It feels as though a monumental effort is needed to help mitigate too many deaths from the global pandemic. Artificial intelligence may have been hyped up-but it also has an established track record when it comes to medicine.
How will machine learning attempt to find a cure for this awful disease?
There’s no shortage of businesses seeking to solve the problem. Oxford-based Exscientia, the first to put an AI-discovered drug into a clinical trial, trawls through 15,000 drugs operated by the California-based Scripps Research Institute.
And Healx, a Cambridge company founded by Viagra co-inventor Dr. David Brown, has updated its AI system for rare disease drug development, which is divided into three parts:
- Trawling through all the current literature on the disease
- studying the DNA and structure of the virus
- Traditionally considered the suitability of different drug discoveries.
“For 45 years I have been doing so and finding 3 drugs,” Brown said to BBC News. Dr. Brown.
Yet AI shows a lot quicker.
“It took several weeks to gather all the data we need, and we also have new information in the last few days and now we are in the critical mass,” said Brown.
“The algorithms ran over Easter and over the next seven days we will have products for the three approaches.” Healx aims to turn this knowledge into a list of drug candidates by May and is already in negotiations with laboratories to carry those forecasts into clinical trials.
There are two choices for those working in the field of AI drug development when it comes to coronavirus:
- find a brand new drug but wait a few years for it to be accepted as safe for use
- repurpose existing drugs.
However, Dr. Brown said, it was highly unlikely that a single drug will be the solution. And for Healx, that means a detailed analysis of the possible 8 million pairs and 10.5 billion triple-drug combinations from the 4,000 drugs approved on the market.