This site has been optimized to work with modern browsers and does not fully support your version of Internet Explorer.

Evaluation of image analysis technology supporting dementia diagnosis

Overall summary

Almost one million people in the UK live with dementia, with an associated economic cost estimated at £25 billion in 2021. This figure is predicted to increase to £94 billion by 2040 as the population ages.

Magnetic Resonance Imaging (MRI) scans are used to detect changes in brain volume and can help to confirm a clinical diagnosis of dementia. However, subtle changes in brain volume can be hard to identify from brain scans alone.

FSL is a software platform which can extract metrics from brain scans. A research team in Oxford is looking to leverage FSL to extract quantitative data of brain structures to support the dementia diagnosis process.

The Oxford AHSN (now Health Innovation Oxford and Thames Valley) carried out a barrier to adoption study to evaluate the potential clinical applications of FSL in the dementia diagnostic pathway in the NHS in England.

What is the challenge?

Dementia mainly affects people over the age of 65, with the likelihood of developing dementia roughly doubling every five years. Dementia can also develop at an earlier age – in the UK, an estimated 70,800 people aged under 65 are living with young onset dementia. Dementia has a huge impact on the life of the person affected, as well as their carer(s) and family. Timely and accurate diagnosis is crucial to provide access to the right care and support.

There are often delays in obtaining a dementia diagnosis. In 2021, the overall waiting time from referral to diagnosis was 17.7 weeks, although this was impacted by the COVID-19 pandemic. In the clinical diagnosis of dementia, structural MRI plays a key role in excluding other pathologies, as well as revealing patterns of brain atrophy. These are currently evaluated qualitatively by neuroradiologists who sometimes use visual rating scales. However, these qualitative ratings are time-consuming, can lack sensitivity and depend on the radiologist’s experience.

What did we do?

FSL can quantify individual brain volumes, which can be compared to big datasets, such as UK Biobank for a healthy, age-specific reference population. This can aid neuroradiologists in interpreting the severity and distribution of brain atrophy in people with suspected dementia.

The Oxford AHSN performed a barrier to adoption study using the lean assessment process methodology and engaged with key stakeholders working in the dementia diagnosis pathway to gain insights into the perceived usefulness, potential clinical benefits, acceptability and barriers to adoption of FSL for this clinical need.

A literature review was also carried out to identify the current pathways for diagnosing people with suspected dementia. The clinical pathway mapping exercise allowed the Oxford AHSN to identify key stakeholders to interview along the pathway. An information sheet on the technology and semi-structured interview questions were compiled. These included qualitative and quantitative questions which allowed clinicians to share their views on the technology.

What has been achieved?

The stakeholders interviewed were very positive about having analysis software to provide quantitative data of brain structures from MRI scans in the dementia diagnosis pathway. One of the key perceived benefits was that it would increase diagnostic confidence and enable better monitoring of the subtle changes in brain volume caused by disease progression. Stakeholders agreed that FSL outputs had the potential to reduce variation and subjectivity compared to existing measures.

What people said

“We have really enjoyed working with the Oxford AHSN and have found the report incredibly useful for our project. The report was very comprehensive and clear, covering all the important aspects of the technology. The AHSN provided an excellent mix of quantitative and qualitative information that can be used to advocate for the importance of the technology, whilst also providing constructive feedback for must-have features. We would love to work with the AHSN in future as the project progresses!” – Ludovica Griffanti, Associate Professor and Alzheimer’s Association Research Fellow, University of Oxford

What next?

The team has successfully secured further funding to build an early stage prototype. They will look to integrate this into the Oxford Brain Health Clinic, a hybrid clinical-research service for patients with memory problems, to test the technology in a real world memory clinic setting.


Florence Serres, Project Manager, Health Innovation Oxford and Thames Valley