The AI technology that could already save your life from diagnosing strokes to spotting heart attacks and cancer

ARTIFICIAL intellig­ence might seem futuristic but experts say it is already helping the NHS save lives every day.

Last year the Government made £21million available to support the use of AI-driven technology in the health service.

GettyArtificial intellig­ence might seem futuristic but experts say it is already helping the NHS save lives every day[/caption]

SuppliedTara Donnelly, the former chief digital officer of NHS England, is now the founder of health advisory firm Digital Care which helps the NHS scale tech such as AI[/caption]

The NHS AI Lab has also been set up to co-ordinate Government, health providers, academics and tech firm any promising innovations.

Tara Donnelly, the former chief digital officer of NHS England, is now the founder of health advisory firm Digital Care, which helps the NHS scale tech.

She told Sun on Sunday Health: “Most weeks there is exciting news of a potential breakthrough in AI in health.

“For example, at Moorfields Eye Hospital in London, the AI system can recommend the correct referral decision for over 50 eye diseases with 94 per cent accuracy, matching world-leading eye experts.”

One reason the NHS is so suited to the use of AI is that it treats almost 1.3million people a day.

That can mean long waiting lists and A&E waits but is also an advantage when it comes to AI, because those patients produce a lot of data.

And it’s data that AI learns from, to make sure it works optimally.

Tara said: “The UK is a great place to train AI for health care purposes because we have the world’s population living in our cities and the more diverse a data set, the better for creating AI that will work for all.

“Some of the most marked impact is in imaging, so we are seeing improvements in stroke care, eye disease and mammography.”

Here she reveals the AI technologies that are saving lives right now and will continue to do so in the future . . . 

STROKE DIAGNOSIS

THIS year, tens of thousands of stroke patients will get treated more quickly, thanks to cutting-edge AI stroke-diagnosis technology.

Patients taken to one of 11 stroke networks across the country can expect to have their brain scans quickly assessed by e-Stroke system The Brainomix.

It uses AI algorithms to shave life-saving minutes off one of the most time-sensitive diagnoses in medicine.

This is thought to have tripled the number of patients recovering from a stroke.

More than 111,000 suspected stroke patients have benefited in just one year.

LUNG CANCER

MORE than 600,000 chest X-rays are carried out on the NHS every month and they are most commonly used to diagnose lung cancer — the leading cause of cancer death in the UK.

Now lung problems that need further investigation can be flagged up immediately when those X-rays are run through AI software, called qXR.

Instead of going home and waiting for a radiographer to review their X-rays, patients identified by AI can be sent for a same-day CT scan which can reveal if they have cancer or not.

RADIOTHERAPY

FOR cancer patients, the time between a radiotherapy referral and treatment is stressful.

That’s because preparing scans can take up to three hours, as doctors need to highlight healthy tissue around the cancer to ensure it’s protected from the radiation.

But by using AI technology Osairis, they can now prepare the scans in just minutes.

The tech reduces the labour-intensive process of manually outlining the health tissue on scans, known as “segmentation”.

HEART ATTACKS

GettyA new AI tool has been trained to spot patients who might be at risk of heart attacks[/caption]

EVERY year around 350,000 NHS patients have a cardiac CT scan, which highlights any dangerous blockages or narrowing of the arteries that can lead to fatal heart attacks.

But the CT scan doesn’t always pick up minute narrowings, which may break up if they are inflamed and can lead to a heart attack in the years following.

A new AI tool has now been trained to spot patients who may be at risk of this, by using information on changes in the fat around inflamed arteries.

It generates a risk score for the patient, and research has found if the technology was used in the NHS it could lead to more than 20 per cent fewer heart attacks.

PREGNANCY

GettyThe Tommy’s app uses an AI algorithm that has learned from data from more than 120,000 pregnancies that can help assess a woman’s risk of complications[/caption]

The algorithm processes information from hormone blood tests and scan resultsTommy’s

THERE are 25,000 cases of the pregnancy complication pre-eclampsia, 3,000 stillbirths and 60,000 preterm births per year in the UK.

Now a woman’s risk of developing this dangerous condition can be identified by The Tommy’s app, an AI algorithm that uses data from more than 120,000 pregnancies.

Hormone blood tests and scan results from a patient’s pre-natal appointments are run through the algorithm, flagging those in need of treatment and careful monitoring.

AI is also being used in IVF clinics across Europe, Asia and South America to detect the most viable embryos.

Clinics using the software have reported a 30 per cent increase in their success rates.

BRAIN SURGERY

GettyNHS surgeons are now working with an AI system trained on hundreds of videos of pituitary gland surgery, which is able to help them to train for the real thing[/caption]

PRECISION is everything in brain surgery, and not damaging the pituitary gland during an op to remove a tumour is critical.

NHS surgeons are now working with an AI system trained on hundreds of videos of pituitary gland surgery, which is able to help them to train for the real thing.

Within just ten months, the AI was able to reach the same levels of knowledge that a surgeon with ten years’ experience would have.

BREAST CANCER

GettyThere is currently a 29 per cent shortfall of NHS radiologists but a ground-breaking AI tool could help to plug that gap[/caption]

SCREENING for breast cancer on the NHS is vital for catching the disease early, so helping to save lives, and women aged 50 to 71 are invited for a mammogram every three years.

But there is a 29 per cent shortfall of NHS radiologists, and several breast units across the country have closed.

Now AI could help plug that gap. A trial published in medical journal Lancet Oncology found an AI-supported screening tool could pick out cancer as fast as two radiologists.

The NHS is currently looking at how to put the AI to work in its breast-screening programme