Wearable autism spectrum disorder-related visual function screening system in VR environment

Abstract

A wearable autism spectrum disorder-related visual function screening system in a VR environment, comprising: a VR stimulus presentation module, which presents a designed visual surround suppression stimulus to a user by means of a wearable VR head-mounted display; an electroencephalogram acquisition and transmission module, which acquires the user's real-time electroencephalogram data by means of a portable electroencephalogram smart headband and transmits same wirelessly to the outside; and a data synchronization module, which synchronizes and aligns the time of the VR stimulus with a timing signal acquired by the electroencephalogram headband to facilitate subsequent data analysis and processing. In the wearable autism spectrum disorder-related screening system, a brain-computer interface and a brain signal decoding algorithm are introduced to achieve quantitative evaluation of visual surround suppression; compared with conventional screening methods that rely on subjective scales, the present invention has good objectivity.

Description

A wearable autism visual function screening system in a VR environment TECHNICAL FIELD

[0001] The application belongs to the technical field of human-computer interaction and autism visual function screening, and particularly relates to a wearable autism visual function screening system in a VR environment. BACKGROUND

[0002] Autism Spectrum Disorder (ASD or autism) mainly has clinical behavior characteristics of social communication disorder, narrow interest and repetitive stereotyped behavior, and abnormal sensory response. There are more than 15 million autism people in China, with a prevalence rate of about 1%, and the prevalence rate of 6-12 year-old children is about 0.7%, which is the primary cause of child mental disability in China; at present, the history investigation and behavior observation are mainly used in clinical practice to screen autism patients, however, this method depends on the subjective report of patients or parents and the naked eye observation of medical staff, and has long evaluation time, low efficiency, requires doctors and therapists to have strong professional knowledge and skills, is affected by subjective bias and uneven medical resources, and is difficult to meet the screening demand.

[0003] Sensory abnormalities are a prominent feature of autism, and the proportion of 3-6 year-old autistic children with sensory abnormalities is as high as 95%, involving almost all sensory modalities, including vision, hearing, smell, touch, taste and proprioception; the sensory symptoms such as over-sensitivity to specific sounds, lack of response to mother's call, lack of eye contact, etc. are often the first abnormal performance of autistic children that attract parents' attention; existing research shows that autism patients have abnormal visual surround inhibition mechanism, and when facing a large amount of visual information from the environment, the filtering ability of background interference information is weaker than that of healthy people, resulting in different processing methods of external stimuli between patients and typical development healthy individuals, so that patients are more likely to have perception confusion, information confusion and attention dispersion, and there is a correlation between the behavior symptoms observed clinically, therefore, the quantitative evaluation of the visual surround inhibition function of autism patients can be used as a new means of clinical screening and diagnosis of autism; however, the evaluation of the visual surround inhibition function at present mainly depends on the subjective report of the subjects, and its effectiveness and accuracy cannot be guaranteed, so it is urgent to develop a more objective and reliable quantitative evaluation tool. SUMMARY

[0004] The application aims to provide a wearable autism visual function screening system in a VR environment, which uses precise neurophysiological electrical signals to detect visual function abnormalities of autism people, solves the problems of lack of universally accepted screening tools and high dependence on subjective judgment of professionals, and ensures the portability and practicability of the application based on wearable VR equipment and EEG head ring, and enhances the applicability to children of low age.

[0005] To achieve the above object, the present application provides the following technical scheme: a wearable autism visual function screening system in a VR environment, comprising

[0006] The VR stimulus presentation module presents the designed visual surround inhibition stimulus to the user using a wearable VR headset;

[0007] The EEG acquisition and transmission module acquires real-time EEG data of the user through a portable EEG intelligent head ring and transmits it wirelessly;

[0008] The data synchronization module synchronizes and aligns the VR stimulus time with the timing signal acquired by the EEG head ring, facilitating subsequent data analysis and processing;

[0009] The data processing module pre-processes the acquired EEG data, quantitatively evaluates the induced SSVEP amplitude, and extracts surround inhibition EEG features for evaluation through the developed multi-level task-related component evaluation algorithm.

