Brain-computer interface with high-speed eye tracking features

Inventors

ALCAIDE, Ramses • Padden, Dereck • JANTZ, Jay • HAMET, James • MORRIS, Jr., Jeffrey • Pereira, Arnaldo

Assignees

Neurable Inc

Abstract

Embodiments described herein relate to systems, devices, and methods for use in the implementation of a brain-computer interface that integrates real-time eye-movement tracking with brain activity tracking to present and update a user interface that is strategically designed for high speed and accuracy of human-machine interaction. Embodiments described herein also relate to the implementation of a hardware agnostic brain-computer interface that uses real-time eye tracking and online analysis of neural signals to mediate user manipulation of machines.

Core Innovation

The invention describes an integrated, hardware-agnostic hybrid brain-computer interface that combines real-time video-based eye tracking with online neural signal recording and analysis to mediate selection and activation of actions in a user interface. It uses eye-movement signals to determine a portion of focus of the user and then limits the presented stimulus set to a subset within that focused portion. Based on the eye-movement signals and the presented second set of flash stimuli, the invention identifies an action intended by the user and selects the identified flash stimulus to implement the identified action.

The system integrates oculomotor and neural signal processing to improve the identification of intended actions, including approaches using event related potentials and visually evoked potentials. It describes ensemble processing that forms a set of data including eye-movement inputs and information related to the second set of flash stimuli, and then processes the ensemble set of data to determine an identified action associated with an identified flash stimulus. The processing supports real-time gaze classification, includes lag-less filtering and classifier approaches, and accommodates missing or low-sampling gaze data through model-based estimation.

The user interface functionality includes focus-driven stimulus presentation such as a magnified configuration for the second set of flash stimuli and hold-release activation/deactivation control. It further supports event-driven or asynchronous operation modes and describes artifact handling in the eye-tracking signals, including using detected artifacts as potential control signals. The invention also includes model building and training for user-specific decoding, including generating a gaze-kinematics model to predict gaze and to replace missing data points, thereby enabling continued inference of focused stimulus selection and intended action.

Claims Coverage

The independent claims share a common control loop with two-stage flash stimuli selection driven by a determined portion of focus from eye-movement signals, plus selection of a flash stimulus to implement an intended action. Across the independent claims, the core coverage includes presentation of first and second sets of selectable flash stimuli, focus determination from eye-movement signals, determination of an intended action from eye-movement signals and the second set, and selecting the identified flash stimulus; one independent claim further includes ensemble processing of an ensemble set of data.

The inventive coverage centers on determining a portion of focus from eye-tracker eye-movement signals, presenting a second set of flash stimuli limited to that focus portion, determining an identified action intended by the user from the eye-movement signals and the second set, and selecting the corresponding identified flash stimulus to implement the action; one independent claim further includes generating and processing an ensemble set of data to determine the identified action.

Stated Advantages

Documented Applications

No documented applications found