Deep Learning Fusion: CNN-LSTM Hybrid for Robust COVID-19 Forecasting and Spatial Localization

The corona virus was first identified in Wuhan, China and has spread around the globe. Around 1.8 million people have been infected, with 114,698 people dying as a result. Due to the number of instances has increased rapidly, most of the countries are experiencing a lack of testing kits and resources. Previous prediction method is based only on the historical data. So the prediction accuracy is less, here the external factors are also necessary to analyze the number of cases that will occur. The method proposed is based on by integrating CNN and LSTM framework to assist radiologist and clinicians in recognising COVID-19 cases in X ray images. The main aim of this work is to predict cases from the data sets available in online. The prediction is done by using Convolutional neural network and LSTM approach. The proposed work performs better than existing methods in-terms of accuracy in prediction.