Wireless

Brain sensing (EEG)

• the intracranial EEG (iEEG) sensed by electrodes implanted by surgery inside the brain near the epileptic zone (focal epilepsy),
• scalp EEG (sEEG) sensed by non-invasive electrodes fixed to scalp in appropriate zones.

During the last decade several methods have been proposed for seizure prediction based on EEG signals of two types the intracranial EEG (iEEG) sensed by electrodes implanted by surgery inside the brain near the epileptic zone (focal epilepsy), scalp EEG (sEEG) sensed by non-invasive electrodes fixed to scalp in appropriate zones.

The intracranial iEEG is much richer from the information point of view but it needs surgery to implant electrodes in appropriate regions.

The scalp sEEG has a poorer signal, with a lower signal to noise ratio and much more subjected to other influences like eye and muscles movements, poor electrical contacts, etc. But it is much more comfortable for the patients.

However the rapid evolution of nanotechnology, tissue engineering and organic electronics allow to predict that in some years it will be possible to implant microelectrodes without significant disturbance of the patient.

In both situations the important is to develop algorithms for real time signal and information processing well appropriate for seizure prediction, and that is why in the project the two types of EEG will be considered, profiting from the experience and knowledge of the partners. The project will research further for real-time (iterative) new algorithms: in a first phase based on intracranial (invasive) data and in a second phase in scalp (non-invasive) data using a high sampling rate EEG system to be developed by the project.

The big challenge is to discover information in these signals sufficient to predict, with useful advance, the approaching seizure, to do it in real time, with a low number of failures (high sensitivity) and a low number of false alarms (high specificity).

Wireless communication, comfort for the patient

Actually in hospitals epileptic patients are connected by a high number of wires and cables to data acquisition and analysis systems (principally Electro Encephalogram, EEG), which obliges them to remain in bed. The coming wireless technologies allow to build high speed high sampling rates EEG acquisition devices transportable by the patient, allowing them to move freely around in clinic environments under a network of radio nodes for Bluetooth roaming.

The seizure predictor will be embedded in an evolving software/hardware platform with wireless communication capabilities, using epilepsy knowledge and latest informatics technologies platform – the Brainatics (brain+informatics)- that will have the potential to be used in ambulatory if tailored to individual patients. It will be based on open architectures and open software when possible.

The ambulatory device will be used in patients with an identified type of epilepsy, previously studied in hospital with the prototype devices. It is previewed that in such cases a limited number of electrodes will be needed (for scalp EEG), lowering the computation requirements and allowing the use of a high end notebook as the computational device.

• to develop the Brainatics, a transportable device to follow the dynamical evolution of the brain state and alarming when an epileptic seizure is predicted, based on multisignal, high sampling rates, wireless transmission and a notebook. The study will be conducted firstly with prototype devices in hospital environment and after in ambulatory under clinical supervision.
• to build an European Database on Epilepsy containing multimodal multidimensional annotated data from high number of patients using invasive and non-invasive EEG signals.

The main technological objectives of EPILEPSIAE are to develop the Brainatics, a transportable device to follow the dynamical evolution of the brain state and alarming when an epileptic seizure is predicted, based on multisignal, high sampling rates, wireless transmission and a notebook. The study will be conducted firstly with prototype devices in hospital environment and after in ambulatory under clinical supervision. to build an European Database on Epilepsy containing multimodal multidimensional annotated data from high number of patients using invasive and non-invasive EEG signals.