Mathematics, Analytics and Computing Entity
MACoE
We apply Mathematics for a better tomorrow




Predictive Analytics: COVID-19


Efficient Machine Learning Algorithm that can predict COVID-19 from X-ray images


2019-Novel-Coronavirus (COVID-19) is a term everyone in the world is aware of and this pandemic is an unusual challenge to most of us. Scientists have realized this obstacle and started looking for any knowledge in the world which can be a potential solution to handle this pandemic. But, critically, after a thorough examination of knowledge especially in the direction of antibody and plasma based methods, controlling COVID-19 is puzzling. After a closer scrutiny we also express an obvious contradiction that the majority of the research activities on COVID-19 across the globe is influenced by antibody-based or plasma-based methods. For a large population like India, detecting the virus is a real challenge as the spreading rate is exponential. So, it is necessary to find a solution for early detection of this virus and a solution that is feasible to implement on such populated countries.

In this project we propose to practice some deep learning image classification algorithms namely to identify the COVID-19 patients using the chest X-ray images. It is easy to find publications using deep learning models like GoogLeNet, AlexNet or ResNet etc., which are high end classifiers and they require a huge amount of data and computing infrastructure to learn.



An estimate for India from a global perspective of this pandemic

data as on May 6, 2020


X-Ray images








Preprocessing




Sample images for the machine to infer the reality without confusing with the appearance





The following word clouds are based out of data as on May 6, 2020



If you have similar interest and want to chat with us, feel free to drop in through vpramesh@gmail.com