Hello! Welcome to Shubu Blog


Slowing down the spread of Covid-19: A Data-Driven Approach

26 April 2021 | 9 min Read

Over the past few days, India has seen what we call the second wave of Covid-19. The rise and surge in cases is nothing less than a horror movie and with the talks of implementing a second lockdown are taking place, everyone is realizing that the second wave of Covid is worse than the first wave.

How will we stop the spread is clouding most people’s minds. But, what’s more important is to track where the cases are coming from and how we can control them. The virus is very rapidly mutating, which might lead to the vaccine’s inefficacy. We need to break this chain by following the data-driven approach mentioned below.

Well, many intellectuals around the globe have thought of various ways and have come up with various theories that can help reduce the spread of the virus and eventually stop it. We are nothing short of being at war with Covid.

If you have watched the movie The Imitation Game, which is about the story of Alan Turing's work on the Enigma machine during the 2nd World War shortened the war in Europe by more than 2 years and saved over 14 million lives, you would realize what we are proposing here is something very similar and could make the difference between life and death for millions of people.

Why is the Virus spreading so fast?

One of the key reasons the virus is spreading so fast in the 2nd wave is that the RT-PCR test is not able to detect the virus with 100% certainty. Several people infected with the Virus are asymptomatic but they unknowingly spread the virus to others. One such example was with my relative whose Covid test was negative and he was moving around freely until he started showing symptoms and the CT scan revealed the infection.

How do we identify these silent Covid spreaders?

As with most things in life, we cannot identify them with 100% certainty. That's where the power of Data Science and Machine Learning can help us out. Probability might have the answer.

Solution: Selective Covid Hotspot Lockdown, Testing and Vaccination

The highest probability of the presence of these silent spreaders would be in Covid Hotspots and this data we already have, thankfully!

Rather than locking down the entire city which can wreak havoc on the economy, we should enforce measures to lockdown these hotspots. For gated communities, this can be enforced easily by the community and local governing bodies.

The next step would be to test everyone in the hotspot and get everyone vaccinated. Also, the vaccine would need to be delivered to the individual's apartment rather than the person going out into the public, defeating the purpose. The lockdown would have to be enforced for only 2 months till the vaccine becomes effective.

Practical Implementation

Rather than performing a random vaccination ( brute force approach! ), it would make more sense to take a data-driven approach which should reduce the time horizon of this deadly virus.

Data-Driven Model to prioritize Vaccine delivery

There are several factors to consider when prioritizing the delivery of vaccines. Currently, the government is prioritizing based on age alone. We propose to prioritize based on Geography ( Covid Hotspots in particular ) and then cluster individuals based on their vicinity to covid positive patients.

Another option is to create a web portal for the hotspots and encourage people to enter their demographic information as well as comorbidities. The more vulnerable population would then be prioritized for vaccination based on a correlation study between comorbidities and severe covid infection - this data can be obtained from the government or hospitals or domain experts.

We could build a more sophisticated model if the data was available for comorbidities and travel history, but that would be very difficult to accomplish in a short period.

If there are challenges to delivering vaccines to the hotspot, we should build a simple Whatsapp queue and appointment management application for each Covid centre and strictly provide appointments to the target recipients so that a Covid infected patient would not spread the virus to other individuals during vaccination.