2
$\begingroup$

The ongoing coronavirus pandemic of coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), as of 29 September 2020, has affected many countries and territories, with more than 33.4 million cases of COVID-19 have been reported and more than 1 million people have died. The live statistics can be found at https://www.worldometers.info/coronavirus/ or in the World Health Organization (WHO) site. Although countries have already started quarantines and have adopted extreme countermeasures (such as closing restaurants or forbidding events with multiple people), the numbers of cases and deaths will probably still increase in the next weeks.

Given that this pandemic concerns all of us, including people interested in AI, such as myself, it may be useful to share information about the possible current applications of AI to slow down the spread of SARS-CoV-2, to help infected people or people in the healthcare sector that have been uninterruptedly working for hours to attempt to save more lives, while putting at risk their own.

What are the existing AI technologies (e.g. computer vision or robotics tools) that are already being used to tackle these issues, such as slowing down the spread of SARS-CoV-2 or helping infected people?

I am looking for references that prove that the mentioned technologies are really being used. I am not looking for potential AI technologies (i.e. research work) that could potentially be helpful. Furthermore, I am not looking for data analysis tools (e.g. sites that show the evolution of the spread of coronavirus, etc.)

$\endgroup$
  • 1
    $\begingroup$ Related at Kaggle: kaggle.com/allen-institute-for-ai/CORD-19-research-challenge and possibly a good source of information for anyone answering $\endgroup$ – Neil Slater Mar 18 at 12:17
  • 1
    $\begingroup$ Here are other possibly useful resources Open Source Toolkit for Predicting Covid-19 Vulnerability and Resources and channels to help with COVID-19 research. $\endgroup$ – nbro Mar 19 at 0:24
  • $\begingroup$ In the United Kingdom a major difficulty has been with the allocation of people requiring a test for COVID-19 to a testing station which has tests available. Some cases have been reported where people have been told to travel several hundred miles to be tested. The optimal allocation of people in real-time to available resources constrained to minimise travel time would appear to be ideal for an AI based optimisation technique such as multi objective genetic algorithms. An added problem would be to predict the likely future resource needs as the infection rises and wanes, and feed that in to o $\endgroup$ – Nick Sep 30 at 19:56
3
$\begingroup$

According to the Baidu Research's blog post How Baidu is harnessing the power of AI in the battle against coronavirus (12-03-2020), there are already some artificial intelligence tools or algorithms being used to fight the coronavirus.

Given that I cannot confirm that these AI tools and algorithms I will mention are really being used in practice, I will only quote the parts of the blog post that potentially answer my original question.

To give some context, similar to HIV viruses, the virus that is causing the coronavirus pandemic, SARS-CoV-2 is capable of rapidly mutating, making vaccine development and virus analysis difficult.

AI-powered and non-contact infrared sensor system

Baidu has developed several tools that are effective in building awareness and screening populations, including an AI-powered, non-contact infrared sensor system that provides users with fast multi-person temperature monitoring that can quickly detect a person if they are suspected of having a fever, one of the many symptoms of the coronavirus. This technology is currently being used in Beijing's Qinghe Railway Station to identify passengers who are potentially infected where it can examine up to 200 people in one minute without disrupting passenger flow.

AI-powered pneumonia screening and lesion detection system

By leveraging PaddlePaddle and the semantic segmentation toolkit PaddleSeg, LinkingMed has developed an AI-powered pneumonia screening and lesion detection system, putting it into use in the hospital affiliated with XiangNan University in Hunan Province. The system can pinpoint the disease in less than one minute, with a detection accuracy of 92% and a recall rate of 97% on test data sets.

Automated HealthMap system

The Boston Children's Hospital used an automated HealthMap system that scans online news and social media reports for early warning signs of outbreaks, which led to the initial awareness that COVID-19 was spreading outside China.

Autonomous vehicles carry out non-contact tasks

Access to health care and resources at a moment's notice is vital for battling the spread of the coronavirus. Autonomous vehicles are playing a useful role in providing access to necessary commodities for health-care professionals and the public alike by delivering goods in infected areas and disinfecting hospitals, effectively minimizing person-to-person transmission and alleviating the shortage of medical staff.

Apollo, Baidu's autonomous vehicle platform, partnered with a local self-driving startup called Neolix to deliver supplies and food to the Beijing Haidian Hospital.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Do you have a source regarding the mutation rate of SARS-Cov-2? I couldn't find it on the page you linked to. At least one source seems to suggest otherwise. $\endgroup$ – user76284 Mar 28 at 21:06
  • $\begingroup$ @user76284 I fixed the link. It should have been the link to the Baidu blog post. $\endgroup$ – nbro Mar 28 at 21:07
1
$\begingroup$

Not really something that slows it down but currently Mauricio Santillana at Harvard is working on modeling the pandemic and has shared some of his approaches. He explained that they have used google search trends to try to predict the number of actual cases (there is a delay between people being sick and getting tested). Looking for search terms like, "how to use an inhaler?" can reveal areas affected by the outbreak and is useful for modeling.

Paper can be found here

Towards Data Science has several articles listing potential AI applications for helping in the fight against COVID

Including:

  1. Identify who is most at risk
  2. Diagnose patients
  3. Develop drugs faster
  4. Predict the spread of the disease
  5. Understand viruses better
  6. Map where viruses come from
  7. Predict the next pandemic

How to fight COVID-19 with machine learning

Though there are probably many creative applications that AI can help with I think most of them are modeling related at the moment.

| improve this answer | |
$\endgroup$
  • $\begingroup$ I asked "What are the existing AI technologies that are already being used to tackle these issues", so I wasn't interested in "research work", but actual stuff that is useful. I had also found some research work. $\endgroup$ – nbro Apr 27 at 23:11
0
$\begingroup$

What are the AI technologies currently used to fight the coronavirus pandemic?

You don't define what is AI for you. Let's suppose it means advanced informatics (but perhaps also abstract interpretation).

In France we have polling websites like https://covidnet.fr/ ... Does that count as AI?

And we also have Deep Tech related Covid19 specific call for proposals, e.g. here ... does that count as AI?

Does the French StopCovid platform count as AI? It is an open source digital project (using Bluetooth) to trace potential Covid-infected people. The current political debate is involving intense discussions about its legitimacy and deployement.

At last, development of Covid specific ventilators (e.g. this open hardware / open software project) involves a lot of embedded software which could be analyzed by static analyzers such as Frama-C ? Does that count as AI ?

My employer CEA (a French scaled down equivalent to US DoE) is an applied research institution participating to several Covid related projects.

And several French research institutions (e.g. INSERM, CNRS, INRIA, LIP6, etc...) are participating to Covid related research programmes. Also, major health organizations like AP-HP.

So is the European Union which is partly funding several of them. Most of the Covid related research projects have strong digital aspects. In particular, related to genome decoding and bioinformatics.

PS. My GitHub helpcovid ongoing project (a free software web application related to Covid19) does not claim to be AI, but certainly is a digital application. See also these slides about RefPerSys (which is an ongoing free software ambitious AI project).

| improve this answer | |
$\endgroup$
  • $\begingroup$ By "AI" I didn't mean websites that just keep track of the statistics (that's not even advanced informatics), otherwise, I would have mentioned the sites I am already mentioning as one of the tools. By "AI tools" I mean computer vision, robotics, machine learning, etc., tools. I was also not asking for research work (I know there's already some research work). I was asking for AI-based technology that is already being used to fight the coronavirus. $\endgroup$ – nbro Apr 30 at 12:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.