A synoptic electronic target system and device

Abstract

This invention discloses a synoptophore electronic target system and device, relating to the field of visual training technology. It includes a storage module that retrieves historical neural response datasets from a pre-stored database using a patient identification code, inputs them into a neural plasticity model, and outputs personalized adjustment factors and a baseline dynamic target parameter set. A control module triggers an audio synthesizer to generate a gamma-band sound signal when the neural response intensity decay index exceeds a neural response entrainment threshold; otherwise, it maintains the baseline dynamic target parameter set and continues to monitor the neural response intensity decay index. It records the change in the neural response intensity decay index under the neural entrainment coupling effect and updates the historical neural response dataset. This invention effectively enhances the synchronous firing activity of neurons in the visual cortex by locking the phase of the gamma-band sound signal with the flicker frequency of the electronic target image, thus creating a neural entrainment coupling effect.

Description

Technical Field

[0001] This invention relates to the field of visual training technology, and in particular to a synoptophore electronic target system and device. Background Technology

[0002] Synoptophores, as key medical devices for visual function assessment and training, have evolved from early mechanical optotypes to electronic and digital optotype systems. Current technologies primarily achieve binocular vision training through pre-programmed optotype patterns and fixed stimulus parameters, employing eye-tracking technology to record pupillary movement responses. Some systems have introduced basic data analysis algorithms to extract temporal or spatial features from eye movement trajectories. In recent years, the application of neuroplasticity theory in visual rehabilitation has deepened, promoting the initial integration of visual stimulation and neural feedback mechanisms.

[0003] However, existing electronic visual target systems mostly use static or fixed stimulation parameters, lacking the ability to dynamically adapt to individual patient neural characteristics and real-time fatigue states. Especially during long-term training, the nervous system is prone to response decay to monotonous stimuli, leading to a decline in training effectiveness. Current technologies struggle to maintain the excitability of the nervous system by monitoring neural response intensity in real time and dynamically adjusting stimulation patterns, thus limiting the continuous optimization of visual fusion training results. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a synoptophore electronic optotype system to solve the problem of poor visual fusion maintenance caused by the decay of neural response during synoptophore training.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a synoptophore electronic target system, comprising: a storage module, which retrieves a historical neural response dataset from a pre-stored database via a patient identification code, inputs it into a neural plasticity model, and outputs a personalized accommodation factor and a baseline dynamic target parameter set; a high-precision display unit, which drives the synoptophore to generate electronic target images through the baseline dynamic target parameter set and simultaneously acquires real-time pupillary motion data; a data analysis module, which performs frequency domain decomposition and spatial clustering on the real-time pupillary motion data based on the personalized accommodation factor, and calculates the binocular fusion error value and the neural response intensity attenuation index; an accommodation mechanism, which dynamically sets a neural response entrainment threshold based on the binocular fusion error value and the personalized accommodation factor; a control module, which triggers an audio synthesizer to generate a gamma-band sound wave signal when the neural response intensity attenuation index exceeds the neural response entrainment threshold; otherwise, it maintains the continuous operation of the baseline dynamic target parameter set and continues to monitor the neural response intensity attenuation index; and a communication module, which locks the phase of the gamma-band sound wave signal with the flicker frequency of the electronic target image to form a neural entrainment coupling effect; records the change in the neural response intensity attenuation index under the action of the neural entrainment coupling effect, and updates the historical neural response dataset.

[0008] As a preferred embodiment of the synoptophore electronic target system of the present invention, the historical neural response dataset includes neural response data recorded in each treatment session; the neural response data includes fatigue sensitivity, stimulation frequency, and therapeutic improvement rate.

[0009] As a preferred embodiment of the synoptophore electronic target system of the present invention, the specific steps for outputting the personalized adjustment factor and the reference dynamic target parameter set are as follows:

[0010] Preprocess the historical neural response dataset;

[0011] Use machine learning algorithms to train a neural plasticity model and learn patterns in neural response data;

[0012] Based on a trained neuroplasticity model, the system inputs neural response data corresponding to the patient identification code and outputs personalized regulatory factors and a set of baseline dynamic target parameters.

