Method for prediction of clinical response to VNS therapy in epileptic patients
| Authors | |
|---|---|
| Year of publication | 2020 |
| Type | Patent |
| MU Faculty or unit | |
| Publisher | European Patent Office |
| State | Germany |
| Patent's number | EP3437692B1 |
| Web | odkaz na stránku Evropského patentového úřadu |
| Description | The present invention provides a method for determining of clinical response of a patient suffering from epilepsy to chronic vagal nerve stimulation (VNS) therapy, which comprises the steps of: - obtaining theta, alpha, beta and gamma frequency bands for at least one, preferably at least 10 scalp electrodes from scalp EEG data obtained from the patient using an EEG recording protocol, said EEG data comprising at least one rest interval, preferably at the beginning of the EEG data, at least one open eyes/close eyes interval, at least one photic stimulation interval, at least one hyperventilation interval, and optionally at least one additional open eyes/close eyes interval and optionally at least one rest interval having the duration of at least 30 s at the end of the EEG data, wherein one interval is selected as a baseline interval, - obtaining absolute mean powers as mean values of passband power envelope inside at least one discriminative interval and inside the baseline interval, and obtaining the relative mean power of the at least one discriminative interval as the ratio of the absolute mean power of said discriminative interval relative to the absolute mean power of the baseline interval, - determining from the relative mean powers of the at least one discriminative interval for at least one discriminative electrode whether the patient is a responder or a non-responder to vagal nerve stimulation therapy based on a responder pattern or a non-responder pattern, wherein the responder and non-responder patterns, and optionally the discriminative electrodes and/or the discriminative intervals, are determined by statistical analysis of EEG data recorded using the said EEG recording protocol for a group of known responders and non-responders. The method of the invention is the first method of prediction of VNS therapeutic outcome using pre-VNS EEG data. |