Did AI anticipate coronavirus pandemic outbreak?

Image Source: https://www.acfcs.org/

In an article titled "Can AI flag disease outbreaks faster than humans? Not quite" that was published on 20 February 2020, Associated Press reported that the first public alert outside China about the novel coronavirus came on 30 December 2019 from the automated HealthMap system at Boston Children’s Hospital. However,  it took human intelligence to recognize the significance of the outbreak and then awaken response from the public health community.

A follow up article by The Next Web details the timeline about how the first wave of global pandemic alerts unfolded as below:

Earlier reports had suggested that a Canadian epidemiologist had raised the first warnings of the outbreak (on 31 Dec 2019), using an algorithm called BlueDot that scanned news reports and airline ticketing data to predict the spread of the disease.

Associated Press reporters Christina Larson and Matt O’Brien were dubious about the claim, and decided to draw up a timeline of when global alert systems noticed the signals.

They determined that the first warning outside China of the virus came from the automated HealthMap system at Boston Children’s Hospital, which scans online news and social media reports for signals of spreading disease. 

At 11:12PM local time on December 30, it sent an alert about unidentified pneumonia cases in Wuhan — but only ranked its seriousness as a three out of five.

Half an hour after the HealthMap system had sent its alert, the human volunteer-led Program for Monitoring Emerging Diseases (ProMed) produced a more detailed warning. Marjorie Pollack, ProMed’s deputy, had first noticed the signs four hours before the HealthMap alert.

She had received an email informing her of a Chinese social media post discussing a Wuhan health agency notice of “unexplained pneumonia” and sent her team to investigate. Their analysis led their coronavirus alert to be sent out slightly later than the AI, but it also meant that they could provide a more comprehensive warning.

Both the systems were credited with producing their alerts before the BlueDot warning (that was sent out one day later).

Associated Press also pointed out that ProMed reports prepared by human experts are often incorporated into other outbreak warning systems, including those run by the World Health Organization, the Canadian government and the Toronto startup BlueDot. World Health Organisation also pools data from HealthMap and other sources.

MIT Technology Review on the other hand suggested on 12 March 2020 that AI could help with the next pandemic—but not with this one.

"It was an AI that first saw it coming, or so the story goes. On December 30 (Note 1), an artificial-intelligence company called BlueDot, which uses machine learning to monitor outbreaks of infectious diseases around the world, alerted clients—including various governments, hospitals, and businesses—to an unusual bump in pneumonia cases in Wuhan, China. It would be another nine days before the World Health Organization officially flagged what we’ve all come to know as Covid-19.

BlueDot wasn’t alone. An automated service called HealthMap at Boston Children’s Hospital also caught those first signs. As did a model run by Metabiota, based in San Francisco. That AI could spot an outbreak on the other side of the world is pretty amazing, and early warnings save lives."

But how much has AI really helped in tackling the current outbreak? That’s a hard question to answer. Companies like BlueDot are typically tight-lipped about exactly who they provide information to and how it is used. And human teams say they spotted the outbreak the same day as the AIs. Other projects in which AI is being explored as a diagnostic tool or used to help find a vaccine are still in their very early stages. Even if they are successful, it will take time—possibly months—to get those innovations into the hands of the health-care workers who need them.

The hype outstrips the reality. In fact, the narrative that has appeared in many news reports and breathless press releases—that AI is a powerful new weapon against diseases—is only partly true and risks becoming counterproductive. For example, too much confidence in AI’s capabilities could lead to ill-informed decisions that funnel public money to unproven AI companies at the expense of proven interventions such as drug programs. It’s also bad for the field itself: overblown but disappointed expectations have led to a crash of interest in AI, and consequent loss of funding, more than once in the past.

So here’s a reality check: AI will not save us from the coronavirus—certainly not this time. But there’s every chance it will play a bigger role in future epidemics—if we make some big changes in three main areas where AI could help: prediction, diagnosis, and treatment. 

