Approaches to Improving the Visualization Accuracy of the Technology of Integrating Functional Near-infrared Spectroscopy with the Brain-computer Interface

The article provides a comparative analysis of existing approaches aimed at improving the accuracy of visualization using functional near-infrared spectroscopy when integrated with the brain-computer interface. Various aspects of functional near-infrared spectroscopy data are being investigated, including signal acquisition, preprocessing, and feature extraction. A comparative analysis of signal filtering technologies at the preprocessing stage and signal processing methods at the feature extraction stage is carried out. The features of the application of such methods as the discrete Fourier transform, the wavelet transform, and the HilbertHuang transform for data processing are presented.

I. Introduction

Currently, neurovisualization techniques play an important role in the study of brain functions. They can show which specific areas of the brain are involved in various cognitive and behavioral processes, allow for a differential study of “thin” brain structures, establish patterns of structural rearrangements of the brain, and quantify metabolism and blood flow.