# SpaRG: Sparsely Reconstructed Graphs for Generalizable fMRI Analysis

arXiv: https://arxiv.org/abs/2410.07201 · HTML: https://mlsanity.com/paper/2410.07201/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 30 references that are in the map.
- First posted on arXiv 24 Sep 2024.

## Builds on (most cited first, 10 of 10)

- [BrainGB: A Benchmark for Brain Network Analysis with Graph Neural Networks](https://mlsanity.com/paper/2204.07054/) (2022)
- [FBNETGEN: Task-aware GNN-based fMRI Analysis via Functional Brain Network Generation](https://mlsanity.com/paper/2205.12465/) (2022)
- [Spatio-Temporal Graph Convolution for Resting-State fMRI Analysis](https://mlsanity.com/paper/2003.10613/) (2020)
- [Graph Neural Network for Interpreting Task-fMRI Biomarkers](https://mlsanity.com/paper/1907.01661/) (2019)
- [Fine-grain atlases of functional modes for fMRI analysis](https://mlsanity.com/paper/2003.05405/) (2020)
- [Self-Supervised Transformers for fMRI representation](https://mlsanity.com/paper/2112.05761/) (2021)
- [An Explainable Geometric-Weighted Graph Attention Network for Identifying Functional Networks Associated with Gait Impairment](https://mlsanity.com/paper/2307.13108/) (2023)
- [Sparse Overlapping Sets Lasso for Multitask Learning and its Application to fMRI Analysis](https://mlsanity.com/paper/1311.5422/) (2013)
- [Preserving Specificity in Federated Graph Learning for fMRI-based Neurological Disorder Identification](https://mlsanity.com/paper/2308.10302/) (2023)
- [Comparison of Matrix Norm Sparsification](https://mlsanity.com/paper/2201.12874/) (2022)

## Abstract (arXiv)

Deep learning can help uncover patterns in resting-state functional Magnetic Resonance Imaging (rs-fMRI) associated with psychiatric disorders and personal traits. Yet the problem of interpreting deep learning findings is rarely more evident than in fMRI analyses, as the data is sensitive to scanning effects and inherently difficult to visualize. We propose a simple approach to mitigate these challenges grounded on sparsification and self-supervision. Instead of extracting post-hoc feature attributions to uncover functional connections that are important to the target task, we identify a small subset of highly informative connections during training and occlude the rest. To this end, we jointly train a (1) sparse input mask, (2) variational autoencoder (VAE), and (3) downstream classifier in an end-to-end fashion. While we need a portion of labeled samples to train the classifier, we optimize the sparse mask and VAE with unlabeled data from additional acquisition sites, retaining only the input features that generalize well. We evaluate our method - Sparsely Reconstructed Graphs (SpaRG) - on the public ABIDE dataset for the task of sex classification, training with labeled cases from 18 sites and adapting the model to two additional out-of-distribution sites with a portion of unlabeled samples. For a relatively coarse parcellation (64 regions), SpaRG utilizes only 1% of the original connections while improving the classification accuracy across domains. Our code can be found at github.com/yanismiraoui/SpaRG.