An Auto-learning System for the Classification of Fetal Heart Rate Decelerative Patterns

The classification of decelerations of the Fetal Heart Rate signal is a difficult and crucial task in order to diagnose the fetal state. For this reason the development of an automatic classifier would be desirable. However, the low incidence of these patterns makes it difficult. In this work, we present a solution to this problem: an auto-learning system, that combines self-organizing artificial neural networks and a rule-based approach, able to incorporate automatically to its knowledge each new pattern detected during its clinical daily use.