چكيده به لاتين
During the last two decades, brain computer interface has received increasing attention. The reason being, that brain signals recorded with multiple electrodes pasted on scalp or implanted inside the brain, are regarded as a communication channel with high potential for the control of prosthetic devices.
In this study, the force variable is decoded by electrodes implanted in the rat's motor cortex during motion task. Simultaneously with the recording of brain signals by an electrode array, the real force of the rat's hand is recorded separately by a force sensor which is embedded on the moving arm.
The steps of this study include preprocessing such as applying different filters to reduce noise and prepare the signal for feature extracting phase.Then, the firing rates of the all neurons are extracted as feature vectors. Finally, two decoding algorithms including linear regression and one-layer perceptron neural network have been used and the efficiency of the methods has been evaluated and compared with the two metrics, coefficient of correlation and coefficient of determination. In this study, the coefficient of correlation and coefficient of determination between the desired force and predicted force which is recorded by the electrodes implanted in the motor cortex, in linear regression method, in average of three sessions for three rats, is equal to r=65.0 and 2 6 06 for the first pattern and r=65.. and 2 6 06 for the second pattern respectively. The results of these two metrics are calculated using the one-layer perceptron neural network method in average of three sessions for three rats are equal to r=6504 and 2 6 44 for the first pattern and r= 6506 and 2 6 40 for the second pattern respectively
