Elektrik ve Enerji Bölümü Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12514/174
Browse
Browsing Elektrik ve Enerji Bölümü Koleksiyonu by WoS Q "Q2"
Now showing 1 - 8 of 8
- Results Per Page
- Sort Options
Article Classification and analysis of epileptic EEG recordings using convolutional neural network and class activation mapping(2021) Yildiz, Abdulnasir; Zan, Hasan; Said, SherifElectrical bio-signals have the potential to be used in different applications due to their hidden nature and their ability to facilitate liveness detection. This paper investigates the feasibility of using the Convolutional Neural Network (CNN) to classify and analyze electroencephalogram (EEG) data with their time-frequency representations and class activation mapping (CAM) to detect epilepsy disease. Several types of pre-trained CNNs are employed for a multi-class classification task (AlexNet, GoogLeNet, ResNet-18, and ResNet-50) and their results are compared. Also, a novel convolutional neural network architecture comprised of two horizontally concatenated GoogLeNets is proposed with two inputs scalograms and spectrogram of the eplictic EEG signal. Four segment lengths (4097, 2048, 1024, and 512 sampling points) with three time-frequency representations (short-time Fourier, Wavelet, and Hilbert-Huang transform) are statistically evaluated. The dataset used in this research is collected at the University of Bonn. The dataset is reorganized as normal, interictal, and ictal. The maximum achieved accuracies for 4097, 2048, 1024, and 512 sampling points are 100 %, 100 %, 100 %, and 99.5 % respectively. The CAM method is used to analyze discriminative regions of time-frequency representations of EEG segments and networks' decisions. This method showed CNN models used different time and frequency regions of input images for each class with correct and incorrect predictions.Article Citation - WoS: 23Citation - Scopus: 36Classification and analysis of epileptic EEG recordings using convolutional neural network and class activation mapping(Biomedical Signal Processing and Control, 2021) Zan, Hasan; Yıldız, Abdulnasir; Said, SherifElectrical bio-signals have the potential to be used in different applications due to their hidden nature and their ability to facilitate liveness detection. This paper investigates the feasibility of using the Convolutional Neural Network (CNN) to classify and analyze electroencephalogram (EEG) data with their time-frequency representations and class activation mapping (CAM) to detect epilepsy disease. Several types of pre-trained CNNs are employed for a multi-class classification task (AlexNet, GoogLeNet, ResNet-18, and ResNet-50) and their results are compared. Also, a novel convolutional neural network architecture comprised of two horizontally concatenated GoogLeNets is proposed with two inputs scalograms and spectrogram of the eplictic EEG signal. Four segment lengths (4097, 2048, 1024, and 512 sampling points) with three time-frequency representations (short-time Fourier, Wavelet, and Hilbert-Huang transform) are statistically evaluated. The dataset used in this research is collected at the University of Bonn. The dataset is reorganized as normal, interictal, and ictal. The maximum achieved accuracies for 4097, 2048, 1024, and 512 sampling points are 100 %, 100 %, 100 %, and 99.5 % respectively. The CAM method is used to analyze discriminative regions of time-frequency representations of EEG segments and networks' decisions. This method showed CNN models used different time and frequency regions of input images for each class with correct and incorrect predictions.Article Citation - WoS: 32Citation - Scopus: 32Development software program for extraction of photovoltaic cell equivalent circuit model parameters based on the Newton–Raphson method(SpringerLink, 2022) Adak, Süleyman; Cangi, Hasan; Yılmaz, Ahmet Serdar; Arifoğlu, UğurFinding the equivalent circuit parameters for photovoltaic (PV) cells is crucial as they are used in the modeling and analysis of PV arrays. PV cells are made of silicon. These materials have a nonlinear characteristic. This distorts the sinusoidal waveform of the current and voltage. As a result, harmonic components are formed in the system. The PV cell is the smallest building block of the PV system and produces voltages between 0.5 V and 0.7 V. It serves as a source of current. The amount of radiation hitting the cell determines how much current it produces. In an ideal case, a diode and a parallel current source make up the equivalent circuit of the PV cell. In practice, the addition of a series and parallel resistor is made to the ideal equivalent circuit. There are many equivalent circuits in the literature on modeling the equivalent circuit of a PV cell. The PV cell single–diode model is the most used model due to its ease of analysis. In this study, the iterative method by Newton–Raphson was used to find the equivalent circuit parameters of a PV cell. This method is one