An adaptive CSP filter to investigate user independence in a 3-class MI-BCI paradigm

• Published in: Computers in biology and medicine Vol. 103; pp. 24 - 33
• Main Authors: Costa, Ana.P., Møller, Jakob.S., Iversen, Helle.K., Puthusserypady, Sadasivan
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.12.2018; Elsevier Limited
• Subjects: Adaptation; Adaptive algorithms; Adaptive filters; Algorithms; Brain; Brain-computer interface (BCI); Calibration; Clinical trials; Common spatial patterns (CSP); Computer applications; Data processing; Diagonal loading (DL) CSP (DLCSP); Electroencephalography; Electroencephalography(EEG); Feedback; Human-computer interface; Imagery; Implants; Internal Medicine; Kalman filters; Mental task performance; Methods; Motor imagery (MI); Other; Recursive least squares (RLS); Recursive methods; Regularization; Rehabilitation; Stroke; Stroke rehabilitation; Motor imagery (MI); Recursive least squares (RLS); Stroke rehabilitation; Diagonal loading (DL) CSP (DLCSP); Common spatial patterns (CSP); Brain-computer interface (BCI); Electroencephalography(EEG)
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2018.09.021

This paper describes the implementation of a Brain Computer Interface (BCI) scheme using a common spatial patterns (CSP) filter in combination with a Recursive Least Squares (RLS) approach to iteratively update the coefficients of the CSP filter. The proposed adaptive CSP (ACSP) algorithm is made more robust by introducing regularization using Diagonal Loading (DL), and thus will be able to significantly reduce the length of training sessions when introducing new patients to the BCI system. The system is tested on a 4-class multi-limb motor imagery (MI) data set from the BCI competition IV (2a), and a more complex single limb 3-class MI dataset recorded in-house. The latter dataset is produced to mimic an upper limb rehabilitation session, e.g., after stroke. The findings indicate that when extensive calibration data is available, the ACSP performs comparably to the CSP (kappa value of 0.523 and 0.502, respectively, for the 4-class problem); for reduced calibration sessions, the ACSP significantly improved the performance of the system (up to 4-fold). The proposed paradigm proved feasible and the ACSP algorithm seems to enable a user or semi user independent scenario, where the need for long system calibration sessions without feedback is eliminated. •Adaptation in common spatial pattern filters reduces the need for long offline sessions.•A Recursive Least Squares approach is used for adaptation.•Diagonal Loading is used to stabilize the adaptation.•With abundant training data, this method performs as well as non-adaptive methods.•On small training data, it significantly outperforms non-adaptive methods.