Deep Learning

---

Can help AI to find a cure for the corona virus?

It feels As If a superhuman effort is required to easily get the global pandemic to kill many so.

Artificial Intelligence may have been hyped - But when It Comes to medicine, it is already a proven success record.

So the machine can learn to find this challenge, a cure for this terrible disease?

There is no shortage of companies to solve the trying to The Dilemma .

Oxford-based Exscientia, The First AI instead discovered the drug in the human trial, is the trawl fisheries More Than 15,000 drugs, the Scripps Research Institute in California.

And Healx, a Cambridge company that has of Viagra-co-inventor Dr. David Brown, a new function must be assigned to an AI system designed to search for drugs for rare diseases.

possible drug candidates to rigorous testing in the laboratory

The System is divided into three parts:

Drug discovery is traditionally slow.

"I've been doing this for 45 Years , and I have three medicines on The Market ," Dr. Brown told Bbc News .

But the AI is proving to be very much faster.

"It took some weeks to collect all the data that we need, and we also have a lot of new information in The Last couple of days, so We Are now Said to be a Critical Mass ," Dr. Brown,.

"The algorithms ran over Easter and we have the output for the three methods in The Next Seven Days . "

Healx hopes to turn that information in to a list of drug candidates to may, and is already in talks with labs, these predictions in clinical studies.

For those in The Field of AI drug discovery, there are two options when It Comes to the Corona-Virus:

It is likely it is a combination of drugs, which defeats coronavirus

But , Dr. Brown Said it was extremely unlikely that a single drug is The Answer .

And for Healx, the detailed analysis of eight million possible pairs and 10 means. 5 billion of three-drug combinations, consisting of the 4,000 approved drugs on The Market .

Professor Ara Darzi, Director of The Institute of Global Health Innovation at Imperial College, told Bbc News : "AI remains to reach one of our strongest paths, which is a tangible solution, But it is a fundamental need for high quality, large and clean data sets.

"Until today, many of these information are in isolated, individual companies such as big pharma, or are lost in the intellectual property-and old-lab space within the University.

"Now, More Than ever, there is a need for standardization of these different drug discovery data sources is to allow AI researchers for their innovative machine-learning techniques to generate new treatments for Covid-19 as soon as possible. "

In the United States, a partnership between Northeastern University is Barabasi Labs, Harvard Medical School , the Stanford University Network Science Institute and biotech start-up Schipher medicine treatments in search of drugs, which are quickly used as Covid-19 -.

amazing results

Normally, only you can get all the work together would take "a year of paperwork," Said Schipher chief executive Alif Saleh.

not to mention But a number of Zoom-calls with a "group of people with an unprecedented determination to get things done, a lot of time on your hands", to speed things up.

"The Last three weeks, usually half a year. Everyone is all like," He Said .

their research revealed surprising results, including:

Schipher medicine combines AI with something it calls, Network Medicine - a method which the views of a disease, The Complex interactions between molecular components.

"A phenotype of the disease is rarely Said by a malfunction of a gene or protein on its own - Nature is not so easy - But the result of a cascade in a network of interactions between multiple proteins," Mr Saleh.

With Network Medicine ", " AI ", and a fusion of the two 81 potential drugs that could help, to identify the consortium to.

"can do AI, a little better, not alone to the higher order correlations But little independent information, which could miss the traditional Network Medicine ," Said Prof Albert-Laszlo Barabasi.

But the AI alone would not have worked, they needed all three approaches.

"with the Different tools look at different perspectives, But together are very powerful," he added.

claimed a number of AI companies already isolated medications that may help.

BenevolentAI has noted, Baricitinib, a drug that is already approved for The Treatment of rheumatoid arthritis, as a possible treatment to prevent, to infect with The Virus in lung cells.

And it is now in a controlled study with the US National Institute of Allergy and Infectious Diseases.

the Chinese company Alibaba claims developed to have a KI test, that can accurately detect coronavirus

Meanwhile, scientists from South Korea and the United States with the help of Deep Learning to investigate the potential for commercially available antiviral drugs have proposed the use of atazanavir for The Treatment of Aids, could be a good candidate.

Other companies are using AI for other purposes, to relieve, such as the analysis of the scans of radiologists and help to predict which patients are most device will probably need a respirator.

Chinese technology giant Alibaba, for example, announced an algorithm that it says can accuracy diagnosing cases within 20 seconds, with 96%.

But some experts warn of the AI systems are likely to have been trained on data that advanced infections, so that they are less effective in the detection of early signs of The Virus .

It took to convince a global effort of policy, the major pharmaceutical companies to join forces with smaller drug-stores data, science and research, non-profit associations, the pool of data resources, Prof Darzi Said .

"The Time has never been more important for drug-to-open discovery data, to help his secrets for AI, in the fight against Covid-19," He Said .

coronavirus pandemic, artificial intelligence

Source of news: bbc.com