Elif Derya Übeyli
TOBB Economics and Technology University, Department of Electrical and Electronics Engineering
Analysis of time-varying biomedical signals
Spectral analysis methods are used for extracting features from time-varying biomedical signals. Decision making is performed in two stages: feature extraction by spectral analysis methods and classification using the classifiers trained on the extracted features. The inputs of the classifiers (expert systems) composed of diverse or composite features are chosen according to the network structures. The entire process of methodologies developed for automated diagnosis can generally be subdivided into a number of disjoint processing modules: preprocessing, feature extraction/selection, and classification. The classification module is the final stage in automated diagnosis. It examines the input feature vector and based on its algorithmic nature, produces a suggestive hypothesis. Feature is a distinctive (sets it appart) or characteristic (its make-up) measurement, transform, structural component made on a segment of a pattern. Features are used to represent patterns with minimal loss of important information. The feature vector, which is comprised of the set of all features used to describe a pattern, is a reduced-dimensional representation of that pattern.
The classical, model-based, time-frequency analysis, eigenvector methods will be presented for feature extraction from the time-varying biomedical signals. The combined neural network, mixture of experts, modified mixture of experts trained on composite or diverse features will be defined.