Empowering Functional Neuroimaging: A Pre-trained Generative Framework for Unified Representation of Neural Signals
This addresses cost and fairness problems in brain-computer interface decoding, representing a new paradigm in the field.
The paper tackles the high cost and fairness issues in multimodal functional neuroimaging for brain-computer interfaces by proposing a generative AI framework that maps data into a unified representation space, enabling generation of data for constrained modalities and underrepresented groups, which improves downstream task performance and enhances model fairness.
Multimodal functional neuroimaging enables systematic analysis of brain mechanisms and provides discriminative representations for brain-computer interface (BCI) decoding. However, its acquisition is constrained by high costs and feasibility limitations. Moreover, underrepresentation of specific groups undermines fairness of BCI decoding model. To address these challenges, we propose a unified representation framework for multimodal functional neuroimaging via generative artificial intelligence (AI). By mapping multimodal functional neuroimaging into a unified representation space, the proposed framework is capable of generating data for acquisition-constrained modalities and underrepresented groups. Experiments show that the framework can generate data consistent with real brain activity patterns, provide insights into brain mechanisms, and improve performance on downstream tasks. More importantly, it can enhance model fairness by augmenting data for underrepresented groups. Overall, the framework offers a new paradigm for decreasing the cost of acquiring multimodal functional neuroimages and enhancing the fairness of BCI decoding models.