Spatial Graph Signal Interpolation with an Application for Merging BCI Datasets with Various Dimensionalities

TL;DR

This paper introduces a spatial graph signal interpolation method to merge diverse BCI Motor Imagery datasets with different electrode configurations, enhancing data utilization for deep learning.

Contribution

The work presents a novel graph-based interpolation technique for electrodes, enabling effective merging of heterogeneous datasets for improved neural network training.

Findings

The proposed method outperforms spherical splines in interpolation accuracy.

It successfully homogenizes datasets with different electrode setups.

Experiments demonstrate improved generalization in neural network models.

Abstract

BCI Motor Imagery datasets usually are small and have different electrodes setups. When training a Deep Neural Network, one may want to capitalize on all these datasets to increase the amount of data available and hence obtain good generalization results. To this end, we introduce a spatial graph signal interpolation technique, that allows to interpolate efficiently multiple electrodes. We conduct a set of experiments with five BCI Motor Imagery datasets comparing the proposed interpolation with spherical splines interpolation. We believe that this work provides novel ideas on how to leverage graphs to interpolate electrodes and on how to homogenize multiple datasets.