Article
Chemistry, Analytical
Md Shafiqul Islam, Keshav Thapa, Sung-Hyun Yang
Summary: This study proposes a dynamic method using a deep learning model to detect epileptic seizures. The method achieves high accuracy through the use of dense convolutional blocks, feature attention modules, residual blocks, and hypercolumn technique.
Article
Neurosciences
Wei Zeng, Liangmin Shan, Bo Su, Shaoyi Du
Summary: This study aims to develop a new approach to automatically recognize seizures using EEG recordings. By constructing a new deep neural network (DNN) model, feature extraction is performed on raw data, and the resulting deep feature maps are fed into different shallow classifiers to detect anomalies. The analysis of EEG Epilepsy dataset and the Bonn dataset for epilepsy shows that the proposed method is effective and robust. The results of this study indicate that our methodology outperforms other up-to-date approaches and has potential application in clinical practice as well.
FRONTIERS IN NEUROSCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Kazi Mahmudul Hassan, Md Rabiul Islam, Thanh Thi Nguyen, Md Khademul Islam Molla
Summary: The study introduces a novel method for epileptic seizure detection that combines multiple techniques, showing significant performance improvement in comparison to current state-of-the-art methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Interdisciplinary Applications
Satarupa Chakrabarti, Aleena Swetapadma, Prasant Kumar Pattnaik
Summary: The study proposes an automatic epileptic seizure detection system based on a long short-term memory network, which achieves high sensitivity and low false positive rate, and is simple and effective. The convergence analysis shows that the proposed model has reliability and accuracy in detecting epileptic seizures.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2021)
Article
Computer Science, Information Systems
Chintalpudi S. L. Prasanna, Md Zia Ur Rahman, Masreshaw D. Bayleyegn
Summary: The study implemented a Brain Epilepsy Seizure Detection Network using deep learning and recurrent learning properties to automatically identify brain seizures through EEG signals, achieving high precision, sensitivity, F1-Score, accuracy, and specificity values for disease detection. This proposed method outperformed existing technologies in terms of performance and efficiency.
Article
Engineering, Mechanical
Zhenxi Song, Bin Deng, Yulin Zhu, Lihui Cai, Jiang Wang, Guosheng Yi
Summary: The study presents a method called Recursive State-Space Neural Network (RSSNN), that can automatically detect EEG signals by inferring the intrinsic geometry and identifying different epileptic patterns. Validated on a public EEG dataset, RSSNN achieves an accuracy of 99.79% at the EEG segment level and 100% at the subject level.
NONLINEAR DYNAMICS
(2023)
Article
Computer Science, Interdisciplinary Applications
Salim Rukhsar, Anil Kumar Tiwari
Summary: This study introduces a novel lightweight Convolution Transformer model that can effectively detect seizures in cross-patient learning, enhancing performance through the inclusion of inductive biases and attention-based pooling.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2023)
Article
Computer Science, Information Systems
Gopal Chandra Jana, Keshav Swami, Anupam Agrawal
Summary: The objective of this study is to propose an approach using Capsule Neural Network (CapsNet) to classify seizure and non-seizure EEG signals through subject specific and cross-subject training and testing. The proposed approach achieved high accuracy in both subject specific and cross-subject experiments, showing the advantage of CapsNet in certain data scenarios. Overall, the performance of the proposed approach improved compared to existing approaches.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Acoustics
Waqar Hussain, Muhammad Tariq Sadiq, Siuly Siuly, Ateeq Ur Rehman
Summary: This study proposes a deep learning-based method for seizure detection, which combines convolutional neural networks and long short-term memory models to generate customized features that can automatically generalize across different patients for real-time detection. Testing and comparing on 21 subjects, the decomposition of signals into time, frequency, and time-frequency features significantly improves the accuracy of seizure detection.
Article
Computer Science, Artificial Intelligence
Mrutyunjaya Sahani, Susanta Kumar Rout, Pradipta Kishore Dash
Summary: In this study, an optimized method combining variational mode decomposition, reduced deep convolutional neural network, and multi-kernel random vector functional link network was proposed for epileptic seizure recognition using EEG signals. The method achieved superior performance in classifying seizure epochs compared to other state-of-the-art methods.
