Approved Research

Exploring brain aging through deep learning and Mendelian randomization with imaging genomics

Lay summary

The aging population is an important challenge in today's society, followed by an increasing number of brain related degenerative diseases such as Parkinson's and Alzheimer's disease. This type of disease is closely related to aging in the human brain and brings enormous pain and economic burden to patients and their families. Brain magnetic resonance imaging is the most convenient non-invasive method for studying the brain. We adopt deep learning modeling to better explore the depth information contained in images and explore their potential as biological biomarkers. And we use deep learning attention or other visualization techniques to deeply explore the areas that play a crucial role in the aging process. At the same time, the newly developed imaging biomarkers of the brain were further included, and genetic data was combined to further explore the mysteries of physiological and mental dual "aging". This study is of great significance for exploring the mechanisms of neurodegenerative changes in the public interest, especially for exploring the correlation between parameters in relevant brain regions and the underlying mechanisms of aging. Mendelian randomization is a scientific statistical method that can help us determine the relationships between many things. Finally, the Mendelian randomization method introduces potential drug targets that can be explored for many brain aging related diseases diseases or mental disorders.

This study will last for 3 years. For our investigation, we will require the whole cohort of genetic, brain imaging, and clinical information accessible in UK Biobank. After its completion, this study will provide neuro-science professionals with strong innovative instruments as well as fresh perceptions of aging of the brain.