[0013] As a preferred embodiment of the synoptic electronic target system of the present invention, the reference dynamic target parameter set includes the flashing frequency, spatial coordinates, dynamic change mode, and duration parameters of the electronic target image.

[0014] In a preferred embodiment of the synoptophore electronic optotype system of the present invention, the specific steps for calculating the binocular fusion error value and the neural response intensity attenuation index are as follows:

[0015] A personalized adjustment factor is used to filter real-time pupil motion data;

[0016] Frequency domain decomposition was performed on the filtered real-time pupil motion data to extract frequency components;

[0017] Spatial clustering of frequency components is used to identify binocular motion patterns, and binocular fusion error values ​​are calculated.

[0018] Based on the continuous change trend of binocular fusion error value over time, the attenuation amplitude of neural response intensity is extracted and a neural response intensity attenuation index is generated.

[0019] As a preferred embodiment of the synoptophore electronic optotype system of the present invention, the step of dynamically setting the neural response entrainment threshold based on the binocular fusion error value and the personalized adjustment factor includes mapping the binocular fusion error value to the range of neural response entrainment threshold through a mapping function, and adjusting the mapping function using a personalized adjustment factor.

[0020] As a preferred embodiment of the synoptic electronic visual target system of the present invention, wherein: the triggering of the audio synthesizer to generate the γ-band sound wave signal means that when the neural response intensity attenuation index is detected to exceed the neural response entrainment threshold, a trigger command is issued and the audio synthesizer is started to output the γ-band sound wave signal.

[0021] In a preferred embodiment of the synoptophore electronic target system of the present invention, the specific steps for forming the neural entrainment coupling effect are as follows:

[0022] Acquire phase information of gamma-band acoustic signals; acquire the flashing frequency of electronic visual targets;

[0023] The phase information of the gamma-band acoustic signal is synchronized with the flashing frequency of the electronic visual target image through a phase-locked loop, maintaining the phase-locked state and forming a neural entrainment coupling effect.

[0024] As a preferred embodiment of the synoptophore electronic target system of the present invention, wherein: the updated historical neural response dataset refers to recording the change in the neural response intensity attenuation index under the neural entrainment coupling effect; the change in the neural response intensity attenuation index is written into the historical neural response dataset and stored in chronological order.

[0025] Secondly, the present invention also proposes a synoptophore electronic target device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of the aforementioned synoptophore electronic target system.

[0026] The beneficial effects of this invention are as follows: By locking the phase of the gamma-band sound wave signal with the flicker frequency of the electronic visual target image, a neural entrainment coupling effect is formed, which can effectively enhance the synchronous firing activity of neurons in the visual cortex. Furthermore, the neural entrainment coupling effect utilizes a multi-sensory synergistic stimulation mechanism to improve the efficiency of neural plasticity induction and inhibit the decay trend of neural response intensity during training. Simultaneously, by recording the changes in the neural response intensity decay index in real time and updating the historical dataset, data-driven optimization is provided for subsequent treatment sessions, thereby achieving continuous and stable regulation of neural excitability. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a flowchart of a synoptophore electronic target system.

[0029] Figure 2 A flowchart for preprocessing historical neural response datasets.

[0030] Figure 3 A flowchart for calculating the binocular fusion error value and the neural response intensity attenuation index.

[0031] Figure 4 A flowchart for updating the historical neural response dataset. Detailed Implementation

[0032] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0033] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0034] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0035] Reference Figures 1-4As one embodiment of the present invention, this embodiment provides a synoptic electronic target system, comprising the following steps:

[0036] The storage module retrieves historical neural response datasets from a pre-stored database using the patient identification code, inputs them into the neural plasticity model, and outputs personalized modulatory factors and a baseline dynamic target parameter set. The historical neural response dataset includes neural response data recorded in each treatment session. The neural response data includes fatigue sensitivity, stimulation frequency, and efficacy improvement rate.

[0037] Preprocess the historical neural response dataset.

[0038] Specifically, for each record of neural response data, invalid data points are identified and removed; invalid data points are defined as records that cannot be parsed; records that cannot be parsed specifically refer to abnormal data entries in the historical neural response dataset that are malformed, non-numerical, or do not meet the requirements of the data structure.

