I was worried about COVID-19 way before most people. Back in January, reports of a new virus tearing through China Wuhan wouldn’t go away. I know about the history of pandemics. I know it’s unlikely that China will be able to contain a highly communicable virus within its own borders. 

This isn’t some cheap “I Told You So.” Maybe, like a lot of engineers, I’m just very risk-aware! But I’ve been following the statistical projections around this closely. Particularly fascinating are:  

Epidemiology involves a ton of math. It’s about rigorous mathematical and statistical models. Like many engineers, I’m drawn to numerical projections like this.  

One of the ways that the world is attempting to deal with COVID-19 is through artificial intelligence (AI). At Opsani, it’s my job to refine the algorithms that can make cloud application optimization faster, more reliable, and more accurate. This involves not only the development of optimization algorithms but also various other algorithms that can assist in boosting this core purpose.

Opsani can only help with the COVID-19 outbreak secondarily – by helping internet services that are facilitating the response to run more smoothly during a time of increased usage. But other applications of AI can help more directly. We, Opsani engineers, are evangelists of artificial intelligence, machine learning, deep reinforcement learning, and everything else in the world of AI. It’s been super exciting to see how AI is helping medical professionals and governments battle the pandemic. 

There are two excellent AI applications we are seeing. Number one: detecting the nature of the disease prevalence. For example, Behold.ai, who has developed an algorithm for detecting lung cancer from X-rays, can use their algorithm to identify chest X-rays from COVID-19 patients as ‘abnormal’. This could rapidly speed up diagnosis. Insilico Medicine is using algorithms to produce new molecules that might restrict COVID-19’s ability to replicate within human cells. Their researchers have published a paper reporting that “a total of 28 machine learning models generated molecular structures and optimized them with reinforcement learning.”

AI is also supercharging drug research. We won’t beat COVID-19 until we have a vaccine. Google DeepMind has sprung into action here. How? By leveraging deep learning to uncover new information about the structure of proteins associated with COVID-19. Speaking of earlier trials, DeepMind’s team said: 

“We confirmed that our system provided an accurate prediction for the experimentally determined SARS-CoV-2 spike protein structure shared in the Protein Data Bank, and this gave us confidence that our model predictions on other proteins may be useful.” 

Cautious, but optimistic. 

There are tons of positive applications for AI. Zipline and Matternet are using drones to deliver life-saving medical supplies and blood units; Touch Surgery has launched the world’s first dynamic artificial intelligence (AI) system to guide surgical teams through complex procedures; The Ocean Cleanup uses satellite imaging and machine learning to rid the world’s oceans of plastic.

AI companies jumping into action to fight COVID-19 continues a proud tradition of using cutting-edge technology to do good in the world. AI is a powerful tool to help us save lives, and drag ourselves out of this terrible crisis.

If you’re interested in reading more about how the world is utilizing AI to help address this pandemic, here are a couple of interesting articles:

AI (Artificial Intelligence) Companies That Are Combating The COVID-19 Pandemic

Artificial Intelligence against COVID-19: An Early Review