Relationship of brain functional, structural and vascular patterns to cognition in ageing and dementia

Cognitive decline with neurodegeneration is a major healthcare burden. To tackle this burden, it is necessary to better understand the brain-cognition relationship in ageing and neurodegenerative diseases especially at an earlier stage, ideally before cognitive impairment. Measurements from neuroimaging can be useful to identify the complex processes of neurodegeneration which involve structural, cerebrovascular and functional changes. In this dissertation, I investigated the following questions: 1) Can we explain the relationship between brain architecture and cognition by the integration, or dissociation, of structural connectivity and functional connectivity? 2) Do the patterns of functional network topography, brain structure and cerebral blood flow have independent or synergistic effects on cognitive function in healthy ageing? 3) Do the patterns of functional network topography, brain structure and cerebral blood flow have independent or synergistic effects on cognitive function at the presymptomatic stage of dementia? To address question 1, I investigated whether the increased reliance on maintaining functional connectivity for good cognition in old age is facilitated by structural network connectivity or independent of structural network connectivity. I used multivariate integrative approaches to assess the relationship to cognition across the adult lifespan of (i) shared signals between structural and functional connectivity and (ii) unique signals in functional connectivity that are independent of differences in structural connectivity. I found that the maintenance of cognitive functions in older people depends on functional connectivity supported by strong structural connectivity. This dependency was related to education level. To address questions 2 and 3, I combined multimodal neuroimaging to investigate the differences associated with cognitive decline across the adult lifespan and in genetic frontotemporal dementia. On healthy ageing subjects from the Cam-CAN cohort, I applied linked independent component analysis which allows for simultaneous characterization of MRI-based structural, cerebrovascular, and functional measurements. The purpose was to integrate all of these neuroimaging signals to better understand their contribution to cognitive decline in ageing. I then applied the same method to presymptomatic carriers of genetic mutations associated with frontotemporal dementia using the GENFI cohort. I found that among these brain patterns, functional network integrity, particularly the frontoparietal network integrity, was the most sensitive to the effects of genetic mutations and significantly correlated with cognitive performance both in healthy ageing and presymptomatic frontotemporal dementia mutation carriers. Together the findings suggest that brain structural, cerebrovascular and functional networks show cognition-related alternations while the individuals are still cognitively normal. This process can be affected by age, genetic mutation, education and the interplay between patterns within the brain. Functional network integrity and connectivity are particularly important for the maintenance of normal cognition, especially in older subjects and in those approaching the likely year of dementia onset. Integrating multiple neuroimaging measurements could be a promising approach to stratify phenotypic and genotypic heterogeneity in otherwise-fuzzy cognitive variability in ageing and neurodegenerative diseases.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2023

Author dc:creator

• Liu, Xulin

Advisor dc:contributor.advisor

• Rowe, James

• ageing
• atrophy
• cerebral perfusion
• dementia
• frontotemporal dementia
• functional networks
• multimodal
• neuroimaging

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/6261c75d-e123-45ee-a54d-992590b82ea3/download
• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

eng

Author Identifier

0000-0002-8219-2848

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/364972