Article
Computer Science, Information Systems
Atiaf A. Rawi, Murtada K. Elbashir, Awadallah M. Ahmed
Summary: This paper presents a novel method for feature extraction and classification of electrocardiogram (ECG) signals using a convolutional neural network (CNN) and extreme gradient boosting (XGBoost). Experimental results demonstrate that this technique outperforms standalone CNN or XGBoost methods in ECG signal classification and achieves better performance than state-of-the-art models.
Article
Computer Science, Information Systems
Junqing Zhang, Yushi Zheng, Weitao Xu, Yingying Chen
Summary: Wireless body area network is a crucial component for connected healthcare, but its security and trustworthiness have been compromised due to recent cyberattacks. This paper proposes a heartbeat-based key generation framework to enhance the security of body area networks. By utilizing electrocardiography (ECG) and photoplethysmography (PPG) sensors, heartbeat signals are captured and processed to generate cryptographic keys. The performance of the proposed algorithm is evaluated using ECG signals from an online public database and PPG signals collected from a testbed, showing its robustness and effectiveness.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Information Systems
Zeeshan Ahmad, Anika Tabassum, Ling Guan, Naimul Mefraz Khan
Summary: This paper proposes two computationally efficient multimodal fusion frameworks for ECG heart beat classification, achieving high classification accuracy and demonstrating superiority in experiments.
Article
Computer Science, Information Systems
Liang-Hung Wang, Yan-Ting Yu, Wei Liu, Lu Xu, Chao-Xin Xie, Tao Yang, I-Chun Kuo, Xin-Kang Wang, Jie Gao, Pao-Cheng Huang, Shih-Lun Chen, Wei-Yuan Chiang, Patricia Angela R. Abu
Summary: This study proposes a novel classification method for arrhythmia using THML ECG data, which can decrease the shortage of medical resources. By utilizing a 1D-CNN model combined with a priority model integrated voting method, the classification accuracy is improved. The practicality and effectiveness of THML ECG data are demonstrated through experiments.
Article
Engineering, Biomedical
Zhenqin Chen, Mengying Wang, Meiyu Zhang, Wei Huang, Hanjie Gu, Jinshan Xu
Summary: Electrocardiography (ECG) is a standard method for diagnosing cardiovascular disease due to its minimal risk, affordable price, and simple application. This study proposes a post-processing refined ECG delineation method that utilizes the morphological information of a heartbeat cycle to accurately locate the boundaries of different waveforms. The testing results on two public ECG databases, LUDB and QTDB, show satisfactory delineation performance.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Mathematics
Adel A. A. Ahmed, Waleed Ali, Talal A. A. Abdullah, Sharaf J. J. Malebary
Summary: Blood circulation relies on electrical activation, and any disruption to the heart's propagating wave can cause arrhythmias. Electrocardiograms (ECG) are commonly used for diagnosing arrhythmias, but their susceptibility to noise and the randomness of arrhythmic events can lead to misdiagnosis. This study proposes a deep learning model, specifically a one-dimensional convolutional neural network (1D-CNN), to address these limitations and achieve accurate and automatic classification of cardiac arrhythmias.
Article
Computer Science, Artificial Intelligence
Zhaoyang Ge, Xiaoheng Jiang, Zhuang Tong, Panpan Feng, Bing Zhou, Mingliang Xu, Zongmin Wang, Yanwei Pang
Summary: The paper proposes an ECG abnormal event detection model based on multi-label correlation guided feature fusion, which calculates the correlation of different levels of ECG abnormalities and integrates features to achieve better experimental results.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Engineering, Biomedical
Viraj Rawal, Priyank Prajapati, Anand Darji
Summary: Electrocardiography (ECG) is used as a diagnostic tool for heart diseases, and this work proposes a safety-critical hardware system for early arrhythmia diagnoses, particularly Atrial Fibrillation (AF). Two software CNN structures, Supreme CNN Architecture (SCA) and Software-Selected CNN Architecture (SSCA), achieve high accuracy for AF classification. The proposed hardware CNN architecture (HCA) achieves Atrial fibrillation classification accuracy with lower power consumption.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Information Systems
Abdallah Benhamida, Miklos Kozlovszky
Summary: This paper introduces the basics and applications of electrocardiograms, emphasizing the importance of daily monitoring. It proposes an automated solution for abnormal ECG signal detection and presents an algorithm for ECG pre-annotation and beat-to-beat separation using Autoencoders.