[0039] For partially missing neural response data (such as the treatment improvement rate of a certain session not being recorded), missing value processing is performed. Specifically, the mean imputation method is used to calculate the feature mean of all available records in the historical neural response dataset and replace the missing values ​​with the feature mean.

[0040] For example, if there are 10 session records in the historical neural response dataset, and the efficacy improvement rate is missing in 2 of them, then calculate the average efficacy improvement rate of the remaining 8 sessions and fill in the missing parts.

[0041] Feature normalization was performed on the neural response data after data integrity checks and missing value processing. Specifically, min-max normalization was applied to map fatigue sensitivity, stimulation frequency, and therapeutic improvement rate to... Intervals; the neural response data after feature normalization is used as the preprocessed historical neural response dataset.

[0042] Use machine learning algorithms to train a neural plasticity model and learn patterns in neural response data.

[0043] Specifically, the preprocessed historical neural response dataset is divided into a training set and a validation set in chronological order. The training set is used for training the neural plasticity model, and the validation set is used to evaluate the performance of the neural plasticity model.

[0044] It should be noted that when dividing the training set and validation set, a time-order partitioning method is used, and all treatment session records in the historical neural response dataset are arranged in ascending order according to the treatment timestamp;

[0045] In the sorted historical neural response dataset, conversation records from earlier time periods are assigned to the training set, while those from later time periods are assigned to the validation set (e.g., the first 80% of conversation records are assigned to the training set, and the remaining 20% ​​to the validation set). This preserves the time-series characteristics of the data and better simulates the ability of neural plasticity models to process newly generated data in real-world applications.

[0046] It should be noted that the 80% training set and 20% validation set split is based on a common data splitting convention in the field of machine learning. This split ensures a sufficient amount of training data for the neuroplasticity model while providing enough and balanced validation samples to evaluate the performance of the neuroplasticity model.

[0047] Gradient boosting decision trees were chosen as the machine learning algorithm because they are suitable for small-scale data and can effectively handle nonlinear relationships. The neuroplasticity model's inputs are normalized fatigue sensitivity, stimulation frequency, and therapeutic improvement rate; the outputs are personalized adjustment factors and a baseline dynamic target parameter set (including flicker frequency, spatial coordinates, dynamic change pattern, and duration parameters). During training, the neuroplasticity model optimizes its parameters by minimizing the error between the predicted output and historical records.

[0048] Evaluate the trained neuroplasticity model on a validation set and calculate the deviation between the predicted output and the actual records on the validation set. If the deviation exceeds an acceptable range, adjust the hyperparameters of the neuroplasticity model (such as decision tree depth or learning rate) and retrain until the prediction accuracy meets the requirements.

[0049] It should be noted that the acceptable range refers to the mean absolute error between the personalized adjustment factor and the baseline dynamic target parameter set predicted by the neural plasticity model on the validation set and the actual values ​​in the historical records, which does not exceed the average fluctuation range of the same indicators in the historical neural response dataset. Meeting the prediction accuracy requirement means that the error between the personalized adjustment factor and the baseline dynamic target parameter set output by the neural plasticity model on the validation set and the corresponding parameters actually used in the historical records reaches convergence after calculation using the loss function built into the gradient boosting decision tree, and no longer significantly decreases with increasing training epochs.

[0050] Based on a trained neuroplasticity model, the system inputs neural response data corresponding to the patient identification code and outputs personalized regulatory factors and a set of baseline dynamic target parameters.

[0051] Specifically, the patient's neural response data is retrieved from a pre-stored database by querying the patient's identification code. If the neural response data is not preprocessed, a normalization method (using the minimum-maximum value of the historical neural response dataset) is applied for real-time normalization.

[0052] The normalized neural response data is input into the trained neural plasticity model; the neural plasticity model performs forward propagation and outputs personalized regulation factors and a baseline dynamic target parameter set.

[0053] Furthermore, the process by which the neural plasticity model outputs personalized adjustment factors and a baseline dynamic target parameter set after performing forward propagation is as follows: Inputting normalized neural response data, the neural plasticity model performs feature weighting and combination calculations through the structure of a gradient boosting decision tree, passing it layer by layer to the output node, generating continuous values ​​of personalized adjustment factors and four components of the baseline dynamic target parameter set; the four components include the numerical dimension of the flicker frequency, the two-dimensional position descriptor of the spatial coordinates, the behavioral feature encoding of the dynamic change pattern, and the time dimension of the duration parameter.

