Granted Patent B2

US 12,558,017 · App. 18/739,741 · Granted Feb 24, 2026

Inventors: Tan Le (San Francisco, CA); Geoffrey Ross Mackellar (Sydney, AU)

Assignee: Emotiv Inc.

A61B5/369A61B5/316G16H50/50

Abstract

Claims (36)

1 . A computer-implemented method, comprising:

for each user of a set of users:

recording a set of EEG signals from the user, wherein recording the set of EEG signals from the user comprises: at each time point of a set of time points, at a neuroheadset of the user, recording an EEG signal; and

determining a set of synchronization pattern features for each EEG signal in the set of EEG signals;

determining a synchronization pattern trend based on the set of synchronization pattern features;

for the set of users:

organizing the synchronization pattern trends into a set of groups; and

assigning a tag to each group of the set of groups, wherein each tag is associated with a user state;

determining a model for each group of the set of groups based on the tagged set of groups;

using the determined models, determining a predicted tag for a new user; and

recommending an activity for the new user to perform based on the predicted tag.

2 . The method of claim 1 , further comprising, for the new user:

recording a set of new EEG signals from the new user, wherein recording the set of new EEG signals from the new user comprises: at each time point of a new set of time points, at a neuroheadset of the new user, recording a new EEG signal;

determining a new synchronization pattern trend based on the set of new EEG signals;

determining a set of correlations based on the new user synchronization pattern trend and the determined models; and

determining the predicted tag based on the set of correlations.

3 . The method of claim 1 , wherein each user state is associated with a focused task.

4 . The method of claim 1 , further comprising, for the new user:

recording a set of new EEG signals from the new user, wherein recording the set of new EEG signals from the new user comprises: at each time point of a new set of time points, at a neuroheadset of the new user, recording a new EEG signal;

determining a new synchronization pattern trend based on the set of new EEG signals;

adding the new synchronization pattern trend to a group of the set of groups; and

updating the model associated with the group based on the new user synchronization pattern trend.

5 . The method of claim 4 , wherein the model is updated using machine learning methods.

6 . The method of claim 1 , wherein each tag comprises a neurological status tag.

7 . The method of claim 1 , wherein determining the synchronization pattern trend for each user of the set of users comprises: filtering a first EEG signal of the set of EEG signals of the user in a first band to generate a first band-filtered EEG signal; filtering the first EEG signal of the set of EEG signals of the user in a second band to generate a second band-filtered EEG signal; and extracting a set of phase locking episodes between the first band-filtered EEG signal and the second band-filtered EEG signal.

8 . The method of claim 1 , wherein the synchronization pattern trend is nonlinear.

9 . A computer-implemented method, comprising:

for each user of a set of users:

recording a set of EEG signals from the user, wherein recording the set of EEG signals from the user comprises: at each time point of a set of time points, at a neuroheadset of the user, recording an EEG signal; and

determining a synchronization pattern trend based on the set of EEG signals, wherein determining the synchronization pattern trend comprises: generating a filtered first EEG signal and a filtered second EEG signal upon filtering each of a first and a second EEG signal of the set of EEG signals of the user, respectively; analyzing the filtered first and second EEG signals via spectral analysis; and identifying stable phase difference episodes between the filtered first and second EEG signals via statistical identification of phase-locking synchrony;

for the set of users:

organizing the synchronization pattern trends into a set of groups; and

assigning a tag to each group of the set of groups, wherein each tag is associated with a user state;

determining a model for each group of the set of groups based on the tagged set of groups;

using the determined models, determining a predicted tag for a new user; and

recommending an activity for the new user to perform based on the predicted tag.

Assignments (1)

Continuity (5)

References Cited (175)