One of the biggest reasons the coronavirus has been so destructive is its definition as a novel virus. According to epidemiologist Larry Brilliant in an article with WIRED, the virus itself is "new," which means "there is no human being in the world that has immunity as a result of having had it before. That means it's capable of infecting 7.8 billion." Technology is doing its part to make sure the extreme numbers in the models are dramatically reduced.
We've got to rely on numbers to help us lower the curve now more than ever since the idea of infection of that magnitude is unthinkable. But there's also something else to consider: With a virus that's spreading as rapidly as COVID-19, real-time data is more critical than ever, as traditional forecasting models rely on years of data — and that's the one thing that no one has.
Roni Rosenfeld, professor of computer science at Carnegie Mellon, has been steadily working on making epidemiological forecasting with machine learning as "universally accepted and useful as weather forecasting." So when the CDC called wanting his hand in predicting the COVID-19 spread, he went straight for AI.
"We have another methodology for forecasting, called the 'wisdom of crowds,'" said Rosenfeld in an interview with Vox. "That's when you gather at least several dozen people and ask them each individually to make a subjective assessment of what the rest of the flu season will look like. What we've learned from experience is that any one of them on their own is not very accurate, but their aggregate tends to be quite accurate."
In addition to crowdsourcing (which has expanded to include everyone, even you, if you're interested), Rosenfeld's model is also taking into account several other data points. Those points include: frequency of the illness mentioned on social media, Google queries for related terms, access to CDC pages and relevant Wikipedia pages, retail purchasing patterns for things like anti-fever medications and thermometers, and health records while removing items that may seem high-volume for social anxiety versus actual illness. This modeling has already been used on predicting how the seasonal flu affects populations, and it's worked. Now, he's saying that we'll see a peak in April or May unless social distancing and shelter-in-place measures are stepped up to help mitigate the more severe timelines.
At the same time, supercomputers are part of the mix — tackling the virus's very deadly ability to spread exponentially with the power of computing at one billion times the speed of a standard laptop. IBM's Summit is that supercomputer, and it's been used to identify traits that lead to opioid addiction, identify patterns that lead to Alzheimer's, and more, in the past. Now, it's being trained on determining how the virus proteins react to certain compounds. So far, it's ranked 77 potential compounds that could potentially help fight the virus, and the longer it runs, the closer we get to reducing the time and cost of clinical trials that arrive at a treatment.
More Technology Combating the Virus Already
As a strained supply chain struggles to meet the exponential needs of frontline doctors, nurses, and medical staff, 3D printers have become literal life savers. Manufacturers are teaming up to produce vital 3D-printed medical equipment such as nasal swabs, face shields, ventilator parts, and more. One Ohio factory, which typically produces 3D printers, is utilizing 250 of its printers “to crank out up to 100,000 nasal swabs for Covid-19 tests every day.” Even on a local level, this Virginia student is helping combat the virus at home with a 3D printer by creating masks for his at-risk uncle and donating others to his community.
Here are a few other examples of how businesses are leveraging the power of tech during this critical time:
- South Korean company Seegen (among other molecular biology companies in SK) started on a test four days before they even had a confirmed case, and is the reason they have had free and easily-accessible tests available for everyone who needs it.
- Another South Korean company, Deargen, is using AI to test molecular sequences against COVID-19 to identify antivirals that may stop the spread of the virus and get us to a treatment faster.
- California's SRI International and Iktos in Paris also recently announced their collaboration to accelerate the discovery of new viral therapies by combining their AI for new drug development and research facilities. Their goal is to expedite the identification of drug candidates that can help treat the virus.
- On the diagnosing front, Infervision launched the Coronavirus artificial intelligence solution, which goes straight to the frontline and helps clinicians diagnose and monitor COVID-19 in patients.
- Amazon Web Services, Google Cloud, and others have offered researchers access to open datasets and analytics tools to help them develop COVID-19 solutions.
- Multiple organizations are trying to use Artificial Intelligence (AI) and data analytics to race to solutions.
- At MIT, mobile technologies is being used to develop privacy-first contact tracing. Health care providers can download the names of those who were in close proximity to the infected individual during the relevant time frame.
- Drones are now in the mix. Terra Drone is helping people in China keep up with their social distancing while still getting the medical supplies they need.
- SnapNurse, a gig economy company with a software platform that connects available nurses and med techs to jobs, is helping get more people into our healthcare system.
- Robots are helping out, too. BlueOcean Robotics is using ultraviolet light to kill bacteria and viruses for several different tasks, and Pudu Technology has re-rerouted its catering robots to assist medical facilities in China.
- Healthcare startup Loyal Health in Atlanta has developed an AI driven chat bot to help assess and direct those with suspected symptoms to care.
Everywhere the world has rallied behind the capabilities of where technology can lend a hand in combating this novel disease, and there are plenty more places where COVID-19 is being countered with data science, analytics, AI, and other technologies.
Photo by Dennis Kummer on Unsplash
Comments?
Leave us your opinion.