[0054] Superiorly, compared to traditional linear regression or basic time series models, using gradient boosting decision trees to model historical neural response data can more effectively capture the complex nonlinear relationships and time-dependent patterns, thereby achieving precise adaptation to the patient's neural characteristics through the generated personalized regulatory factors.

[0055] The high-precision display unit drives the synoptophore to generate electronic target images through a reference dynamic target parameter set, and simultaneously collects real-time pupil movement data.

[0056] Specifically, the four components of the reference dynamic target parameter set are encoded into a binary instruction stream according to the data format and transmission order specified in the synchro hardware communication protocol, and then sent to the synchro via a serial communication interface.

[0057] After receiving the driving command, the synchro unit renders the basic target pattern at the specified position on the screen according to the binary command stream; it periodically switches the brightness state of the target (bright → dark → bright) according to the flashing frequency.

[0058] The viewpoint position or shape is updated according to the dynamic change mode: when the dynamic change mode is linear movement, the viewpoint is translated by a fixed displacement vector each frame; when the dynamic change mode is rotation, the viewpoint is rotated by a fixed angle increment each frame. An internal timer is started, and the display stops and the image position is reset when the duration parameter reaches the set value.

[0059] It should be noted that the duration parameter refers to the duration of a single continuous display of an electronic target image as defined in the baseline dynamic target parameter set; the set value is a specific time value pre-configured in the duration parameter (such as 5 seconds), and when the internal timer accumulates to this value, it triggers the display stop and position reset operation.

[0060] Furthermore, at the same time as the electronic optotype images begin to display, a high-speed infrared camera is triggered to capture the video stream of the patient's eyes; the camera lens maintains a fixed distance from the patient's eyes to ensure complete coverage of the range of eye movements;

[0061] Furthermore, each frame of the patient's binocular video stream is processed: the color image is converted to grayscale to enhance the contrast between the pupil and iris; an adaptive thresholding method is used to determine the pupil boundary, and the segmentation threshold is dynamically adjusted according to the local brightness of the image; morphological closing operations are performed on the segmented pupil region to eliminate noise interference and fill internal gaps. The segmentation threshold is used to distinguish the grayscale value critical point between the pupil region and non-pupil region in the image, and is specifically set by calculating the mean and standard deviation of the grayscale values ​​of the local image region.

[0062] Within the identified pupil region, the set of pixel coordinates of the connected regions is extracted. The set of pixel coordinates of the connected regions refers to the two-dimensional coordinate set of all interconnected pupil region pixels identified by image processing algorithms (such as 8-neighborhood connected component analysis), including the row number (Y coordinate) and column number (X coordinate) of each pixel in the image.

[0063] Calculate the arithmetic mean of the x and y coordinates of all pixels within the connected region, and then calculate the geometric center of the set of pixel coordinates, which will be used as the pupil center coordinates of the current frame.

[0064] The pupil center coordinates of the left and right eyes are recorded separately. The start time of the electronic visual target picture is taken as the time zero point. The relative time offset is calculated based on the camera frame rate and frame number. The time offset is added to the corresponding pupil coordinate value. The processing results of all frames are integrated in chronological order to output real-time pupil motion data. The real-time pupil motion data includes time stamps, left eye pupil X coordinate, left eye pupil Y coordinate, right eye pupil X coordinate, and right eye pupil Y coordinate.

[0065] It should be noted that during the synchronous acquisition of real-time pupil motion data, the amplitude of pupil coordinate changes between adjacent frames needs to be continuously checked. If the coordinates change abruptly exceeds the reasonable physiological motion range, it is judged as data anomaly. Abnormal data segments are marked and discarded, and a re-acquisition process is triggered. The reasonable physiological motion range refers to the upper limit of the image pixel displacement converted from the maximum movement speed of the human pupil during normal visual tracking (usually not exceeding 700 degrees of visual angle per second). The upper limit is dynamically calculated through camera calibration parameters (such as pixel / angle conversion coefficient).

