A preliminary cooperative genetic fuzzy proposal for epilepsy identification using wearable devices

Abstract

The epilepsy is one of the neurological disorders that affects people of all socioeconomic groups and ages. An incorrect treatment or a lack in monitoring might produce cognitive damage and depression. In previous work we presented a preliminary method for learning a generalized model to identify epilepsy episodes using 3DACC wearable devices placed on the dominant wrist of the subject. The model was based on a Fuzzy Finite State Machines to detect the epilepsy episodes in 3DACC time series. The learning model applied was a classical Genetic Fuzzy Finite State Machine. The goal of the present work is to adapt the previous learning scheme to a Cooperative Coevolutionary Genetic Fuzzy Finite State Machine to improve the classification results. The obtained results show that a Cooperative proposal outperform moderately the results of the original proposal.