Skip to navigation Skip to main content Skip to footer

Understanding brain development with imaging genomics

Understanding brain development with imaging genomics

Principal Investigator: Professor Hualou Liang
Approved Research ID: 57854
Approval date: March 9th 2020

Lay summary

Major psychiatric illnesses are increasingly understood as disorders of brain development, which has led to large-scale imaging genetic studies that combine imaging and genetic data with clinical phenotyping. Together, such data have emphasized the promise of objective 'growth charts' of brain development. However, synergies across major efforts remains unrealized due to use of different clinical instruments, different scanning protocols, challenges in informatics, and difficulties in data integration. In this project, we will overcome these obstacles by leveraging recent advances in data science and machine learning, especially deep learning, to build imaging genomics-based growth charts of human brain development. First, we will perform association study to discover how brain imaging phenotypes provided by disparate imaging modalities, clinical and genetic factors affect ageing rate. Second, we will use the UK Biobank data to build imaging genomics-based biomarkers of an individual's brain health in relation to measures of cognitive function and psychopathology symptoms for clinical relevance. We have recently developed an innovative deep learning approach for supporting the use of neuroimaging-based 'brain age' as a biomarker of an individual's brain health. We will employ our deep learning model to predict chronological age as a proxy for brain age using data from the UK Biobank. We will interpret the predictive model by investigating the contribution of biomarkers to the prediction of physiological ageing, and compare these results to our early study. Finally, we will link developmental abnormalities within specific brain networks and changes in gene and epigenetic expression to dimensions of psychopathology. The tools developed for this study will be evaluated and refined through simulations and other valuable real datasets such as the human connectome project (HCP).

This study will last for 3 years. The full cohort of genetic, neuroimaging, and clinical data available in UK Biobank will be needed for our project. Upon completion, this project will provide powerful new tools for the neuroscience community, as well as novel insights regarding brain development and early psychosis-spectrum symptoms.