Note 1:  Most of the sources disputed 30 December 2019 as the date when BlueDot has sent out its first alert, instead, it was suggested 31 December 2019 to be the correct date.

The Brookings Institution’s Artificial Intelligence and Emerging Technology (AIET) Initiative summarizes AI's capability in predicting unprecedented rare event in a report titled "A guide to healthy skepticism of artificial intelligence and coronavirus" as follows:

AI is far better at minute details than big, rare events

Wired ran a piece in January titled “An AI Epidemiologist Sent the First Warnings of the Wuhan Virus” about a warning issued on Dec. 31 by infectious disease surveillance company, BlueDot. One blog post even said the company predicted the outbreak “ before it happened.” However, this isn’t really true. There is reporting that suggests Chinese officials knew about the coronavirus from lab testing as early as Dec. 26. Further, doctors in Wuhan were spreading concerns online (despite Chinese government censorship) and the Program for Monitoring Emerging Diseases, run by human volunteers, put out a notification on Dec. 30.

That said, the approach taken by BlueDot and similar endeavors like HealthMap at Boston Children’s Hospital aren’t unreasonable. Both teams are a mix of data scientists and epidemiologists, and they look across health-care analyses and news articles around the world and in many languages in order to find potential new infectious disease outbreaks. This is a plausible use case for machine learning and natural language processing and is a useful tool to assist human observers. So, the hype, in this case, doesn’t come from skepticism about the feasibility of the application, but rather the specific type of value it brings.

Even as these systems improve, AI is unlikely to build the contextual understanding to distinguish between a new but manageable outbreak and an emerging pandemic of global proportions. AI can hardly be blamed. Predicting rare events is just very hard, and AI’s reliance on historical data does it no favors here. However, AI does offer quite a bit of value at the opposite end of the spectrum—providing minute detail.

For example, just last week, California Gov. Gavin Newsom explicitly praised BlueDot’s work to model the spread of the coronavirus to specific zip codes, incorporating flight-pattern data. This enables relatively precise provisioning of funding, supplies, and medical staff based on the level of exposure in each zip code. This reveals one of the great strengths of AI: its ability to quickly make individualized predictions when it would be much harder to do so individually. Of course, individualized predictions require individualized data, which can lead to unintended consequences.

More recently, CNBC dissects AI's limited roles to date in tackling coronavirus crisis in an article that was published on 29 April 2020. CNBC reports that AI’s role in this pandemic is likely to be more nuanced than some may have anticipated such as making people’s lives easier and more fun while they’re in lockdown, autonomous robots trained with deep learning techniques can be used to disinfect hospitals, AI tutors can support parents with the burden of home schooling, AI driven CCTV could track whether the citizens are wearing masks and etc . Rather than being mainstream, AI is likely to play peripheral roles in responding to various Covid-19 driven scenarios. AI isn’t about to get us out of the woods any time soon.

Why hasn’t AI had more impact?

Catherine Breslin, a machine learning consultant who used to work on Amazon Alexa explained that AI researchers rely on vast amounts of nicely labeled data to train their algorithms, but right now there isn’t enough reliable coronavirus data to do that.

“AI learns from large amounts of data which has been manually labeled — a time consuming and expensive task,” said Catherine Breslin.

“It also takes a lot of time to build, test and deploy AI in the real world. When the world changes, as it has done, the challenges with AI are going to be collecting enough data to learn from, and being able to build and deploy the technology quickly enough to have an impact.”

Breslin agrees that AI technologies have a role to play. “However, they won’t be a silver bullet,” she said, adding that while they might not directly bring an end to the virus, they can make people’s lives easier and more fun while they’re in lockdown.

Of course, there are a few useful AI projects happening here and there.

