ABSTRACT
In the midst of the COVID-19 pandemic, new requirements for clean air supply are introduced for heating, ventilation, and air conditioning (HVAC) systems. One way for HVAC systems to efficiently remove airborne viruses is by filtering them. Unlike disposable filters that require repeated purchases of consumables, the electrostatic precipitator (ESP) is an alternative option without the drawback of reduced dust collection efficiency in high-efficiency particulate air (HEPA) filters due to dust buildup. The majority of viruses have a diameter ranging from 0.1 μm to 5 μm. This study proposed a two-stage ESP, which charged airborne viruses and particles via positive electrode ionization wire and collected them on a collecting plate with high voltage. Numerical simulations were conducted and revealed a continuous decrease in collection efficiencies between 0.1 μm and 0.5 μm, followed by a consistent increase from 0.5 μm to 1 μm. For particles larger than 1 μm, collection efficiencies exceeding 90% were easily achieved with the equipment used in this study. Previous studies have demonstrated that the collection efficiency of suspended particles is influenced by both the ESP voltage and turbulent flow at this stage. To improve the collection efficiency of aerosols ranging from 0.1 μm to 1 μm, this study used a multi-objective genetic algorithm (MOGA) in combination with numerical simulations to obtain the optimal parameter combination of ionization voltage and flow speed. The particle collection performance of the ESP was examined under the Japan Electrical Manufacturers' Association (JEMA) standards and showed consistent collection performance throughout the experiment. Moreover, after its design was optimized, the precipitator collected aerosols ranging from 0.1 μm to 3 μm, demonstrating an efficiency of over 95%. With such high collection efficiency, the proposed ESP can effectively filter airborne particles as efficiently as an N95 respirator, eliminating the need to wear a mask in a building and preventing the spread of droplet infectious diseases such as COVID-19 (0.08 μm–0.16 μm). © 2023 by the authors.
ABSTRACT
The COVID-19 pandemic has resulted in large scale of generation of big data. This big data is heterogeneous and includes the data of people infected with corona virus, the people who were in contact with an infected person, demographics of infected persons, data on corona testing, a huge amount of GPS data of people location, and a large amount of unstructured data about prevention and treatment of COVID-19. Thus, the pandemic has resulted in producing several Zettabytes of structured, semi-structured, and unstructured data. The challenge is to process this big data, which has the characteristics of very large volume, brisk rate of generation and modification, and large data redundancy in a time-bound manner to take timely predictions and decisions. Materialization of views for Big data is one of the ways to enhance the efficiency of processing of the data. In this paper, Big data view selection problem is addressed as a bi-objective optimization problem using multi-objective genetic algorithm.
ABSTRACT
Purpose: In early 2020, the world is amid a significant pandemic due to the novel coronavirus disease outbreak, commonly called the COVID-19. Coronavirus is a lung infection disease caused by the Severe Acute Respiratory Syndrome Coronavirus 2 virus (SARS-CoV-2). Because of its high transmission rate, it is crucial to detect cases as soon as possible to effectively control the spread of this pandemic and treat patients in the early stages. RT-PCR-based kits are the current standard kits used for COVID-19 diagnosis, but these tests take much time despite their high precision. A faster automated diagnostic tool is required for the effective screening of COVID-19. Methods: In this study, a new semi-supervised feature learning technique is proposed to screen COVID-19 patients using chest CT scans. The model proposed in this study uses a three-step architecture, consisting of a convolutional autoencoder based unsupervised feature extractor, a multi-objective genetic algorithm (MOGA) based feature selector, and a Bagging Ensemble of support vector machines based binary classifier. The proposed architecture has been designed to provide precise and robust diagnostics for binary classification (COVID vs.nonCOVID). A dataset of 1252 COVID-19 CT scan images, collected from 60 patients, has been used to train and evaluate the model. Results: The best performing classifier within 127 ms per image achieved an accuracy of 98.79%, the precision of 98.47%, area under curve of 0.998, and an F1 score of 98.85% on 497 test images. The proposed model outperforms the current state of the art COVID-19 diagnostic techniques in terms of speed and accuracy. Conclusion: The experimental results prove the superiority of the proposed methodology in comparison to existing methods.The study also comprehensively compares various feature selection techniques and highlights the importance of feature selection in medical image data problems.