ABSTRACT
We train a deep learning algorithm to flag potential covid-19 infected in chest x-rays. The deep learning algorithm used is a Convolutional Neural Network that is 121 layers deep. Due to the lack of a large open-source of covid-19 infected x-ray images, we combine data from five different sources. Combined, the dataset has 17,194 images that are used for training procedure. The model classifies a given chest X-ray image as either a "Normal", "Covid-19", or a 'Pneumonia"infection. The trained model has a 0.93 F1 Score and 93.496% accuracy. © 2022 IEEE.
ABSTRACT
In this paper, we are predicting and forecasting the COVID-19 outbreak in India based on the machine learning approach, where we aim to determine the optimal regression model for an in-depth analysis of the novel Coronavirus in India. We are implementing the two regression models namely linear and polynomial and evaluating the two using the R squared score and error values. The COVID-19 dataset for India is being used to serve the research of this paper. The model is predicting the number of confirmed, recovered, and death cases based on the data available from March 12 to October 31,2020. For forecasting the future trend of these cases, we are utilizing the time series forecasting approach of tableau. Furthermore, the time series forecasting method is being employed to forecast the total count of confirmed cases in the future. © 2021 IEEE