June 20, 2022

expert reaction to study looking at using predictive modelling and a single MRI scan to diagnose Alzheimer’s Disease

A study published in Communications Medicine looks at the use of an MRI brain scan and predictive modelling to detect Alzheimer’s disease.

Dr Rosa Sancho, Head of Research at Alzheimer’s Research UK, said:

“Alzheimer’s disease is the most common cause of dementia. We desperately need to see better ways to tackle Alzheimer’s, and this requires progress on a number of fronts. A key part of turning the tide is improving how we identify and diagnose people with the disease. This will enable patients to access support and available treatments, but also help to address some of the significant challenges in recruiting to dementia research, particularly clinical trials.

“Doctors may request an MRI for people with suspected Alzheimer’s, but these scans cannot conclusively show whether or not someone has the disease. At the moment an MRI can help to rule out other potential causes of memory and thinking problems, such as a brain tumor, or to see if there are signs of brain shrinkage that could help distinguish between different diseases that cause dementia.

“In this study, scientists developed a computerised approach to predict if someone has Alzheimer’s disease. The researchers analysed MRI brain scans captured routinely in hospitals and then used a computer algorithm to generate a complex picture of the brain and predict Alzheimer’s.

“This is not the first-time using computer technology like this has shown promise, outperforming scans that look for a single measure of brain health alone. However, the computational effort to process the information from the brain scan is large and future research is needed to understand how the process can be made more efficient.

“Through our own research we know the public support having brain scans to help know their own risk of developing the disease. These findings will need to be further developed before we know how it could benefit people undergoing diagnosis in the clinic. We need to see sustained funding and ambition for dementia research to turn promising discoveries like this into real world breakthroughs that are crucial to improving the diagnostic pathway and preparing the NHS for future treatments.”

Dr Richard Oakley, Associate Director of Research at Alzheimer’s Society said:

“We estimate there are around 900,000 people living with dementia in the UK, with over 250,000 people with dementia living without a diagnosis. Having a diagnosis is vital as it opens the door to treatment to manage symptoms and support, and helps individuals and families plan for the future.

“This new MRI-based analysis method could simplify the Alzheimer’s disease diagnosis process. This analysis does not require an expert to run, uses MRI brain scanning technology which is already available, and is 98% accurate in distinguishing brain changes due to Alzheimer’s disease. This could help clinicians come to an accurate Alzheimer’s disease diagnosis more quickly and easily. 

“As promising as this research is, the technology only distinguishes Alzheimer’s disease and not other diseases that cause dementia. We hope future studies will expand the MRI computer programme so it can differentiate brain changes linked to other diseases that cause dementia.”

Prof Rob Howard, Professor of Old Age Psychiatry, University College London (UCL), said:

“While this is an interesting study that adds value to the application of MR imaging to support the diagnosis of Alzheimer’s disease dementia, it is really important to say that such a diagnosis can never be made on the basis of a brain scan alone.

“A diagnosis of dementia is life-changing and should always be made after consideration of the patient’s history, the clinical examination and the results of tests such as brain scans. Over-reliance on brain imaging has been shown to be associated with dementia misdiagnoses and I have learned to be cautious with patients who tell me that their dementia was diagnosed from a brain scan.”

Dr Tom Russ, Director of the Alzheimer Scotland Dementia Research Centre; Consultant Psychiatrist & Honorary Clinical Senior Lecturer, University of Edinburgh, said:

“This study presents an interesting approach to analysing brain scans using a computer algorithm rather than a doctor examining the scan. It shows promise for an automated approach to analysing brain scans for evidence of Alzheimer disease. However, many people have Alzheimer disease in their brains who never develop any symptoms of dementia. As such, this approach will not be able to replace the process of getting a dementia diagnosis in a memory clinic. It may prove to be a useful development of how we investigate people with possible dementia, though.”

Dr Charles Marshall, Clinical Senior Lecturer and Honorary Consultant Neurologist, Wolfson Institute of Population Health, Queen Mary University of London, said:

“It is exciting to see the development of technologies like this that may be able to detect subtle changes in the brain better than the human eye can. This type of approach has the potential to revolutionise early and accurate diagnosis of Alzheimer’s disease. For patients to benefit, we now need to evaluate how well machine learning technologies can detect Alzheimer’s disease in real-world clinical settings rather than using carefully curated research data.”

‘A predictive model using the mesoscopic architecture of the living brain to detect Alzheimer’s disease’ by Marianna Inglese et al. is published in Communications Medicine

DOI: https://doi.org/10.1038/s43856-022-00133-4

Declared interests

Prof Rob Howard: “No relevant conflicts of interest.”

Dr Tom Russ: “No conflicts in relation to this study.”

Dr Charles Marshall: “No relevant declarations.”

For all other experts, no reply to our request for DOIs was received.