Indian study on probability attributed to COVID-19 coronavirus infection using machine learning algorithm

Abstract

As a result of the SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) outbreak, the present COVID-19 public health crisis has resulted in the loss of human life as well as significant economic disruptions. Using a machine-learning algorithm, we can determine if a patient is more likely to survive than die, or the other way around, for a particular case. We use previous data, such as medical records, demographics, and information specific to the COVID-19, to fine-tune this algorithm. Data on confirmed and probable COVID-19 infections in India, which the Indian Government has collated and made publicly accessible, is used in this report. We show that the suggested technique can predict high-risk patients with high accuracy in each of the four indicated clinical phases, hence enhancing hospital capacity planning and prompt treatment. Our approach may be expanded to produce optimum estimations for hypothesis-testing procedures that are extensively used in biological and medical statistics. With the present epidemic, we hope that the information we've gathered may help doctors make better, more timely decisions about patient treatment.