Estimating Reproducible Functional Networks Associated with Task Dynamics using Unsupervised LSTMs

TL;DR

This paper introduces an unsupervised LSTM-based method to estimate more reproducible and task-related functional networks from fMRI data, outperforming existing approaches in consistency and association strength.

Contribution

The study presents a novel unsupervised LSTM approach for extracting functional networks from fMRI data that are more strongly linked to task activity and more reproducible across subjects and datasets.

Findings

LSTM-derived networks show stronger association with task activity.

Networks are more reproducible across subjects and datasets.

Outperforms other decomposition methods in identifying task-related networks.

Abstract

We propose a method for estimating more reproducible functional networks that are more strongly associated with dynamic task activity by using recurrent neural networks with long short term memory (LSTMs). The LSTM model is trained in an unsupervised manner to learn to generate the functional magnetic resonance imaging (fMRI) time-series data in regions of interest. The learned functional networks can then be used for further analysis, e.g., correlation analysis to determine functional networks that are strongly associated with an fMRI task paradigm. We test our approach and compare to other methods for decomposing functional networks from fMRI activity on 2 related but separate datasets that employ a biological motion perception task. We demonstrate that the functional networks learned by the LSTM model are more strongly associated with the task activity and dynamics compared to other approaches. Furthermore, the patterns of network association are more closely replicated across subjects within the same dataset as well as across datasets. More reproducible functional networks are essential for better characterizing the neural correlates of a target task.