Patients could soon be diagnosed by artificial intelligence, NHS leaders say

Computers may soon be able to read X-rays and analyse diseased tissue, the NHS’s top doctor has said.

Artificial intelligence (AI) could potentially mean doctors will be replaced by machines within the next few years.

NHS England plans to invest more of its £120 billion budget in AI to speed up its application to the mental health service, especially in analysing huge amounts of information collected from patients about their symptoms.

AI is already being used around the world in machines that can help identify patients who have autism or Alzheimer’s disease.

Bruce Keogh, NHS England’s National Medical Director, said: “We know from a number of studies that have been done that in, certain circumstances, AI is better than doctors at diagnosing certain conditions.

“It’s possible that certain types – and there are many different types of AI – will be able to read X-rays. I’ve been told by people who are developing this sort of stuff that’s within a four-year timeframe. We also know that this kind of approach can be used for reading histopathology slides.

“All of this takes us into very new territory and it’s not a long way over there, it’s actually here now.”

Simon Stevens, NHS chief executive, said: “We have a great opportunity to get smarter about the way we are using AI and machine learning with datasets to improve the quality of clinical care.

“The Lancet pointed out that over the course of a career a radiologist would probably read over 10 million images, a dermatologist might analyse over 200,000 skin lesions and a pathologist will review nearly 100,000 specimens – whereas there is great potential to actually automate huge swathes of what is actually happening in radiology and to some extent pathology and dermatology.”

Jeremy Hunt, Health Secretary, told the NHS expo that computers could be routinely diagnosing health conditions even before they display symptoms by the time NHS turns 80 in 2028.

Hunt said: “The changes in medical innovation are likely to transform humanity by as much in the next 25 years as the internet has in the last 25 years.

“So what might medicine look like when the NHS is 80? Well the first thing is we may well not be going to doctors for a diagnosis, we might be going to computers instead, who will be looking at the 300,000 biomarkers in every single drop of blood, mashing that with big data information about everyone else’s biomarkers.

“When we do go to a doctor, he or she may be comparing your medical record with our fully sequenced genome and giving us much more accurate predictive information.

“We may be in a world where if we show any symptoms of a disease we consider that a sign of failure – because the name of the game will be to catch diseases when we are asymptomatic and we have a much better chance of being able to nip illness in the bud.”