Principal Investigator: Dr. Gordon Waiter
Department: Aberdeen Biomedical Imaging Centre
Institution: University of AberdeenTags: 24089, ageing, diffusion, MRI, template, white matter
University of Aberdeen’s Roland Sutton Academic Radiology Trust grant.
1a: Progress towards understanding changes in brain structure in diseases associated with ageing, such as Alzheimer’s Disease (AD), are hampered by a lack understanding of “normal” ageing. Characterising “normal” brain ageing would allow the focus to be shifted to areas that appear abnormal.
- Stratify individuals according to markers of “normal” cognitive ageing from UKBioBank data.
- Apply Tract-Based Automatic Analysis (TBAA) to characterize “normal” white-matter microstructural features.
- Identify white matter regions where age significantly associates with diffusion-relevant microstructural features.
- Incorporate macroscopic and microscopic structural information to create age specific normal adult brain templates.
1b: This research seeks to use the clinical, cognitive and imaging data from UKBioBank to study the mechanisms of age related changes in brain structure and use them as a platform to better diagnosis. This aim is completely consistent with UKBioBank’s aims. Providing this mechanistic information will help to identify new therapeutic interventions and possible lifestyle changes. Stratifying age related changes in white matter structure into more homogenous categories will provide better ‘disease’ targets for mechanism based research because there will be less aggregation of individuals with diverse aetiologies within the same heterogeneous category of ageing.
1c: We will create digital atlases of the brain using MRI scans from people without disease so these may be used to better highlight subtle changes in the brain that are associated with conditions like Alzheimer’s Disease. We will align the brains and measure features in the white matter, the electrical circuitry of the brain, along each cable bundle. The output of these measurements is a set of data for each participant that can be compared. We will investigate the relationships of these white matter features with respect to age, sex and behavioral variables such as cognitive performance.
1d: We are interested in the subgroup of UKBioBank data that includes brain imaging (MRI) data. From that data we will identify cognitively normal adults across the lifespan for further analysis.
Our original aims are to:
- Stratify normal individuals according to age and markers of normal cognitive ageing from UK Biobank data.
- Apply the Tract-Based Automatic Analysis (TBAA) algorithm (Chen et al., 2015) on the diffusion data of the normal individuals to standardize the white matter microstructural features.
- Identify the white matter regions where the age significantly associates with the diffusion-relevant microstructural features.
- Incorporate macroscopic and microscopic structural information to create age specific normal adult brain atlases/ templates.
To achieve these aims we need to create a “clean” normal dataset by excluding subjects who were diagnosed with neurodegenerative, neurovascular or psychiatric disease at any of the assessment visits. This data is available in the Main and Secondary Diagnoses fields 41202 and 41203 respectively. Subsequently we can then look at these ‘special subjects’ to see if their tract integrity really falls beyond the range of our normal template. Damage to the white matter, as evidenced by white matter hyperintensities visible in T2 FLAIR images, will have an adverse effect on our Tract-Based Automatic Analysis (TBAA) algorithm used to standardize white matter microstructural features. We therefore request the T2 FLAIR structural brain images in NIFTI format, field 20253, to allow us to create masks of normal appearing white matter. One aspect of macroscopic structural information that is important in ageing is structural complexity. We would like to incorporate a measure of structural complexity into our age specific normal adult brain atlases/ templates that we can derive from the T1 scans we already have. This was not explicitly mentioned in the original application but we believe it is within the scope of the original application but we wanted confirmation that this is the case.
Last updated Jul 5, 2017