of the most widely used methods for determining the roots of nonlinear equations in numerical analysis. In this study, five unknown parameters (Iph, Io, Rs, Rsh and m) of the PV cell equivalent circuit were quickly discovered with the software program prepared based on the Newton–Raphson method in MATLABArticle Local Pattern Transformation-Based convolutional neural network for sleep stage scoring(2023) Zan, Hasan; Yıldız, AbdulnasırSleep stage scoring is essential for the diagnosis and treatment of sleep disorders. However, manual sleep scoring is a tedious, time-consuming, and subjective task. Therefore, this paper proposes a novel framework based on local pattern transformation (LPT) methods and convolutional neural networks for automatic sleep stage scoring. Unlike in previous works in other fields, these methods were not employed for manual feature extraction, which requires expert knowledge and the pipeline behind it might bias results. The transformed signals were directly fed into a CNN model (called EpochNet) that can accept multiple successive epochs. The model learns features from multiple input epochs and considers inter-epoch context during classification. To evaluate and validate the effectiveness of the proposed approach, we conducted several experiments on the Sleep-EDF dataset. Four LPT methods, including One-dimensional Local Binary Pattern (1D-LBP), Local Neighbor Descriptive Pattern (LNDP), Local Gradient Pattern (LGP), and Local Neighbor Gradient Pattern (LNGP), and different polysomnography (PSG) signals were analyzed as sequence length (number of input epochs) increased from one to five. 1D-LBP and LNDP achieved similar performances, outperforming other LPT methods that are less sensitive to local variations. The best performance was achieved when an input sequence containing five epochs of PSG signals transformed by 1D-LBP was employed. The best accuracy, F1 score, and Kohen's kappa coefficient were 0.848, 0.782, and 0.790, respectively. The results showed that our approach can achieve comparable performance to other state-of-the-art methods while occupying fewer computing resources because of the compact size of EpochNet.Article Citation - WoS: 12Citation - Scopus: 14Local Pattern Transformation-Based convolutional neural network for sleep stage scoring(ScienceDirect, 2023) Zan, Hasan; Yildiz, AbdulnasırSleep stage scoring is essential for the diagnosis and treatment of sleep disorders. However, manual sleep scoring is a tedious, time-consuming, and subjective task. Therefore, this paper proposes a novel framework based on local pattern transformation (LPT) methods and convolutional neural networks for automatic sleep stage scoring. Unlike in previous works in other fields, these methods were not employed for manual feature extraction, which requires expert knowledge and the pipeline behind it might bias results. The transformed signals were directly fed into a CNN model (called EpochNet) that can accept multiple successive epochs. The model learns features from multiple input epochs and considers inter-epoch context during classification. To evaluate and validate the effectiveness of the proposed approach, we conducted several experiments on the Sleep-EDF dataset. Four LPT methods, including One-dimensional Local Binary Pattern (1D-LBP), Local Neighbor Descriptive Pattern (LNDP), Local Gradient Pattern (LGP), and Local Neighbor Gradient Pattern (LNGP), and different polysomnography (PSG) signals were analyzed as sequence length (number of input epochs) increased from one to five. 1D-LBP and LNDP achieved similar performances, outperforming other LPT methods that are less sensitive to local variations. The best performance was achieved when an input sequence containing five epochs of PSG signals transformed by 1D-LBP was employed. The best accuracy, F1 score, and Kohen’s kappa coefficient were 0.848, 0.782, and 0.790, respectively. The results showed that our approach can achieve comparable performance to other state-ofthe-art methods while occupying fewer computing resources because of the compact size of EpochNet.Article Multi-task learning for arousal and sleep stage detection using fully convolutional networks(2023) Zan, Hasan; Yıldız, AbdulnasırObjective. Sleep is a critical physiological process that plays a vital role in maintaining physical and mental health. Accurate detection of arousals and sleep stages is essential for the diagnosis of sleep disorders, as frequent and excessive occurrences of arousals disrupt sleep stage patterns and lead to poor sleep quality, negatively impacting physical and mental health. Polysomnography is a traditional method for arousal and sleep stage detection that is time-consuming and prone to high variability among experts. Approach. In this paper, we propose a novel multi-task learning approach for arousal and sleep stage detection using fully convolutional neural networks. Our model, FullSleepNet, accepts a full-night single-channel EEG signal as