AI found signs of coronavirus spread on December 30, may help contain it in future
On December 30, researchers using artificial intelligence systems to comb through media and social platforms detected the spread of a unique flu-like illness in Wuhan, China.
It could be days prior to the World Health Organisation released a risk assessment and a complete month prior to the UN agency declared a global public health emergency for the novel coronavirus.
Could the AI systems have accelerated the process and limited, or even arrested, the extent of the COVID-19 pandemic?
Clark Freifeld, a Northeastern University computer scientist dealing with the global disease surveillance platform HealthMap, among the systems detecting the outbreak, said it's hard to tell the severity of the illness.
Dataminr, a real-time risk detection technology firm, said it delivered the earliest warning about COVID-19 on December 30 predicated on eyewitness accounts from inside Wuhan hospitals, pictures of the disinfection of the Wuhan seafood market where the virus originated and a warning by a Chinese doctor who later died from the virus himself.
“Among our biggest challenges is we have a tendency to be reactive in these circumstances, it’s human nature,” said Kamran Khan, founder and leader of the Toronto-based disease tracking firm BlueDot, one of the early systems that flashed warning flags in December over the epidemic.
“Whenever you’re dealing with a fresh, emerging disease, you don’t have all the answers. Time is your most effective resource; you cannot obtain it back.”
Khan, who's also a professor of medicine and public health at the University of Toronto, told AFP by telephone the info showed “echoes of the SARS outbreak 17 years earlier, but we didn’t know was how contagious this is.”
Nevertheless, AI systems have tested to be valuable in tracking epidemics by scouring a varied array of sources which range from airline bookings, Twitter and Weibo messages to news reports and sensors on linked devices.
Humans in the loop
Still, Freifeld said AI systems have limits, and the big decisions must be made by humans.
“We utilize the AI system as a force multiplier, but we are focused on the concept of having humans in the loop,” he said.
AI and machine learning systems will probably help the battle in a number of ways, from tracking the outbreak itself to speeding up drug testing.
“We are able to run simulations unlike we’ve ever done before, we understand biological pathways unlike we’ve ever understood before, and that’s all because of the energy of AI,” said Michael Greeley of the equity firm Flare Capital Partners, which has committed to several AI medical startups.
But Greeley said it remains challenging to apply these technologies to sectors like drug delivery where in fact the normal testing time could be years.
“There is extraordinary strain on the industry to begin using these tools despite the fact that they may not be equipped for prime time,” he said.
According to Khan, AI is helping in the containment phase with systems which used “anonymised” smartphone location data to track the progression of the disease and find hotspots, and also to determine if persons are following “social distancing” guidelines.
Andrew Kress, CEO of medical technology firm HealthVerity, said it remains challenging to gather medical data for disease outbreaks while complying with patient privacy.
It’s possible to find trends with signals such as for example pharmacy visits and sales of certain medications or even online searches, Kress said, but aggregating which has privacy implications.
“We need to have a genuine discussion about balance and utility around specific use cases and potentially the proper kind of research to keep to determine new methods to leverage many of these nontraditional data sources,” Kress said.
Data mining
AI systems are also being put to work to scour the a large number of research studies for clues on what treatments could possibly be effective.
The other day, researchers joined the White House in order to provide some 29,000 coronavirus research articles which can be scanned for data mining.
Your time and effort brought together the Allen Institute for AI, Chan Zuckerberg Initiative, Microsoft, Georgetown University and others.
Through Kaggle, a machine learning and data science community owned by Google, these tools will be openly designed for researchers all over the world.
“It’s difficult for folks to manually proceed through a lot more than 20,000 articles and synthesise their findings,” said Kaggle CEO and co-founder Anthony Goldbloom.
“Recent advances in technology can be helpful here. We’re putting machine-readable versions of the articles before our community greater than four million data scientists. Our hope is that AI can be used to greatly help find answers to an integral set of questions about COVID-19.”
