama Kurtosis Adaptif Filtre Algoritmasi](Institute of Electrical and Electronics Engineers Inc., 2016) Mengüç E.C.; Acir N.In this study, a complex-valued least mean Kurtosis (CLMK) adaptive filter algorithm is designed for processing complex-valued signals. The performance of the designed algorithm is tested on a complex-valued system identification and compared the complex-valued least mean square (CLMS) and complex-valued normalized least mean square (CNLMS) algorithms. As a result, the CLMK algorithm shows a higher performance than the other algorithms in terms of the convergence rate, mean square error (MSE) and mean square deviation (MSD). © 2016 IEEE.Öğe Deep neural network training with iPSO algorithm [IPSO algoritmasi ile derin sinir agi egitimi](Institute of Electrical and Electronics Engineers Inc., 2018) Kosten M.M.; Barut M.; Acir N.Deep learning-based methods are frequently preferred in many areas in recent years. Another issue, which is as important as deep neural networks applications, is the training of deep neural networks. Although many techniques are proposed in the literature for the training of deep nets, most of these techniques use gradient descent based approaches. In this study, differently from the conventional gradient method, Improved Particle Swam Optimisation (IPSO) algorithm is used for the training of deep neural networks. LeNet-5 network is preferred as network structure and MNIST is utilized as data set. Depending on the number of particles, a performance of up to 96.29% was achieved. In the cases after 20 particles, the average performance was over 90%. © 2018 IEEE.Öğe Design of a nonlinear adaptive infinite impulse response filter [Dogrusal olmayan sonsuz darbe cevapli adaptif filtre tasarimi](2007) Acir N.This study introduces a wavelet network based adaptive IIR filtering system satisfying asymptotic stability in the sense of Lyapunov. The proposed system is also integrated with the advantages of the time-frequency specific properties of wavelet networks in its dynamics. Two well known experiments are proposed. The results are verified that the proposed filtering system not only performs better than conventional gradient descent based algorithms but also performs as good as other stability theory based optimization algorithms.Öğe Kurtosis-Based CRTRL Algorithms for Fully Connected Recurrent Neural Networks(Institute of Electrical and Electronics Engineers Inc., 2018) Menguc E.C.; Acir N.In this paper, kurtosis-based complex-valued real-time recurrent learning (KCRTRL) and kurtosis-based augmented CRTRL (KACRTRL) algorithms are proposed for training fully connected recurrent neural networks (FCRNNs) in the complex domain. These algorithms are designed by minimizing the cost functions based on the kurtosis of a complex-valued error signal. The KCRTRL algorithm exploits the circularity properties of the complex-valued signals, and this algorithm not only provides a faster convergence rate but also results in a lower steady-state error. However, the KCRTRL algorithm is suboptimal in the processing of noncircular (NC) complex-valued signals. On the other hand, the KACRTRL algorithm contains a complete second-order information due to the augmented statistics, thus considerably improves the performance of the FCRNN in the processing of NC complex-valued signals. Simulation results on the one-step-ahead prediction problems show that the proposed KCRTRL algorithm significantly enhances the performance for only circular complex-valued signals, whereas the proposed KACRTRL algorithm provides more superior performance than existing algorithms for NC complex-valued signals in terms of the convergence rate and the steady-state error. © 2012 IEEE.Öğe Wavelet denoising of middle latency response in auditory evoked potentials(2011) Erkan Y.; Acir N.This paper presents a new approach for obtaining Middle Latency Response (MLR) in auditory evoked potentials. In this study, we first generated synthetic single trial MLR data at some specified noise levels by using gamma-tone function technique and then applied denoising procedure to the noisy MLR synthetic data. We have also compared the denoised MLR and ensemble averaged MLR at different noise levels. Simulation results have been demonstrated that denoised MLR with wavelet transform can be used to reduce single trial numbers for obtaining MLR data. So, the measurement time in clinics will be reduced. © 2011 IEEE.