[0010] As a preferred technical scheme of the present application, the VR stimulus presentation module includes a head-mounted VR display and a stimulus interface, wherein the stimulus interface includes a foreground flicker stimulus and a static background.

[0011] As a preferred technical scheme of the present application, the foreground flicker stimulus includes 5 circular flicker keys with snowflake patterns, distributed around the central "+" sign on the interface, and flickers at a fixed frequency of 20Hz to induce the user's EEG signal with SSVEP component; the static background can present snowflake pattern and blank, corresponding to two test conditions with / without background inhibition.

[0012] As a preferred technical scheme of the present application, the EEG intelligent head ring adopts a patch dry electrode acquisition method, which can reduce the impedance between the skin and the electrodes by applying physiological saline and scrub cream, and can be flexibly attached to the forehead and / or occipital position.

[0013] As a preferred technical scheme of the present application, the EEG acquisition and transmission module can be flexibly arranged at the positions of the visual occipital channels Oz, O1 and O2 under the 10-20 electrode distribution rule through the head ring elastic band, realizing the coverage of the visual occipital brain area.

[0014] As a preferred technical scheme of the present application, the task-related component analysis algorithm based on multi-level spatial filtering optimizes the SSVEP component extraction efficiency under similar stimulation conditions, strengthens the differences between SSVEP extraction under different stimulation conditions, and improves the surround inhibition detection precision.

[0015] As a preferred technical scheme of the present application, spatial filters are constructed for brain electrical signals induced by various stimuli, so that the characteristic components of SSVEP under the same stimulation condition can be efficiently extracted.

[0016] As a preferred technical scheme of the present application, the data synchronization module receives a stimulation timestamp from the VR stimulus presentation module and real-time electroencephalogram data from the electroencephalogram acquisition and transmission module, synchronizes and aligns the two, and transmits them to the data processing module in a wireless manner.

[0017] As a preferred technical scheme of the present application, the acquired electroencephalogram data is preprocessed, including filtering, noise reduction, and ocular artifact removal steps.

[0018] As a preferred technical scheme of the present application, the data processing module includes an electroencephalogram signal preprocessing module, a feature extraction module, a feature selection module, a steady-state visual evoked potential quantitative evaluation module, and a surround suppression quantitative evaluation module.

[0019] Compared with the prior art, the present application has the following advantages:

[0020] 1) The wearable autism screening system based on steady-state visual evoked potential in the VR environment disclosed in the present application has the advantage of introducing a brain-computer interface and a brain signal decoding algorithm to realize quantitative evaluation of visual surround suppression, which is more objective than the previous screening method relying on subjective scales.

[0021] 2) The visual surround suppression electroencephalogram decoding algorithm disclosed in the present application can better identify the induced SSVEP amplitude with / without background interference, and realize more accurate evaluation of surround suppression effect.

[0022] 3) The wearable VR+ electroencephalogram intelligent head ring integrated system disclosed in the present application has the advantages of convenient wearing, controllable cost, no need to wash hair, wireless transmission, and can be used to realize lightweight detection of electroencephalogram indicators. BRIEF DESCRIPTION OF DRAWINGS

[0023] Fig. 1 is a schematic diagram of the overall framework structure of the wearable autism visual function screening system in the VR environment of the present application;

[0024] Fig. 2 is a VR visual stimulation interface of the present application;

[0025] Fig. 3 is a flowchart of the visual surround suppression electroencephalogram decoding algorithm of the present application. DETAILED DESCRIPTION

[0026] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application. Embodiment 1

[0027] Please refer to FIG. 1-FIG. 3, which is the first embodiment of the present application, which provides a wearable autism visual function screening system in VR environment, comprising