Article
Computer Science, Information Systems
Taki Hasan Rafi, Young Woong-Ko
Summary: Cardiovascular disease is a major cause of morbidity and mortality, and Electrocardiogram (ECG) is a reliable tool for monitoring cardiovascular health. To address the lack of data in rare medical diseases, a new generative adversarial network-based deep learning method called HeartNet was developed. This method tackles the data insufficiency problem by synthesizing additional training samples through a generative adversarial network, significantly improving classification performance.
Article
Computer Science, Artificial Intelligence
Yusra Obeidat, Ali Mohammad Alqudah
Summary: This study introduces a hybrid lightweight 1D deep learning model that combines CNN and LSTM methods for accurate, fast, and automated beat-wise ECG classification. The hybrid model utilizes CNN for deep feature extraction and LSTM for contextual time information, achieving high accuracy and sensitivity. Suitable for embedded systems design, it provides faster and more efficient monitoring of heart diseases in clinical applications.
TRAITEMENT DU SIGNAL
(2021)
Article
Engineering, Biomedical
Carl Boeck, Peter Kovacs, Pablo Laguna, Jens Meier, Mario Huemer
Summary: The study introduces a model based on Hermite and sigmoid functions, combined with piecewise polynomial interpolation, for accurate segmentation and low-dimensional representation of ECG beats. Results demonstrate better denoising and segmentation performance compared to state-of-the-art algorithms.
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING
(2021)
Article
Computer Science, Information Systems
Chih-Ta Yen, Sheng-Nan Chang, Cheng-Hong Liao
Summary: This study proposes a cuffless blood pressure estimation method based on PPG and ECG signals using a deep learning model. The developed model shows stable performance and accurately predicts blood pressure and heart rate after processing the signals and conducting ten-fold cross-validation.
Article
Computer Science, Artificial Intelligence
Md Shofiqul Islam, Khondokar Fida Hasan, Sunjida Sultana, Shahadat Uddin, Pietro Lio', Julian M. W. Quinn, Mohammad Ali Moni
Summary: We propose a hybrid hierarchical attention-based bidirectional recurrent neural network with dilated CNN (HARDC) method for arrhythmia classification. This method fully exploits the dilated CNN and bidirectional recurrent neural network unit (BiGRU-BiLSTM) architecture to generate fusion features, improving the model's performance for prediction. By combining the fusion features with a dilated CNN and a hierarchical attention mechanism, the trained HARDC model showed significantly improved classification results and interpretability of feature extraction on the PhysioNet 2017 challenge dataset.
Article
Computer Science, Information Systems
Eduardo M. M. Vasconcellos, Bruno Georgevich Ferreira, Jorge S. Leandro, Baldoino F. S. Neto, Filipe Rolim Cordeiro, Idagene A. Cestari, Marco A. Gutierrez, Alvaro Sobrinho, Thiago D. Cordeiro
Summary: The Electrocardiogram (ECG) is a low-cost exam commonly used for diagnosing abnormalities in the cardiac cycle. This study proposes a Siamese Convolutional Neural Network (SCNN) model for classifying 12-Lead ECG heartbeats with limited training samples. The model achieves high accuracy and AUC values.
Article
Thermodynamics
Javad Zareei, Kourosh Ghadamkheir, Seyed Alireza Farkhondeh, Azher M. Abed, Maria Jade Catalan Opulencia, Jose Ricardo Nunez Alvarez
Summary: This research focuses on the conversion of port fuel-injected engines to direct-injection in two main stages. The first stage involves modifying the injection mechanism of CNG direct-injection engines. The second stage investigates the effects of hydrogen as a gaseous additive on CNG. The results show that a CNG direct-injection engine can significantly reduce CO2 emissions and increase NOx emissions compared to a gasoline port injection system. Additionally, the use of hydrogen in the fuel mixture, along with increased injection pressure, improves engine power, torque, and fuel conversion efficiency. The emission reduction for CO is observed when using a blend of HCNG with hydrogen, while NOx emissions have a reverse correlation with hydrogen concentration.