In March, DeepMind announced that it had used a machine-learning technique called “free modelling” to detail the structures of six proteins associated with SARS-CoV-2, the coronavirus that causes the Covid-19 disease. Elsewhere, Israeli start-up Aidoc is using AI imaging to flag abnormalities in the lungs and a U.K. start-up founded by Viagra co-inventor David Brown is using AI to look for Covid-19 drug treatments.

Verena Rieser, a computer science professor at Heriot-Watt University, pointed out that autonomous robots can be used to help disinfect hospitals and AI tutors can support parents with the burden of home schooling. She also said “AI companions” can help with self isolation, especially for the elderly.

“At the periphery you can imagine it doing some stuff with CCTV,” said former director of machine learning at Amazon Cambridge Neil Lawrence, adding that cameras could be used to collect data on what percentage of people are wearing masks.

Separately, a facial recognition system built by U.K. firm SCC has also been adapted to spot coronavirus sufferers instead of terrorists. In Oxford, England, Exscientia is screening more than 15,000 drugs to see how effective they are as coronavirus treatments. The work is being done in partnership with Diamond Light Source, the U.K.’s national “synchotron.”

But AI’s role in this pandemic is likely to be more nuanced than some may have anticipated. AI isn’t about to get us out of the woods any time soon.

On another note, World Economic Forum emphasizes on the significance of human experts' role in driving data driven AI to combat COVID-19 global pandemic crisis:

Artificial intelligence (AI) has the potential to help us tackle the pressing issues raised by the COVID-19 pandemic. It is not the technology itself, though, that will make the difference but rather the knowledge and creativity of the humans who use it.

Indeed, the COVID-19 crisis will likely expose some of the key shortfalls of AI. Machine learning, the current form of AI, works by identifying patterns in historical training data. When used wisely, AI has the potential to exceed humans not only through speed but also by detecting patterns in that training data that humans have overlooked.

However, AI systems need a lot of data, with relevant examples in that data, in order to find these patterns. Machine learning also implicitly assumes that conditions today are the same as the conditions represented in the training data. In other words, AI systems implicitly assume that what has worked in the past will still work in the future.

What does this have to do with the current crisis? We are facing unprecedented times. Our situation is jarringly different from that of just a few weeks ago. Some of what we need to try today will have never been tried before. Similarly, what has worked in the past may very well not work today.

Humans are not that different from AI in these limitations, which partly explains why our current situation is so daunting. Without previous examples to draw on, we cannot know for sure the best course of action. Our traditional assumptions about cause and effect may no longer hold true.

Humans have an advantage over AI, though. We are able to learn lessons from one setting and apply them to novel situations, drawing on our abstract knowledge to make best guesses on what might work or what might happen. AI systems, in contrast, have to learn from scratch whenever the setting or task changes even slightly. 

Finally, in his latest article published on WIRED magazine, world-leading artificial intelligence expert and former president of Google Greater China Dr. Kai-Fu Lee made the following assessment on the impact of AI against Covid-19 pandemic for the recent 4 months: 

These snapshots of AI in action against Covid-19 provide a glimpse of what will be possible in the various aspects of health care in the future. We have a long way to go. Truth be told, AI has not had a particularly successful four months in the battle of the pandemic. I would give it a B-minus at best. We have seen how vulnerable our health care systems are: insufficient and imprecise alert responses, inadequately distributed medical supplies, overloaded and fatigued medical staff, not enough hospital beds, and no timely treatments or cures.

Health care systems around the world—even the most advanced ones—are some of the most complicated, hierarchical, and static institutions in society. This time around, AI has been able to help in only pockets of excellence. The reasons for this are simple: Before Covid-19 struck, we did not understand the importance of these areas and act accordingly, and, crucially as far as AI is concerned, we did not have the data to deliver the solutions.