[0028] VR stimulus presentation module: using wearable VR headsets to present the designed visual surround inhibition stimulus to the user; further description: the module includes a head-mounted VR display and a stimulus interface, wherein the stimulus interface includes two parts of foreground flicker stimulus and static background; the foreground flicker stimulus contains 5 circular flicker keys with snowflake pattern, which are distributed around the central "+" sign of the interface, and flicker at a fixed frequency of 20 Hz to induce the user's EEG signal with SSVEP component; the static background can present snowflake pattern and blank, corresponding to two test conditions of with / without background inhibition respectively;

[0029] EEG acquisition and transmission module: through a portable EEG intelligent head ring, the real-time EEG data of the user is acquired and wirelessly transmitted; further description: the multi-channel EEG intelligent head ring used adopts a patch dry electrode acquisition method, without the need for glue, and can be flexibly attached to the forehead and / or occipital lobe position by applying physiological saline and scrub cream to reduce the impedance between the skin and the electrodes;

[0030] Data synchronization module: synchronizing and aligning the VR stimulus time with the timing signal acquired by the EEG head ring, facilitating subsequent data analysis and processing;

[0031] Data processing module: pre-processing the acquired EEG data, quantitatively evaluating the induced SSVEP amplitude, and extracting surround inhibition EEG features for evaluation through the developed multi-level task-related component evaluation algorithm; further description: the present application proposes a task-related component analysis algorithm based on multi-level spatial filtering, which optimizes the SSVEP component extraction efficiency under similar stimulus conditions, strengthens the differences between SSVEP under different stimulus conditions (such as with / without background inhibition), and is used to improve the surround inhibition detection accuracy;

[0032] According to the idea of TRCA, the multi-channel EEG signal induced by the task state ( is mainly composed of two parts of source signals: the part related to the task or induced by the task ; The part that is not related to the task, that is, noise ; Therefore, the EEG signal obtained after multi-channel spatial filtering can be written as follows:

[0033] (1)

[0034] Where a and b are the coefficients of constructing EEG signals from two source signals, For channel The corresponding spatial filtering coefficient; the goal of TRCA is to maximize the covariance between different trials under the same stimulation conditions, that is, to maximize the correlation between different trials after spatial filtering; ideally, only task-related components are retained after spatial filtering. , the signals between different trials under the same stimulation condition will be highly similar; Indicates the first The covariance between the EEG signals of trial h1 and h2 can be calculated using the following formula:

[0035] (2)

[0036] in and The channels under this stimulation condition are and The corresponding spatial filter coefficient; for all trials under the same stimulation condition, the covariance between each two is calculated according to the above formula, and then the sum is obtained:

[0037] (3)

[0038] Where S is the sum of the covariances between different channels in paired trials under the same stimulus condition, which can be defined as follows:

[0039] (4)

[0040] To ensure that Formula 3 has a finite solution, the variance of , that is, the overall covariance between different channels (including all trials), needs to be constrained according to the following conditions:

[0041] (5)

[0042] Finally, the extraction of task-related EEG features under the same stimulation conditions is transformed into the following optimization problem:

[0043] (6)

[0044] The first-level spatial filter, i.e. the solution to the optimization problem, can be obtained by the eigenvector corresponding to the largest eigenvalue of the matrix The size of the eigenvalue reflects the consistency between different trials under the same stimulus condition.

[0045] According to the above steps, the spatial filter is constructed for the brain electrical signals induced by multiple stimuli, which can realize efficient extraction of SSVEP characteristic components under the same stimulus condition. However, the first-level spatial filter only considers the consistency between different trials under the same stimulus, ignoring the feature difference between different stimulus conditions. Therefore, this project considers adding a second-level spatial filter based on the first level to strengthen and highlight the differentiated features of SSVEP induced under excitation and inhibition conditions. First, the brain electrical signals under the excitation condition after the first-level spatial filtering are denoted as , and the brain electrical signals under the background inhibition condition are denoted as ; the and are paired and subtracted according to the trial order to obtain the excitation-inhibition difference signal :

