BCI (brain-computer interface) method for multi-modal signals

Abstract

The invention discloses a BCI (brain-computer interface) method for multi-modal signals. The BCI method for multi-modal signals comprises a calibration stage and an identification stage. At the calibration stage, synchronously collected EEG and near infrared optical brain signals are pretreated respectively, so as to obtain signals in three modes; characteristics of the signals in the three modes are extracted respectively; the characteristic vector is adopted to train a classifier 1, a classifier 2 and a classifier 3; then, output signals of the three trained classifiers are adopted to train a classifier 4; at the identification stage, synchronously collected EEG and near infrared optical brain signals are pretreated; characteristics of the synchronously collected EEG and near infrared optical brain signals are extracted; the characteristic vectors of the signals in the three modes are input to the classifier 1, the classifier 2 and the classifier 3 respectively; then, the classification results of the three classifiers are input to the classifier 4; lastly, the brain-computer interface for the multi-modal signals outputs results. The BCI method for the multi-modal signals has the advantages of improving the precision of the BCI for single-modal signals and effectively overcoming the illiteracy phenomenon of the BCI for single-modal signals.

Description

technical field [0001] The present invention relates to the technical field of brain-computer interface, in particular to a brain-computer interface method for multimodal signals. Background technique [0002] The brain-computer interface is a system that converts neurophysiological signals in the thinking process into control signals to control external machines without relying on the peripheral nervous system and muscles. Brain-computer interface technology has important application prospects in rehabilitation engineering, virtual reality, game entertainment, aerospace, military and other fields. Brain-computer interface technology is divided into invasive brain-computer interface technology and non-invasive brain-computer interface technology. The invasive brain-computer interface technology adopts the implanted electrode technology of the cerebral cortex, which has the advantages of high signal-to-noise ratio and precise control. However, invasive brain-computer interf...

AI Technical Summary

Problems solved by technology

There is currently no method for synchronously collecting EEG and other modal signals in the prefrontal cortex, motor cortex, and posterior parietal cortex to form a multimodal BCI

Images