From diagnostics to listening in to your GP appointments, all the ways artificial intelligence is set to shake-up health

It’s the revolutionary technology that proponents and critics alike claim will change the world upturning every industry or potentially killing us all. 

But Prime Minister Rishi Sunak this week hailed AI as a way to ‘help millions’ of Brits as he spoke ahead of a global summit on artificial intelligence in South Korea. 

He was speaking about a new NHS trial that will see thousands of British women have their mammograms assessed by an AI ‘doctor’.

It’s hoped the results will not only lead to more accurate diagnoses but also speed up waiting times as stretched staff will have more time for other tasks. 

This is just one way, however, that proponents claim the tech will help revolutionise healthcare.   

Here, MailOnline explores some of the other aspects of AI-driven health care that could be coming to a hospital, GP practice or even the palm of your hand…

Reading your scans and test results

Mr Sunak’s flagship announcement was an example of perhaps the easiest way AI could be incorporated into healthcare. 

Alongside his South Korean counterpart, the PM hailed a collaboration between the NHS and Korean firm Lunit on using AI to improve the speed and accuracy of breast cancer diagnosis

‘We all know how vital early detection is — so just imagine what improvements here could mean for millions of women and their families,’  they wrote in a joint article in The i.

In the trial, medics will use AI to analyse mammograms, scans of the breast taken to spot signs of breast cancer, at a screening clinic operated by the Leeds Teaching Hospitals NHS Trust.

Click here to resize this module

Normally, this analysis is done by two radiologists who each separately look at a scan, to double-check no signs of cancer are missed. 

But in the trial one of these medics with be substituted with the AI.

Organisers anticipate 2,575 scans will be looked at using the system with the results compared when the trial concludes next year. 

It’s not the first time the technology has been hailed as a game changing diagnostics tool. 

AI’s ability to spot tiny subtle changes in tissue, almost impossible to spot with the human eye, is being used in studies and trials to detect a variety of diseases. 

Most of these relate to cancers and involve using AI to analyse, or help analyse scans like X-rays, MRI and CTs or blood tests.

The technology has been tried on lung and bowel cancers, as well as incredibly hard to detect forms of the disease like pancreatic cancer.

Previous studies using millions of old patient scans have found AI can diagnosed conditions at least as accurately as radiologists 94 per cent of the time.

AI can also be used to improve cancer care using the same principles that allow it to detect subtle signs of the disease.

Radiotherapy, a form of cancer treatment that uses intense blasts of targeted radiation to kill tumours can cause side effects like nausea, fatigue, sore skin and infertility as healthy tissue is caught in the crossfire.

But using AI clinicians can further narrow down the sections of the body needing to be treated with radiotherapy potentially reducing the severity of side effects. 

Some AI proponents have even claimed the technology can be used to detect conditions like autism with ‘100 per cent accuracy’ though independent experts have urged caution over the figure.

Boost productivity of time-poor NHS staff

It’s no secret that NHS staff are stretched for time, at least from a patient perspective.

Agonising waits in A&E, record busting backlogs, continually missed targets for timely cancer and mental health treatment, are just some of the ways this is reflected in reality. 

But AI could provide a solution, at least in part.

While a medic can never be in two places at once many of the same ways AI could help improve diagnostic capability also touts to help save staff time.

As per Mr Sunak’s breast cancer example, using AI to replace one of two radiologists interpreting scans effectively doubles the number of staff able to carry out such analyses, increasing the number of staff that can be seen. 

Or, as per using AI to target areas for radiotherapy, a task that used to be done by medics by hand could be sped up, again boosting productivity. 

All of this is, naturally, contingent on the AI actually be effective.

NHS staff aren’t going to be more productive if they have to spend any additional time they would save fixing algorithms’ mistakes. 

However, some plans to use AI to boost productivity have created patient privacy and safety concerns.

One Government proposal, labelled ”creepy’ by critics, touted using the technology to listen in to patients’ private medical appointments.

Announcing the scheme, Health Secretary Victoria Atkins said having AI generate patients notes for the medic the background improve productivity by cutting the time they spent doing paperwork. 

However, privacy and patient safety campaigners said they had ‘huge concerns’ over the safety of patients’ data.

They also warned people could come to harm because they may be too embarrassed to discuss medical issues freely whilst being recorded. 

The much maligned NHS ‘penguin’ robots 

NHS penguin-bots were mocked relentlessly in some NHS circles as a Government white elephant for a service which faces chronic understaffing and some cases crumbling infrastructure. 

Affectionally nicknamed ‘Milton’, the robot uses similar AI to self-driving vehicles to transport medicines around hospitals.

The ‘helper bot’ is being used to carry and deliver prescriptions and other items around Milton Keynes University Hospital, helping to relieve pressure on human staff. 

There were plans to potentially roll out the robots to other NHS hospitals in the future based on this trial.  

But it’s not the only robot, programmed with AI, to potentially slide into healthcare.

Earlier this year, experts released the results of a trial using ‘social assistance’ robots in health care settings. 

These were designed to answer patients’ questions freeing up time for medics.

These mechanical medical assistants were also able to engage in ‘general chit-chat’ with patients – as well as offering quizzes, riddles and even calming breathing exercises.

Other very high-tech robots, such as the 2ft tall Japanese nurse-bot Aeo have been hailed as a potential solution to the NHS workforce crisis. 

In addition to the functionalities of its more penguinesque cousins, Aeo can also check on patients and alert staff if something is amiss, deliver blood samples and speak to patients. 

Could AI therapists help people beat depression and eating disorders?

One in four Brits are believed to experience a mental health problem every year, so could digital therapists prove the way forwards? 

Such are the wait times to see an NHS mental health specialist that thousands of sufferers are showing up at A&E looking for help, the Royal College of Psychiatrists have even said. 

The smartphone app Wysa has already been made available to thousands of teenagers in West London to help them cope with mental illness.

When a user logs on, the app asks how their day is going. 

If they’re feeling anxious, the chatbot guides them through meditation and breathing exercises, for example, to help ease their state of mind with language designed to portray empathy and support.

The app is also being used in a £1million trial for patients on the NHS mental health waiting list in North London and Milton Keynes, comparing their wellbeing with other patients on the waiting list without access to the app.

But the use of digital therapists has triggered alarm among experts who fear some patients who need proper psychiatric care may resort to the apps instead of getting the help they need.

Others have warned the lack of human involvement could even exacerbate mental health issues in vulnerable people.

Last year National Eating Disorders Association in the US was forced to pull the plug on Tessa, a chatbot it developed to replace counsellors. 

It followed claims by former eating disorder sufferer Sharon Maxwell, from San Diego, that the bot had told her a good way to cope was to weigh herself regularly and even measure her body fat with callipers.

And in December, NHS Dorset came under fire after it published a list of free apps it hoped could save young people from deteriorating mental health.

The library, curated by an app review firm, could prevent people from needing other treatments, the trust said. 

However, it cautioned the apps were not an alternative to conventional therapy.