[0046] (7)

[0047] where h represents the order of trial, and Nc is the number of channels. Then, the excitation-inhibition difference signal is substituted into formula 6 to replace the original brain signal , and the obtained eigenvector is the weight coefficient of the second-level spatial filter, and the eigenvalue reflects the consistency of the excitation-inhibition difference between different trials. Further, the canonical correlation analysis CCA is used to quantitatively evaluate the SSVEP features induced by different stimulus conditions after two-level spatial filtering. Specifically, the brain electrical signals form a matrix , and the ideal sine and cosine signal templates are constructed according to the following formula :

[0048]

[0049] where f is the stimulus flicker frequency, and i is the f harmonic frequency included in the analysis. After linear transformation of the brain electrical matrix X and the template matrix Y by the canonical correlation analysis, the correlation coefficient is maximized by optimizing the coefficient matrix and , i.e. solving the following optimization problem:

[0050]

[0051] wherein To evaluate the correlation coefficient obtained, the larger the value represents the stronger SSVEP induced by flickering stimulation; finally, by comparing the SSVEP intensity induced with / without background inhibition, the precise quantitative evaluation of visual surround inhibition effect can be realized, and whether the user has visual function abnormality related to autism can be judged according to the set threshold condition. Embodiment 2

[0052] Please refer to FIG. 1-FIG. 3, as the second embodiment of the present application, the embodiment provides a wearable autism visual function screening system in VR environment, comprising

[0053] The VR stimulation presentation module adopts a head-mounted VR device, the screen refresh rate is not less than 60Hz, and Unity is used as the development platform to realize the excitation-inhibition stimulation paradigm in the VR environment; the visual stimulation paradigm mainly induces the generation of electroencephalogram steady-state visual evoked potential (SSVEP) through flickering stimulation, and the flickering frequency is 20Hz; wherein the foreground stimulation appears on the circumference with the screen center as the center and a radius of 5 degrees of visual angle, the user does not need to make any active behavioral response, only needs to fixate on the central screen, and the remaining light accepts the raster flickering stimulation, which acts on the area where the retinal rod cells are located and activates the occipital lobe fissure; in the absence of background interference, SSVEP can be effectively induced in the visual cortex; when there is static background interference, the SSVEP induced by the foreground stimulation will be inhibited; and the time stamp related to the stimulation flickering time sequence is transmitted to the data synchronization module through Bluetooth;

[0054] The electroencephalogram acquisition and transmission module can be flexibly arranged at the positions of the visual occipital lobe channels Oz, O1 and O2 under the 10-20 electrode distribution rule through the head ring elastic band, so as to realize the coverage of the visual occipital lobe brain area; the patch dry electrode does not need to be glued, and the data is transmitted to the data synchronization module through a wireless manner;

[0055] The data synchronization module receives the stimulation time stamp from the VR stimulation presentation module and the real-time electroencephalogram data from the electroencephalogram acquisition and transmission module, synchronizes and aligns the two, and transmits to the data processing module through a wireless manner;

[0056] The data processing module analyzes the collected electroencephalogram signals according to the following electroencephalogram decoding algorithm:

[0057] Firstly, the original EEG data collected is preprocessed, including filtering, noise reduction, and EOG removal, and then the data of multiple trials under the same background is stacked; the preprocessed data is subjected to a first spatial filtering according to the task-related component analysis algorithm, so that the covariance between different trials under the same stimulus condition is maximized, i.e., the correlation between different trials after spatial filtering is maximized; in an ideal case, only the task-related components are retained after spatial filtering , and the signals between different trials under the same stimulus condition are highly similar; then, a second spatial filter is added on the basis of the first level, for strengthening and highlighting the differentiated features of SSVEP induced under excitation and inhibition conditions; specifically, the EEG signal under the excitation condition after the first spatial filtering is denoted as , and the EEG signal under the background inhibition condition is denoted as ; the pairs of and are subtracted in the order of trials to obtain the excitation-inhibition difference signal ; then, the excitation-inhibition difference signal is replaced by the original brain signal , and the task-related component analysis spatial filtering is performed again, and the obtained feature vector is the weight coefficient of the second spatial filter, and the feature value reflects the consistency of the excitation-inhibition difference between different trials; then, the canonical correlation analysis CCA is used to quantitatively evaluate the SSVEP features induced under different stimulus conditions after two-stage spatial filtering; finally, the SSVEP features under the background inhibition condition are compared with those without the background inhibition condition, to obtain the quantitative evaluation result of the user's visual surround inhibition, and whether the user has the visual function abnormality related to autism is determined according to the set threshold condition.

