Hybrid Brain-Computer Interface Instant Messaging System For Patients With Physical Disabilities And Language Impairments

Brain computer interface(BCI)is an information exchange channel established between the brain and the external environment,which can help people achieve "mind" control without the involvement of nerves and muscles.BCI plays an important role in the field of medical assistance.Patients with physical disabilities or other dysfunctions can communicate with the outside world through BCI,their living standards and self-care ability have been greatly improved.Hybrid BCI is a BCI that uses one electroencephalogram(EEG)signal and another(or more)EEG signals(or non-EEG signals)as the control signal source.Studies have shown that Hybrid BCI has better performance than single-mode BCI.Groups of different types of disabilities have different physical conditions,and a targeted design of the BCI system can help them in a better way.In order to help the disabled to achieve communication with the outside world better,this article uses P300,voice signals,and electrooculogram(EOG)signals to design two sets of instant messaging systems,through which users can complete chat and email communications.The main work of this paper is as follows:(1)A set of P300 and voice-based hybrid brain-computer interface instant messaging system has been designed for groups with only physical disabilities.Most patients with physical disabilities cannot directly control mobile phones and computers.BCI can help them solve this problem.However,BCI also has shortcomings,the efficiency of control only through brain signals is relatively low,so the voice signal is introduced to form a hybrid BCI to improve the control efficiency.The system interface is designed using the single-character stimulus paradigm,the speech recognition software development kit of i Flytek Corporation is used to recognize and process the speech signal,Bayesian Linear Discriminant Analysis(BLDA)is used to classify and recognize P300 signals.The experimental results of 6subjects show that the system has high recognition accuracy and low false trigger rate.At the same time,the load survey results also show that the system has a certain practicability.(2)A set of P300 and EOG-based hybrid brain-computer interface instant messaging system has been designed for groups with physical disabilities and language dysfunction.EOG is a kind of bioelectric signal with obvious characteristics and stable waveform,and most disabled patients have good eyesight and blinking ability,so the use of EOG can help groups with speech dysfunction effectively.This system uses the threshold judgment method to realize the waveform detection of a single blinking action,and adopts the decision-making idea of multiple rounds of waveform detection to solve the problem of unconscious blinking in EOG.Aiming at different subjects,EOG calibration experiments are designed to obtain threshold parameters that meet their own conditions.The comparative experiment verifies that the decision-making idea of multiple rounds of waveform detection can improving the accuracy of EOG detection and reducing the false touch rate effectively.The results of online experiments further prove the feasibility of the system.