Image copyright
Getty Images
The NHS in England is setting up a national artificial intelligence laboratory to enhance care of patients and research.
The Health Secretary, Matt Hancock, said AI had “enormous power” to improve care, save lives and ensure doctors had more time to spend with patients.
He has announced £250m will be spent on boosting the role of AI within the health service.
However, AI will pose new challenges including protecting patient data.
What can AI do in health?
The advanced computer software is already showing its potential to revolutionise medicine in fields ranging from diagnosing patients, gleaning new insights into diseases and improving how hospitals run.
Artificial intelligences thrive when an algorithm can learn from huge amounts of data to spot patterns.
Medical imaging – where an AI can be trained on thousands of scans – has led the charge.
Clinical trials have proven AIs are as good as leading doctors at spotting lung cancer, skin cancer, and more than 50 eye conditions from scans.
This has the potential to let doctors focus on the most urgent cases and rule out those that do not need treatment.
Other tools have been developed that can predict ovarian cancer survival rates and help choose which treatment should be given.
Meanwhile, University College London has developed an AI that flags up which patients are more likely to miss an appointment and could be targeted with reminder phone calls.
However, these tools are not yet being used routinely across the NHS.
“The power of artificial intelligence to improve medicine, to save lives, to improve the way treatments are done, that power is enormous,” Mr Hancock told BBC News.
“In this country, we’ve got the opportunity to be one of the leading countries in the world at using this new technology.”
Artificial intelligence also has the potential to unlock fresh insights into diseases.
It has been used to identify a new way of diagnosing atrial fibrillation, a common condition that causes the heart to beat irregularly.
“I want the NHS, through its AI lab, to actually be searching itself for new insights that are going to save lives,” said Mr Hancock.
What are the challenges?
Increasing the use of AI will also pose challenges for the health service – from training staff to enhancing cyber-security and ensuring patient confidentiality.
The Royal Free Hospital, in north London, was criticised in 2017 for sharing 1.6 million patient data records with Google’s AI division, DeepMind.
The Information Commission ruled the hospital had not done enough to safeguard patient data.
In May 2017, the NHS was hit by a large-scale cyber-attack that disrupted hospital and GP appointments.
The other challenge with an AI is it can only ever be as good as the data it learns from.
But most medical research is on people of white ancestry, meaning AIs trained on such data could be less effective for people of black and minority ethnic backgrounds.
Mr Hancock said it was “incredibly important” that AI’s “represented society”.
Dr Nicola Strickland, president of the Royal College of Radiologists, said: “I expect radiologists to be leaders in using AI algorithms to assist them, provided they can see evidence that these AI algorithms have been developed using large enough, properly curated data and rigorously validated and tested.”
Follow James on Twitter.