Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The paper introduces TABLeT, a method for modeling long-range fMRI dynamics by using a pre-trained 2D image autoencoder to tokenize 3D fMRI volumes, thus reducing dimensionality. This allows for efficient processing of long temporal sequences with a Transformer encoder. TABLeT outperforms existing voxel-based methods on UKB, HCP, and ADHD-200 datasets while significantly reducing computational and memory costs.
Compressing fMRI data with image autoencoders unlocks Transformer-based modeling of brain dynamics at scales previously impossible.
Modeling long-range spatiotemporal dynamics in functional Magnetic Resonance Imaging (fMRI) remains a key challenge due to the high dimensionality of the four-dimensional signals. Prior voxel-based models, although demonstrating excellent performance and interpretation capabilities, are constrained by prohibitive memory demands and thus can only capture limited temporal windows. To address this, we propose TABLeT (Two-dimensionally Autoencoded Brain Latent Transformer), a novel approach that tokenizes fMRI volumes using a pre-trained 2D natural image autoencoder. Each 3D fMRI volume is compressed into a compact set of continuous tokens, enabling long-sequence modeling with a simple Transformer encoder with limited VRAM. Across large-scale benchmarks including the UK-Biobank (UKB), Human Connectome Project (HCP), and ADHD-200 datasets, TABLeT outperforms existing models in multiple tasks, while demonstrating substantial gains in computational and memory efficiency over the state-of-the-art voxel-based method given the same input. Furthermore, we develop a self-supervised masked token modeling approach to pre-train TABLeT, which improves the model's performance for various downstream tasks. Our findings suggest a promising approach for scalable and interpretable spatiotemporal modeling of brain activity. Our code is available at https://github.com/beotborry/TABLeT.