DeepJIVE: Learning Joint and Individual Variation Explained from Multimodal Data Using Deep Learning

TL;DR

DeepJIVE is a deep learning method that effectively uncovers shared and unique structures in high-dimensional multimodal data, including nonlinear patterns, demonstrated on synthetic, real-world datasets, and Alzheimer's disease imaging data.

Contribution

It introduces a novel deep learning framework for joint and individual variation analysis that overcomes limitations of traditional methods, handling high-dimensional and nonlinear data structures.

Findings

DeepJIVE successfully uncovers joint and individual variations in multimodal datasets.

It identifies biologically plausible covariation patterns in Alzheimer's disease imaging data.

The method outperforms traditional approaches in handling high-dimensional, nonlinear data.

Abstract

Conventional multimodal data integration methods provide a comprehensive assessment of the shared or unique structure within each individual data type but suffer from several limitations such as the inability to handle high-dimensional data and identify nonlinear structures. In this paper, we introduce DeepJIVE, a deep-learning approach to performing Joint and Individual Variance Explained (JIVE). We perform mathematical derivation and experimental validations using both synthetic and real-world 1D, 2D, and 3D datasets. Different strategies of achieving the identity and orthogonality constraints for DeepJIVE were explored, resulting in three viable loss functions. We found that DeepJIVE can successfully uncover joint and individual variations of multimodal datasets. Our application of DeepJIVE to the Alzheimer's Disease Neuroimaging Initiative (ADNI) also identified biologically plausible covariation patterns between the amyloid positron emission tomography (PET) and magnetic resonance (MR) images. In conclusion, the proposed DeepJIVE can be a useful tool for multimodal data analysis.