The Library
Analysis of neuroimaging with big data to understand brain systems involved in emotion
Tools
Wan, Zhuo (2021) Analysis of neuroimaging with big data to understand brain systems involved in emotion. PhD thesis, University of Warwick.
|
PDF
WRAP_Theses_Wan_2021.pdf - Submitted Version - Requires a PDF viewer. Download (4Mb) | Preview |
Official URL: http://webcat.warwick.ac.uk/record=b376647
Abstract
The development of neuroimaging technologies and open-access large-scale datasets provides the opportunity to explore biological big data, by developing computational models for the brain network, aiming to find biomarkers for mental disorders and investigate brain mechanisms. Due to the high dimensionality and massive sample size of neuroimaging data, exploring new computational methods and modelling approaches for neuroimaging data with various behavioural patterns is of great importance. This will lead to more stable and reliable models to understand the pattern of human behaviours and the underlying brain mechanism.
The research interest of this thesis is to explore new data analysis methods and novel modelling approaches of the brain network related to mental disorders including addiction, anxiety and depression, and brain functions related to emotion, personality, and cognitive performance with large scale neuroimaging data.
First, a novel prediction model was developed based on the elastic net regression of the sensation-seeking personality from brain functional connectivity in a large-scale study (Wan et al., 2020). This provides a novel way to investigate the relationship between behavioural measures and brain functional connectivity, replacing the usual correlation analysis with a prediction model. Furthermore, the prediction model examines groups of functional connectivity links, instead of an individual link, which is usual in correlation analysis; this indicates the relationship between behaviour measures and a group of links as community. Biologically, the sensation-seeking score was found to be optimally predicted from the functional connectivity mostly between the medial orbitofrontal cortex and the anterior cingulate cortex brain areas. This discovery helps to show how this group of links in part of the medial orbitofrontal cortex reward system plays a role in sensation-seeking.
Second, the relationship between risk-taking and worrier/anxious feeling was investigated in an advanced approach with over 30,000 participants from a massive open-access dataset, the UK Biobank (Rolls et al., 2022). Instead of performing the traditional correlation between behaviour scores, the association pattern of behaviour measures with functional connectivity was analysed. In this project, the significantly associated functional connectivity links between risk-taking and anxiety involved similar regions of the brain, including the medial orbitofrontal cortex, VMPFC, and the parahippocampal gyrus, but in the opposite direction (p<0.001, FDR corrected). This investigation revealed, to my knowledge for the first time, that risk-taking individuals were not normally worriers, and the medial orbitofrontal cortex, which is a key area of the reward system, was associated positively with risk-taking, and negatively with anxiety.
In addition, childhood traumatic events in relation to cognitive performance, various psychological disorders, including anxiety, depression and addiction in adulthood was investigated with the UK Biobank dataset. This investigation focused on
the long-lasting relationship of childhood trauma with other behaviours over 30 years later, while most childhood trauma studies are limited to childhood or early adolescence. Moreover, with the massive datasets of various behaviour measures available, the relationships were investigated with nine mental health measures and three cognitive measures in adults, and the associated patterns of functional connectivity. These findings highlight the long-lasting relationship between childhood traumatic events and a wide range of mental health problems and cognition in later life. They also provide insights into the neural mechanisms of the long-lasting relationship, including brain areas involved in executive function, emotion, face processing, and memory.
By exploring novel computational methods and modelling approaches to large-scale neuroimaging data, efficient models were developed in different cases. Moreover, significant progress has been made in understanding the brain mechanism of human behaviours and mental disorders, including impulsivity (sensation-seeking in Chapter 4, risk-taking in Chapter 6), verbal intelligence (Chapter 5), and childhood trauma with mental health problems and cognitive performances (Chapter 7).
Item Type: | Thesis (PhD) | ||||
---|---|---|---|---|---|
Subjects: | Q Science > QA Mathematics > QA76 Electronic computers. Computer science. Computer software Q Science > QC Physics Q Science > QP Physiology R Medicine > RC Internal medicine R Medicine > RC Internal medicine > RC0321 Neuroscience. Biological psychiatry. Neuropsychiatry |
||||
Library of Congress Subject Headings (LCSH): | Brain -- Imaging -- Data processing, Big data, Emotions -- Physiological aspects, Machine learning, Support vector machines, Magnetic resonance imaging | ||||
Official Date: | October 2021 | ||||
Dates: |
|
||||
Institution: | University of Warwick | ||||
Theses Department: | Department of Computer Science | ||||
Thesis Type: | PhD | ||||
Publication Status: | Unpublished | ||||
Supervisor(s)/Advisor: | Feng, Jianfeng ; Thomson Rolls, Edmund | ||||
Format of File: | |||||
Extent: | 13 unnumbered leaves, 153 leaves : illustrations | ||||
Language: | eng |
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |
Downloads
Downloads per month over past year