Introduction to Brain Computer Interface Seminar Topic:
This paper discussed about how to measure brain function and importance of different neuroimaging modalities. As per researches, the best tool to measure brain function is EEG (electroencephalographic). In this tool, by using both invasive and non-invasive real time techniques, activities from different parts of the brain can be measured. These EEG signals will get applied to signal processing unit for further processing.
Overview:
The main task involved in this tool is the classification of EEG patterns with respective to its frequency and amplitude in different consciousness states. EEG signals can be obtained from brain by both invasive and non-invasive techniques. In Invasive technique, EEG signals obtained either with the help of needle electrode or by a chip which implanted in the brain. In non-invasive technique, user needs to wear an electrode cap which records EEG activity from the brain and identify brain waves.
There is no need of any surgery or implantation in this technique. But in this technique, there are some limitations exist while recording from the scalp using electrode cap. Multi-Layer Perceptron Neural Network (MLP-NN) based method is one of the simpler and most efficient method under EEG classification. EEG signals are used to control the output devices and this control can be achieved either by using tracking system or software technique.
Conclusions:
In BCI, the machine should learn how to differentiate brain activity patterns and the user should learn how to perform different tasks in order to yield distinct brain signals. In real-time there is a need to execute few tasks by the user for an effective working of BCI and the tasks are like Parameter extraction, pattern recognition and classification and generation of neurofeedback. EEG patterns characteristic changes can be produced by performing few mental tasks like mental cube rotation or attention versus relaxation. Initial training, offline analysis and classifier generation, training with neuro-feedback and classifier update are the four steps involved in the whole training process of BCI.
Download Brain Computer Interface Seminar Topic with Report for CSE Students .