Cookies: We use cookies to give you the best possible experience on our site. By continuing to use the site you agree to our use of cookies. Find out more
House of Lords Hansard
x
NHS: Artificial Intelligence
24 April 2018
Volume 790

Question

Asked by

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

To ask Her Majesty’s Government what steps they are taking to encourage the adoption of artificial intelligence in the National Health Service.

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

My Lords, the Government believe that artificial intelligence, or AI, has the potential to transform health and care services. Our work to support this includes £4.2 billion of spending to update hospital IT infrastructure as well as the digital health technology catalyst, which provides funding to help small and medium-sized enterprises turn innovative ideas into reality. In addition, the Industrial Strategy Challenge Fund is supporting AI programmes in digital pathology and radiology, with winners of the wave 3 competition announced next month. We will continue to ensure that our regulatory frameworks protect patients while enabling the benefits of AI to transform care.

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

My Lords, does my noble friend agree that the potential benefits to the National Health Service from AI go far further than just the clinical setting? For example, DNAs currently cost the service in excess of £1 billion. What more can be done to generate proofs of concept and work, not just in London but across the country, to ensure that AI machine learning, with all its benefits, can be put into making the NHS the greatest health service, not only in patient care but in taking up all the advantages of the fourth industrial revolution?

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

My Lords, I thank my noble friend for his question and congratulate him and the rest of the committee on the excellent report, AI in the UK: Ready, Willing and Able?, which has a substantial chapter on AI’s application in healthcare. The potential to transform every element of health and care is susceptible to artificial intelligence. A couple of areas outside the clinical setting that I would highlight are workforce planning and triaging patients between different forms of care. As for support, in addition to the items in my first Answer, I highlight the work of the Topol review, which is designed to make sure that staff are fully equipped and trained to take advantage of these technologies as they come through the system, rather than letting them sit with a few early adopters and not becoming more widespread in the NHS.

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

My Lords, it is quite clear that the use of big data and AI will have transformative outcomes for patients. There are at least two challenges. The first is investment, which the Minister has already mentioned. What framework of accountability and transparency is in place to deal with that level of investment? How will we know whether it is being sensibly invested? The second is safeguarding and protecting data, and I use my local hospital as an example. A partnership between Google DeepMind and the Royal Free Hospital trust resulted in a breach of the Data Protection Act and the personal data of more than 1.6 million patients was transferred to the Google subsidiary as part of the creation of Streams, an app to diagnose and detect acute kidney injury—which we would, of course, all support. This suggests inexperienced procurement and negotiation skills in the NHS and the potential for the Googles of this world to run rings round them, to all our detriment. What are the Government doing to safeguard patients and their data?

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

The case the noble Baroness highlighted brings to the fore both the potential benefits and risks. There are tremendous benefits in having personalised healthcare, and we all want to see that delivered. At the same time, if data is not used safely and securely we lose the public’s trust. If we do not have that trust, we will not be able to get the changes that we want. The Government respect the decisions made by the Information Commissioner and National Data Guardian in their judgments about poor practice at the Royal Free. I am pleased to say that the hospital has responded well to these. We are doing a couple of things to make more systematic changes. First is implementing the proper data standards of the GDPR in one month’s time. We will also make sure that National Data Guardian’s 10 data standards are written into every NHS contract so that, when it comes to procurement, there is understanding about the kind of things they should and should not be doing to safeguard data.

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

My Lords, does the Minister agree with the recommendations of the AI Select Committee, regarding NHS data, that a framework for the sharing of data is now urgent or needs to be delivered by the end of 2018? Does he support the need to digitalise, in consistent formats, by 2022? Evidence received by the committee suggests that failure to do so risks our missing out on profound opportunities from AI, because the current approach to data storage—especially among different NHS trusts—is outdated and piecemeal.

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

The noble Baroness has highlighted two of the recommendations from the report. I support the proposal for a regulatory framework; it is a piece of work that I have kicked off in the department. I cannot put a timing on that, but I understand the need to provide a safe operating environment so that people who want to get into this field, whether from NHS trusts or businesses, can do so knowing that they are operating on a legal basis. That is something that we are working on.

On digitalisation, she is quite right: the £4 billion programme known as Personalised Health and Care 2020 is trying to deliver before 2020—as the name suggests—the kind of digitalisation that will enable AI to bring those benefits across every corner of the health and care systems.

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

My Lords, is the Minister aware that many parts of the world envy Britain’s strengths and opportunities in AI, particularly in the health area, and that government procurement could turn this early lead into a golden opportunity for the UK?

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

Yes, I absolutely agree with that. As the report highlights, we have a unique opportunity because of the nature of the way that the NHS was set up and its potential for realising a comprehensive data set of 65 million people. It is not just about those procurement rules; we have talked about having the right framework. It is about providing reassurance within the system—at a time when the public are beginning to understand just what data can do for good and for bad—that the NHS will use their data safely, securely and legally so that they can trust that it is being used for proper purposes from which they will benefit.

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

My Lords, I declare my interest as chairman of University College London Partners. Does the Minister believe that there is a sufficiently robust mechanism for the diffusion of the innovation associated with digitalisation and artificial intelligence across the NHS? In particular, what role does the Minister think the academic health science networks should play in that process?

The edit just sent has not been saved. The following error was returned:
This content has already been edited and is awaiting review.

Of all the innovations, diffusion is probably one of the greatest challenges that the NHS faces, as the noble Lord knows very well. We are doing a couple of things. First, we are supporting the global digital exemplars, which are providing that digitalisation at trust level, to make sure that they have the infrastructure there. Secondly, he talks about academic health and science networks. They have just been relicensed and are now to have a national remit to promote innovation. AI is absolutely part of the work that we are expecting them to do.