Wanna know more about data science? Make sure to check out my events and my webinar What it's like to be a data scientist and What’s the best way to become a data scientist !
COVID-19: A global crisis
The world is currently facing an unprecedented health crisis. The coronavirus known as COVID-19 has already taken the lives of thousands of people worldwide and has infected many more. At the time of writing, the crisis is not over yet, and there are concerns that it might lead into a new financial crisis.
At these times of crisis, humanity needs any weapon at its disposal. So, what can AI and data science do in order to help with this crisis? Apparently, many organisations have started taking the initiative.
AI and data science against COVID-19
One of the biggest moves is coming from the White House and the Allen Institute for AI. They have released a database of 29,000 research articles on the coronavirus family. The research articles have been preprocessed in order to be machine readable by natural language processing libraries. The hope is that through the use of AI, researchers will be able to extract new insights from this huge body of knowledge about vulnerabilities of the virus.
This dataset is also listed in a Kaggle challenge. At the time of writing there are more than 100 Kaggle kernels that analyse the data. Kaggle has outlined 10 questions that need to be answered (outlined below)
- What is known about transmission, incubation, and environmental stability?
- What do we know about COVID-19 risk factors?
- What do we know about virus genetics, origin, and evolution?
- Help us understand how geography affects virality.
- What do we know about vaccines and therapeutics?
- What do we know about non-pharmaceutical interventions?
- What has been published about medical care?
- What do we know about diagnostics and surveillance?
- What has been published about information sharing and inter-sectoral collaboration?
- What has been published about ethical and social science considerations?
AI can also be used to detect patients with high vulnerability. Medical Home Network, an organization serving patients in the Chicago area, is using AI to identify individuals who have a heightened vulnerability to severe complications from COVID-19. This helps them prioritise patients that are most in need.
AI has also been used to generate vaccines. This is a fairly new and experimental application, but it is very promising. A team of medical researchers from Australia’s Flinders University have designed an AI that has, for the first time, created a drug entirely by itself. While the technology might not be mature enough to fight COVID-19 at this stage, one can easily imagine it being rolled out at a subsequent pandemic.
Also, AI can also be used in some simple but effective ways, in order to reduce the stress on health systems. For example, chatbots can be used, in place of real doctors, in order to help patients understand whether they might be suffering from COVID-19 or simply the common cold. A bot called AskSophie is one such chatbot.
And let’s not forget traditional predictive analytics. Israeli researchers are using AI in order to identify where the next breakout will take place and prevent it. The Israeli health system is not in possession of a large number of tests, so they have to be very careful with how they are used. Hence, they’ve decided to simply hand out daily questionnaires to citizens, and then use data science in order to infer which clusters of people seem to be infected. In this way, they can save valuable resources from their health service.
Data science and AI in the service of humanity
It is clear that there many different ways in which AI and data science can help national health systems and countries fight off this horrible virus. Some of the initiatives mentioned in this article are just the beginning. We are surely going to see many more applications coming out in the near future. I am personally a big believer in using AI to develop vaccines, but another huge application could be predicting the next outbreak before it even takes place.
Maybe some of the next AI unicorns will come out of this area! In the meantime, let’s hope for the best and that this crisis will soon be over!