Machine Learning Against a Pandemic

In the race against COVID-19, scientists have been adapting to technology faster than ever to create the most effective solution during this crisis. During an interview at Microsoft, Kevin Scott – the chief technology officer, revealed that Microsoft is using AI Technology to map the immune system. This would give them a better understanding of the internal structure when a virus enters the system. Specifically, he says they are using this technology to help see progression of the antibodies when they enter the bloodstream. 

One major machine learning skill mentioned was neural networks. Neural networks for natural language processing can be used to run molecular simulations to “ identify therapies by SARS Clubs 2 Spike Production” according to Scott. Neural networks are primarily used for recognizing patterns. This is important because one virus that affects so many different types of people can provide a lot of statistics and figures as to what parts of the body it affects and whether there are predetermined factors or underlying illnesses that increase one’s risk of getting corona. For example, USA Today revealed evidence that of the patients who died, 55% suffered from hypertension, 37% of diabetes and the rest from respiratory disease and dementia. Neural networks would have been used to cluster these people into one of these categories based on what is found in their molecular simulations. Below is a short video to explain how neural networks are used to benefit the scientific community when it comes to interacting with a foreign agent entering the body.



Through the use of machine learning, we are saving months of our quarantine by speeding the process of discovering a potential treatment for COVID-19 cases. The ML club encourages you to stay home and stay safe! If you need something to do this quarantine you can read our other blog posts and try an exciting ML challenge.



Leave a Comment

Your email address will not be published. Required fields are marked *