Biomedical Signal Processing / Medical Signal Processing / Biosignal Processing
Aref Einizade; Sepide Hajipour Sardouie
Volume 14, Issue 3 , October 2020, , Pages 221-233
Abstract
The brain electrical signal has been widely used in clinical and academic research, due to its ease of recording, non-invasiveness, and precision. One of the applications can be emotion recognition from the brain's electrical signal. Generally, two types of parameters (Valence and Arousal) are used to ...
Read More
The brain electrical signal has been widely used in clinical and academic research, due to its ease of recording, non-invasiveness, and precision. One of the applications can be emotion recognition from the brain's electrical signal. Generally, two types of parameters (Valence and Arousal) are used to determine the type of emotion which in turn indicate "positive or negative" and "level of extroversion or excitement" for a specific emotion. The significance of emotion is determined by the effects of this phenomenon on daily tasks, especially in cases where the person is confronted with activities that require careful attention and concentration. In the emotion recognition problem, firstly, using proper emotion stimuli, different emotions are created for the subjects under study and the brain signals corresponding to each stimulus are recorded. The two main steps for solving the emotion recognition problem are extracting suitable features and using appropriate classification or regression methods. In previous studies, different visual and auditory have been used and various linear and nonlinear features and classifiers have been investigated. In this paper, the main goal was the improvement of linear regression algorithms to estimate the criteria for recognizing human emotions more efficiently. For this purpose we proposed a new algorithm that uses the sparseness of the mixing vector along with the linear regression cost function. The effectiveness of the proposed algorithm on simulated data has been investigated and its superiority to linear regression algorithms such as PLS, LASSO, SOPLS and Ridge was shown. Also, to apply the proposed algorithm on EEG data corresponding to emotion recognition, the DEAP dataset was used and the AR coefficients were extracted from the EEG signals. The results obtained from the proposed algorithm were compared with those of the other linear regression algorithms, which in total showed the relative superiority of the proposed method.
Neuro-Muscular Engineering
Tahmineh Sadati; Mohammad Reza Daliri
Volume 12, Issue 1 , June 2018, , Pages 1-10
Abstract
A brain-computer interface is a system which works based on the neural activity created by the brain and it has attracted the attention of many researchers in recent years. These interfaces are independent of the usual pathway of peripheral and muscular nerves and are very important because of their ...
Read More
A brain-computer interface is a system which works based on the neural activity created by the brain and it has attracted the attention of many researchers in recent years. These interfaces are independent of the usual pathway of peripheral and muscular nerves and are very important because of their ability to provide a new dimension in communication or control of a device for the disabled persons. The neural activity used in the brain-computer interface can be recorded by various invasive and non-invasive methods and can be converted to the desired signal by different decoding algorithms. In this study, 3 rats were used to perform a movement task which was pressing a key and receiving a drop of water by a mechanical arm for corrected trials. By implanting a 16-channel microelectrode array in the rat's motor cortex during an invasive process, the brain signals are recorded during the task, and simultaneously the signal received by the force sensor is also stored. By performing the necessary preprocessing on spikes and extracting the firing rates of signal as a feature vector by convolving a Gaussian window with the spike trains, the necessary inputs for the decoding algorithm, which is linear regression here, are obtained. Two patterns have been used for cross validation. The first pattern considers 60% of the data from the beginning of the signal as a train set and the remaining 40% of the signal as a test set and the second pattern is the opposite of the first one. Several methods have been used to evaluate the decoding algorithm used in the studies. In this paper, the correlation coefficient and coefficient of determination have been used. The correlation coefficient and coefficient of determination between the desired force and predicted force in linear resgression method, in average of three sessions for three rats, are equal to r=0.56 and =0.20 for the first pattern and r=0.55 and =0.30 for the second pattern respectively. These results show that firing rates of neurons are proper features to predict continous variables such as force. Besides, it can be concluded that linear regression is a suitable method for decoding a motor variable like force and follows the desired signal properly.