Article
Computer Science, Artificial Intelligence
Lokesh Malviya, Sandip Mal
Summary: This study proposes a hybrid deep learning model for detecting stress levels in humans using EEG signals. By utilizing DWT for signal denoising and decomposition, and combining CNN and BLSTM for feature selection and classification, the proposed method achieves high accuracy in detecting stress levels.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Chemistry, Analytical
Jammisetty Yedukondalu, Lakhan Dev Sharma
Summary: This study presents an automated method for removing electrooculogram (EOG) artifacts from electroencephalogram (EEG) signals. The approach decomposes the contaminated signals into intrinsic mode functions (IMFs) using Circulant Singular Spectrum Analysis (CiSSA) and removes the artifact components using 4-level discrete wavelet transform (DWT). The proposed technique effectively eliminates EOG artifacts while preserving low-frequency EEG information.
Article
Computer Science, Artificial Intelligence
N. J. Sairamya, M. S. P. Subathra, S. Thomas George
Summary: A computer-aided diagnosis method using RLNDiP technique for Schizophrenia was proposed in this study, achieving a maximum accuracy of 100% with the fusion of alpha brain rhythm and TD features in an artificial neural network, surpassing existing methods in classification performance by selecting effective connectivity features.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Dustin Carrion-Ojeda, Rigoberto Fonseca-Delgado, Israel Pineda
Summary: This study focuses on the analysis of factors influencing the performance of a biometric system based on electroencephalogram signals. Different classifiers were used to compare decomposition levels and examine the significance of recording time, showing that SVM and AdaBoost are the most effective for this specific problem. The study highlights the unique nature of EEG signals and the potential for their use in developing robust biometric systems.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Multidisciplinary Sciences
Jothi Letchumy Mahendra Kumar, Mamunur Rashid, Rabiu Muazu Musa, Mohd Azraai Mohd Razman, Norizam Sulaiman, Rozita Jailani, Anwar P. P. Abdul Majeed
Summary: This study evaluates the effectiveness of different TL models in extracting features for wink-based EEG signal classification, ultimately identifying Inception ResNetV2 with an optimized RF pipeline as the best combination, achieving a 100% classification accuracy on both training and validation datasets. Implementation of this proposed architecture in a BCI system could potentially improve the quality of life for post-stroke patients.
Article
Computer Science, Artificial Intelligence
Hong He, Xinyue Liu, Yong Hao
Summary: The PDWC, based on DWT and RF, mimics the progressive object identification process of human beings with recognition cycles to enhance wavelet energy features and improve recognition accuracy through cascade RF classifiers. It outperforms traditional schemes and deep learning methods with a mean accuracy of 0.9914 for diverse EEG signals.
ARTIFICIAL INTELLIGENCE IN MEDICINE
(2021)
Article
Medicine, General & Internal
Gopal Chandra Jana, Anupam Agrawal, Prasant Kumar Pattnaik, Mangal Sain
Summary: Brain Computer Interface technology is used for seizure detection through the analysis of EEG signals. This study proposes a DWT-EMD feature level fusion-based approach for seizure detection using both multi and single channel EEG signals. The performance of different classifiers is evaluated and the DWT-EMD feature fusion method shows improved results compared to individual DWT and EMD features. Quantification results are provided in the Results section.
Article
Engineering, Biomedical
Tengzi Liu, Muhammad Zohaib Hassan Shah, Xucun Yan, Dongping Yang
Summary: The highly individual-dependent EEG pattern of seizure activities requires experienced specialists to annotate. Visual identification of seizure activities in EEG signals is time-consuming and error-prone. We propose a novel unsupervised learning approach, DBM_transient, to represent EEG signals in a 2D feature space and visually cluster seizure and non-seizure events.
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2023)
Article
Neurosciences
Jisu Elsa Jacob, Sreejith Chandrasekharan, Gopakumar Kuttappan Nair, Ajith Cherian, Thomas Iype
Summary: This study explores the discriminatory power of individual features and their combination in diagnosing encephalopathy using EEG signals. Feature selection is done using Gini impurity score to improve discriminative power, and three nonparametric classifiers are employed for disease classification. The models created are analyzed and evaluated to find an optimal model for automated diagnostic applications.
NEUROSCIENCE LETTERS
(2021)
Article
Computer Science, Interdisciplinary Applications
Himika Sharma, Rajnish Raj, Mamta Juneja
Summary: Yoga and meditation are effective strategies for reducing stress, with the combination of yoga and Sudarshan Kriya used as an alternative therapy. Through EEG signal analysis, it was found that Support Vector Machine (SVM) provided the highest classification accuracy between meditator and non-meditator states.
