Other methods
Short descriptions of other methods used for comparison.
Methods used for comparison
The results on healthy individuals based on our proposed techniques were compared to the results based on the following methods:
M1
Method 1 shows the results based on a tangent space-based transfer learning technique by Gaur et al [1].
M2
Method 2 shows the results obtained from the evaluation session from a study which uses a subject-specific multivariate empirical mode decomposition-based filtering method (SS-MEMDBF) for preprocessing and implements a Riemannian geometry (RG) framework designed separately for each subject for classification[2]. It is also interesting to note that the implementation of M2 involves an overhead of an SS-MEMDBF for preprocessing, which increases the computational complexity and makes it challenging to perform online.
M3
Method 3 implements CSP and uses detection of the covariate shift and adaptive learning [3].
M4
Method 4 uses CSP on bandpass filtered EEG between 8 and 30 Hz before calculating the log variance from three pairs of filters for extraction of features and uses LDA for a binary classification problem [4].
P. Gaur, K. McCreadie, R. B. Pachori, H. Wang, G. Prasad, "Tangent space features-based transfer learning classification model for two-class motor imagery brain–computer interface," in International journal of neural systems, vol. 29, no. 10, Dec. 2019, Art. no. 1950025, doi: 10.1142/S0129065719500254. ↩︎
P. Gaur, R. B. Pachori, H. Wang, G. Prasad, "A multi-class EEG-based BCI classification using multivariate empirical mode decomposition based filtering and Riemannian geometry," in Expert Systems with Applications, vol. 95, pp. 201-211, 2018, doi: 10.1016/j.eswa.2017.11.007. ↩︎
H. Raza, H. Cecotti, Y. Li, and G. Prasad, "Adaptive learning with covariate shift-detection for motor imagery-based brain–computer interface", in Soft Computing, vol. 20, pp. 3085–3096, 2016, doi: 10.1007/s00500-015-1937-5. ↩︎
F. Lotte and C. Guan, "Regularizing Common Spatial Patterns to Improve BCI Designs: Unified Theory and New Algorithms," in IEEE Transactions on Biomedical Engineering, vol. 58, no. 2, pp. 355-362, Feb. 2011, doi: 10.1109/TBME.2010.2082539. ↩︎