AI technology can be used to help bring researchers closer to a cure. It can do this by scouring through the world’s massive reservoir of data to find potential treatments.
AI may prove useful in bolstering health information technology (HIT), offering “intelligent” HIM (health information management) and even an HIE (health information exchange) to better contend with the corona-virus outbreak.
The Canadian startup BlueDot, which specializes in surveillance, was among the first in the world to accurately identify the risk–and spread–of the coronavirus. In late December, its AI software discovered a cluster of unusual pneumonia-like cases in Wuhan, China. It then predicted where the virus might go next. Systems equipped with AI are more capable of dealing with highly-granular data-sets; and so are best suited to identify the most important factors driving differences in transmission rates across different cities.
Key stakeholders in the community must engage in global collaboration. AI technology can help in knowledge-sharing via programs based on open-access / open-source data. This was done by DarwinAI, an operation that launched COVID-Net as an open-source system for fighting the disease.
Companies like BenevolentAI in London are using AI technology to find drugs that target the coronavirus. Part of this is identifying drugs that were developed to fight other diseases yet might be repurposed. By studying the molecular setup of existing drugs, systems equipped with AI may be able to find which ones effectively fight COVID-19.
BenevolentAI began focusing on the novel coronavirus in late January. Its AI-powered system (a “knowledge graph”) can parse a massive catalog of scientific literature and cull huge amounts of biomedical research to find potential links between the genetic / biological properties of a disease and the composition of certain drugs. The company was able to reprogram its AI to work on COVID-19 by feeding it the latest research on the virus. It was able to machine-read large amounts of documents, adapting its “knowledge graph” to account for the latest information about COVID-19.
The COVID-19 Open Research Dataset (CORD-19), an initiative building on Seattle’s Allen Institute for Artificial Intelligence: “AI2”. The “Semantic Scholar” team is behind the CORD-19 data-set at AI2. They use natural language processing to analyze tens of thousands of scientific research papers at an unprecedented pace. This means of discovery has tremendous potential to inform vaccine and treatment development–now crucial during COVID-19 pandemic.
“Semantic Scholar” was created based on the theory that cures for many ills are buried in scientific literature…waiting to be culled. Literature-based discovery has tremendous potential to inform vaccine and treatment development, which is a critical next step in the COVID-19 pandemic. But only AI technology is equal to this formidable task.
Computer models can be used to map out infected cells. Coronaviruses invade cells through “spike proteins,” but they take on different shapes in different coronaviruses. Analyzing the shape of the spike protein in SARS-coronavirus-2 that impels the contagion is crucial to figuring out how to target the virus; as it will reveal how it works…and thus what its weaknesses are. Dozens of research papers related to spike proteins are in the CORD-19 Explorer. Recently, the University of Washington’s Institute for Protein Design mapped out 3D atomic-scale models of the corona-virus-2 spike protein that mirror those first discovered in a University of Texas lab in Austin.
More than forty organizations are developing a COVID-19 vaccine, including three that have made it to human testing. Advanced mapping technology, powered by AI, will help us accomplish this important task.