Machine learning comes of age as AI enters the life-saving business

Machine learning is on the verge of becoming a vital tool for saving lives, with AI fighting cancer,  AI in the detection of diseases and AI employed in the battle against superbugs finding antibiotics.

Machine learning is not quite ready to save millions of lives, but the day that it can be applied to detect the early development of cancer, without seeing a specialist and the day when it can hand us a decisive victory over superbugs, potentially saving hundreds of millions of lives, is not far off.

There is a narrative that AI is more about escalating a marketing arms race than doing anything truly worthwhile. Sure, machine learning applied to customer data can give one company an edge over rivals. But will it really change the world? The answer is yes, and there are many reasons why machine learning can be a force for good, but here is one of them; AI can save lives.
Here is the evidence; or at least some of the evidence

AI against cancer

One of the benefits of AI in the fight against cancer is in image recognition. For example, in one classic study led by Andre Esteva and Brett Kuprel from Stanford, AI was found to be at least as effective as dermatologists in identifying between malignant melanomas and benign moles. This led to their revolutionary suggestion that smartphones could be used in skin cancer detection beyond the clinic. They conclude: "It is projected that 6.3 billion smartphone subscriptions will exist by the year 2021 and can therefore potentially provide low-cost universal access to vital diagnostic care."

We are entering the pharmaceutical golden age
This is just the beginning; further studies reveal that machine learning can be applied in identifying "lymph node metastases in breast cancer, rivalling human performance, " suggests a report in Nature.

Covid

And that brings us to Covid. A team of researchers claim to have a tool that can predict the likelihood of a Covid patient developing severe symptoms or even dying. Their conclusion, using machine learning techniques, "an interactive tool was developed that rapidly and accurately provides the probability of an individual patient's progression to severe illness or death on the basis of readily available clinical information."

Antibiotics

Forget Covid; the overuse of antibiotics means we could be losing the battle against superbugs.

It has been estimated that the accidental discovery of penicillin in 1928 by Alexander Fleming saved between 80 million and 200 million lives.  

Superbugs pose a much bigger threat to our long-term health than a Coronavirus.

AI could be the solution.

Take Halicin, a team of researchers from MIT were able to discover a new antibiotic they called Halicin by applying machine learning.  To cite Nature: "The researchers trained its neural network to spot molecules that inhibit the growth of the bacterium Escherichia coli, using a collection of 2,335 molecules for which the antibacterial activity was known. This includes a library of about 300 approved antibiotics, as well as 800 natural products from plant, animal and microbial sources."

And at Apple

Meanwhile, Apple has been busy recruiting "data scientists to develop machine learning algorithms for the early detection of certain diseases or disease conditions."

The future

Machine learning is not quite ready to save millions of lives, but the day that it can be applied to detect the early development of cancer, without seeing a specialist and the day when it can hand us a decisive victory over superbugs, potentially saving hundreds of millions of lives, is not far off.