Automated seizure detection
AssigneeMindsigns Health Inc.
InventorsJustin DAUWELS, Yuanyuan YAO, Wei Yan PEH
Life of this patent
Estimating the term…
Abstract
A robust seizure detector to detect seizures in a non-patient-specific matter without fine-tuning (if any) is provided. A multi-modular cascading deep and machine learning-based pipeline to detect seizures from channel- to segment- to EEG-level is utilized. For channel-level detection, one of three models is utilized: convolution neural network (CNN), CNN with belief matching (BM) framework (CNN+BM), and Transformer-CNN with BM (CNN+Transformer+BM). Regional features are extracted from the channel-level detector output and a machine learning module for the segment-level detector is applied. For the EEG-level detector, segment-level output is utilized and a convolutional-based postprocessing module is used for seizure detection.
Claims20
An automated method of seizure detection using a method selected from scalp electroencephalogram (EEG), intracranial EEG (iEEG) and a combination thereof, the method comprising:
- (a) applying a transformer with a convolutional neural network (CNN) and a belief matching (BM) loss to acquire subject data from at least one single-channel segment (channel-level), acquiring channel-level data;
- (b) extracting from the channel-level data at least two regional features, thereby acquiring data from multi-channel segments (segment level);
- (c) from the multi-channel segment data, predicting seizure probability in the multi-channel segments;
- (d) repeating (c) at least once, acquiring successive segment-level outputs;
- and
- (e) applying at least one convolutional-based postprocessing module on the successive segment-level outputs, thereby detecting the seizure.
Dependencies are inferred from the claim text.