COVID-19 Diagnosis by Gray-Level Cooccurrence Matrix and PSO

Introduction

The spread of COVID-19 has had a profound impact on international politics and diplomatic relations. The pandemic has hampered economic development in many countries, so most are looking for ways to grow alongside it. However, the epidemic remains serious as a result of different prevention and control policies and measures adopted by different governments, uneven vaccine distribution and the rapid mutation of the Novel Coronavirus. Researchers have continued to study the coronavirus from multiple perspectives, such as the number of casualties worldwide, possible sequelae and complications to patients (Elnaggar et al., 2022; Onoyama et al., 2022), and the development of vaccines or drugs based on the biology of the virus (Hotez & Bottazzi, 2022). The paper (Pourkarim et al., 2022) summarizes the literature on the mechanism of action, safety, efficacy and clinical trials of molnupiravir (Imran et al., 2021) in patients with COVID-19. The results suggest that molnupiravir may be effective against COVID-19 in animal-based trials and will need to be studied in more elaborate randomized clinical trials in the future. The paper (Group, 2022) argues that Tuberculosis should be considered a risk factor for serious COVID-19 disease. TB patients should be prioritized in COVID-19 prevention efforts. As well as concentrating on therapeutic and preventive studies and clinical work, doctors also have a role in researching the psychology of patients. The paper(Durankus & Aksu, 2022) uses an anonymous questionnaire distributed and voluntarily returned by the respondents to find out the anxiety levels of pregnant women affected by the COVID-19 pandemic on depression and depressive disorders. The paper (Shanahan et al., 2022) showed that participants experienced greater stress and anger over the course of the pandemic. Economic and lifestyle interruptions and despair associated with COVID-19 are the biggest cause of suffering for young people.

Before diagnosing COVID-19 using gray level cooccurrence matrix and particle swarm optimization algorithm, we reviewed the current advanced COVID-19 detection methods based on lung CT medical images of COVID-19 positive patients. This paper (Bhuyan, 2021) introduces a COVID-19 diagnosis system based on deep learning technology. Detect and segment infected areas from lung X-ray or CT scan images, use CNN to identify specific infected areas, and use FrCN (Sasank & Venkateswarlu, 2022) to identify COVID-19 patients. Finally, use quadruple cross-validation tests to generate estimates. The framework proposed in the paper (Shaik & Cherukuri, 2022) was Vision The Transformer architecture serves as the backbone. The Siamese encoder in the network is divided into two parts for processing the raw image and the enhanced image respectively. The input image is divided into patches and fed through the encoder. The results show that the scores of several evaluation criteria are better than the latest method. The paper (Purohit, 2022) reports a multi-image enhancement technique based on CNN. This technique uses discontinuous information obtained from filtered images to increase the number of valid samples. The accuracy of CT image classification can reach 95.38%. The paper (Pi, 2021) first finds the textural characteristics of the preprocessed images using a grey level cooccurrence matrix and then classifies them using a Schmidt neural network. The model in this paper (Wang, 2022) is based on wavelet entropy and Cat Swarm Optimization for the classification of COVID-19 chest CT images. The paper (Pi & Lima, 2021) uses a mixed model to classify images combined with gray level cooccurrence matrix and extreme learning machine. This paper(Wang, 2021) is rooted in wavelet entropy, one-hidden layer FNN and Jaya algorithm with K fold cross validation..