Brain imaging predictors of cognitive ability - a multimodal investigation

Supervisors: Alison D Murray, Edwin Jacques Rudolph van Beek, Keith W Muir, Anca Sandu Giuraniuc

Project description:

Multimodal investigation based on Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) have facilitated our understanding of brain ageing and pathologies which cause cognitive decline, although there remain many unanswered questions. Distinguishing normality from asymptomatic abnormalities and identifying factors that endow us with the resilience to preserve normal cognitive ability (cognitive reserve) will help develop future diagnostic methods.

The Aberdeen Children of the Nineteen Fifties (ACONF) cohort comprises individuals born in Aberdeen between 1950 and 1956. This is a resource with exceptional longitudinal phenotyping, imaging and multiple ‘omics data’ in a fully NHS data-like sample, who recently had comprehensive assessment including 3T brain MRI as part of Welcome Trust funded STRADL study (300 participants). The plan for this PhD proposal is to perform 7T MRI, amyloid PET MR and ApoE genotyping to provide the greatest detail currently available in these informative participants. This project has the advantages of both accessing national facilities (7T MRI in Glasgow and PET MR in Edinburgh) using a highly informative sample from Aberdeen and providing rich brain imaging data that will define normal brain ageing.

The participants are now at an age when, there is increased inter-individual variability, which remains within the normal range, but is reduced for that individual compared to earlier measures.  Based on cognitive change between midlife (54-60 years) and late-midlife (60-65 years) we will invite those with diverging cognitive profiles for multimodal imaging (PET and MRI). The pressing problem is to pinpoint the start of these changes in this rather short interval. The digit symbol test, which measures information processing speed, has less practice effect, is sensitive to age, and is thought to be a fundamental variable that underlies intelligence, will be used to measure cognitive trajectory.

Brain amyloid deposition is present in up to 28% of healthy subjects age 65-70[1], is increased in those with pathology and precedes neurodegeneration. High brain iron may predispose to neurodegeneration[2] and cerebral small vessel disease frequently co-exists with other pathologies and is ubiquitous in older people. The hypothesis is that 1. amyloid burden, 2. brain iron  3. small vessel disease and 4. brain atrophy are distinct pathological burdens with different risk factors, but that all need to be quantified to give a complete picture of the negative burden on cognition.

The PhD candidate will have a unique opportunity to access state-of-the-art multimodal images collected from PET MR and 7T MRI scanners. The MR protocol will include the following sequences: 3D T1-weighted to measure brain volumes and shape; 3D T2 and fluid attenuation inversion recovery (FLAIR) to quantify hyperintensities; susceptibility weighted imaging (SWI) to quantify brain iron; and resting state fMRI -to investigate brain networks.

PET-MRI multimodal analyses will be performed in order to identify precisely the location of amyloid deposition on co-registered high resolution MR images, co-varying for ApoE genotype, accessible through Generation Scotland. Image analysis expertise will be enhanced by co-supervision and 6 month placement with David Dickie Analytics (DDA).


[1] Cecchin D, Barthel H, Poggiali D, Cagnin A, Tiepolt S, Zucchetta P, Turco P, Gallo P, Frigo AC, Sabri O, Bui F. A new integrated dual time-point amyloid PET/MRI data analysis method. Eur J Nucl Med Mol Imaging. 2017 Jul 4.

[2] Ayton S, Faux NG, Bush AI; Alzheimer's Disease Neuroimaging Initiative. Ferritin levels in the cerebrospinal fluid predict Alzheimer's disease outcomes and are regulated by APOE. Nat Commun. 2015;6:6760.