Nevertheless, Dr Kai-Fu Lee is optimistic that as the pandemic rolled around the planet, more and more innovative applications of AI have cropped up in many different locations including South Korea's location-based messaging that is capable to alert citizens when they are near a confirmed case, China's Alibaba AI algorithm that it says can diagnose suspected cases within 20 seconds with 96 percent accuracy, Robots in China’s Hubei and Guangdong provinces delivered food, medicine, and goods to patients in hospitals or quarantined families,  and the use of AI models and algorithms for new drug discovery and medical breakthroughs in genomic sequencing, stem cells and Crispr.

He further elaborates that there are two grounds for optimism:

The first is that data, always the lifeblood of AI, is now flowing. Kaggle, a machine learning and data science platform, is hosting the Covid-19 Open Research Dataset. CORD-19, as it is known, compiles relevant data and adds new research into one centralized hub. The new data set is machine readable, making it easily parsed for AI machine learning purposes. As of publication, there are more than 128,000 scholarly articles on Covid-19, coronavirus, SARS, MERS, and other relevant terms.

The second is that medical scientists and computer scientists across the world are now laser-focused on these problems. Peter Diamandis, founder of the XPrize Foundation, estimated that up to 200 million physicians, scientists, nurses, technologists, and engineers are now taking aim at Covid-19. They are running tens of thousands of experiments and sharing information “with a transparency and at speeds we’ve never seen before.”

Conclusion

The neural network based machine learning and deep learning models driven applications can only accurately model and predict the fast changing dynamic world when it is fed with continuous incoming data stream from the relatively normal operating world. In rare abrupt development that breaks the normal pattern and terminates the supply of reliable training data, Machine Learning and Deep Learning models can no longer accurately predict the world.

Under such circumstances, we might need to let human centric augmented intelligence to take over the driver seat before the AI models get adjusted with the new normal. Besides being more flexible and agile, human subject matter experts also have clear advantage over AI in drawing immediate lessons learned from the unprecedented situation, forming hypothesis and conducting experiments to look for the best remedies; while AI models need to be trained with new dataset with different distributions (maybe also with different parameters and different parameter weightings) from scratch in order to make sense of the new situation.

Further Readings:

[1] Associated Press: Can AI flag disease outbreaks faster than humans? Not quite
https://apnews.com/100fbb228c958f98d4c755b133112582

[2] The Brookings Institution’s Artificial Intelligence and Emerging Technology (AIET) Initiative: A guide to healthy skepticism of artificial intelligence and coronavirus
https://www.brookings.edu/research/a-guide-to-healthy-skepticism-of-artificial-intelligence-and-coronavirus/

[3] AI could help with the next pandemic—but not with this one
https://www.technologyreview.com/2020/03/12/905352/ai-could-help-with-the-next-pandemicbut-not-with-this-one/

[4] World Economic Forum: AI can help with the COVID-19 crisis - but the right human input is key
https://www.weforum.org/agenda/2020/03/covid-19-crisis-artificial-intelligence-creativity/

[5] The Next Web: AI sent first coronavirus alert, but underestimated the danger
-- The humans were 30 minutes slower but faster to recognize the significance of the outbreak
https://thenextweb.com/neural/2020/02/21/ai-sent-first-coronavirus-alert-but-underestimated-the-danger/

[6] CNBC: A.I. can’t solve this: The coronavirus could be highlighting just how overhyped the industry is
https://www.cnbc.com/2020/04/29/ai-has-limited-role-coronavirus-pandemic.html

[7] Wired: An AI Epidemiologist Sent the First Warnings of the Wuhan Virus
-- The BlueDot algorithm scours news reports and airline ticketing data to predict the spread of diseases like those linked to the flu outbreak in China.
https://www.wired.com/story/ai-epidemiologist-wuhan-public-health-warnings/

[8] Wired: Covid-19 Will Accelerate the AI Health Care Revolution
-- Disease diagnosis, drug discovery, robot delivery—artificial intelligence is already powering change in the pandemic’s wake. That’s only the beginning.
https://www.wired.com/story/covid-19-will-accelerate-ai-health-care-revolution/