Application of Genetic Algorithm and Fuzzy Vector Quantization on EEG-based automatic sleep staging by using Hidden Markov Model

The Genetic Algorithm (GA) and Fuzzy Vector Quantization (FVQ) are combined in this paper to improve the performance of sleep staging. We use GA to train a codebook for Hidden Markov Model (HMM) and use FVQ to model HMM to improve the performance of the HMM. This paper adopts the sleep features of EEG based on 1968's R&K rules as well as the features used in other research for sleep staging. All the selected features are used to train HMM model and then are fed into the HMM model for recognition. In the previous researches, the modeling of HMM is independent of the special properties of the sleep stage transition. In this study, the HMM modeling is designed to meet the special properties of sleep stage transition. The experimental results in this paper show that the proposed method greatly enhances the recognition rate compared with those in other existing researches.