COMPUTER METHODS IN BIOMECHANICS AND BIOMEDICAL ENGINEERING
(2022)
Article
Biology
Asghar Zarei, Babak Mohammadzadeh Asl
Summary: A novel algorithm was developed for automatic seizure detection from EEG signals using DWT and OMP techniques, which improved detection accuracy by extracting signal coefficients, calculating nonlinear features, and statistical features. The proposed OMP-based technique with SVM classifier showed good performance in different classification tasks according to the experimental results.
COMPUTERS IN BIOLOGY AND MEDICINE
(2021)
Article
Computer Science, Information Systems
Arpan Phukan, Deepak Gupta
Summary: This study utilizes deep neural networks and feature extraction techniques for emotion classification from EEG signals. By comparing the results of different classifiers, it is found that random forest ensemble performs the best in terms of accuracy and other relevant metrics.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Engineering, Biomedical
Parul Madan, Vijay Singh, Devesh Pratap Singh, Manoj Diwakar, Avadh Kishor
Summary: This paper introduces a new algorithm called STWaTV, which utilizes total variation method and a bivariate shrinkage rule in the stationary wavelet transform domain for thresholding of ECG signals to remove noise. Experimental results show that STWaTV can effectively denoise ECG signals without altering the amplitude of the original signals.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Computer Science, Information Systems
Rafid Mostafiz, Mohammad Shorif Uddin, Nur-A-Alam, Md. Mahfuz Reza, Mohammad Motiur Rahman
Summary: This paper proposes an intelligent approach to detect Covid-19 from chest X-ray images by hybridizing deep CNN and DWT features. Experimental results demonstrate that the approach outperforms existing methods with an overall accuracy of over 98.5%.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Engineering, Biomedical
B. Indira Priyadarshini, D. Krishna Reddy
Summary: An optimized Adaptive Neuro Fuzzy Inference System (OANFIS) classifier is proposed in this paper to automatically detect seizures, aiming to increase classifier accuracy while reducing computational complexity. By selecting optimal features using the Binary Particle Swarm Optimization (BPSO) algorithm, the proposed system achieves a classification accuracy of 99.25% and consumes only 2.018 mu W power.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Automation & Control Systems
Cosimo Gentile, Francesca Cordella, Cesar Ramos Rodrigues, Loredana Zollo
Article
Computer Science, Information Systems
Luis Henrique Rodovalho, Orazio Aiello, Cesar Ramos Rodrigues
Article
Engineering, Electrical & Electronic
Rafael Sanchotene Silva, Afonso Roberto Plantes Neto, Jefferson Luiz Brum Marques, Omid Kavehei, Cesar Ramos Rodrigues
Summary: The study introduces a novel QRS detection system using compact and low-power blocks, aiming to reduce power consumption and silicon area in wearable or implantable ECG processing systems. Through circuit design, simulations, and tests, the system demonstrates comparable performance to state-of-the-art algorithmic approaches, making it a competitive option for integrated front-ends of implantable or wearable biomedical devices.
MICROELECTRONICS JOURNAL
(2021)
Article
Computer Science, Information Systems
Luis Henrique Rodovalho, Cesar Ramos Rodrigues, Orazio Aiello
Summary: This paper presents a single-stage single-ended inverter-based OTA with improved composite transistors for ultra-low-voltage supplies, achieving high performance while maintaining small-area, high power-efficiency and low output signal distortion. Through post-layout simulations in TSMC 180 nm technology process, the proposed OTA demonstrates significant improvements in differential voltage gain, gain-bandwidth product, power consumption and area occupation.
Article
Engineering, Electrical & Electronic
Anabeth P. Radunz, Thiago L. T. da Silveira, Fabio M. Bayer, Renato J. Cintra
Summary: This work proposes low-computational cost approximations for the KLT that are suitable for image and video compression. Extensive computational experiments on blocklengths of 4, 8, 16, and 32 demonstrate the effectiveness of these low-complexity transforms.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2022)
Article
Engineering, Electrical & Electronic
Thiago L. T. da Silveira, Diego Ramos Canterle, Diego F. G. Coelho, Vitor A. Coutinho, Fabio M. Bayer, Renato J. Cintra
Summary: The discrete cosine transform (DCT) is a relevant tool in signal processing applications, known for its good decorrelation properties. Recent research has focused on low-complexity approximations of the DCT, which are important for real-time computation and low-power consumption applications. This paper presents a new multiparametric transform class and its associated fast algorithm. By solving an optimization problem, four novel low-complexity DCT approximations are obtained. Experimental results show that these new transforms perform as well as or better than current state-of-the-art DCT approximations.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Interdisciplinary Applications
Dennis G. Balreira, Thiago L. T. da Silveira, Juliano A. Wickboldt
Summary: This paper examines the impact of adopting C and Python programming languages in introductory programming courses for engineering students. Through analyzing student data and conducting surveys, the study found that Python was more favorable in terms of learning confidence, motivation, and general programming design decisions, while C was preferred for its ease of learning data structures. This study can provide insights for institutions considering designing or redesigning introductory programming courses.
