Extraction of stationary components in biosignal discrimination

摘要: Biosignal recordings are widely used in the medical environment to support evaluation and diagnosis of pathologies. Nevertheless, main difficulty lies non-stationary behavior biosignals, difficulting obtention patterns characterizing changes physiological or pathological states. Thus, stationary components a biosignal is still an open issue. This work proposes methodology detect time-homogeneities based on time-frequency analysis with aim extract biosignal. Results show increase stationarity distance between classes reconstructions from enhanced representations. The extracted proposed approach can be solve classification problems.