Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing EEG source localization. One of the methods includes obtaining brain data comprising: EEG data comprising respective channel data corresponding to each of a plurality of electrodes of an EEG sensor, and fMRI data comprising respective voxel data corresponding to each of a plurality of voxels; identifying, in a three-dimensional coordinate system, a respective location for each electrode; generating, using the respective identified locations of each electrode, data representing a location in the three-dimensional coordinate system of each voxel; determining, for each electrode, a region of interest in the three-dimensional coordinate system; and identifying, for each electrode, one or more corresponding parcellations in the brain of the subject, wherein each parcellation that corresponds to an electrode at least partially overlaps with the region of interest of the electrode.

This application provides a user attention detection method and system. The method includes: collecting an electroencephalogram signal of a user from an ear side by using an ear-side wearing apparatus (1100); when it is determined that the ear-side wearing apparatus (1100) can collect electroencephalogram signals from both a left ear canal and a right ear canal of the user, performing differential processing on the electroencephalogram signals from the left ear canal and the right ear canal of the user to obtain an electroencephalogram signal; and detecting an attention type of the user based on the electroencephalogram signal. According to the method and the system, the electroencephalogram signals of the user can be obtained from the ear canals more conveniently and quickly, and an attention status of the user can be detected anytime and anywhere.

This application provides a user attention detection method and system. The method includes: collecting an electroencephalogram signal of a user from an ear side by using an ear-side wearing apparatus (1100); when it is determined that the ear-side wearing apparatus (1100) can collect electroencephalogram signals from both a left ear canal and a right ear canal of the user, performing differential processing on the electroencephalogram signals from the left ear canal and the right ear canal of the user to obtain an electroencephalogram signal; and detecting an attention type of the user based on the electroencephalogram signal. According to the method and the system, the electroencephalogram signals of the user can be obtained from the ear canals more conveniently and quickly, and an attention status of the user can be detected anytime and anywhere.

A machine learning system for aggregating electroencephalographic (EEG) data, as well as external data, in preparation for downstream analysis via further machine learning models. Machine learning models can be used to assist in diagnosis of various mental health conditions, brain-computer interface, mood detection systems, or other biometric functions. Implementations of the present disclosure, employ a portion of the transformer network (the attention encoder stack) to aggregate EEG trials or EEG data segments, in a data-driven way, by ensuring the important content of each trial is not lost. Each EEG trial to be aggregated is converted into an input embedding, or a vector which numerically represents the data in the trial.

A machine learning system for aggregating electroencephalographic (EEG) data in preparation for downstream analysis via further machine learning models. Machine learning models can be used to assist in diagnosis of various mental health conditions, brain-computer interface, mood detection systems, or other biometric functions. Implementations of the present disclosure, employ a portion of the transformer network (the attention encoder stack) to aggregate EEG trials or EEG data segments, in a data-driven way, by ensuring the important content of each trial is not lost. Each EEG trial to be aggregated is converted into an input embedding, or a vector which numerically represents the data in the trial.

A neurostimulation device includes a plurality of electrodes adapted to be electrically connected to a subject to receive multichannel electrical signals from the subject's brain, a multichannel seizure detection unit electrically connected to the plurality of electrical leads to receive the multichannel electrical signals, and a neurostimulation unit in communication with the multichannel seizure detection unit. The plurality of electrodes are at least three electrodes such that the multichannel electrical signals are at least three channels of electrical signals, and the multichannel seizure detection unit detects a presence of a seizure based on multichannel statistics from the multichannel electrical signals including higher order combinations than two-channel combinations.

The invention relates to methods and systems for monitoring and teaching children with autistic spectrum disorders and can be used for effectively managing special educational work with children with autistic spectrum disorders (ASDs). According to the invention, the system comprises a remote server, personal computer devices of parents and specialists that are connected by an integrated network to the remote server, and a neuro-interface module for tracking a child's brain activity, said module being placed on the child and being connected by means of the integrated network to the remote server and comprising EEG sensors, wherein the neuro-interface module comprises an accelerometer and a gyroscope, and sensors for detecting a gaze direction, and the remote server is capable of collecting and analyzing visual data about the child's activity.

One aspect of the present disclosure relates to a system that can identify a focal area of abnormal brain interactions in a subject. Time series data can be received that corresponds to recordings from a plurality of regions in a brain of the subject during a resting period. Based on the time series data, an information inflow associated with each of the plurality of regions can be determined. The focal area of the abnormal brain interactions can be identified as one of the plurality of regions having a maximum information inflow.

A method is implemented by a computing device for helping a particular user use a user interface (UI). Electroencephalography (EEG) data is obtained that indicates brain activity of a particular user during a period in which that user views the UI and/or interprets help information that describes how to use the UI. Based on the EEG data, the computing device selects, from among multiple predefined cognitive states, the one or more cognitive states that characterize the particular user during the period. The computing device assists the particular user to use the UI by customizing the help information for the particular user based on the one or more selected cognitive states. A complementary computing device and computer program product are also disclosed.

A detection report data generation method including acquiring event type information of an electrocardiogram event corresponding to electrocardiogram event data, wherein the event data has one or more pieces of event type information; screening the event data according to signal quality evaluation indexes so as to obtain report conclusion data and report entry data; carrying out quality assessment on an event segment included in the event data according to the signal quality evaluation indexes, and determining a pre-selected sample segment according to a quality assessment result; determining position information of an event heart beat in the pre-selected sample segment, and determining segment interception parameters; carrying out interception processing on the pre-selected sample segment according to the segment interception parameters so as to obtain a typical data segment; generating report graphic data according to the typical data segment; and outputting the entry data, the graphic data and the conclusion data.