COMPUTER APPLICATIONS IN ENGINEERING EDUCATION
(2023)
Article
Computer Science, Hardware & Architecture
Luis Henrique Rodovalho, Cesar Ramos Rodrigues, Orazio Aiello
Summary: This work introduces two-stage single-ended operational transconductance amplifiers (OTA) that have very high voltage gain and rail-to-rail output voltage excursion. It achieves this by using improved composite transistors (ICT) with safe forward-body-biasing, allowing the amplifier to be used with typical I/O supply voltages and high-VT thick-oxide transistors without significant parasitic substrate current. Two versions of the OTA were designed and simulated using the open-source Skywater 130 nm PDK. The first version achieves an 84 dB voltage gain, 1.33 MHz GBW, and 60 degrees phase margin, while the proposed ICT OTA version has the same specifications except for a 33 dB increase in voltage gain, reaching 121 dB.
ANALOG INTEGRATED CIRCUITS AND SIGNAL PROCESSING
(2023)
Article
Computer Science, Information Systems
Thiago L. T. da Silveira, Paulo G. L. Pinto, Thiago S. Lermen, Claudio R. Jung
Summary: This paper proposes a novel omnidirectional 2.5D representation of volumetric chest CTs, which allows efficient exploration of 2D deep learning architectures with volume-level annotations only. The method utilizes a siamese feature extraction backbone applied to each lung and combines these features with a combination of Squeeze-and-Excite strategies and Class Activation Maps in a classification head. Experimental results show that the method provides better or comparable prediction quality, accurately distinguishing COVID-19 infections from other types of pneumonia and healthy lungs.
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION
(2023)
Article
Engineering, Electrical & Electronic
Rafael Sanchotene Silva, Luis Henrique Rodovalho, Orazio Aiello, Cesar Ramos Rodrigues
Summary: This paper proposes a new technique to improve the DC voltage gain and maintain high linearity in symmetrical OTA bulk-driven topology by combining enhanced forward-body-biasing self-cascode current mirror and source degeneration. Simulations and Monte Carlo analysis show that the proposed OTA achieves high voltage gain, CMRR, and linearity with low power consumption, making it an attractive solution for implementing OTA-C filters in wearable devices and bio-sensing applications.
JOURNAL OF LOW POWER ELECTRONICS AND APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Adriano Q. de Oliveira, Thiago L. T. da Silveira, Marcelo Walter, Claudio R. Jung
Summary: The study introduces a novel DIBR pipeline for view synthesis that effectively tackles artifacts arising from 3D warping, while maintaining structural characteristics of the scene through an image superpixel algorithm. Comparative analysis demonstrates superior performance in average results across common assessment metrics, with visual results highlighting the technique's potential for real-world applications.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Pedro H. C. Avelar, Anderson R. Tavares, Thiago L. T. da Silveira, Cliudio R. Jung, Luis C. Lamb
2020 33RD SIBGRAPI CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI 2020)
(2020)
Proceedings Paper
Mathematical & Computational Biology
Alex A. Schmidt, Alice J. Kozakevicius, Dia Zeidan, Stefan Jakobsson
INTERNATIONAL CONFERENCE ON NUMERICAL ANALYSIS AND APPLIED MATHEMATICS ICNAAM 2019
(2020)
Article
Engineering, Electrical & Electronic
Diego Ramos Canterle, Thiago L. T. da Silveira, Fabio M. Bayer, Renato J. Cintra
Proceedings Paper
Computer Science, Interdisciplinary Applications
Ricardo S. Rodrigues, Marcia Pasin, Alice Kozakevicius, Vinicius Monego
2019 IEEE 43RD ANNUAL COMPUTER SOFTWARE AND APPLICATIONS CONFERENCE (COMPSAC), VOL 1
(2019)