Artificial intelligence could diagnose dementia in one day

·3 min read
Abstract image of making AI brain by low poly wire frame network.
Artificial intelligence could spot signs of dementia after a single brain scan. (Stock, Getty Images)

Artificial intelligence (AI) could diagnose a suspected dementia patient the day they are assessed.

The disease currently has no set test, with medics generally relying on cognitive assessments and brain scans.

With it sometimes taking years to reach a diagnosis, scientists from the University of Cambridge are developing an AI system that could spot signs of the disease after a single brain scan.

The system is "trained" to compare a suspected patient's brain scan against thousands of confirmed cases, with the algorithm potentially identifying signs of the disease that a neurologist could not spot.

Read more: Dementia set to triple worldwide by 2050

Although the technology is still in a trial stage, it could diagnose dementia years before symptoms emerge. This would enable patients to receive treatments that ease their condition before it becomes advanced.

Hands holding brain with puzzle paper cutout, autism, Epilepsy and alzheimer awareness, seizure disorder, world mental health day concept
Dementia is an umbrella term for a loss of brain function. (Stock, Getty Images)

Dementia is an umbrella term for a loss of brain function, with Alzheimer's being the most common form.

In the UK alone, 850,000 people live with dementia, which is expected to increase to 1.6 million by 2040.

Read more: Tooth loss raises dementia risk

"If we intervene early, the treatments can kick in early and slow down the progression of the disease and at the same time avoid more damage, and [therefore] it's likely symptoms occur much later in life or may never occur," said trial investigator Professor Zoe Kourtzi, as reported by the BBC.

The AI system compares suspected dementia brain scans against those of confirmed patients and the latter's medical records.

The algorithm then identifies patterns in the scans, matching them against the outcomes of confirmed patients.

Early stage trials suggest the technology could diagnose dementia years before symptoms develop, even when there is no obvious sign of damage on a patient's scan.

The Cambridge scientists are now testing whether the system works in a clinical setting, alongside conventional ways of diagnosing dementia.

Read more: Professional rugby players have 'concerning' brain changes

The results will be sent to the participants' doctors, who may use the findings to guide their patients' treatment.

The AI system may also predict whether a patient's condition will remain stable for many years, slowly deteriorate or requires urgent treatment.

"When I am delivering this information to a patient, 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 to be able to do," said lead investigator Dr Tim Rittman.

Watch: Severe coronavirus linked to Alzheimer's onset

Around 500 suspected dementia patients are expected to take part in the trial in its first year, one of whom is Denis Clark, 75.

Clark's wife Penelope noticed his memory was deteriorating in 2020. The couple are concerned he may be developing dementia and they could have to sell their home to fund his care.

An early diagnosis would allow the pair to plan financially.

"We would know whether as a couple we could have a few holidays before things get so bad that I can't take Denis on holiday," said Penelope.

Fellow participant Mark Thompson, 57, underwent "at least four scans" before he was diagnosed with dementia.

Thompson, who started having memory lapses 10 months ago, endured mental health problems as a result of his uncertain diagnosis, as he worried he might have a tumour.

A more certain diagnosis earlier in his illness may have put Thompson's mind at rest.

Watch: Improved air quality lessens dementia risk

Our goal is to create a safe and engaging place for users to connect over interests and passions. In order to improve our community experience, we are temporarily suspending article commenting