ESTformer: Transformer Utilizing Spatiotemporal Dependencies for EEG Super-resolution

• URL: http://arxiv.org/abs/2312.10052v1
• Date: Sun, 3 Dec 2023 12:26:32 GMT
• Title: ESTformer: Transformer Utilizing Spatiotemporal Dependencies for EEG Super-resolution
• Authors: Dongdong Li, Zhongliang Zeng, Zhe Wang, Hai Yang
• Abstract summary: ESTformer is an EEG framework utilizingtemporal dependencies based on the Transformer. The ESTformer applies positional encoding methods and the Multi-head Self-attention mechanism to the space and time dimensions.
• Score: 14.2426667945505
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Towards practical applications of Electroencephalography (EEG) data, lightweight acquisition devices, equipped with a few electrodes, result in a predicament where analysis methods can only leverage EEG data with extremely low spatial resolution. Recent methods mainly focus on using mathematical interpolation methods and Convolutional Neural Networks for EEG super-resolution (SR), but they suffer from high computation costs, extra bias, and few insights in spatiotemporal dependency modeling. To this end, we propose the ESTformer, an EEG SR framework utilizing spatiotemporal dependencies based on the Transformer. The ESTformer applies positional encoding methods and the Multi-head Self-attention mechanism to the space and time dimensions, which can learn spatial structural information and temporal functional variation. The ESTformer, with the fixed masking strategy, adopts a mask token to up-sample the low-resolution (LR) EEG data in case of disturbance from mathematical interpolation methods. On this basis, we design various Transformer blocks to construct the Spatial Interpolation Module (SIM) and the Temporal Reconstruction Module (TRM). Finally, the ESTformer cascades the SIM and the TRM to capture and model spatiotemporal dependencies for EEG SR with fidelity. Extensive experimental results on two EEG datasets show the effectiveness of the ESTformer against previous state-of-the-art methods and verify the superiority of the SR data to the LR data in EEG-based downstream tasks of person identification and emotion recognition. The proposed ESTformer demonstrates the versatility of the Transformer for EEG SR tasks.

Related papers

• DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs [70.91804882618243]
  This paper proposes DSMoE, a novel approach that achieves sparsification by partitioning pre-trained FFN layers into computational blocks. We implement adaptive expert routing using sigmoid activation and straight-through estimators, enabling tokens to flexibly access different aspects of model knowledge. Experiments on LLaMA models demonstrate that under equivalent computational constraints, DSMoE achieves superior performance compared to existing pruning and MoE approaches.
  arXiv  Detail & Related papers  (2025-02-18T02:37:26Z)
• Estimation of System Parameters Including Repeated Cross-Sectional Data through Emulator-Informed Deep Generative Model [5.3060535072023844]
  In politics, economics, and biology, available data are often independently collected at distinct time points from different subjects. Conventional optimization techniques struggle to accurately estimate DE parameters when RCS data exhibit various heterogeneities. We propose a new estimation method called the emulator-informed deep-generative model (EIDGM) EIDGM integrates a physics-informed neural network-based emulator that immediately generates DE solutions and a Wasserstein generative adversarial network-based parameter generator.
  arXiv  Detail & Related papers  (2024-12-27T08:19:23Z)
• ART: Artifact Removal Transformer for Reconstructing Noise-Free Multichannel Electroencephalographic Signals [0.10499611180329801]
  Artifact removal in electroencephalography (EEG) significantly impacts neuroscientific analysis and brain-computer interface (BCI) performance. This study presents an innovative EEG denoising model employing transformer architecture to adeptly capture the transient millisecond-scale dynamics characteristic of EEG signals. Our evaluations confirm that ART surpasses other deep-learning-based artifact removal methods, setting a new benchmark in EEG signal processing.
  arXiv  Detail & Related papers  (2024-09-11T15:05:40Z)
• EEG-SSM: Leveraging State-Space Model for Dementia Detection [21.67998806043568]
  State-space models (SSMs) have garnered attention for effectively processing long data sequences. This study introduces EEG-SSM, a novel state-space model-based approach for dementia classification using EEG data.
  arXiv  Detail & Related papers  (2024-07-25T06:20:03Z)
• Geodesic Optimization for Predictive Shift Adaptation on EEG data [53.58711912565724]
  Domain adaptation methods struggle when distribution shifts occur simultaneously in $X$ and $y$. This paper proposes a novel method termed Geodesic Optimization for Predictive Shift Adaptation (GOPSA) to address test-time multi-source DA. GOPSA has the potential to combine the advantages of mixed-effects modeling with machine learning for biomedical applications of EEG.
  arXiv  Detail & Related papers  (2024-07-04T12:15:42Z)
• Generative AI Enables EEG Super-Resolution via Spatio-Temporal Adaptive Diffusion Learning [14.96787832363301]
  HD devices improve the spatial resolution of the EEG by placing more electrodes on the scalp. This technique faces challenges such as high acquisition costs and limited usage scenarios. In this paper, adaptive diffusion models (STAD) are proposed to pioneer use of diffusion models for achieving spatial SR reconstruction.
  arXiv  Detail & Related papers  (2024-07-03T13:26:31Z)
• Physics-informed and Unsupervised Riemannian Domain Adaptation for Machine Learning on Heterogeneous EEG Datasets [53.367212596352324]
  We propose an unsupervised approach leveraging EEG signal physics. We map EEG channels to fixed positions using field, source-free domain adaptation. Our method demonstrates robust performance in brain-computer interface (BCI) tasks and potential biomarker applications.
  arXiv  Detail & Related papers  (2024-03-07T16:17:33Z)
• vEEGNet: learning latent representations to reconstruct EEG raw data via variational autoencoders [3.031375888004876]
  We propose vEEGNet, a DL architecture with two modules, i.e., an unsupervised module based on variational autoencoders to extract a latent representation of the data, and a supervised module based on a feed-forward neural network to classify different movements. We show state-of-the-art classification performance, and the ability to reconstruct both low-frequency and middle-range components of the raw EEG.
  arXiv  Detail & Related papers  (2023-11-16T19:24:40Z)
• Convolutional Monge Mapping Normalization for learning on sleep data [63.22081662149488]
  We propose a new method called Convolutional Monge Mapping Normalization (CMMN) CMMN consists in filtering the signals in order to adapt their power spectrum density (PSD) to a Wasserstein barycenter estimated on training data. Numerical experiments on sleep EEG data show that CMMN leads to significant and consistent performance gains independent from the neural network architecture.
  arXiv  Detail & Related papers  (2023-05-30T08:24:01Z)
• Dynamic Mode Decomposition in Adaptive Mesh Refinement and Coarsening Simulations [58.720142291102135]
  Dynamic Mode Decomposition (DMD) is a powerful data-driven method used to extract coherent schemes. This paper proposes a strategy to enable DMD to extract from observations with different mesh topologies and dimensions.
  arXiv  Detail & Related papers  (2021-04-28T22:14:25Z)
• Reinforcement Learning for Adaptive Mesh Refinement [63.7867809197671]
  We propose a novel formulation of AMR as a Markov decision process and apply deep reinforcement learning to train refinement policies directly from simulation. The model sizes of these policy architectures are independent of the mesh size and hence scale to arbitrarily large and complex simulations.
  arXiv  Detail & Related papers  (2021-03-01T22:55:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.