Is artificial intelligence ready to personalise care for people with MS?
In 2024, our biennial conference for researchers MS Frontiers hosted a debate about artificial intelligence and MS.
MS Frontiers is the UK's largest biennial conference for MS research professionals. During each conference, we hold a debate on a hot topic in MS research.
This year, Professor Nikos Evangelou from the University of Nottingham brought together researchers to debate whether artificial intelligence (AI) is ready to personalise care for people with MS.
How could AI help people with MS?
The debate focused on whether AI could help people with MS by developing personalised care. Personalised care allows people to have choice and control over their care, tailored to their individual needs and circumstances. Rather than a ‘one-size-fits-all’ approach. This has potential to help people with MS because everyone can have different lived experiences of the condition.
At the start of the session, a vote showed that only 23% of the audience believed that AI is ready to personalise care for people with MS.
The results of a conference debate alone won’t change practice. But it provides a platform for our experts to share ideas and knowledge, explore and challenge different viewpoints.
AI is ready...are we?
Dr Kate Petheram, a consultant neurologist at Tyneside and Sunderland NHS Foundation Trust, was first to the floor to fight her case that AI is ready.
Kate described several areas where AI is already primed to personalise care for people with MS. One example is diagnosis. There has already been a huge amount of research into using AI to diagnose MS from MRI scans. And it has been shown that it can help to identify subtypes of MS.
She highlighted that AI can also quickly find important information from patient records for doctors during appointments. This is already being used at St Michael’s Hospital in Canada, where neurologists are using an AI tool to quickly learn about new patients, prepare for appointments, and plan future care. This saves time for medical staff and allows them to instead spend their time on meeting their patients’ needs.
And to achieve personalised care, it is vital that people with MS have access to, and can understand, the information they need to make informed choices and decisions. Kate argued that AI can be used to tailor patient information leaflets to individual reading levels, making it easier for people to understand their options.
AI is the future...but it’s not ready yet
Dr Arman Eshaghi is based at University College London, where he studies machine learning in MS. He argued that AI is not yet ready. Arman agreed that AI would one day be used by doctors to support the diagnosis and monitoring of MS, and to predict how someone’s MS might progress. But he also highlighted some reasons why we are not quite there yet.
There is a lot of research going on into how AI can benefit people with MS, particularly for analysing scans. But Arman argued that these techniques are still a long way from being routinely used in the clinic.
He highlighted that AI technologies still need to go through a rigorous testing process, similar to clinical trials for treatments. This testing takes time. We currently don’t have any published studies that show that AI can improve quality of life for people with MS. And we don’t yet have the information to see if AI is cost-effective for use in the NHS.
Panel discussion
After hearing from Kate and Arman, Nikos opened the floor for audience questions and a panel discussion. This featured MS Society Ambassador Stuart Nixon MBE, neurologist Professor Helen Ford from the University of Leeds and Courtney Kremler, a PhD student from the University of Cambridge researching machine learning.
The panel agreed that AI would never replace clinicians. But they do believe that it has the potential to significantly improve care. They raised concerns about digital exclusion and how these technologies could be implemented within the NHS. And highlighted the need for more diverse data to improve AI accuracy.
The votes are in: AI is ready
After hearing from the panel, we put the question back to the audience to make their final decision. Support for the motion increased to 45%. So, Kate was able to successfully convince a further 22% that AI is ready to personalise care for people with MS.
AI is a fast-moving area of research, and we’re interested to see how it develops over the coming years.
You may also be interested in
