Get smart

AI’s role in the COVID vaccination process

Since its inception, artificial intelligence has represented a great hope for medical research. Once taught, AI can simply perform complex and repetitive tasks much faster than humans can. It filters information and recognizes patterns much better than the vast majority of us can. COVID-19 has presented the challenge that is putting to the test how much our current level of AI can aid in medical breakthroughs. First, it was used to help develop vaccines. Now, it’s helping to get vaccines into arms safely.

Side Effects

In the UK, the government has contracted software company Genpact UK to develop AI to process people’s adverse reactions to COVID-19 vaccines. While approved vaccines have undergone lengthy, multistage clinical trials and have been deemed safe by accredited doctors, there are side effects to any drug. There is always a small percentage of people who have negative physical reactions to shots, and governments are planning to vaccinate literally billions of people before the year is out. Not everything will go perfectly.

As a Medicines and Healthcare Regulatory Authority spokesperson told Healthcare IT News, ““Our past experience with other new immunization campaigns is that we tend to receive around 1 Yellow Card report per 1,000 doses administered and we are preparing our surveillance systems on that basis.

“The purpose of the AI tool we are introducing as part of our Yellow Card system is to help us rapidly evaluate such reports after approval, and not as part of the approval process.”

In the US, the FDA awarded prizes to companies, including Enigma, for designing algorithms to spot trends in adverse reactions. Thus far, the FDA has not announced it will employ these algorithms, but the Biden administration may very well do so as it aims to distribute 100 million vaccines in its first 100 days.

Effectiveness

Racial minorities have borne a proportionally heavier burden of COVID-19 death and illness than white people. Not without good reason a higher percentage of minority populations are wary of vaccines. Thus, they were underrepresented in clinical trials and as a result COVID-19 vaccines might not be as effective in some minorities’ immune systems as it will be in the rest of the population.

A team at MIT”s Computer Science and Artificial Intelligence Lab used AI to study the effectiveness of a form of vaccine modeled on the ones Pfizer and Moderna created. They found that people of Asian descent are 20 times more likely than whites to have their immune systems not respond robustly.

“There are obviously many other factors to consider, but our preliminary results suggest that, on average, people of Black or Asian ancestry could have a slightly increased risk of vaccine ineffectiveness,” MIT professor David Gifford, senior author of a new paper outlining the findings, said. “Our work shows that clinical trials need to carefully consider ancestry in their study designs to ensure that efficacy is measured across an appropriate population.”

The lab is using these findings to continue building on its OptiVax design system, which seeks to augment or design vaccines to be effective in people of diverse ethnic backgrounds.

Targeted Distribution

It’s no secret the vaccine rollout started slower than expected in the US. Operation Warp Speed fell well short of its goal of vaccinating 20 million Americans by the end of 2020. AI could get things rolling faster by identifying people whose health history and living conditions make them most vulnerable to COVID. Getting those populations vaccinated first would go a long way to slowing the spread as others await their turn in the vaccine line.

Cogitativo has been doing just that throughout the pandemic and working with Blue Shield of California to provide custom care for the most vulnerable populations. Writing in Fortune, CEO Gary Velasquez said: “Relying solely on age or making judgment calls about whether employees in a particular industry are at greater risk will result in vulnerable populations being overlooked and put at unnecessary risk or worse.”

Mental health, proximity to a grocery store, and financial status are just as if not more important in determining outcomes for COVID patients, Cogitativo’s risk-finder learned.

AI can also analyze where large outbreaks are likely to occur—based on factors such as mask compliance and recent large public gatherings—and manage flexible supply chains that deliver vaccine does to those places.

We’re in a unique situation with vaccines here but the worst of the pandemic is raging. The goal of vaccinating people is to limit suffering and bring the pandemic under control. AI can play a key role in doing that.