Search
  • Darshan Pandya

Combating COVID-19 with Artificial Intelligence

In order to predict and monitor the propagation of the novel coronavirus, scientists use technologies from Dynamic real-maps to advanced forecasting algorithms. Solutions like using artificial intelligence to handle a vast quantity of data from various sources in different countries.


How AI is helping fight COVID-19
Combating COVID-19 with Artificial Intelligence

Many businesses and organizations have brought together software to combat the coronavirus, as scientists believe data collection systems and applied artificial intelligence approaches will contribute to a comprehensive infectious disease database that can in effect help-policy decision-making. It, however, presents difficulties for scientists as global data standards still need to be established.


"There is a tremendous volume of internet evidence and conversations mostly not from official outlets but from news and chat rooms and forums and social media. We need to sift through the knowledge, identify it, sift through the noise, geocode, and then give a glimpse of emerging diseases," says John Brownstein, Harvard Medical School's Chief Technology Officer, and Professor.

"Generally, data modeling and disease detection activities arise in silos with minimal coordination of approaches and evidence between model creators and end-users," note the authors of an epidemic modeling report by Nature. "Generally, the analysis of' cross-discussion' between actors within and within countries is often constrained and mostly exists within a framework of legal and ethical ambiguity."


Nine days before the World Health Organization cautioned about the novel coronavirus, a researcher based in Toronto noticed the threat faced by the virus-caused outbreak, COVID-19. BlueDot, a $9.5 million growing corporation, used data from hundreds of thousands of outlets, such as comments from official public health agencies, newspapers, health surveys and trends, as well as airline statistics. The company focused on the experience of its co-founder in treating SARS patients as an epidemiologist and medical practitioner.


"Our climate is evolving rapidly, and consequently diseases that endanger human health, welfare, and stability are emerging with greater frequency and magnitude," BlueDot founder and CEO Kamran Khan said last August.


In the meantime, Brownstein's HealthMap, an artificial intelligence tool developed by Harvard Medical School that detects infectious diseases, also reported early signs of coronavirus outbreak in December in Wuhan, China.


"The system offers an early detection but we are now using the technologies to establish the scenario image of what's evolving, collecting data sources and attempting to develop a global map of cases that can be used to power modeling and forecasting, spread influence on various demographic resources," he said in an interview.


Many companies, like WeBank in China, have taken AI even further and are using it to forecast the Chinese economy's coming rebound from the COVID-19 crisis.


In particular, automation and artificial intelligence may be extended to emergency medical care systems, with computers used to forecast the transmission of a common illness, patient tracking, and emergency room department operations. Scientists around the world can use algorithms to identify and cluster knowledge, browse the internet and understand language associated with the disease being studied, examine pictures and deploy robots for a variety of medical tasks that doctors wouldn't feel safe to do.


"In this period, there is a lot of discussion about technologies to create new solutions, particularly in developed world settings, AI is likely to support EM (emergency medicine) through its digitization capacity and storage in information," two Pakistani and Oman medical researchers wrote in a paper.


Yet using AI for coronavirus in patient care and health care alerts isn't difficult, Brownstein states. The main problem is to build a convincing and accurate global virus dataset and its distribution based on numerous national datasets of differing metrics.


"There are various systems and taxonomies, forms people respond to the disease, different cultures, cultural meaning, words people use around the disease that may be used for certain types of items, and there is a lot more analysis that is required and it takes a tremendous amount more machine testing to get to the point that these resources are being useful."


Did you find this article useful? Please let us know in the comments section below.

150 views0 comments

Recent Posts

See All

It's FREE!

It's FREE!