(205) Feasibility of an ambulatory HD EEG system for home monitoring in Epilepsy Patients

Around 50 million people worldwide are affected by epilepsy, and there is need for early diagnosis, routine monitoring, and therapy. Non-invasive HD EEG (up to 256 channels) in epilepsy patients have been shown to be effective for source localization using advanced patient specific head models. To make it clinically feasible and easier for doctors, technicians and patients, there is a need for a wearable, ambulatory and in-home use HD-EEG system using an easy to apply HD-EEG sensor net.

Here, we show the feasibility of a HD-EEG system with the geodesic truncated icosahedron design of the sensor net connected to a mobile and wearable Explorer Quest HD EEG system from Ripple Neuromed.

Methods:

Two prototypes, (i) 16-channel and (ii) 256-channel HD-EEG sensor webs (Geodesic Head Web, GHW) were constructed using the truncated icosahedron design (a geodesic polyhedron, used in soccer ball) with a stretchable polyurethane material. This geometrical design gives a better approximation to the sphere and thus a better fit across different ethnic head shapes. The electrodes were made up of Ag/AgCl, which has been designed to hold sponges (for quick short-term recordings) and paste (for long term recordings).
The 16-channel GHW was connected to the wireless, wearable, Explorer Quest HD EEG system, with miniaturized analog to digital front ends as close to the sensors as possible (~10” from neck) to reduce the signal path and common mode noise. The data was sampled at 2 KHz. The 256-channel GHW was connected to the NetAmps 400 (Magstim-EGI) with the data acquired at 250 Hz. Few minutes of EEG data was collected (eyes open, eyes closed, eye blinks) from one healthy adult male subject from the two prototypes. The EEG sensor web was self-applied by the subject on his own head. To compare the data quality, EEG was also collected with the (iii) FDA cleared HydroCel Geodesic Sensor Net with NetAmps 400 in the same subject. The EEG data was band pass filtered from 0.5 Hz to 40 Hz. Before collecting EEG data, electrode to scalp impedances were measured in all sessions. Bad channels were ignored in the analysis.

Results: The truncated icosahedron design was a good fit to the head, especially the warping 3D design makes the electrodes more stable at the back of the head. Most of the prototype GHW electrode noise was less than 1 uV RMS. The electrode to scalp impedances in all the three sessions were approximately 50 K Ohms in most of the electrodes. The EEG data quality (eyes closed, eyes open and eye blinks) were good in all the three scenarios. Clear alpha waves were seen during the eyes closed condition.

Conclusions: We have shown here a feasibility of a low-noise, wireless, wearable HD-EEG system, which can be used for ambulatory and in-home use for epilepsy monitoring.

Funding: Please list any funding that was received in support of this abstract.: No funding.