Usefulness of data in Covid Times*QGrqe9hpHFxSkIVG

Original Source Here

Usefulness of data in Covid Times

Renu Chaturvedi
ISB Institute of Data Science, Hyderabad, India

Photo by Markus Winkler on Unsplash

It has been over 15 months since the world has been reeling under the Covid-19 pandemic. What started as an isolated incident in Wuhan (China) rapidly spread worldwide to cause health, social and economic upheaval. There is hardly anyone who has been left untouched by this disaster. While we can debate endlessly about what we could have done better to deal with the situation, one of the bright spots in humanity’s fight against the disease has been the use of data and data sciences. AI and ML, once seen as science fiction, have become life-saving agents in combating Covid. With allied technologies like Big Data, ML, and data science, AI has offered quick information access to front-line caregivers and resources to researchers & drug developers. In this post, we examine how AI and ML have helped battle Covid-19 and will continue to assist us in recovering from this pandemic of crisis proportions.

Speed of Vaccine Development

Vaccines have historically been developed through extensive, time-consuming rounds of clinical trials. However, with the rapid spread of the Coronavirus, Covid vaccine development had to move forward at lightning speed. AI & ML helped in accelerating vaccine development. Researchers developed the most effective formulas of medications through pattern recognition and simulation to help the body build antigens and immunity against the virus.

Moderna (now a household name) was a biotech startup born in the Amazon Web Services (AWS) cloud in 2010. It developed its complete drug discovery and manufacturing processes around digital tools infused by artificial intelligence. On January 11, 2020, researchers from China published the genetic sequence of the coronavirus. Just two days later, Moderna’s team and NIH scientists finalized the targeted genetic sequence they would use in the vaccine. (

The first clinical-grade batch of the vaccine was released on February 7, and the first subject was administered the dose on March 16- only 65 days from sequence to dosing.

Using the coronavirus’ genetic sequence, which Chinese researchers published on January 11, 2020, Dr. Sahin — CEO and founder of BionTech, designed ten different candidates on his computer that weekend. The one candidate that got selected for more extensive trials was the one the FDA has authorized. Dr. Sahin had used the power of ML and AI to design a vaccine in less than one day. (

Contact Tracing

Contact tracing allows healthcare providers and officials to identify infected patients and asymptomatic carriers. With this information, they can isolate Covid-positive patients and deliver healthcare solutions to the targeted segment.

Using applications like Arogya Setu ( and models like SIR (Susceptible, Infectious, and Recovered) that leverage AI and IoT, caregivers have been able to seamlessly trace contacts, identify vulnerable regions and clusters, announce containment zones, deploy resources, and much more. The app had reached 100 million installs in 40 days of its launch. Until June 29, 2021, The Arogya Setu App has now been downloaded by approximately 195 million users, recorded about 410 million samples collected, and about 2 million samples tested. (

As of 30th June, 2021; The Arogya Setu website displayed 30 million total cases of which there were about 500k active cases, and about 400k covid deaths so far in India.

Co-Win is an app developed by the Government of India for collecting data on the Covid vaccination drive. The popularity of the Co-win has been such that about 50 countries, including Canada, Mexico, Nigeria and Panama, have shown interest in having a Co-WIN like system to run their vaccination drive, said Dr R S Sharma, the chairman of the empowered group for Covid-19 vaccine administration. (

Photo by Ivan Diaz on Unsplash

Modeling Spread of the Virus

In addition to offering prescriptive solutions, AI has also been used to predict positivity and mortality rates, probable mutations of viruses and their reflections on symptoms, and even arrive at dates and times when the contagion will peak. With data-driven statistics and credible AI modules, officials have been able to proactively take measures like announcing lockdowns and shelter in place protocols, procuring vaccines, oxygen cylinders, PPE kits, testing apparatus, and more. The forecasts that big data generated, pre-empted people to take necessary precautions and in a way; impeded the speed of the spread of the virus.

The Susceptible, Undetected, Tested (positive), and Removed Approach (SUTRA) model ( developed at IIT uses big data and ML tools to predict the virus’s spread within 10% of projection accuracy. Predictions for the USA, India, and Italy from this model closely match the observed outcomes. (


Healthcare centers & institutions are overburdened like never before. Precise diagnostic chatbots have helped provide the first line of treatment. Arogya Setu app has a built-in self-diagnosis tool. Technology concepts such as Natural Language Processing (NLP), Paginemediche (an Italian startup) have rolled out chatbots that offer a highly accurate diagnosis of Covid-19 through data fed to it by users. Based on responses to questions, the chatbots retrieve and offer guidelines, diagnosis, and solutions from the most credible resources and suggest if a patient needs to be isolated or if he needs to seek medical attention. Chatbots can also help the patient understand if their infection is a common flu and not Covid-19. In effect, fewer patients need to go to hospitals and healthcare centers. (

Data Interpretation

Data from large clinical trials can often be hard to interpret. There is a lot of data from clinical trial experiments where AI and ML can find patterns that a human brain might not spot by looking at all the parameters. As vaccine candidates advance to the second and third phases of clinical testing, thousands of patients will be involved. AI systems will be critical in rapidly analyzing big clinical data.

And as more researchers add their studies to the ever-increasing body of literature on the novel coronavirus, scientists will need help sorting through those papers. The Allen Institute developed a resource called CORD-19 that provides more than 130,000 scholarly articles on COVID-19 in a machine-readable format. The Kaggle community, among other groups,

leveraged the data set to create multiple AI systems to help researchers keep up with literature and answer high-priority research questions. (

Caution on AI and ML

Vaccines are first tested in the lab on cells and animals and then on vast numbers of people in clinical tests. Several thousand trial volunteers receive a vaccine before regulators approve it.

AI tools, as we see them now; cannot replace these time-consuming steps. AI might predict which antigens the immune system will see, but what the immune system will do in a living human being is beyond the capabilities of today’s computers. “The human body is so complex that our models cannot necessarily predict with reliability what this molecule or this vaccine will do for the body,” says Oren Etzioni, CEO at the Allen Institute for Artificial Intelligence. “That’s why we have these slow and painful trials — our predictive models aren’t good enough to give you reliable data.” (

#AL #ML #bigdata #vaccines #covid-19 #coronavirus #ArogyaSetu #Moderna #BionTech #Sutra #IIDS #ISBAMPBA


Trending AI/ML Article Identified & Digested via Granola by Ramsey Elbasheer; a Machine-Driven RSS Bot

%d bloggers like this: