Abstract
An overview of methods that use adaptive filters to cancel out the periodic interference appearing in biomedical signals is presented. Periodic interference frequently disturbs the measurement of small signals and causes problems. This issue is particularly critical when measuring bioelectric signals such as EEG and ECG signals using electrodes attached to the surface of the human body. Especially, in the case of electroretinogram (ERG), the periodic interference included in the signal presents a significant hindrance, because the electric potential of the response is minute and includes the fundamental frequency of the AC power source and its several harmonic frequencies. Although the commonly used notch-filter cancels such periodic interference, it also destroys the objective signal. Adaptive filters are thought to have the potential to cancel such interference with less damage to objective signals. However, we have pointed out that adaptive filters may also produce a “self-cancelling component.” Thus, the cancellation of periodic interference is still an open problem. Some specific solutions have been given so far. We review several practical applications using such new technologies.
