Control of External Devices with Thoughts by Using BCI (Brain Computer Interface) for IOT-EEG Calibration System
A Brain-Computer Interface (BCI) is a device that translates neuronal information into commands capable of controlling external software or hardware such as a computer, a wheelchair, robotic arm. BCI is used to treat neurological disorders and thus helps in the restoration of sensory and motor functions.
BCIs are used with assisted living devices for individuals suffering from motor or sensory impairments. With BCI, a person with all kinds of disabilities but having a functioning brain can do many things that otherwise are impossible to do. With BCI, a person without performing any muscle movement can do various activities using thoughts in the brain. BCI interface is a direct communication pathway between a human or animal brain and an external device.
BCI devices are of two types: invasive and noninvasive. Here, we have used a noninvasive BCI headset device that uses Electroencephalography (EEG) to monitor and measure brain activity. In the market, devices are available that work on Functional Magnetic Resonance Imaging (fMRI) and Magnetoencephalography (MEG). All the existing devices lack an efficient way to train mental commands, maintain different human brain profiles, and a standard way to integrate with IoT devices.
We have done the work to address the above problem. BCI software for EEG headsets is developed, which trains mental commands with calibration and provides easy integration with IoT devices. The BCI software developed offers an interface which can accurately classify the signals of the brain using advanced techniques from machine learning and deep learning domains. The person is able to control external devices with his thoughts. The EEG headset is used to capture the brain waves of the person (signal acquisition phase) and then it is processed using a machine learning model and deep learning model to classify and interpret it. (in signal processing phase).
We have built a brain controlled computer-mouse that integrates with EEG headset. The person will be able to control the mouse clicks on the desktop with human thoughts.The EEG dataset is examined using a deep learning-based approach and machine learning-based approach. The comparison of the accuracy obtained with various algorithms is shown