SLEEP CLASSIFICATION IN INFANTS BY DECISION TREE-BASED NEURAL NETWORKS

This paper presents an AI-based approach to automatic sleep stage scor ing. The system TBNN (Tree-Based Neural Network) uses a decision-tree generator to provide knowledge that defines the architecture of a back propagation neural network, including feature selection and initialisa tion of the weights. The case study reports a successful application t o the data from polygraphic all-night sleep of 8 babies aged 6 months. The teaching input was provided by a medical expert in accordance wit h the rules of Guilleminault and Souquet. The performance of TBNN is c ompared with 5 other methods and the results are discussed.