APPLIED SOFT COMPUTING
(2021)
Article
Engineering, Biomedical
Xuanjie Qiu, Fang Yan, Haihong Liu
Summary: This paper proposes a new hybrid deep learning model DARLNet for epileptic seizure detection. The model uses a combination of Residual Neural Network (ResNet) and Long Short-Term Memory Network (LSTM) to capture spatial correlations and temporal dependencies respectively. It also incorporates a difference layer to mine additional epileptic seizure information and a channel attention module to focus on seizure-relevant information.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Artificial Intelligence
Hong Peng, Cancheng Li, Jinlong Chao, Tao Wang, Chengjian Zhao, Xiaoning Huo, Bin Hu
Summary: This study proposes a novel sparse representation-based epileptic seizure classification method based on dictionary learning, which is evaluated on public EEG databases. The new method shows higher automation and recognition rates compared to traditional methods.
Article
Chemistry, Analytical
Kunpeng Song, Jiajia Fang, Lei Zhang, Fangni Chen, Jian Wan, Neal Xiong
Summary: In this paper, an intelligent epileptic prediction system based on Synchrosqueezed Wavelet Transform (SWT) and Multi-Level Feature Convolutional Neural Network (MLF-CNN) was proposed for smart healthcare IoT network. The system achieved high accuracy and sensitivity on both public and private datasets.
Article
Computer Science, Artificial Intelligence
Jian Lian, Fangzhou Xu
Summary: Feature extraction is crucial in epilepsy detection and recognition. In this study, a graph convolutional neural network-based framework is proposed to capture the spatial enhanced pattern of multichannel EEG signals, characterizing the behavior of EEG activity and visualizing salient regions. This approach can also be used as a novel classifier for distinguishing different types of EEGs, achieving high sensitivity, specificity, and accuracy.
INTERNATIONAL JOURNAL OF NEURAL SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Abhijeet Bhattacharya, Tanmay Baweja, S. P. K. Karri
Summary: The electroencephalogram (EEG) is an effective technique to study epilepsy, and automated screening using a combination of signal processing and deep learning shows superior performance and potential.
INTERNATIONAL JOURNAL OF NEURAL SYSTEMS
(2022)
Article
Computer Science, Interdisciplinary Applications
Erdem Yavuz, Can Eyupoglu
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING
(2020)
Article
Materials Science, Textiles
Can Eyupoglu, Seyda Eyupoglu, Nigar Merdan
Summary: This study investigated the use of polyester fabric made from microfibers as siding material in the construction industry, exploring the UV absorbance capacity and effects of dyeing process on the samples. By applying UV absorber and utilizing MLP-ANN model, the UV protection properties of polyester microfiber fabric were accurately predicted, demonstrating high regression values for all properties.
JOURNAL OF THE TEXTILE INSTITUTE
(2021)
Article
Computer Science, Information Systems
Erdem Yavuz
Summary: This study introduces a novel parallel processing architecture for accelerating image encryption based on chaos, utilizing multiple chaotic ciphers running concurrently to process partitions and optimizing encryption speed. The proposed architecture incorporates powerful output mixing logic and simple operations to ensure data diffusion, while leveraging loop-level parallelism to execute blending operations on independent encryption threads concurrently.
JOURNAL OF INFORMATION SECURITY AND APPLICATIONS
(2021)
Article
Materials Science, Textiles
Can Eyupoglu, Seyda Eyupoglu, Nigar Merdan
Summary: This study explores the use of ascorbic acid as a substitute for improving fastness properties of natural dyes, with the mordanting process using microwave energy. The results show that the color strength, washing, light, and rubbing fastness of dyed mohair fiber improve slightly with the premordanting process and by adding ascorbic acid. In addition, a machine learning-based model using artificial neural network (ANN) was developed for the prediction of dyeing properties of mohair fiber dyed with natural dyes.
JOURNAL OF NATURAL FIBERS
(2022)
Article
Materials Science, Textiles
Seyda Eyupoglu, Cengiz Karabulut, Serdar Erdem Gul, Ahmet Tamer Esener, Firat Yilmaz, Mazyar Ahrari, Can Eyupoglu, Dilek Kut
Summary: This study investigated the wrinkle resistance of cellulose acetate-based fabrics treated with a resin-based treatment. Oxygen plasma pre-treatment was applied to enhance the wrinkle resistance, resulting in improved wrinkle resistance angle and tensile strength. The surface morphology and functional groups of the samples were analyzed. The results showed that the plasma treatment generated microscopic grooves and micro-cracks on the sample surfaces, while no significant alterations in functional groups were observed. The study concluded that the combination of plasma pre-treatment and wrinkle-resistance resin can provide both eco-friendly and high-quality finishing for textile materials.
