Healthcare is entering a new era, with Artificial Intelligence (AI) playing an increasingly important role.
The UK government has already committed to a £250 million investment in AI technology, which includes the setting up of a National Artificial Intelligence Laboratory to help develop new solutions for the NHS. AI technologies will play an important role in helping the NHS to achieve more with its resources.
AI is seen as a valuable, strategic tool that can assist the skilled work of clinicians, rather than replace them. Crucially, it is hoped that AI will be able to reduce the burden on medical staff and free up more time for them to spend on face to face patient care. No wonder then that an estimated 52% of NHS trusts are already deploying AI technologies.
Kelvyn Hipperson, Chief Information Officer, Cornwall Partnership NHS Foundation Trust and Royal Cornwall Hospitals NHS Trust, says: “Fundamentally, the role for AI is to help clinicians deliver more effective patient care. AI has been emerging for many years – but now the rate of adoption is accelerating. It is set to be a real game changer.”
“AI is already making a difference across healthcare settings. For example, we have recently upgraded our voice recognition system and the improved capability means we now have better resilience in our clinical documentation processes. Whether in the community, GP surgeries or hospitals, AI will soon be helping deliver improved healthcare in the UK.”
Here are five ways in which AI is beginning to transform the way that healthcare is delivered:
1. Speed and accuracy of screening
Examining thousands of x-rays or mammograms is painstaking work and the signs of illnesses like cancer can be difficult to identify. So there is currently worldwide interest in using AI to support this analysis.
Recently, the journal Nature published the results of a trial that used an AI system to analyse 29,000 mammograms. The software had been taught to identify the presence of tumours. The research found that the AI system was as successful in identifying cancers as using two trained doctors to carry out the work, which is how mammograms are currently reviewed. The next challenge is to see how this approach could be upscaled to a clinical setting.
AI has the potential to enhance our testing for a wide range of cancers and other life-threatening conditions, including cardiovascular disease.
2. New preventative treatments
AI can provide new techniques for pro-actively analysing the wealth of data that we hold on patients, including the results of DNA testing. This could be used to identify those who are at risk of developing specific chronic illnesses in the future. As a result, a new approach to precision medicine could be developed, focused on prevention rather than cure. This would have huge implications for the way that the NHS operates.
The British Heart Foundation is helping to fund the development of an AI tool that will help doctors identify those patients that are at future risk of suffering a heart attack or stroke. A team of researchers led by the University of Cambridge will use the long-term health records of over two million people in the UK to develop a new machine learning algorithm to automatically predict peoples’ risk.
3. Advanced management systems
AI is opening up new ways of managing complex medical facilities, taking into account the vast amount of data that now needs to be processed in order to aid management decision making.
Bradford Royal Infirmary has become the first hospital in Europe to launch an AI-powered command centre. Rather like an air traffic control system, it gives an instant, real-time overview across the 800-bed hospital and helps staff make quick and informed decisions on how to best manage patient care. Advanced algorithms help staff to anticipate and resolve bottlenecks in care delivery before they occur, enabling more responsive treatment and better allocation of resources.
4. New diagnostic tools
Working with AI systems gives clinicians the potential to develop completely new ways of diagnosing conditions, leading to earlier treatment and better outcomes.
For five years, Moorfields Eye Hospital has been working in partnership with AI specialist, DeepMind. Together, they have successfully trained an AI system to accurately identify signs of eye disease and recommend how patients should be referred for care. The system can deliver the correct referral decision for over 50 eye diseases with 94% per cent accuracy, which Moorfields say is comparable to world-leading eye experts. This means that severe eye diseases can be identified and treated before there is irreversible damage.
5. Remote healthcare
AI-driven systems are opening up new ways of providing healthcare and advice remotely. Not only is this more convenient for patients, but it can also help to significantly reduce pressure on GP surgeries and A&E departments. It has the further benefit of getting patients more pro-actively involved in the management of their health.
For example, Lifelight is a software technology that measures blood pressure, heart rate, respiration and oxygen saturation in just 40 seconds simply by a patient looking into the camera on a standard smartphone or tablet. No wearables or contact is needed. Lifelight’s algorithms are trained using data from an 8,500 patient clinical study at Portsmouth Hospitals Trust. Not only does Lifelight allow fast, contactless ward observations and it could also offer a patient at home the ability to measure their own vital signs with their smartphone as part of a 111 call or remote consultation.
Making it happen
While AI has the power to transform many aspects of the NHS over the coming years, it is clear that a number of challenges will need to be met in order for it to achieve its full potential. These include:
- Scaling up from trials to full systems
- Proving the long-term benefits of AI systems versus costs
- Understanding the impact on clinical workflows
- Training staff to understand and use the new technology
- Reassuring and educating patients so that they accept the new approach
Kelvyn Hipperson adds: “Clinical safety and IT safety have to work together. As a technologist, one of the big challenges is managing the introduction and trial of new technology and balancing this with the need to manage long term sustainability. We need to trial innovative new technology – but simultaneously have one eye on what the exit strategy will be if a new technology does not work out and we need to replace it. How can we manage this, whilst not risking clinical safety or compromising patient care? This is a constant challenge and one that requires agility on the part of technologists and clinicians alike.”
As technologists and healthcare providers rise to these challenges, we will see more and more exciting applications of AI emerging over the next five years.