Methods, systems, and non-transitory computer-readable mediums for SSVEP detection
Inventors
Kim, Insoo • Shishavan, Hossein Hamidi • Golzari, Kia • Farooq, Muhamed
Assignees
Denso Corp • Toyota Motor Corp • Mirise Technologies Corp • University of Connecticut Health Center
Publication Number
US-11684301-B1
Publication Date
2023-06-27
Expiration Date
2042-01-14
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
In accordance with one embodiment of the present disclosure, a method includes generating a plurality of icons, wherein each icon has a target frequency unique from each other, receiving brain activity data based on an epoch, generating a reference signal based on the epoch, calculating correlation coefficients between the brain activity data and the reference signal, wherein the correlation coefficients are calculated in a window that is within ±0.5 Hz of the target frequencies, including endpoints, determining a confidence score based on the correlation coefficients and the epoch, and determining a selected icon among the plurality of icons based on the correlation coefficients in response to the confidence score surpassing a threshold confidence score.
Core Innovation
The invention relates to methods, systems, and non-transitory computer-readable mediums for steady-state visually evoked potentials (SSVEP) detection utilized in brain-computer interfaces (BCI). The method involves generating multiple icons each flickering at unique target frequencies, receiving brain activity data over an epoch, generating a reference signal based on the epoch, and calculating correlation coefficients between the brain activity data and the reference signal. The calculation of correlation coefficients is performed within a frequency window of ±0.5 Hz around each target frequency to examine brain responses corresponding to specific flicker rates of icons.
This approach narrows the frequency analysis window to increase computational efficiency compared to traditional methods which analyze a broad range of frequencies with fine resolution, resulting in substantial time and resource consumption. A confidence score is determined from the correlation coefficients and epoch data, and a selected icon is identified once the confidence score exceeds a predefined threshold, thereby enabling reliable detection of user interaction with a particular icon based on brain activity.
The problem addressed is the inefficiency of traditional canonical correlation analysis (CCA)-based SSVEP detection methods which require calculating correlation coefficients over a large frequency range at fine increments, consuming significant computational resources and time. The invention solves this by optimizing the detection process through focused analysis on the target frequencies and their harmonics within specified frequency bands, as well as incorporating a confidence scoring mechanism to expedite selection once sufficient evidence of user focus is obtained.
Claims Coverage
The patent includes three independent claims covering a method, a system, and a non-transitory computer-readable medium for improved SSVEP detection.
Focused calculation of correlation coefficients within frequency windows around target frequencies
The correlation coefficients between brain activity data and reference signals are calculated only within a window of ±0.5 Hz around each unique target frequency for the plurality of icons, rather than across a broad frequency range.
Determining confidence score based on correlation coefficients and epoch data
A confidence score is calculated from the correlation coefficients derived within the specified frequency windows and the corresponding epoch to assess the reliability of the detected frequency.
Icon selection based on confidence score surpassing threshold
A selected icon is determined from the plurality of icons based on correlation coefficients once the calculated confidence score exceeds a predetermined threshold confidence score.
Consideration of harmonic frequencies in correlation coefficient calculations
The method/system also determines, for each target frequency, whether a harmonic frequency is stronger and calculates correlation coefficients within ±0.5 Hz windows of harmonic frequencies accordingly, including identifying multiple harmonic frequencies and incorporating them in the analysis.
Confidence score determination using maximum or sum of correlation coefficients
Confidence scores may be computed based on either the maximum correlation coefficient or the sum of correlation coefficients within the frequency windows, accounting for target frequencies and harmonics.
Outputting control signals based on selected icon
Once an icon is selected based on correlation data and confidence score, the system outputs a control signal corresponding to the selected icon for further processing.
The independent claims collectively cover a comprehensive method, system, and computer-readable medium for efficient and accurate SSVEP-based icon selection by focusing correlation coefficient calculations around target and harmonic frequencies within narrow windows, applying confidence scoring to expedite detection, and enabling output control signals based on selections.
Stated Advantages
Improves computational efficiency by restricting correlation coefficient calculations to frequency windows of ±0.5 Hz around target frequencies, thus reducing time and resource consumption.
Enhances execution speed by employing a confidence scoring system that allows icon selection once a predetermined confidence threshold is surpassed, avoiding unnecessary prolonged processing.
Accounts for imprecisions inherent in the SSVEP paradigm by including harmonic frequencies in analysis windows, resulting in more accurate detection.
Documented Applications
Application in brain-computer interface (BCI) systems using electronic displays presenting user interfaces with multiple flickering icons for user interaction via SSVEP detection.
Particularly applicable to user interfaces displayed on heads-up displays (HUDs) in automobiles and personal computers.
Interested in licensing this patent?