Article
Chemistry, Physical
Javad Zareei, Abbas Rohani, Jose Ricardo Nunez Alvarez
Summary: This study investigates the effect of adding hydrogen to natural gas and EGR ratio on a diesel engine using AVL Fire multi-domain simulation software. The results show that increasing EGR reduces thermal efficiency, engine power, and specific fuel consumption while increasing cumulative heat release, but it has insignificant effect on cylinder pressure. Adding hydrogen to natural gas increases combustion temperature and subsequently NOx emissions, but reduces CO and HC emissions. The optimal values of EGR and HCNG are determined using Gaussian process regression and multi-objective genetic algorithm.
INTERNATIONAL JOURNAL OF HYDROGEN ENERGY
(2022)
Article
Automation & Control Systems
J. S. Xia, Mohamad Khaje Khabaz, Indrajit Patra, Imran Khalid, Jose Ricardo Nunez Alvarez, Alireza Rahmanian, S. Ali Eftekhari, Davood Toghraie
Summary: This paper uses an Artificial Neural Network (ANN) to investigate the influence of rolling parameters on the rolling force, rolling power, and slip of tandem cold rolling. Real data collected from a practical tandem rolling line is used to train and test the network. The best topology of the ANN is determined, and the results show accurate prediction of the cold rolling parameters considered in this study.
Article
Green & Sustainable Science & Technology
Oriza Candra, Abdeljelil Chammam, Jose Ricardo Nunez Alvarez, Iskandar Muda, Hikmet S. Aybar
Summary: Growing population and limited energy resources have driven the importance of energy sources, particularly renewable energy, for economic growth and environmental sustainability. This study compares the effects of increasing renewable energy use on economic growth and greenhouse gas emissions in middle income and high income countries. The results show that renewable energy production has positive economic effects in both the short-term and long-term, with varying contributions in different countries.
Article
Engineering, Chemical
Javad Zareei, Jose Ricardo Nunez Alvarez, Yolanda Llosas Albuerne, Maria Rodriguez Gamez, Angel Rafael Arteaga Linzan
Summary: The number of injector holes and the fuel-injection pressure have an impact on engine performance and emissions. This study investigates the effects of injection pressure, injector hole numbers, compression ratio, and hydrogen enrichment on engine performance and emissions. The results show that increasing the number of injector holes improves engine performance and reduces COemissions.
Article
Multidisciplinary Sciences
Yeizabet Napoles-Baez, Guillermo Gonzalez-Yero, Ruben Martinez, Yunier Valeriano, Jose R. Nunez-Alvarez, Yolanda Llosas-Albuerne
Summary: This study aims to identify, model, and control the dynamic behavior of the hydraulic actuator in a Ladle Furnace. Three models were obtained using Pseudo-Random Binary input signals and black box models, allowing a better understanding of the system behavior and evaluation of the electrode weight variation and controller tuning.
Article
Engineering, Electrical & Electronic
Luis B. Corrales-Barrios, Juan C. Fernandez-Blanco, Jose R. Nunez-Alvarez, Herminio Martinez-Garcia, Felix H. Hernandez-Gonzalez
Summary: Motors with high voltage and power consumption generate harmful harmonics to the electro-energy system, affecting its correct operation. This study analyzed the electrical energy quality of a motor with these characteristics by measuring various parameters. The results showed that the main problem was the excessive total current harmonic distortion, which exceeded the permissible limit. However, the total voltage harmonic distortion remained within the allowable range. Passive and active filters were designed and simulated using MATLAB R2015a and LabView 2014 software to mitigate the harmful effects.
ELECTRICAL ENGINEERING
(2023)
Article
Green & Sustainable Science & Technology
Wen-Cheng Wang, Ngakan Ketut Acwin Dwijendra, Biju Theruvil Sayed, Jose Ricardo Nunez Alvarez, Mohammed Al-Bahrani, Anibal Alviz-Meza, Yulineth Cardenas-Escrocia
Summary: The internal components of a smart building interact through a compatible fabric and logic to create a dynamic and cost-efficient environment. Smart buildings reduce cooling and heating load, lowering operating costs and energy consumption without sacrificing comfort. This research aims to develop a data transmission model for routing IoT data in smart buildings.
