New diagnosis methods

A new hi tech early dementia assessment service that could slash the time it takes for doctors to diagnose Alzheimer’s disease from 18 months to three months is being developed, thanks to underpinning research at Imperial College London.
The assessment service integrates several cognitive tests with computerised examinations of magnetic resonance imaging (MRI) brain scans. It could provide medics with a way of detecting dementia much earlier than is currently possible using conventional technology. At the heart of the assessment service’s software is an algorithm created by Dr Robin Wolz and Professor Daniel Rueckert from the Biomedical Image Analysis Group in the Department of Computing at Imperial.

The new system could slash the time it takes for doctors to diagnose Alzheimer?s disease from 18 months to three months

The researchers say the 15 months of diagnosis time the new system could save are critical to the effectiveness of treating diseases such as Alzheimer’s. This is due to the fact that existing drug treatments are most effective in these diseases’ earliest stages.
Dr Wolz, who received his Ph.D. from Imperial College London in medical image analysis, says:
“The size of the parts of the brain important for memory such as the hippocampus and amygdala steadily decrease as Alzheimer’s progresses. Doctors use this as an indicator of how far Alzheimer’s has progressed in patients. Currently, doctors have to work out what is happening to the structure of the hippocampus using an MR scan. This is a painstaking, costly and time-consuming process, where doctors have to analyse the three-dimensional scan slice by slice, so is only ever done in highly specialized centres. Our automated technology can provide this information to all doctors, seamlessly integrated into current processes, and by adding information and increasing confidence, speed up the diagnosis process. This could lead to quicker treatments for patients to improve their quality of life.”
Dr Wolz’s method, Learning Embeddings for Atlas Propagation (LEAP), removes the need for a doctor to map out the brain structure by hand. LEAP can examine MRI brain scans and measure the volume of specific brain structures like the hippocampus and amygdala rapidly and accurately.
As part of the project, GPs will start testing the memory of patients who potentially have dementia using the memory-testing software. Patients who test abnormally will take a second cognitive test and receive an MRI brain scan. The scans will run through a computer program that uses Dr Wolz’s method to assess signs of brain shrinkage. The results from these two tests will be combined with an assessment of blood vessel damage in the brain and given to the GP who will decide on further treatment.
This project is a collaboration between Cambridge Cognition, a creator of cognitive tests, and the medical imaging company IXICO Ltd., which is an Imperial College London spinout company founded in 2004 by academics from Imperial, University College London and King’s College London.
In early November, the partners were awarded a grant for £2.1 million towards the cost of this £3.3m project from the Government-funded Biomedical Catalyst to establish The Brain Health Centre. The aim of the Centre is to make testing for dementia by the NHS quick, cost-effective and seamlessly integrated into the existing patient pathway and health care structure. It will also enable the team to collect valuable data on dementia, which will be used in the development of the assessment system.
Dr Wolz adds: “It is exciting to be able to contribute to this ongoing project. It is very motivating to see my research applied in such a setting. I think that the project has the potential to revolutionise the diagnosis of dementia in day to day practice.”