[0066] The data analysis module, based on personalized adjustment factors, performs frequency domain decomposition and spatial clustering on real-time pupil motion data, and calculates binocular fusion error and neural response intensity attenuation index.

[0067] A personalized adjustment factor is used to filter the real-time pupil motion data; the filtered real-time pupil motion data is then decomposed in the frequency domain to extract frequency components.

[0068] Specifically, the personalized adjustment factor (scalar value) is converted into the cutoff frequency of the Butterworth low-pass filter; the cutoff frequency is used to perform a fourth-order Butterworth low-pass filter on the real-time pupil motion data, and the smoothed pupil trajectory is output.

[0069] Perform an FFT on the smoothed pupil trajectory (e.g., the X-coordinate of the left pupil) to obtain the frequency amplitude spectrum, expressed as:

[0070] ;

[0071] In the formula, Represents frequency variable The corresponding amplitude spectrum, This represents the modulo (absolute value) operation. The summation symbol indicates the summation of discrete-time indices. from arrive Summation, The total length of the signal, i.e., the total number of sampling points for real-time pupil movement data on the time axis. Indicates a point in time At this point, the X-axis coordinates of the center of the left pupil after filtering are shown. The kernel function represents the discrete Fourier transform. It is a natural constant. The imaginary unit, Pi;

[0072] Identify the peak values ​​of the frequency amplitude spectrum and record the peak frequencies and corresponding amplitudes as the dominant frequency components.

[0073] Spatial clustering of frequency components is used to identify binocular motion patterns, and binocular fusion error is calculated.

[0074] Specifically, the dominant frequency components of both eyes are organized into feature vectors, and two core patterns are predefined during the training phase of the historical dataset.

[0075] Mode 1 (fusion state): Both eyes move in the same direction, frequency difference Hertz, with zero error;

[0076] Mode 2 (Separated State): Eye movement opposite direction or frequency difference hertz;

[0077] It needs to be explained that the settings Hertzian grading is used as a classification criterion for binocular motion patterns based on the frequency domain synchronization characteristics of visual fusion in neurophysiology. When the frequency difference between binocular motions is less than... At Hertz, the visual cortex of the brain can maintain the temporal integration mechanism of binocular signals. At this time, the disparity information is within the phase-locked range of the primary visual neurons, which is consistent with the physiological basis of binocular single vision.

[0078] For the separated state, calculate the angle between the lines of sight based on the filtered coordinates. The expression is:

[0079] ;

[0080] In the formula, It is the arctangent function, i.e., the arctan function. For operations on the Euclidean norm (or L2 norm), Interpupillary distance, This represents the X-coordinate of the center of the left pupil after filtering. This represents the X-coordinate of the center of the right pupil after filtering. This represents the Y-coordinate of the center of the left pupil after filtering. This represents the Y-coordinate of the center of the right pupil after filtering. This is a spatial coordinate identifier representing the center of the left pupil. This is the spatial coordinate identifier for the center of the right pupil;

[0081] Based on the angle of the visual axis The angle-error conversion method is used to convert the line of sight angle. The binocular fusion error values ​​are converted into dimensionless binocular fusion error values. The conversion process uses a preset proportional relationship to map the angle values ​​to standardized error values. After the conversion is completed, the binocular fusion error values ​​corresponding to all time points are integrated and output in chronological order using a time series aggregation method to form a binocular fusion error value sequence.

[0082] It should be noted that the preset proportional relationship refers to the conversion rule that linearly maps the visual axis angle to a standardized error value by a fixed multiple; the fixed multiple is determined by the sensitivity characteristics of the visual nerve to changes in angle.

[0083] Based on the continuous change trend of binocular fusion error value over time, the attenuation amplitude of neural response intensity is extracted and a neural response intensity attenuation index is generated.

[0084] Specifically, the binocular fusion error value sequence is divided into M overlapping sub-windows. Least squares fitting is performed on the binocular fusion error value sequence within the M overlapping sub-window, and the fitting slope is... M represents the total number of overlapping sub-windows;

[0085] If the slope A value greater than zero indicates that the fusion error continues to increase, which is determined to be a decline in neural response.

