Abstract:In this chapter we describe new neural-network techniques developed for visual mining clinical electroencephalograms (EEGs), the weak electrical potentials invoked by brain activity. These techniques exploit fruitful ideas of Group Method of Data Handling (GMDH). Section 2 briefly describes the standard neural-network techniques which are able to learn well-suited classification modes from data presented by relevant features. Section 3 introduces an evolving cascade neural network technique which adds new input nodes as well as new neurons to the network while the training error decreases. This algorithm is applied to recognize artifacts in the clinical EEGs. Section 4 presents the GMDH-type polynomial networks learnt from data. We applied this technique to distinguish the EEGs recorded from an Alzheimer and a healthy patient as well as recognize EEG artifacts. Section 5 describes the new neural-network technique developed to induce multi-class concepts from data. We used this technique for inducing a 16-class concept from the large-scale clinical EEG data. Finally we discuss perspectives of applying the neural-network techniques to clinical EEGs.
Abstract:The neural networks have trained on incomplete sets that a doctor could collect. Trained neural networks have correctly classified all the presented instances. The number of intervals entered for encoding the quantitative variables is equal two. The number of features as well as the number of neurons and layers in trained neural networks was minimal. Trained neural networks are adequately represented as a set of logical formulas that more comprehensible and easy-to-understand. These formulas are as the syndrome-complexes, which may be easily tabulated and represented as a diagnostic table that the doctors usually use. Decision rules provide the evaluations of their confidence in which interested a doctor. Conducted clinical researches have shown that iagnostic decisions produced by symbolic rules have coincided with the doctor's conclusions.