input and produces segmentation masks for arousal and sleep stage labels. FullSleepNet comprises four modules: a convolutional module to extract local features, a recurrent module to capture long-range dependencies, an attention mechanism to focus on relevant parts of the input, and a segmentation module to output final predictions. Main results. By unifying the two interrelated tasks as segmentation problems and employing a multi-task learning approach, FullSleepNet achieves state-of-the-art performance for arousal detection with an area under the precision-recall curve of 0.70 on Sleep Heart Health Study and Multi-Ethnic Study of Atherosclerosis datasets. For sleep stage classification, FullSleepNet obtains comparable performance on both datasets, achieving an accuracy of 0.88 and an F1-score of 0.80 on the former and an accuracy of 0.83 and an F1-score of 0.76 on the latter. Significance. Our results demonstrate that FullSleepNet offers improved practicality, efficiency, and accuracy for the detection of arousal and classification of sleep stages using raw EEG signals as input.Article Citation - WoS: 11Citation - Scopus: 13Multi-task learning for arousal and sleep stage detection using fully convolutional networks(IOP Publishing, 2023) Zan, Hasan; Yıldız, AbdulnasirObjective: Sleep is a critical physiological process that plays a vital role in maintaining physical and mental health. Accurate detection of arousals and sleep stages is essential for the diagnosis of sleep disorders, as frequent and excessive occurrences of arousals disrupt sleep stage patterns and lead to poor sleep quality, negatively impacting physical and mental health. Polysomnography is a traditional method for arousal and sleep stage detection that is time-consuming and prone to high variability among experts. Approach: In this paper, we propose a novel multi-task learning approach for arousal and sleep stage detection using fully convolutional neural networks. Our model, FullSleepNet, accepts a full-night single-channel EEG signal as input and produces segmentation masks for arousal and sleep stage labels. FullSleepNet comprises four modules: a convolutional module to extract local features, a recurrent module to capture long-range dependencies, an attention mechanism to focus on relevant parts of the input, and a segmentation module to output final predictions. Main results: By unifying the two interrelated tasks as segmentation problems and employing a multi-task learning approach, FullSleepNet achieves state-of-the-art performance for arousal detection with an area under the precision-recall curve of 0.70 on Sleep Heart Health Study and Multi-Ethnic Study of Atherosclerosis datasets. For sleep stage classification, FullSleepNet obtains comparable performance on both datasets, achieving an accuracy of 0.88 and an F1-score of 0.80 on the former and an accuracy of 0.83 and an F1-score of 0.76 on the latter. Significance: Our results demonstrate that FullSleepNet offers improved practicality, efficiency, and accuracy for the detection of arousal and classification of sleep stages using raw EEG signals as input.Article Physical properties of solution processable n-type Fe and Al co-doped ZnO nanostructured thin films: Role of Al doping levels and annealing(ELSEVIER, 2018) BOZ, İbrahim; GÖKTAŞ, Abdullah; ASLAN, Ferhat; YEŞİLATA, BülentThe role of annealing temperature and Fe and Al co-doping on structural, optical, electrical and magnetic properties of solution processable ZnO thin films were investigated. ZnO:Fe thin films fixed with 2% of typical ferrous component were obtained to examine the role of 1–10% Al doping. X-ray diffraction analyses clearly indicates that the films to be polycrystalline and preferentially oriented along the c-axis of the hexagonal wurtzite structure. The film thickness, homogeneous distribution and decreasing/increasing of grain size dependence on Al content/annealing temperature (TA) were assessed by scanning electron microscopy. X-ray photoelectron spectroscopy revealed that Al3+ and Fe2+ ions to substitute for Zn2+ without changing the wurtzite structure. A slight decrease in the optical band gap of ZnO at fixed Fe dopant and a considerable increase of the optical band gap with increased Al doping concentrations and TA were observed. The refractive index increases with the fixed Fe dopant level and then decreases by Al doping levels, whereas the extinction coefficient clearly increases depended on both of Fe and Al concentrations. The refractive index and extinction coefficient both decrease with TA. Hall measurements show n-type conductivity and the increase of charge carrier concentration by Al doping levels and TA. Magnetic studies indicate room temperature ferromagnetism in Al and Fe co-doped ZnO thin films, whereas no room temperature ferromagnetism for the Fe-doped ZnO thin films was observed. An enhanced room temperature ferromagnetism in Al and Fe co-doped ZnO thin films was observed to depend on TA.
