FAIR-ESI: Feature Adaptive Importance Refinement for Electrophysiological Source Imaging
This addresses the problem of precise ESI for diagnosing brain disorders, representing an incremental improvement over existing model-based and deep learning methods.
The paper tackles the challenge of accurate feature selection and refinement in electrophysiological source imaging (ESI) for brain disorder diagnosis by proposing FAIR-ESI, a framework that adaptively refines feature importance across spectral, temporal, and patch-wise views, with experiments on simulation and clinical datasets validating its efficacy.
An essential technique for diagnosing brain disorders is electrophysiological source imaging (ESI). While model-based optimization and deep learning methods have achieved promising results in this field, the accurate selection and refinement of features remains a central challenge for precise ESI. This paper proposes FAIR-ESI, a novel framework that adaptively refines feature importance across different views, including FFT-based spectral feature refinement, weighted temporal feature refinement, and self-attention-based patch-wise feature refinement. Extensive experiments on two simulation datasets with diverse configurations and two real-world clinical datasets validate our framework's efficacy, highlighting its potential to advance brain disorder diagnosis and offer new insights into brain function.