‘Super scans’ give hope in battle against dementia

Dr Hilary issues warning about missed dementia diagnoses

We use your sign-up to provide content in ways you’ve consented to and to improve our understanding of you. This may include adverts from us and 3rd parties based on our understanding. You can unsubscribe at any time. More info

The breakthrough could mean patients get the results in one day and save the NHS millions. Dr Timothy Rittman, the Cambridge scientist leading the study, said: “This is exciting because there are a lot of machine learning algorithms that have been applied to neuroimaging but none of them have really been put to the test in the real world.” The technology may also predict how a confirmed dementia sufferer’s difficulties with memory and thinking are likely to progress.

UK scientists say the revolutionary AI system can detect minute patterns invisible to the naked eye.

Researchers hope earlier diagnosis and treatment might result in fewer admissions to hospital and a huge cut in social care bills. The computer program itself has also been relatively low-cost to develop as it uses brain scans already taken as part of routine care.

Consultant neurologist Dr Rittman, leading the study with colleagues at Cambridge University, told the BBC the approach is a “fantastic development”.

He continued: “These set of diseases are really devastating for people. Anything I can do to be more confident about the diagnosis, to give them more information about the likely progression of the disease to help them plan their lives is a great thing.”

Dr Rittman added: “At the moment the route to diagnosis can be long and can involve coming back to the clinic multiple times.

“And we don’t have any idea of being able to predict prognosis, what’s going to happen.

“The machine learning algorithm basically gives us a score which not only helps with the diagnosis but also gives an idea of how far someone is moving away from the ‘normal’ state towards the disease and dementia state. That gives us an idea of the prognosis.”

More than 850,000 people in the UK have dementia and that toll is tipped to hit 1.6 million by 2040.

Currently, testing for dementia often involves a barrage of tests and brain scans which are assessed by radiologists for signs of disease.

But using AI to analyse the images can help spot miniscule changes that may not be visible to the human eye.

The algorithm compares the scan with those from thousands of dementia patients to identify similar patterns. In pre-clinical studies, it was able to predict if people without symptoms would develop dementia.

The approach is being trialled at Addenbrooke’s Hospital in Cambridge and NHS memory clinics nationwide. Dr Rittman said yesterday: “I’ve done a clinic this morning and there were a couple of patients who I’m not sure whether they’ve got dementia or not.

“I’ve had to send them away to have extra scans or to come back in six months or a year’s time.

“If we could use this algorithm to be more confident about the diagnosis at the first assessment that would really help people.”

A quick diagnosis can reduce the anxiety caused by uncertainty for patients experiencing problems with their memory and thinking.

It would also help spot patients who may benefit from joining trials of drugs intended to tackle Alzheimer’s disease.

Prof Zoe Kourtzi, of Cambridge University and a Fellow at The Alan Turing Institute, developed the algorithm. She said: “If we intervene early, the treatments can kick in early and slow down the progression and at the same time avoid more damage happening to the brain.

“It’s likely that then symptoms might occur much later in life or they may actually never occur.” Around 500 trialists are being recruited in the first year.

Preliminary results are due in around 12 months and researchers hope the technology may be rolled out in around five years.

The system would not replace existing memory clinics but would provide doctors with an extra diagnostic tool.

Dr Rittman said analysis indicated the system might save the NHS half a billion pounds over five years, by reducing hospital admissions and social care costs.

He added: “Anything that enables us to be more confident about the diagnosis is really important.

“Being able to talk to people more confidently about prognosis – what’s going to happen over the next few years – would really be the revolutionary part that we just can’t do at the moment.” Dr Laura Phipps, from Alzheimer’s Research UK, said that the AI technology could have a huge impact on people with dementia – and on their families too.

She continued: “Artificial intelligence systems like this are currently being applied to many aspects of healthcare, drawing on the insights from huge datasets to help doctors make more informed decisions about diagnosis, treatment and care.”

Denis Clark, 75, is taking part in the trial in the hope it will reveal whether he is developing dementia and, if so, how it may progress.

His wife Penelope noticed last year that he was beginning to struggle with his memory. The couple are worried they may have to sell their home to pay for his care if Denis’s health deteriorates.

Penelope told BBC News that having an early diagnosis and indication of how his dementia is likely to progress would help the couple prepare for the future.

She said: “Then we could plan financially as well as as a couple, to be able to have a few holidays.”

Source: Read Full Article