Research

How artificial intelligence can help people affected by dementia

What is artificial intelligence?

Artificial intelligence (or machine learning) may sound like something from a science fiction novel, but today it’s a feature of everyday life.  
Artificial intelligence (AI) is computer technology that allows programmes to simulate human intelligence and problem-solving abilities. 

These programmes can rapidly learn and analyse information, make decisions, and self-correct when doing a task that might normally be carried out by a person. In some cases, they might even understand and respond to what you say or ask it to do, for example when using voice-activated devices and applications in your home or mobile phone. 

AI in research

AI is designed to learn information incredibly fast and carry out analysis and find patterns or solutions much quicker than a person can. For example, analysing a brain scan or discovering new pathways and mechanisms based on biological data. 

There are many ways that researchers can use the power of AI, ranging from creating innovative devices to help people living with dementia, to developing computer programmes that would help them learn and analyse information at a much faster rate.  

AI in dementia research and technology is highly adaptable. As dementia is progressive, AI and machine learning solutions allow programmes to learn about a person, understand when their condition is in decline, and change the support provided.  

Ultimately this technology can support people to be more independent and avoid crises that might see a person hospitalised, as well as allowing people to continue with the activities that make life worth living.

Artificial intelligence is a fast-moving field in dementia research. Alzheimer’s Society is working hard to make sure AI is used to make a positive impact for everyone affected by dementia. 

Ways artificial intelligence can help people affected by dementia

Earlier diagnosis 

One of the challenges researchers face, is being able to accurately and quickly diagnose dementia. Depending on the type of dementia, brain changes can begin up to 20 years before symptoms appear, by which time many brain cells have died. 

Researchers across the globe are developing computer programmes to spot the earliest signs of dementia. These teams are using huge amounts of data to identify changes to the physical structure, energy use and chemical make-up of the brain in detailed scans that can’t be seen by the human eye.

This technology can also be used to spot patterns or changes to memory and thinking.  

Technology like this is used in an ongoing project by Professor Geoff Parker (University College London) funded by Alzheimer’s Society. Professor Parker’s research is working to develop a faster way of carrying out accurate MRI scans using machine learning to diagnose Alzheimer’s disease. This could result in a more cost effective, efficient, and widely available method of early diagnosis.  

The length of time for a person to have an MRI being shortened and the images being better would really make a difference to people's experience, those in the machines and those interpreting the findings

- Alzheimer's Society Research Network Volunteer

Understanding how dementia symptoms develop

Another challenge for researchers and clinicians is that dementia is a progressive condition. It affects people differently over time. 

Researchers want to use AI to predict how the condition will affect the brain physically over time and, as a result, how symptoms will change and develop in the future. This is vital for looking at the true impact of new treatments being tested in people with the condition, and for people with dementia and families to plan more effectively for the future.

Researchers are also beginning to develop use artificial intelligence to analyse our genetic profiles and create models to predict who might be more likely to develop later in life.  

This is why we funded research by Dr Riccardo Marioni, who investigates whether it is possible to predict dementia years before it is diagnosed. 

To do so, Dr Marioni will use machine learning to analyse data and give us an insight into: the genes that cause Alzheimer’s disease, how lifestyle affects the chance of developing dementia, how to identify high-risk individuals, and how to use all this information to design more efficient clinical trials tailored to individuals. 

Discovery of new treatments

Machine learning methods and AI can be used in dementia drug discovery and development.  

AI-based technology can help researchers analyse large datasets to find potential new drug targets, predicting if there may be safety issues with a drug, monitoring potential side effects, developing new drugs, and improving the design of new clinical trials.  

An important feature of clinical trials is monitoring symptoms of participants before and after treatment, for example by monitoring changes in memory. These are usually carried out through standard questionnaires and assessments. 

In an Alzheimer’s Society funded research, Dr Magdalena Jones (Imperial College London) aims to find new ways to accurately monitor new and existing dementia medication and their side effects. To help with this research, Dr Jones uses an AI-based technology called Cognitron to carry out patient assessments which would be more reliable than questionnaires. 

Living at home for longer and maintaining independence

Technology like artificial intelligence will never replace person-to-person care but used in the appropriate circumstances, could enhance care and support people affected by dementia to live at home for longer and avoid crises situations.  

Researchers are exploring the possibility of using sensors around the house to intelligently monitor a person's behaviour. This will allow them to spot when they are doing something risky or if there are signs their health might be deteriorating. They can then take action to help the person in real time. 

The Care and Technology centre of the UK Dementia Research Institute, part funded by Alzheimer’s Society, is using AI to automatically integrate patient’s information and flag any unexpected changes. For example, highlighting a change in walking pattern that might suggest a patient is at risk of a fall.

We’re supporting researchers at the University of Oxford who are using smartphones to collect huge amounts of data through the GameChanger project. The research will help us understand healthy changes to the brain as we age.

Innovative AI products to help people living with dementia 

Alzheimer’s Society is proudly collaborating with Innovate UK to fund cutting-edge technology through the Longitude Prize on Dementia, which will be delivered by Challenge Works.  

The Longitude Prize on Dementia is a £4.42 million prize to drive the creation of personalised, technology-based tools that are co-created with people living with the early stages of dementia, helping them live independent, more fulfilled lives and being able to do the things they enjoy.

Currently, there are 24 teams competing on the prize, and the winning solution will use the latest advances in technology, artificial intelligence and machine learning to provide personalised support for people living with dementia. 

Looking forward

AI is a growing area of dementia research. It holds huge potential for improving care, and accelerating a more accurate, early diagnosis of dementia, as well as predicting its progression.

Whilst AI could be a useful tool, it is also essential to look at it with a critical eye. We must ensure that AI does not replace the human contact that is so important in care instead enhancing care in a way that is beneficial to people affected by dementia. 

AI-based technology should be co-designed alongside people with dementia and their carers to ensure that the devices used were designed and located so that they respect the privacy and dignity of the people researchers are supporting.    

We look forward to seeing how this exciting field grows and develops in the coming years. We will continue to support research in this fast-paced, ever-changing field. 

This article was first written in 2019 and most recently updated in July 2024.