From Wharton.
A new machine learning approach to COVID-19 testing has produced encouraging results in Greece. The technology, named Eva, dynamically used recent testing results collected at the Greek border to detect and limit the importation of asymptomatic COVID-19 cases among arriving international passengers between August and November 2020, which helped contain the number of cases and deaths in the country.

The findings of the project are explained in a paper titled “Deploying an Artificial Intelligence System for COVID-19 Testing at the Greek Border,” authored by Hamsa Bastani, a Wharton professor of operations, information and decisions and affiliated faculty at Analytics at Wharton; Kimon Drakopoulos and Vishal Gupta from the University of Southern California; Jon Vlachogiannis from investment advisory firm Agent Risk; Christos Hadjicristodoulou from the University of Thessaly; and Pagona Lagiou, Gkikas Magiorkinis, Dimitrios Paraskevis and Sotirios Tsiodras from the University of Athens.

The analysis showed that Eva on average identified 1.85 times more asymptomatic, infected travelers than what conventional, random surveillance testing would have achieved. During the peak travel season of August and September, the detection of infection rates was up to two to four times higher than random testing.

“Our work paves the way for leveraging [artificial intelligence] and real-time data for public health goals, such as border control during a pandemic,” the paper stated. With the rapid spread of a new coronavirus strain, Eva also holds the promise of maximizing the already overburdened testing infrastructure in most countries.

“The main issue was, given the fixed budget for tests, whether we could conduct the tests in a smarter way with dynamic surveillance to identify more infected travelers,” said Bastani. One of the biggest challenges governments face in dealing with COVID-19 is the inability of the testing infrastructure at their national borders to realistically check every arriving passenger. Such comprehensive testing would be both costly and time-consuming, which is why most countries screen either arriving passengers from specific countries or conduct random testing for COVID-19.

“The main issue was, given the fixed budget for tests, whether we could conduct the tests in a smarter way with dynamic surveillance to identify more infected travelers.”–Hamsa Bastani

Eva also allowed Greece to identify when a country was exhibiting a spike in COVID-19 infections a median of nine days earlier than what would have been possible with machine learning-based algorithms using only publicly available data.

The underlying technology of Eva is a “contextual bandit algorithm,” a machine-learning framework built for “sequential decision-making,” taking into account various practical challenges like time-varying information and port-specific testing budgets, Bastani explained. The algorithm balances the need to maintain high-quality surveillance estimates of COVID-19 prevalence across countries and the allocation of limited testing results to catch likely infected travelers. Eva is the first instance of that technology being applied to address a public health challenge, although such algorithms have found use in online advertising and A/B testing, she added.

Currently, the state-of-the-art or best performing AI models are almost all based on deep learning models, Professor Sun observes. In deep learning, the model learns to perform recognition tasks from images, text, or sound based on the deep neural network architectures that contain many layers. If the input is an image, for example, the assumption is the image can be described by different spatial scales or layers of features.