Interference cancellation in biosignals using artificial intelligence techniques

Abstract

A great challenge in biomedical engineering is the noninvasive assessment of the physiological changes occurring inside the human body These variations can be measured by electrocardiogram fetal electrocardiogram electroencephalogram and electromyogram These signals are usually weak non stationary and are distorted by noise and interferences Noise combating presents one of the most challenging problems in biosignal processing basically due to the fact that a signal can pick up noise and be distorted such that the information carried by the signal can be misinterpreted Appropriate signal processing techniques are therefore essential in order to recover the required signal from the corrupted potential recordings Basic methods of signal analysis like amplification digitization filtering processing and storage can be applied to biological signals In addition to these common procedures sophisticated digital processing methods are quite common and can significantly improve the quality of the retrieved data newlineThe usual method of estimating a signal corrupted by noise is to pass the composite signal through a filter that tends to suppress the noise while leaving the signal relatively unchanged Filters used for this purpose can be fixed or adaptive The design of fixed filters must be based on prior knowledge of both the signal and the noise newline newline