Fused electroencephalogram and machine learning for precognitive brain-computer interface for computer control
AssigneeX Development LLC
InventorsSarah Ann Laszlo, Philip Edwin Watson, Carl Ferman McCleary Smith, Aysja Johnson
Life of this patent
Estimating the term…
Abstract
A method for analyzing electroencephalogram (EEG) signals is disclosed. EEG signals are received from a sensor coupled to a user. Contextual information from one or both of the user and the user's environment is also received. The EEG signals are processed in real time using a machine learning model to predict an action of the user, which is associated with the contextual information. Output associated with the predicted action is then generated.
Claims20
A method for analyzing electroencephalogram (EEG) signals, comprising:
- receiving EEG signals from a sensor coupled to a user;
- receiving contextual information from one or both of the user and the user's environment;
- processing the EEG signals in real time using a machine learning model to predict a future action of the user, the machine learning model having a mapping function that maps a time series of values corresponding to an amplitude or change in amplitude of the EEG signals to an action and the contextual information;
- and
- generating an output associated with the predicted action.
A method for training a machine learning algorithm to predict a future action using an electroencephalogram (EEG) signal, comprising:
- receiving EEG signals from a sensor coupled to a user, the EEG signals being acquire over a period in which the user selects an action from a plurality of possible actions;
- receiving contextual information from one or both of the user and the user's environment;
- extracting at least one first segment from each of the EEG signals that precedes a moment when the user selects the action;
- and
- providing, to the machine learning algorithm, at least the first segments from each of the EEG signals, the selected action, and the contextual information to train the machine learning algorithm to predict the selected action from the plurality of possible actions given EEG signals corresponding to the first segments and the contextual information.
A system comprising:
- one or more computers;
- and
- one or more storage devices storing instructions that are operable, when executed on one or more computers, to cause the one or more computers to perform the operations of:
- receiving EEG signals from a sensor coupled to a user;
- receiving contextual information from one or both of the user and the user's environment;
- processing the EEG signals in real time using a machine learning model to predict a future action of the user, the machine learning model having a mapping function that maps a time series of values corresponding to an amplitude or change in amplitude of the EEG signals to an action and the contextual information;
- and
- generating an output associated with the predicted action.
One or more non-transitory computer-readable storage mediums comprising instructions stored thereon that are executable by a processing device and upon such execution cause the processing device to perform operations of:
- receiving EEG signals from a sensor coupled to a user;
- receiving contextual information from one or both of the user and the user's environment;
- processing the EEG signals in real time using a machine learning model to predict a future action of the user, the machine learning model having a mapping function that maps a time series of values corresponding to an amplitude or change in amplitude of the EEG signals to an action and the contextual information;
- and
- generating an output associated with the predicted action.
Dependencies are inferred from the claim text.