A pregnant woman was driving in the HOV lane near Dallas.
ΑΤΗΕΝS — Greece's reopening to tourists in the summer of 2020 as the COVID-19 pandemic was raging presented a problem to the government: how to test all incoming travelers to see if they were infected, time-consuming and difficult.
It's now reported that a Greek-American operations researcher who works in data science at the University of Southern California – Kimon Drakopoulos – played a key role in using Artificial Intelligence (AI) to help identify them.
The site Nature reported that a few months into the pandemic he sent Prime Minister Kyriakos Mitsotakis and the head of Greece's advisory panel of doctors and scientists an email to ask if they needed any more advice.
They did – and fast – so within a few hours, the site said, Drakopoulos got his answer. Yes.
At that time, Greece and most countries hadn't developed the capacity to test all visitors and there was no real track-and-trace method on cell phones and rapid tests and screening processes were still being set up.
Between August and November 2020 — with input from Drakopoulos and his colleagues — Greece launched an AI system that used a machine-learning algorithm to determine which tourists should be tested for COVID.
The researchers said it was more effective at finding asymptomatic people than was random testing or testing based on a traveller’s country of origin. They found the system detected two to four times more infected travelers than did random testing, the site said.
You can thank Eva. That's the name given the machine-learning system Greece used.
There were some concerns though, including accuracy and data privacy during the digital age when spyware has proliferated on phones, hackers can find out your personal details and there's suspicion about government surveillance.
In many countries, travelers are chosen for COVID-19 testing at random or according to risk categories but Eva collected not only travel history, but demographic data such as age and sex from the Passenger Locator Forms (PLF) required for entry to Greece.
The characteristics were matched with data from previously tested passengers and results used to estimate an individual’s risk of infection. COVID-19 tests were targeted to travellers calculated to be at highest risk.
The algorithm also issued tests to allow it to fill data gaps, ensuring that it remained up to date as the situation unfolded, said Nature.
It hasn't been more widespread in use – and was pulled in November 2020 after the trial – because of concerns about governments and companies producing the software having people's data stored and shared with researchers.
“It’s also not clear how consent can be obtained to use such personal data, or how to ensure that these data are stored safely and securely,” the site said.
Eva was developed in consultation with lawyers, who ensured that the program abided by the privacy protections under the European Union's General Data Protection Regulation (GDPR) although used by airlines too.
JUST THE FACTS
GDPR requires that security standards must be followed and that consent is needed to store and use the data that would be shared with a public authority even as skepticism has grown over collecting and storing personal information.
Those worries are also cutting into the ability to use data-harvesting software and AI, especially in the EU which is stricter with regulations, the site recommending that there be data-sharing and privacy-protection protocols.
The British technology news website The Register also reported on Eva and said it used reinforcement learning, specifically multi-armed bandit algorithms, to identify which potentially infected, asymptomatic passengers were worth testing and putting into quarantine if necessary.
It also was able to produce up-to-date statistics on infections for officials to analyze, such as early signs of the emergence of COVID-19 hot spots abroad, the site said.
Eva was used at all 40 of Greece's entry points and travelers were asked to fill out a questionnaire detailing the country and region they were coming from as well as their age and gender.
Based on these characteristics, Eva selected whether they should be tested for COVID-19 upon arrival. At its peak, Eva was apparently processing between roughly 30,000 and 55,000 forms a day, each form representing a household, and about 10 to 20 percent of households were tested, the report said.
The software was crafted to home in on high-risk travelers without relying on test figures provided by individual nations, which might under-report infections, suffer from biases, or lag behind the actual spread of the virus, The Reg said.
Eva would use its own fresh, real-time data from people arriving in Greece, and try to keep the infected away from the general population to help mitigate the pandemic while also scouring the asymptomatic.
“First, given current information, Eva seeks to maximize the number of infected asymptomatic travelers identified,” the US-Greece academic team behind the software explained in a paper in Nature.
“Second, Eva strategically allocates some tests to traveler types for which it does not currently have precise estimates in order to better learn their prevalence,” as a determining method.
“There is a very interesting pattern that we observed and report in our study that shows that increases in the prevalence that we measure through our system are followed by a pickup in cases a few weeks later in the corresponding countries,” Drakopoulos and Vishal Gupta, USC Assistant Professors told The Reg.
“We had enough resources to test about 10 per cent of arrivals in the peak tourist season and 20 percent in the off-peak tourist season when arrivals were lower,” they also said.
Eva ended because, “When the tourist season was over, the number of arriving international passengers became very low, and so there was very little benefit to permitting non-essential travel to the country,” said Drakopoulos.
They didn't reveal how many people were tested after being singled out by Eva, citing privacy reasons. “Clearly, having access to more data would improve performance but would compromise people’s privacy,” Drakopoulos and Gupta told the site.
A pregnant woman was driving in the HOV lane near Dallas.
Want to save this article? Get a subscription to access this feature and more!Subscribe