Beating COVID-19 Is a Team Sport
by: Edward Scott
April 23, 2020
Artificial intelligence and machine learning are proving to be meaningful weapons in our arsenal during the coronavirus crisis.
For months we’ve watched the world suffer as the coronavirus pandemic spreads to every corner of the globe. With more than 2.5 million confirmed COVID-19 cases worldwide, and more than 150,000 fatalities, every measure to slow the spread and find both effective treatment as well as a vaccine could save the lives of hundreds of thousands of people.
At the forefront of that movement is artificial intelligence. AI and machine learning have proven to be some of the most powerful, transformative and innovative technologies to define our modern world. And it’s more important than ever that we utilize those technologies to their full potential.
Now is the time for everyone to join the fight, using the full force of their skills, knowledge and technological know-how to aid in the crisis. As scientists and engineers explore every possible option, AI has generated some truly marvelous and promising leads. From using AI algorithms to search for new molecules that will help with treatment, to others that can assist in monitoring the spread and predict outbreak zones, the technology can not only help to contain, but also stop the spread of COVID-19. Behind every corner, people are trying new and innovative uses of AI to help in this fight. Unlike previous pandemics, AI has proven to be a new, effective weapon in our arsenal. However, where I am most confident in AI to make an immediate and meaningful impact is in medical imagery.
Healthcare networks in COVID-19 hot zones are in desperate need of immediate and effective solutions to manage the admittance of patients. Waiting rooms are overflowing as hospitals quickly approach or in some cases, reach capacity, and many patients who are deemed likely to recover without hospital care are being sent home even if they test positive and show signs of infection. If you go to a hospital in a crisis zone today, you may learn that yes, you indeed do have the virus, and no, you will not be admitted to the hospital. Even after all the social distancing, and unprecedented measures to control this virus have been taken, we have not reduced the spread of the disease enough to take the burden off our hospitals systems. Only in this current crisis have we seen our healthcare centers turn away patients who clearly would benefit from medical help. The decision to send someone home is not easy to make, and with the unpredictability of the virus, the lack of reliable information, the inability to study and learn the nuances of care usually afforded our treatment providers—sometimes that decision is not made correctly. Doctors in these hospitals can be found checking their patient lists, looking to see if a patient they sent home yesterday was indeed a wrong call. It’s unfair to ask our front-line workers to not take advantage of the best technology our community has to offer.
AI applied to medical imaging has immense potential to resolve one of these most crucial questions in our hospitals today: Who gets admitted, and who is sent home? An X-ray is one dimension that can help make a diagnosis by determining the extent of damage and even the progression of the disease over time. However, when machine learning is applied, it can help augment a radiologist’s knowledge in several ways, allowing them to focus their attention on those critical, tough calls effectively and with more information.
One particular X-ray solution has been built in for this purpose in cooperation with New York area hospitals as a direct response to current clinical practices. In collaboration with more than 20 radiologists and data scientists, ElectrifAi has aggregated X-ray and CT-scan images of COVID-19 infected lungs. Starting this week, PulmoAi X-ray will now use images of lungs along with the power of cloud compute to give front-line hospital workers help with those critical decisions, with the speed and accuracy they need to manage the crisis.
We are also applying AI algorithms to CT scans, to learn more about coronavirus and the entirely unpredictable and inexplicable progression it has made. PulmoAi CT uses different technology and has been built to work with CT-scans of affected patients through algorithms that will help the medical community better understand how the virus manifests and spreads through the lungs with unprecedented levels of precision. Using these insights, we can work together to investigate how this virus is impacting some patients differently than others. The solution will also enable quick evaluations for drug treatments and changes in treatment protocols.
With the most powerful tools technology can offer at our disposal, it is the responsibility of every tech leader across the world to look closely at their capabilities, and to put their solutions toward the treatment and control of this pandemic. COVID-19 is an adversary unlike any other that our medical community has had to manage and winning will be a full team effort.