Artificial intelligence is coming up against COVID-19

Artificial intelligence is coming up against Covid-19

Artificial intelligence comes in because it is felt that more effort is needed to help alleviate the global epidemic.

Artificial intelligence may have been misleading – but when it comes to medicine, it already has a proven track record.

So can machine learning raise you to the challenge of finding a cure for this terrible disease?

There is no shortage of companies trying to solve a problem.

Oxford-based Exscientia, the first to incorporate the drug found in Artificial intelligence into human trials, is treading on 15,000 drugs stored by the DRM research center, California.

And Healx, a Cambridge company founded by Viagra founder Dr. David Brown, has backed its Artificial intelligence program designed to find rare disease drugs.

The program is divided into three sections:

trawl through all the current literature on the disease

learn about the DNA and the virus

consider the suitability of various drugs

Drug availability has been a slow process.

“I have been doing this for 45 years and have found three drugs to sell,” Dr. Brown told BBC News.

But Artificial intelligence is showing a lot of speed.

“It has taken us a few weeks to gather all the information we need and we have received new information over the past few days, so we are now in a great crisis,” said Dr. Brown.

“The algorithms worked on Easter and we will be rolling out three alternatives in the next seven days.”

Healx hopes to turn this information into a list of drug addicts in May and is already in discussions with labs to take those predictions into clinical trials.

For those working in the field of Artificial intelligence drug discovery, there are two options for coronavirus coverage:

find a completely new drug but wait a few years for it to be approved for safe use

replace existing drugs

However, Dr. Brown said, it was unlikely that a single drug would be the answer.

And for Healx, that means a detailed analysis of about eight million combinations and 10.5 billion combinations from approved drugs in the market.

Professor Ara Darzi, director of the Center for Global Health Innovation, at Imperial College, told BBC News: “Artificial intelligence remains one of our strongest ways to achieve a tangible solution but there is a fundamental need for quality, big and clean data.

“So far, most of this information has been sent to individual companies such as large-scale or lost pharma in the forensic area and in the old laboratory area within universities.

“Now more than ever, there is a need to integrate these different drug discovery data sources to allow Artificial intelligence researchers to apply their Covid-19 new machine learning techniques quickly.”

In the US, a partnership between Northeast University’s Barabasi Labs, Harvard Medical School, the Stanford Network Science Institute and a biotech startup Schipher Medicine is also in the process of seeking drugs that can be recreated as Covid-19 treatments.

A startling discovery

Usually, getting them all together will take “a year of paperwork”, Schipher chief executive Alif Saleh said.

But the Zoom series calls a group of “people with unprecedented determination to get things done, let alone a lot of their time”, things are accelerating.

“The last three weeks are usually half a year. Everyone threw everything,” he said.

Already, their research has yielded surprising results, including:

a suggestion that the virus might invade the brain tissue, which could explain why some people lose the sense of taste or smell)

prediction can invade the reproductive system of men and women

Schipher Medicine combines Artificial intelligence with what it calls a network medicine – a way of looking at a disease through complex interactions between molecular elements.

“The nature of the disease phenotype is not usually caused by a malfunction of one type or of the protein itself – the environment is not an easy one – but the result of the creation of a network of interactions between multiple proteins,” Mr. Saleh.

With the use of network medicine, Artificial intelligence and fusion of these two have resulted in a combination of about 81 drugs that can help.

“Artificial intelligence can do much better, looking not only at the consistency of higher orders but the small amount of independent information that a traditional network medicine might miss,” said Prof Albert-Laszlo Barabasi.

But AI alone wouldn’t work, they needed all three modes.

“Different tools look at different perspectives but collectively they have great potential” he added.

Some AI companies are already claiming that they have unique drugs that can help.

BenevolentAI identified Baricitinib, a drug already approved for the treatment of rheumatoid arthritis, as a possible treatment to prevent the virus from invading the lung cells.

And now a controlled trial has been conducted with the US National Institute of Allergy and Inferior Diseases.

At the time, scientists from South Korea and the United States using intensive studies to investigate the possibility of antiviral drugs have suggested that atazanavir, used for the treatment of Aids, may be a viable candidate.

Some companies are using AI for other purposes, such as scanning to reduce the burden on radiologists and helping to predict which patients most need a ventilator.

Chinese expert Alibaba, for example, announced an algorithm that says it can test cases in 20 seconds, with 96% accuracy.

But some experts warn AI systems may be trained on data about advanced infection, making them less effective at detecting early signs of a virus.

There should have been a global effort from policymakers to urge large pharmaceutical companies to join forces with small drug stores, academics and research aid organizations to use the information sources, said Prof Darzi.

“Time has never been more important for drug discovery data to unlock its AI secrets to help fight Covid-19,” he said.


  1. […] “[Covid-19] will change consumer preferences and actually open up new possibilities for automation.” […]


Please enter your comment!
Please enter your name here