Utilization of Transfer Learning Model in Detecting COVID-19 Cases From Chest X-Ray Images

Diagnosis of COVID-19 pneumonia using patients' chest x-ray images is new but yet important task in the field of medicine. Researchers from different parts of the globe have developed many deep learning models to classify COVID-19. The performance of feature extraction and classifier plays a vital role in the recognizing the different patterns in the image. The pivotal process is the extraction of optimum features from the chest x-ray images. The main goal of this study is to design an efficient hybrid algorithm that integrates the robustness of MobileNet (using transfer learning approach) to extract features and support vector machine (SVM) to classify COVID-19. Experiments were conducted to test the proposed algorithm, and it was found to have a high classification accuracy of 95%.

Effectiveness of BBV152/Covaxin and AZD1222/Covishield vaccines against severe COVID-19 and B.1.617.2/Delta variant in India, 2021: a multi-centric hospital-based case-control study.

OBJECTIVES: India introduced BBV152/Covaxin and AZD1222/Covishield vaccines in January 2021. We estimated the effectiveness of these vaccines against severe COVID-19 among individuals aged ≥45 years. METHODS: We did a multi-centric, hospital-based, case-control study between May and July 2021. Cases were severe COVID-19 patients, and controls were COVID-19 negative individuals from 11 hospitals. Vaccine effectiveness (VE) was estimated for complete (2 doses ≥ 14 days) and partial (1 dose ≥ 21 days) vaccination; interval between two vaccine doses and vaccination against the Delta variant. We used the random effects logistic regression model to calculate the adjusted odds ratios (aOR) with a 95% confidence interval (CI) after adjusting for relevant known confounders. RESULTS: We enrolled 1143 cases and 2541 control patients. The VE of complete vaccination was 85% (95% CI: 79-89%) with AZD1222/Covishield and 71% (95% CI: 57-81%) with BBV152/Covaxin. The VE was highest for 6-8 weeks between two doses of AZD1222/Covishield (94%, 95% CI: 86-97%) and BBV152/Covaxin (93%, 95% CI: 34-99%). The VE estimates were similar against the Delta strain and sub-lineages. CONCLUSION: BBV152/Covaxin and AZD1222/Covishield were effective against severe COVID-19 among the Indian population during the period of dominance of the highly transmissible Delta variant in the second wave of the pandemic. An escalation of two-dose coverage with COVID-19 vaccines is critical to reduce severe COVID-19 and further mitigate the pandemic in the country.

Diagnosis of covid-19 using optimized pca based local binary pattern features

Introduction: COVID-19 is a pandemic disease affecting the global mankind since December 2019. Diagnosing COVID-19 using lung X-ray image is a great challenge faced by the entire world. Early detection helps the doctors to suggest suitable treatment and also helps speedy recovery of the patients. Advancements in the field of computer vision aid medical practitioners to predict and diagnosis disease accurately. Objective: This study aims to analyze the chest X-ray for determining the presence of COVID-19 using machine learning algo-rithm. Methods: Researchers propose various techniques using machine learning algorithms and deep learning approaches to de-tect COVID-19. However, obtaining an accurate solution using these AI techniques is the main challenge still remains open to researchers. Results: This paper proposes a Local Binary Pattern technique to extract discriminant features for distinguishing COVID-19 disease using the X-ray images. The extracted features are given as input to various classifiers namely Random Forest (RF), Linear Discriminant Analysis (LDA), k-Nearest Neighbour (kNN), Classification and Regression Trees (CART), Support Vector Machine (SVM), Linear Regression (LR), and Multi-layer perceptron neural network (MLP). The proposed model has achieved an accuracy of 77.7% from Local Binary Pattern (LBP) features coupled with Random Forest classifier. Conclusion: The proposed algorithm analyzed COVID X-ray images to classify the data in to COVID-19 or not. The features are extracted and are classified using machine learning algorithms. The model achieved high accuracy for linear binary pattern with random forest classifier. © IJCRR.