The science fiction of yesterday seems like it’s becoming science fact today. Fans of Netflix’s Stranger Things will recall the Shadow Monster, who intended to kill everything and everyone on Earth by spreading a virus via humans. Sadly, as the ever-increasing incidents of pandemics over the last decade have shown, narratives such as the Shadow Monster are steadily expanding into the real world.
The monster now has a new name: COVID-19.
With the virus quickly mutating as it’s transmitted from one individual to another, it’s difficult for scientists to keep up. Our ability to create a vaccine or even provide a treatment is taking more time than we’d like. The severity of the COVID-19 threat calls for a proactive approach to building predictive modeling systems to understand the code (RNA, DNA) of future viruses and act accordingly.
Making Use of the Best Tool Available: Data
Fast access to the millions of data points available in an outbreak (thanks to the rise of preprint servers such as bioRxiv) will help, as will predictive analytics for future detection and therapeutic applications. Insilico Medicine, a startup focused on using analytics for disease prevention, is rapidly identifying molecules that could form the basis of an effective treatment. It took Insilico’s predictive system four days to identify thousands of new molecules that could be turned into potential medicinal therapies to combat the virus.
Deep-learning algorithms are also helping to make sense of the gargantuan datasets for steroid identification that are beyond the comprehension of today’s data scientists. The first system to warn of the outbreak was the deep-learning-driven BlueDot. The intelligent system sent an email to its clients warning of the virus a week before the World Health Organization alerted the rest of us.
Akin to the rubber, scrap metal and paper drives of World War II, perhaps we’ll need “protein drives” to better understand the molecular composition of infectious disease. AlphaFold, a deep-learning system, is working to predict the interactions of protein structures with chemical compounds to facilitate new drugs or recommend current medicines. It recently put its resources online, where average people can share their health information to help scientists pinpoint the best, minute locations within the protein strands comprised by the coronavirus from which to subdue it. The researchers have developed a massive-scale distributed computing project, called Folding@home, that harnesses AlphaFold to run simulations.
Arraying Massive Compute Power With Process Modernization
Today’s massively parallel computers and tomorrow’s quantum computers will further enable us to process more data more quickly and precisely to speed new molecule discovery. Already, companies like Kebotix in Cambridge, Mass., are using AI and machine learning to synthesize molecules that scientists might use. In early 2020, Kebotix partnered with the National Center for Advancing Translational Sciences (NCATS) at the National Institutes of Health to use AI and robotics to increase the rate of discovery at lower costs. Previously, the NCATS team needed to execute, by brute force, full complements of 294 experiments.
These types of approaches could also yield breakthroughs in materials science that could help in our fight against climate change. By giving the platform an outcome, such as “create a weed killer that does the job without harming the soil, other plants or water” (or, when it comes to infectious diseases, a virus-killer that doesn’t harm people), AI then runs the tesseract of all permutations of the periodic table of elements to get to the outcome.
On the home front, self-administered processes for prediction, prevention and early detection are all part of the technology prescription, as well. We’ll rapidly see more devices that can screen earlier and offer different approaches to medication therapies. For example, it’s easy to imagine requiring people to undergo an automated, self-administered, Star Trek-like “tricorder” scan (similar to a pre-boarding scan at an airport) before entering any building, and be turned away if symptoms of contagion are detected. After all, a lasting lesson from COVID-19 is that early and extreme caution via mandated self-quarantine practices saves lives.
Process transformation is also critical. We already see today’s manual work processes becoming digitized, allowing for better doctor-to-doctor communications, and helping to eliminate the need for expensive, repeat testing (all while helping to make the phrase “fax me the patient’s documents” as rare as using a leech for bloodletting). Further process advancements in telemedicine and truly interoperable EMR/EHRs that travel with the patient (rather than sitting in disconnected health provider and insurance carrier silos) will also help.
Disease Identification, Miniaturized & Personalized
While no one is doing cardiac ablation by iPhone (yet), we need wider availability of devices resembling “a Fitbit for your physiology” (i.e., sensors that monitor more than just heart rate and steps). Such advancements will undeniably yield better diagnostics, intelligent routing to specialists and home-based triaging to relieve beleaguered doctors and nurses.
Witness tools like AccuVein’s virtual phlebotomy device that – driven by augmented reality – can “see” subcutaneously through the patient’s skin to find a vein, leading to a reported 45% reduction in escalations. How about a “Shazam for your cough,” which could facilitate early detection of asthma, bronchitis or pneumonia, yielding results for immediate interpretation by general practitioners and specialists alike (akin to what’s being worked on by Australia’s Resapp Health)? Or DIY telehealth evaluations like Everlywell’s easy, at-home lab tests, which empower people to live healthier lives through technological advances?
Tackling the Unknown
As of mid-April, over 2.5 million confirmed cases of COVID-19 have been reported worldwide, and nobody knows when the count will stop rising. Simply put, the greatest challenge for humanity is to find a therapy to control the symptoms and ill effects of contracting the virus. Hopefully, a vaccine will be brought to market in the next year to 18 months, but as history has taught us, hope is not a strategy.
The increasingly sophisticated technologies of AI, machine learning, prediction, prevention and deep learning will alter the way drug discoveries are made, and provide us with more effective alternatives for handling future pandemics.
As COVID-19 has taken the world into unknown territory, we urgently need to dig deeper with the responsible use of new technologies that can help us identify ways to manage the monster, and then defeat it.
This blog is part of our special report on the future of infectious disease. Stay tuned for more blogs on this topic.
Visit our COVID-19 resources page for additional insights and updates.
Subscribe to our newsletter and get expert insights straight to your inbox.×
SUBSCRIBE TO OUR NEWSLETTER✖
Thanks for your interest in Digitally Cognizant.
To complete the subscription process, please click the link in the email we just sent you.