AI Versus COVID-19 : A Soldier We Did Not Know We Need

Coronavirus, COVID-19, is the talk of the town over weeks now if not months. The pandemic nightmare continues to terrorize on a global scale. It is bizarre to believe that bustling shopping malls, house full PVR halls, the crowd at Starbucks phase into Mexican Drug Ghost-towns.

As of March 23, 2020, more than 349,000 people have contracted the novel coronavirus and at least 15,308 have died, according to a tally by Johns Hopkins University.  Although the documented cases in terms of total recovery are at 100,165, the number of causalities by Coronavirus is larger than SARS (2002-20040 and Bird Flu of 2013, and is slowly closing to the total deaths in Swine Flu (2009-2010) i.e. 18,036.

So with lives, plunging economy, tourism at stake, Tech Giants have to take a risky bet on Artificial Intelligence. AI predicted the possibility of a pandemic much before the World Health Organization (WHO) and the US Centers for Disease Control and Prevention (CDC). Let us take a quick review on how AI has transformed through its war on virus SARS-COV-2 and how the Indian Government can take notes:

Prediction and Data Sharing:

BlueDot, a Toronto-based health surveillance company launched in 2014, had warned its clients not to travel to Wuhan China when it discovered mysterious pneumonia affecting the people in that area. It used machine learning algorithms to analyze the available data which comprised airline flight routes, livestock health, reports from health organizations, etc. and found similarities to the SARS outbreak.

InferVISION is another AI-based software that claims to show signs of virus caused pneumonia in suspected patients. It is powered by NVIDIA's Clara SDKs.

Researchers at Harvard University's School of Public Health and the National Tsing Hua University, in Taiwan, are working with Facebook to track the travel history of the affected people.

R&D Sector:

Finding a cure for a disease is a mind-numbingly lengthy procedure. Besides it also dents the financial resources.

In the case of AI, it can

•  Develop antibodies and vaccines.

•  Check through pharmaceutical history for an existing drug.

•  Design drugs to combat the current strain or future epidemics.

However, on a silver lining, Google's Deep Mind's AlphaFold System released structure predictions of several proteins associated with the virus. This is believed to guide scientific minds in understanding the virus.

DeepMind's researchers wrote on the AI lab's website. "We confirmed that our system provided an accurate prediction for the experimentally determined SARS-CoV-2 spike protein structure shared in the Protein Data Bank, and this gave us confidence that our model predictions on other proteins may be useful."

In another part of the world, British startup Exscienta became the first company to put an AI-designed drug molecule to human trials earlier this year.

Fighting Misinformation:

Most of the social networking sites and search engines are working to minimize the circulation of misinformation, conspiracy theories, phishing, etc.

All the platforms now have a pop-up link that directs users to official pages containing information and preventive measures to be taken.

Monitoring:

Did you too were pointed a gun-like device, placed on your forehead at a public place by security personnel just to know if you are at risk. Well, you are not alone. Currently, most of the cities around the globe are using traditional methods like thermometer guns, which uses an infrared sensor to measure either your frontal or body heat. Well, we can employ advanced scanner like Badu's AI-powered, non-contact sensor system. This system can do a quick multi-scan in public areas, identifying users with fever. This subtracts the risk of cross-infection, manages the crowd effectively while providing accurate results. It is currently in use at Beijing's Qinghe Railway Station, where it scans 200 people per minute and flags people with temperature over 37.3 degrees Celsius.

SenseTime is an artificial intelligence SaaS company, scans people to check if they are wearing masks.

Due to its contagious nature, most of the medical team is under constant exposure to the virus. AI turns a hero here. China is using robots to provide faster diagnostic checks. In Hangzhou, city ambulances are assisted by AI to speed through traffic.

Jack Ma's Alibaba new AI system can detect coronavirus in CT scans of patients' chests with 96% accuracy within 20 seconds. It can differentiate between a COVID-19 virus and ordinary viral pneumonia.

Information Overload:

Since the virus first made headlines in December 2019, more than 3000 articles and research papers have been published online. To mine insights from the internet can be overwhelming and time taking especially when we are in a race against the clock.  The White House Office of Science and Technology Policy, announced a project called the COVID-19 Open Research Dataset, aka CORD-19. Under this initiative, it collaborated with Microsoft Research, the National Library of Medicine, Kaggle, Semantic Scholar project by Allen Institute for AI (AI2),  the Chan Zuckerberg Initiative,  and others. They used natural language processing to collect and sift over 29,000 papers related to the new virus and the wider coronavirus family, out of which 13,000 of them processed so that computers can read the underlying data, plus information about the authors and their affiliations. This step will enable intercluster communication and help in studying the underlying history of the virus, risk factors and develop a possible vaccine or mass medication option.

Deployment of Automated Vehicles:

Access to healthcare and resources is crucial to combat the spread of the coronavirus. While Drones are used to ensure residents are taking proper precautions, in parts of the world, they have been assigned to conduct thermal imaging, dispatch aid like testing kits, food, medicines to the needy. It reduces person-to-person contact and obviates a shortage of medical staff.

Similarly, the Guangdong Provincial People's Hospital in Guangzhou City is also making use of autonomous delivery robots to transport drugs around the hospital.

This reinforces contactless monitoring of the outbreak too.

Denmark based UVD Robots' are proving to be handy by disinfecting patients without human interference. These bots emit UV light to disinfect the target area and kill viruses.

Although AI is yet to parallel Human Intelligence, they are proving to be very helpful in tracking the outbreak, diagnosing patients, disinfecting areas, and speeding up the process of finding a cure for COVID-19. While the panic and emergence of new hot spots continues, this development by AI has proven to be inspiring and groundbreaking.