Time-varying EEG spectral power predicts evoked and spontaneous fMRI motor brain activity

2025-04-14

AI Summary

This study investigates whether time-varying spectral power from electroencephalography (EEG) can predict functional magnetic resonance imaging (fMRI) activity in the motor network during both task and resting states across days, and quantifies the shared neurophysiological information between these modalities. Method: We employed subject-specific sparse group Lasso modeling, integrated with personalized linear regression and cross-day generalization validation. Statistical significance was assessed via null-model permutation testing and benchmarking against sensorimotor rhythm baselines. Contribution/Results: For the first time, we achieved statistically significant, cross-day, single-subject fMRI signal prediction for both task and resting states. Most participants showed robust prediction—particularly during task conditions—enabling successful decoding of cross-modal coupling in spatial topography, spectral preference (e.g., α/β bands), and hemodynamic delay profiles. These findings establish an interpretable, EEG-based modeling framework with empirical support for clinical translation in neurofeedback applications.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Multimodal Learning
• Humans and AI: Brain-Sensing and Analysis

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Simultaneous EEG-fMRI recordings are increasingly used to investigate brain activity by leveraging the complementary high spatial and high temporal resolution of fMRI and EEG signals respectively. It remains unclear, however, to what degree these two imaging modalities capture shared information about neural activity. Here, we investigate whether it is possible to predict both task-evoked and spontaneous fMRI signals of motor brain networks from EEG time-varying spectral power using interpretable models trained for individual subjects with Sparse Group Lasso regularization. Critically, we test the trained models on data acquired from each subject on a different day and obtain statistical validation by comparison with appropriate null models as well as the conventional EEG sensorimotor rhythm. We find significant prediction results in most subjects, although less frequently for resting-state compared to task-based conditions. Furthermore, we interpret the model learned parameters to understand representations of EEG-fMRI coupling in terms of predictive EEG channels, frequencies, and haemodynamic delays. In conclusion, our work provides evidence of the ability to predict fMRI motor brain activity from EEG recordings alone across different days, in both task-evoked and spontaneous conditions, with statistical significance in individual subjects. These results present great potential for translation to EEG neurofeedback applications.

Problem

• Predict fMRI motor activity from EEG spectral power
• Assess EEG-fMRI shared neural information via interpretable models
• Validate cross-day EEG-based fMRI prediction for neurofeedback

Innovation

• Predicts fMRI from EEG spectral power
• Uses Sparse Group Lasso regularization
• Validates models across different days