Imaging-genetic study for human brain aging based on deep learning
Approved Research ID: 68143
Approval date: August 3rd 2021
The goal of our research project is to identify the complicated morphological patterns inherent in human brain images and to learn dynamic characteristics of morphological changes in aging. Developing an interpretable and explainable deep learning model that helps investigate the relation between genetic information and morphological and longitudinal features of a human brain over ages.
Recent studies have been supporting the importance of identifying the potential relation between genetic information and morphological features for better understanding of a human brain. Meanwhile, the normal aging-related morphological patterns could be informative to further investigate brain diseases and/or disorders. Leveraged by recent advances in AI, especially, deep-learning, it is a paramount problem to develop a deep-learning model that can be used to analyze complex nonlinear patterns inherent in human brain images and to understand their changes over time (i.e., aging).
We will develop a novel deep-learning framework for the following. First of all, to learn and understand aging-related morphological patterns and their changes over time. Secondly, to investigate the relation between genetic information and morphological patterns. Finally, to identify the genetic features that might be related to brain diseases or disorders, e.g., Alzheimer's disease, major depression disorder, autism spectrum disorder.