*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Multimodal Learning Frameworks for Active Subspaces and Variation of Information Complexity in Neuroimaging Data
Committee:
Dr. Calhoun, Advisor
Dr. Dovrolis, Chair
Dr. Anderson
Abstract: The objective of the proposed research is to develop active subspace learning frameworks for neuroimaging data. Various supervised learning approaches focus on diagnostic prediction or feature importance at the level of brain regions, while unsupervised learning methods have been utilized to decompose neuroimaging data into lower dimensional components. However, most learning frameworks either do not consider the target assessment information while extracting brain subspaces, or can extract only high-dimensional importance associations as an ordered list of involved features, making manual interpretation at the level of subspaces difficult. This work presents a novel multimodal active subspace learning framework to understand various subspaces within the brain by performing a decomposition using feature importance to extract robust multimodal subspaces that define the most significant change in a given cognitive or biological trait. Through a rigorous cross-validation procedure on an Alzheimer's disease (AD) dataset, the proposed framework not only uncovers AD-related brain regions in the associated brain subspaces, but also enables automated identification of multiple collectively varying structural and functional sub-systems of the brain. Additionally, the utilization and studying of sub-domains structures in deep learning models for neuroimaging are explored. This work introduces a flexible deep learning framework to effectively incorporate multimodal features while accounting for and exploiting the heterogeneity in the sub-domains of the brain. Experiments with this framework demonstrate that the discriminatory information from structural and functional sub-domains can be better recovered and analyzed if the complexity of the subspace structure in the model can be tuned to reflect the extent of non-linearity with which each sub-domain encodes the information. The upcoming work involves utilizing these branched deep learning architectures for active subspace learning on neuroimaging data and developing ICA-based active subspace learning frameworks.
