The hype around AI is taking over all sectors, but, if we trust the thoughts of some of the most recognised voices, it is raising serious concerns amongst individuals.

In September, Russian President Vladimir Putin told students in an open lecture, according to RT, that the leading actor in AI research and application will become ‘the ruler of the world’. But South African entrepreneur Elon Musk warned, once again, that this competition could lead to a third world war.

China, Russia, soon all countries w strong computer science. Competition for AI superiority at national level most likely cause of WW3 imo.

— Elon Musk (@elonmusk) September 4, 2017

The incredibly transformative potential of new technology has divided opinion in the medical sector, with companies embracing the use of AI now having to specify their software does not intend to replace the work of individual clinicians in any shape or form.

Instead, they say it aims to speed up processes and solve system complexity to simply help clinicians care for their patients.

At the Health and Care Innovation Expo in Manchester earlier this year, NHS England Chief Executive Simon Stevens said they will invest in AI over the next few months, building on recommendations from University of Oxford Regius Professor of Medicine Sir John Bell, included in the Life Sciences Industrial strategy:

“John Bell’s report identified the faster application of AI in the NHS as one of the four big winds that we could potentially support and without doubt we will see as we make the case for public investment and NHS investment in this area, this will be one where NHS England will be backing that with our investment over the course of the next 12 months,” Stevens added.

#Expo17NHS @NHSEngland Simon Stevens: there is a 'clear recognition of the importance of the life sciences industry to the UK'

— BJHC News (@BJHC_news) September 12, 2017

At a recent King’s Fund event, Dr Dominic King, Clinical Lead at Google’s DeepMind Health, gave an example of how an AI-enabled system could change a patient’s experience.

Robert, a patient with poorly controlled diabetes, collapses at home and hits his head, but he is wearing wearable sensors so his doctor is automatically alerted that something is wrong.

“AI is just a much better statistical approach to assessing risk and predicting things than anything we’ve ever had access to before, so these wearable sensors predict that there is some issue or problem with Robert that needs escalation or activation,” King explained.

Seeing the alert, his doctor accesses the medical record on his smartphone and realises the incident cannot be solved in a community setting, therefore he organises an urgent admission.

Another algorithm, King added, could predict what hospital Robert’s care could be best managed at, based on his symptoms, while if he needs a CT a system could ‘intelligently’ trial the order of patients that should be seen first.

After a few other interventions and Robert’s recovery, an AI system could help automate the paper-filling process for his discharge form and ease pressure on doctors and nurses.

An ‘autonomous’ vehicle could even take Robert home at the end, King laughed.

But when will NHS patients see this transformation?

“Nothing I showed you is not actually already happening,” King said.

“In every example there’s wearable technology, imaging technology, surgical targeting technology that is available, it’s just not joined up in any meaningful way at the moment.”

The NHS has a long way to go before AI can be effectively leveraged

Dr Vasilis Argyriou, Associate Professor at Kingston University London’s School of Computer Science and Mathematics, emphasised smartphones and wearable devices already incorporate sensors with a digital health monitoring function:

“These systems are based on AI and machine learning that help to monitor physiological data of individuals and older people with chronic conditions.

“Furthermore, machine learning and AI can be used for safety monitoring (detect falls, seizures and heart attacks in older people), rehabilitation, personalised treatments and early detection of disorders and other conditions (e.g. dementia),” he told BJ-HC.

The NHS, it seems, has a long way to go before it can look at AI becoming an integral part of the system. Analysts argue the ‘paperless’ agenda, with an unclear timeframe, presumed to have moved from 2018, to 2020, and ‘unofficially’ to 2023, is unrealistic at this pace. 

Trusts up and down the country continue struggling with Electronic Patient Record systems, as a recent HSJ story showed, and the consensus is that the current infrastructure cannot support the use of AI.

“The idea that an algorithm from DeepMind or any other organisation is going to sort out some of the issues that are a consequence of the use of outdated technologies is actually fanciful and so this is why so much of our focus is in the work we’re doing on the Streams clinical app,” King told delegates last week.

Dr Argyriou argued that this is a ‘multi-dimensional’ issue, based on the readiness of the technology, the public’s perception, economic factors and many others.

However, the experts agree that AI-enabled care will become a reality in the future, but will the NHS be the ‘last man standing’ as every other sector embraces technology?