[0058] Although the embodiments of the present application have been shown and described in detail, as described above, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A wearable self- autism visual function screening system in a VR environment, characterized in that: Comprising VR stimulus presentation module: using wearable VR headsets to present the designed visual surround inhibition stimulus to the user; EEG acquisition and transmission module: collecting real-time EEG data of the user through a portable EEG intelligent head ring and transmitting it wirelessly; Data synchronization module: synchronize and align the VR stimulus time with the timing signal collected by the EEG head ring, which is convenient for subsequent data analysis and processing; Data processing module: pre-process the collected EEG data, quantify the induced SSVEP amplitude, and extract the surround inhibition EEG features for evaluation through the developed multi-level task-related component evaluation algorithm.

2. The wearable self-contained autism visual function screening system in a VR environment of claim 1, wherein: The VR stimulus presentation module includes a head-mounted VR display and a stimulus interface, which includes a foreground flicker stimulus and a static background.

3. The wearable self-contained autism visual function screening system in a VR environment of claim 2, wherein: The foreground flicker stimulus includes 5 circular flicker keys with snowflake patterns, which are distributed around the central "+" sign, and flicker at a fixed frequency of 20Hz to induce the user's EEG signal with SSVEP component; The static background can present snowflake pattern and blank, corresponding to two test conditions with / without background inhibition.

4. The wearable self-contained autism visual function screening system in a VR environment of claim 1, wherein: The EEG intelligent head ring adopts a patch dry electrode acquisition method, which can reduce the impedance between the skin and the electrode by applying physiological saline and scrub cream, and can be flexibly attached to the forehead and / or occipital position.

5. The wearable self-contained autism visual function screening system in a VR environment of claim 4, wherein: The EEG acquisition and transmission module can be flexibly arranged at the positions of visual occipital channels Oz, O1 and O2 under the 10-20 electrode distribution rule through the head ring elastic band, realizing the coverage of visual occipital brain area.

6. The wearable self-contained autism visual function screening system in a VR environment of claim 1, wherein: The task-related component analysis algorithm based on multi-level spatial filtering optimizes the SSVEP component extraction efficiency under similar stimulus conditions, strengthens the difference between SSVEP extraction under different stimulus conditions, and improves the surround inhibition detection accuracy.

7. The wearable self-contained autism visual function screening system in a VR environment of claim 6, wherein: Spatial filters are constructed for EEG signals induced by multiple stimuli to achieve efficient extraction of SSVEP feature components under similar stimulus conditions.

8. The wearable self-contained autism visual function screening system in a VR environment of claim 1, wherein: The data synchronization module receives the stimulus timestamp from the VR stimulus presentation module and the real-time EEG data from the EEG acquisition and transmission module, synchronizes and aligns them, and transmits them to the data processing module wirelessly.

9. The wearable self-contained autism visual function screening system in a VR environment of claim 1, wherein: The collected EEG data is pre-processed, including filtering, noise reduction, and ocular elimination steps.

10. The wearable self-contained autism visual function screening system in a VR environment of claim 1, wherein: The data processing module includes EEG signal preprocessing module, feature extraction module, feature selection module, steady-state visual evoked potential quantitative evaluation module, and surround inhibition quantitative evaluation module.

Images