Intelligence COVID-19 monitoring framework based on deep learning and smart wearable IoT sensors

Abstract

The World Health Organization (WHO) refers to the 2019 new coronavirus epidemic as COVID-19, and it has caused an unprecedented global crisis for several nations. Nearly every country around the globe is now very concerned about the effects of the COVID-19 outbreaks, which were previously only experienced by Chinese residents. Most of these nations are now under a partial or complete state of lockdown due to the lack of resources needed to combat the COVID-19 epidemic and the concern about overstretched healthcare systems. Every time the pandemic surprises them by providing new values for various parameters, all the connected research groups strive to understand the behavior of the pandemic to determine when it will stop. The prediction models in this research were created using deep neural networks and Decision Trees (DT). DT employs the support vector machine method, which predicts the transition from an initial dataset to actual figures using a function trained on a model. Extended short-term memory networks (LSTMs) are a special sort of recurrent neural network (RNN) that can pick up on long-term dependencies. As an added bonus, it is helpful when the neural network can both recall current events and recall past events, resulting in an accurate prediction for COVID-19. We provided a solid foundation for intelligent healthcare by devising an intelligence COVID-19 monitoring framework. We developed a data analysis methodology, including data preparation and dataset splitting. We examine two popular algorithms, LSTM and Decision tree on the official datasets. Moreover, we have analysed the effectiveness of deep learning and machine learning methods to predict the scale of the pandemic. Key issues and challenges are discussed for future improvement. It is expected that the results these methods provide for the Health Scenario would be reliable and credible.