Paper Title
Early Prediction of Covid-19 using Modified Convolutional Neural Networks

Abstract
COVID-19 virus development is one of the most broadly perceived kindof disease. In this paper, we have accomplished several deep convolutional networks with pre-training techniques for classifying X-ray images into three classes: Normal, Pneumonia, and COVID-19, based on two open-source data sets. Our data set contains 980 X-ray images that belong to infected by COVID-19 patients, and we attempted and experimented to apply various methods to achieve the best possible results. In our research, we introduced some pre-training techniques that helps the network learn better, when we have an unbalanced data set (less cases of COVID-19 along with more cases of other classes). We also propose a modified Convolutional Neural Network that is a convolution is the simple application of a filter to an input that results in activation. Repeated application of the same filter to input results in a map of activation called a feature map, indicating the locations and strength of a detected feature in an input, such as an image concatenation of the neural networks and VGGNet networks. This network achieved the best accuracy by utilizing multiple features extracted by two robust networks. For evaluating our network, we have tested it on 1,960 images to report the actual accuracy achievable in real circumstances. The average accuracy of the proposed network for detecting COVID-19 cases is 96.50%, and the overall average accuracy for all classes is 61.6%. The proposed work accomplished comparative study with parameters like accuracy, time complexity and high performance, reduces Computational cost, working with small amount of training data is better than the existing system. Keywords - Detection, Classification, Neural Networks, CNN, MCNN, COVID-19 virus, Deep Learning, DWI, CAD, Image Processing, Deep learning, Transfer learning, Deep feature extraction, Chest X-ray images.