EEG recording and analysis

Inventors

Elwood, Michael K. • Frankel, Mitchell A. • Lehmkuhle, Mark J. • Wheeler, Jean M. • Lingstuyl, Robert • West, Erin M. • McGrath, Tyler D.

Assignees

Epitel Inc

Abstract

One embodiment provides a method, including: obtaining EEG data from one or more single channel EEG sensor worn by a user; classifying, using a processor, the EEG data as one of nominal and abnormal; and providing an indication associated with a classification of the EEG data. Other embodiments are described and claimed.

Core Innovation

The invention processes electroencephalogram (EEG) sensor data by receiving raw discrete bipolar sensor data from a plurality of discrete wireless EEG sensors positioned on a scalp of a patient. Each discrete wireless EEG sensor includes a housing that supports a first electrode and a second electrode forming a single bipolar channel, and each sensor generates the raw discrete bipolar sensor data independently of the other sensors. The processing then normalizes the raw discrete bipolar sensor data into normalized discrete bipolar EEG sensor data to account for inter-sensor differences in data amplitudes.

After normalization, the invention segments the normalized discrete bipolar EEG sensor data into a plurality of segments of discrete bipolar EEG sensor data. The invention detects a seizure event from the plurality of segments of discrete bipolar EEG sensor data, and seizure detection can be based on segment-level seizure probabilities and one or more features extracted from the plurality of segments.

In some implementations, the processing includes normalization operations such as noise removal or filtering on the raw discrete bipolar sensor data, including low-pass filtering to remove electromyographic (EMG) muscle activity. The invention can further normalize by rejecting electro-ocular artifact. System configurations can use a four-sensor arrangement with sensors positioned at defined locations around the scalp and can support sensor/reference configurations without a common reference electrode.

Claims Coverage

The document contains three independent claims (method, system, and non-transitory computer readable medium). Across these independent claims, the core coverage includes four main inventive features: receiving raw discrete bipolar sensor data from multiple discrete wireless EEG sensors, normalizing to account for inter-sensor amplitude differences, segmenting the normalized data, and detecting a seizure event from the segments, with claim-dependent refinements.

Collectively, the independent claims define a computer-implemented EEG seizure detection pipeline using multiple discrete wireless single bipolar channel sensors whose independently generated data is normalized for inter-sensor amplitude differences, segmented, and then used to detect a seizure event from the segments.

Stated Advantages

Documented Applications