RPA, AI and the Coronavirus

RPA, AI and the Coronavirus

As we sit in the middle of the next serious virus, Covid-19, questions about how to approach this outbreak come flying in from all sides.  Many people are rightly concerned, but how do we quantify the situation?

If you listen to the pundits, who are paid to scare you, I’m sure you read this article from the dungeon in your basement (completely stocked with a years supply of water and other necessities).  I, on the other hand prefer to be pragmatic in my understanding.  First, what does the data say?  What can we infer from prior flu’s and other similar outbreaks?  What was the optimal way that those situations were handled, and what was the outcome?

Let’s not act like this is the first time the flu has come about, nor some earth shattering disease.  I remember my wife thinking the world would end due to Ebola and having to tell my children, if it does end, what’s the point of fretting now (for the record, I didn’t think it was going to end, as I don’t think it will end now).

So let’s look at the information and compare it to other diseases like Mers and SARS.  As you can see, Corona is far more contagious according to the data analyzed to this point..

Something else I infer here is that contracting MERS was bad news.  Another point I would make, that we should consider is that the mortality rate for Coronavirus is heavily skewed to those over 50, even more so over 80.

Graphs can be found: https://www.weforum.org/agenda/2020/02/comparing-outbreaks-coronavirus-mers-sars-health-epidemic/

In terms of picking an example country to analyse, let’s look at South Korea. With more notice as to the virus capability, it provides a better baseline to consider how the virus will truly spread once it hits a country.

As you can see there is a substantial increase in Coronavirus cases, perhaps as it was not quarantined and it could well have been one individual caused the mass spread. ALthough we can, as ever, read statistics anyway, this does lend credibility that the virus is far more contagious than others that have been concerned us.

Last on the mortality rate, you can see in this graph it is highly concentrated on the elderly.

Lastly, here is a final comparison set, excluding the big one “Spanish Flu”

So what do I personally summarise from this? The Coronavirus is not nearly as contagious as H1N1, but is more fatal, but hardly as fatal as many of the others.

This brings me to the pragmatic part, the bit where I apply my professional skill set.  As ever in our data driven existence, the data is or paramount importance in stopping or slowing this spread. We need to know quickly the nature of the virus and its ability to spread. So, how was the information collected, and how can we better this data collection and analysis  in the future ? This is where RPA and Advanced Intelligence come in.

One of the things we can do now is compare and model data sets. Model our way to a version of truth not skewed or affected by a subsection of  the data set. There have been reports of course that China, where the outbreak started perhaps did not report as quickly as they could moreover the numbers are under egged. This we will never know but what we can do , is  look back and historically review how China has reported the information, then surmise the true numbers, likely using that data as a parameter in an AI algorithm.

Further, can we consolidate all the reporting into one location, say that WHO doesn’t have all the information correct, how could one accomplish this?  Leveraging and comparing the data sets provided.  This is exactly what RPA can be used for.  In this case the validity and quantity of the information can be used comparatively to determine if all the stated facts are indeed correct and discovered.  Think about it this way, once you build the AI Model and the rpa data consolidation, you have an engine that can effectively predict an outbreak with a greater certainty than random internet prognosticators.  Which is exactly what we need.

  • Posted by USRDEV2018
  • On March 6, 2020

Leave Reply

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