JOURNAL OF NATURAL FIBERS
(2022)
Article
Materials Science, Textiles
Seyda Eyupoglu, Can Eyupoglu
Summary: This study examined the properties of fibers extracted from beetroot plants using chemical, physical, and instrumental tests, including scanning electron microscopy, energy dispersive X-ray analysis, X-ray diffraction, and thermogravimetric analysis. The results indicate that the beetroot fibers have suitable mechanical behavior and chemical characteristics.
JOURNAL OF NATURAL FIBERS
(2022)
Article
Chemistry, Applied
Zeynep Omerogullari Basyigit, Can Eyupoglu, Seyda Eyupoglu, Nigar Merdan
Summary: In the colouring processes of textile products, alternative methods should be used to protect the environment. This study used ultrasonic dyeing and air vacuum plasma treatment to increase dye absorption in wool fabric, achieving a green and effective process.
COLORATION TECHNOLOGY
(2023)
Article
Chemistry, Applied
Can Eyupoglu, Seyda Eyupoglu, Nigar Merdan, Zeynep Omerogullari Basyigit
Summary: This study investigated the ecological dyeing process of wool fabrics through plasma treatment and extraction from Rubia tinctorum. The effects of plasma treatment and dyeing time on the properties of wool fibres were analyzed using microscopy and infrared analysis. An artificial neural network model was proposed to estimate the dyeing properties of wool fabrics. The experimental results show that the proposed model achieves high regression values for all dyeing properties.
COLORATION TECHNOLOGY
(2023)
Article
Energy & Fuels
Seyda Eyupoglu, Can Eyupoglu, Nigar Merdan
Summary: This study aims to find a sustainable alternative to man-made fibers by using Sambucus ebulus L. stem fibers as reinforcement in polymer composites. The characterization of the stem fibers revealed their chemical composition, fiber diameter, elemental composition, crystalline index, and functional groups. The results showed that Sambucus ebulus L. stem fibers have properties comparable to other natural fibers, suggesting their potential use as a substitute for man-made fibers in composites.
BIOMASS CONVERSION AND BIOREFINERY
(2023)
Article
Computer Science, Information Systems
Burak Cem Kara, Can Eyupoglu
Summary: The study proposes a new data anonymization algorithm that incorporates an outlier data detection mechanism to boost data utility. The algorithm outperforms existing methods in multiple metrics and effectively handles high-dimensional datasets.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Proceedings Paper
Engineering, Aerospace
Can Eyupoglu
Summary: Power systems have a direct impact on the operation of satellites. This study proposes a machine learning-based approach for fault diagnosis of geosynchronous satellite power systems, which is shown to be effective.
2023 10TH INTERNATIONAL CONFERENCE ON RECENT ADVANCES IN AIR AND SPACE TECHNOLOGIES, RAST
(2023)
Proceedings Paper
Engineering, Aerospace
Fatma Gumus, Can Eyupoglu
Summary: This article provides a literature review on the motivation and methods of aircraft cockpit dashboard image processing, highlighting the unique challenges of dynamic flight environments such as lighting and vibration. It also aims to identify methodologies from needle-type instrument reading research that can be potentially applied to this unexplored application area.
2023 10TH INTERNATIONAL CONFERENCE ON RECENT ADVANCES IN AIR AND SPACE TECHNOLOGIES, RAST
(2023)
Proceedings Paper
Computer Science, Theory & Methods
Esra Kutlugun, Can Eyupoglu
2020 5TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND ENGINEERING (UBMK)
(2020)
Proceedings Paper
Engineering, Electrical & Electronic
Ufuk Sanver, Erdem Yavuz, Can Eyupoglu
PROCEEDINGS OF THE 2019 IEEE CONFERENCE OF RUSSIAN YOUNG RESEARCHERS IN ELECTRICAL AND ELECTRONIC ENGINEERING (EICONRUS)
(2019)