Toward the Design of a Wearable System for the Remote Monitoring of Epileptic Crisis

Abstract

A primary or secondary compulsive epileptic crisis (EC) causes a sudden loss of consciousness. During the acute phase, the subject is completely unable to interact with the environment. Electroencepholographic (EEG) signals show a significant frequency spectrum centered at 6.5 Hz during an EC. Telemonitoring of EEG signals could be useful to remotely communicate the occurrence of an EC. The purpose of this work was to design and construct a system based on a frequency algorithm to detect an EC. In addition, the system must remotely communicate the occurrence of the EC and its geographical coordinates through the national global system for mobile communications (GSM) network. A wearable system for the detection of the EC and the communication has been proposed. The EEG signals are multiplexed, amplified, and low pass filtered at 20 Hz by means of the conditioning unit. As soon as an EC event is detected by a signal-processing unit based on the microP DSPic33fj12gp (Microchip Technolgoy, Inc., Chadler, AZ), a trigger is sent to a communication unit based on two separate GPS/GSM FT596K9 (Siemens, Berlin, Germany) units. The communication unit then sends an SMS message with an alarm and the coordinates of the subject. The final weight of the complete system developed using surface-montage-technology was equal to 3,760 mN, with a volume equal to 282 cm(3). Each component of the wearable system was designed separately and integrated. The system was tested in each component part and simulated in a closed loop. In particular, the system as a whole was simulated using 20 EEG rough traces (10 positive and 10 negative) in describing an EC. In both cases, the system did not show failures.