Article
Green & Sustainable Science & Technology
Tzu-Chia Chen, Jose Ricardo Nunez Alvarez, Ngakan Ketut Acwin Dwijendra, Zainab Jawad Kadhim, Reza Alayi, Ravinder Kumar, Seepana PraveenKumar, Vladimir Ivanovich Velkin
Summary: This research focuses on modeling the electrical energy network with renewable energy sources and gas production systems. The model provides a mixed integer linear optimization model that integrates distributed generation sources, energy storage systems, gas power systems, and electric vehicles in an integrated electricity and gas system. The research also considers electric vehicles as a base load, which presents a limitation in optimizing their maximum charging. One of the important findings is that the investment cost in the first scenario was USD 879,340, with purchased electric energy of 37,374 kW and gas of 556,233 m(3) from the respective networks.
Article
Energy & Fuels
Elfizon Elfizon, Jose Ricardo Nunez Alvarez, Abdeljelil Chammam, Ibrahim H. Al-Kharsan, Muhsin J. Jweeg, Patricio Yanez-Moretta, Reza Alayi, Imran Khan, Yung-Cheol Byun, Dag oivind Madsen
Summary: This research proposes the design and use of combined systems for the simultaneous production of water, heat, and energy, and models the multi-effect evaporative desalination and combined heat and power generation system to meet the demands of a hotel. By using an energy storage tank, the combined system can reduce annual costs.
FRONTIERS IN ENERGY RESEARCH
(2023)
Article
Thermodynamics
Wongchai Anupong, Mark Treve, Iskandar Muda, I. B. Sapaev, Julio Francisco Jimenez Arana, Raed H. C. Alfilh, Jose Ricardo Nunez Alvarez, Morteza Almassi
Summary: This paper investigates the thermal performance of uncovered solar collectors using perforated absorbent plates. The study analyzes the impact of parameters such as diameter and step holes, air suction velocity, and solar radiation on the collector's performance. The results show that increasing air suction velocity and solar radiation can enhance the thermal efficiency of the collector.
INTERNATIONAL JOURNAL OF LOW-CARBON TECHNOLOGIES
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
J. C. Fernandez, L. B. Corrales, I. F. Benitez, J. R. Nunez
Summary: This article presents a method of diagnosing faults in MTU-16V-S4000-G81 internal combustion engines using fuzzy logic tools. The method allows for the detection of potential deviations in key parameters based on historical data and offers fuzzy maintenance suggestions to restore normal fuel consumption.
INTELLIGENT COMPUTING SYSTEMS (ISICS 2022)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
J. C. Fernandez, L. B. Corrales, F. H. Hernandez, I. F. Benitez, J. R. Nunez
Summary: The document proposes a method to diagnose multiple incipient faults in a power transformer using fuzzy logic, achieving high classification performance based on historical data. This method outperforms traditional methods in detecting incipient faults and offers a simple and practical solution.
PROGRESS IN ARTIFICIAL INTELLIGENCE AND PATTERN RECOGNITION
(2021)
Article
Engineering, Mechanical
Jose Cabrera Cesar, Jean Caratt Ortiz, Guillermo Valencia Ochoa, Rafael Ramirez Restrepo, Jose R. Nunez Alvarez
Summary: A study was conducted on a solar collector and storage tank coupled with a single effect LiBr-H2O absorption refrigeration system using MATLAB's built-in App Designer for assessment. The model was developed based on balances of mass, energy, and species conservation in the components, considering the effect of external streams. Simulations and validation procedures showed a maximum relative error of 2.65% on the energy analysis, with the generator identified as the biggest source of irreversibility in the inner cycle.
Article
Computer Science, Information Systems
Daily Milanes Hermosilla, Rafael Trujillo Codorniu, Rene Lopez Baracaldo, Roberto Sagaro Zamora, Denis Delisle-Rodriguez, Yolanda Llosas-Albuerne, Jose Ricardo Nunez Alvarez
Summary: This study proposes a MI classification scheme using Convolutional Neural Networks, enhancing the performance by capturing event-related desynchronization/synchronization (ERD/ERS) and achieving superior results compared to existing methods.