Diffusion-based translation between unpaired spontaneous premature neonatal EEG and fetal MEG

2025-07-16

AI Summary

This study addresses the modality gap between preterm neonatal electroencephalography (EEG) and fetal magnetoencephalography (fMEG) to advance mechanistic understanding of prenatal brain development and early neurodevelopmental pathology. To tackle the challenging problem of unpaired cross-modal neural signal translation—where no aligned EEG–fMEG pairs exist—we propose a novel unpaired diffusion-based translation framework built upon a dual-diffusion bridging architecture, augmented with a numerical integration optimization strategy. This design significantly reduces computational overhead while enhancing time-frequency fidelity. Crucially, the method avoids the mode collapse inherent in generative adversarial networks (GANs): it achieves ~5% lower time-domain mean squared error (MSE) than state-of-the-art GAN-based approaches and near-perfect frequency-domain signal reconstruction. It establishes the first high-fidelity, interpretable cross-modal translation paradigm for fMEG–EEG mapping, enabling robust, quantitative fetal neurofunctional assessment.

Technology Category

Natural Language Processing: Language Grounding & Multi-modal NLPMachine Learning: Multimodal LearningComputer Vision: Multi-modal Vision

Application Category

Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web dataGraph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsSearch and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Background and objective: Brain activity in premature newborns has traditionally been studied using electroencephalography (EEG), leading to substantial advances in our understanding of early neural development. However, since brain development takes root at the fetal stage, a critical window of this process remains largely unknown. The only technique capable of recording neural activity in the intrauterine environment is fetal magnetoencephalography (fMEG), but this approach presents challenges in terms of data quality and scarcity. Using artificial intelligence, the present research aims to transfer the well-established knowledge from EEG studies to fMEG to improve understanding of prenatal brain development, laying the foundations for better detection and treatment of potential pathologies. Methods: We developed an unpaired diffusion translation method based on dual diffusion bridges, which notably includes numerical integration improvements to obtain more qualitative results at a lower computational cost. Models were trained on our unpaired dataset of bursts of spontaneous activity from 30 high-resolution premature newborns EEG recordings and 44 fMEG recordings. Results: We demonstrate that our method achieves significant improvement upon previous results obtained with Generative Adversarial Networks (GANs), by almost 5% on the mean squared error in the time domain, and completely eliminating the mode collapse problem in the frequency domain, thus achieving near-perfect signal fidelity. Conclusion: We set a new state of the art in the EEG-fMEG unpaired translation problem, as our developed tool completely paves the way for early brain activity analysis. Overall, we also believe that our method could be reused for other unpaired signal translation applications.

Problem

• Translate premature neonatal EEG to fetal MEG signals
• Improve understanding of prenatal brain development
• Overcome data quality and scarcity in fMEG

Innovation

• Unpaired diffusion translation method
• Dual diffusion bridges integration
• Improved signal fidelity and efficiency