Chapter 1 - Introduction and Background

Abstract

Recently many achievements in the biomedical signal analysis and classification have been realized. Biomedical signals contain several information, which cannot be observed directly but hidden in the signal's structure. This hidden information can be converted into meaningful interpretations by utilizing different signal processing and machine learning techniques. Moreover, biomedical signals represent biological system properties to identify numerous pathological conditions. Therefore, the biomedical signal analysis using diverse signal processing and machine learning methods becomes a vital instrument to extract clinically significant information hidden in the signal. Biomedical signal analysis is used to develop automated diagnostic systems for decision support. Usually, biomedical signals are assessed manually or visually by expert physicians, but this might result in unreliable or inefficient diagnostic. Consequently, the aim of the automated biomedical signal analysis is to reduce the subjectivity of manual assessment of biomedical signals. The computer-aided biomedical signal analysis for assessing different signal characteristics help to make objective decision by improving accuracy. Additionally, in order to reduce subjectivity of assessment, biomedical signal processing is employed for the feature extraction and dimension reduction to support characterizing and understanding of the information exist in a biomedical signal.