Adaptive base line correction of EEGs using warped polynomials

This paper presents an adaptive method for removing base line wander due to eye movements in electroencephalograms (EEGs). The EEG is adaptively low-pass filtered to extract the base line, which is then subtracted. The adaptive filter is new and is called the time-warped polynomial filter (TWPF). The TWPF has some important advantages over existing adaptive techniques for filtering biomedical signals: it reacts immediately to changes in the signal´s properties, independently of the relative amplitudes of the EEG and its baseline; it does not require a reference signal and can be applied to non-periodical signals. An example illustrates the superiority of TWPF for base line correction over traditional procedures