Research On The Voice Communication Nursing System Of Stroke Patients Based On Brain-computer Interface

There are so many stroke patients suffering from speech disorders,and some functional disorders caused by stroke influenced their quality of life seriously.Therefore,they strongly hope that the inconvenience caused by stroke can be improved,their quality of life can be improved and the impact of the disease can be reduced.Brain computer interface(BCI)technology that establishes communication between humans and external devices through EEG offers hope to those who are disabled by peripheral nerve or muscle damage.This technology not only enables those patients with motor dysfunction to regain the ability to exercise or manipulate foreign objects,but also enables patients with speech dysfunction to regain the ability to communicate with others.Therefore,the above technology can be used to improve the communication status of stroke patients with speech dysfunction.First of all,this paper expounds the domestic and foreign research of BCI speller system.From the research progress of these technologies,it can be clearly found that the BCI technology based on Steady-state visual evoked potential(SSVEP)is becoming more and more mature,and it is gradually showing more and more advantages in the field of speller based on BCI.Therefore,this paper mainly studies the visual stimulation and signal analysis algorithms of SSVEP-BCI.Secondly,based on the related research on BCI speller system and SSVEP-BCI,this paper designs a voice communication nursing system based on brain-computer interface for stroke patients.By adding real-time detection of Alpha wave before SSVEP signal analysis in this system,an asynchronous BCI system is formed.The asynchronous working mode allows the patient to choose the time to start the system autonomously,which is a more humane and independent way to break the system’s constraints on patients.The input of pinyin and tone of the patient’s gaze can be achieved by using the filter bank canonical correlation analysis(FBCCA)algorithm to analyze the SSVEP signal.A voice output module is added to the traditional BCI character input system,and the conversion of Chinese pinyin and tone to Chinese voice is realized through the voice output module.The voice mode breaks the limitation on the distance between caregivers and patients,facilitates the care of patients by family members and caregivers,and also helps patients’ lives and recovery.Finally,the feasibility and effectiveness of the proposed method are verified by offline and online experiments.In the experiment,a total of 5 subjects participated in the experiment.The experimental results show that the subjects can use the Alpha wave to control the starting of the system,and the asynchronous performance accuracy reaches 100%.After the start of the system,pinyin or tone selection works by looking at the corresponding stimulus target on the visual stimulation interface,and the overall analysis accuracy of the system is more than 90%.