[0086] Calculate all slopes The window is used to calculate the average attenuation intensity value. The expression is:

[0087] ;

[0088] ;

[0089] In the formula, Represents a set The number of elements in the middle. Indicates that belonging to the set All Sum the corresponding values. For the index of the time window, It is all that satisfy the slope Time window index The resulting subset is used to filter out time periods in which the intensity of neural responses shows a decreasing trend;

[0090] The average decay intensity value is exponentially normalized to output the neural response intensity decay index. .

[0091] The superior frequency domain decomposition and spatial clustering algorithm, compared with the traditional eye-tracking analysis method that mainly relies on spatial coordinate differences, can extract richer binocular coordination information from time-frequency domain features. It uses the frequency synchronization of binocular visual cortex signals as the core criterion for fusion state, providing a more explanatory quantitative evaluation dimension at the neurophysiological level.

[0092] The adjustment mechanism dynamically sets the neural response entrainment threshold based on the binocular fusion error value and personalized adjustment factors.

[0093] The binocular fusion error value is mapped to the neural response entrainment threshold range through a mapping function, and the mapping function is adjusted using a personalized modulation factor.

[0094] Specifically, the neural response entrainment threshold range is preset to... ;

[0095] Define the basic mapping function as:

[0096] ;

[0097] In the formula, The entrainment threshold of the basic neural response, and These represent the preset minimum and maximum values ​​of the neural response entrainment threshold, respectively. This represents the mean value of binocular fusion error during the current treatment session. and These represent the minimum and maximum binocular fusion errors recorded in the historical neural response dataset, respectively.

[0098] It needs to be explained that, and The range of neural response entrainment thresholds for all patients in the historical neural response dataset is determined by statistical distribution. Fixed boundary values ​​are set by analyzing the percentile distribution of the population data. The population data refers to the set of neural response entrainment thresholds from the historical treatment session records of all patients in the pre-stored database, which includes the actual measurement values ​​of different individuals at different treatment stages.

[0099] Adjust the underlying mapping function using a personalized adjustment factor when When the mapping result is the same as the underlying mapping, when... When, compress the mapping range (suitable for highly sensitive patients); when At this time, the mapping range is expanded (for patients with low sensitivity).

[0100] Check the adjusted neural response entrainment threshold Does it exceed the preset effective range? ;

[0101] like Less than This will entrain the threshold in the neural response. Set as ;

[0102] like Less than This will entrain the threshold in the neural response. Set as ;

[0103] like exist If it is within the range, then use it directly. As a neural response entrainment threshold ;

[0104] It should be noted that the final effective value of the neural response entrainment threshold is , These are merely transitional variables in dynamic calculations.

[0105] The control module triggers the audio synthesizer to generate a gamma-band sound signal when the neural response intensity decay index exceeds the neural response entrainment threshold. Otherwise, it maintains the baseline dynamic target parameter set and continues to monitor the neural response intensity decay index.

[0106] When the detection of the neural response intensity decay index exceeding the neural response entrainment threshold is detected, a trigger command is issued and the audio synthesizer is started to output a gamma band sound wave signal.

[0107] Specifically, comparing the attenuation index of neural response intensity. Relationship with the magnitude of the neural response entrainment threshold;

[0108] like If the neural response decay is determined to be beyond the acceptable range, a trigger command is generated.

[0109] like If the neural response is determined to be in a stable state, the audio synthesizer is not triggered, the baseline dynamic target parameter set is maintained, and the process returns to step S2 to continue collecting real-time pupil movement data.

[0110] When the condition is met, a digital trigger signal (TTL pulse) is sent to the audio synthesizer.

[0111] After receiving the trigger digital trigger signal, the audio synthesizer generates a sine wave signal with a frequency in the γ band (e.g., 40 Hz); the duration of the sine wave signal is aligned with the duration parameter in the reference dynamic target parameter set.

[0112] The amplitude of the sine wave signal is dynamically adjusted according to a personalized adjustment factor. The dynamically adjusted sine wave signal is output in real time through headphones or speakers, while the trigger timestamp is recorded. At the same time as triggering, the reference dynamic target parameter set continues to run to ensure the spatiotemporal synchronization of the sound wave and the electronic target picture.

[0113] The preferred approach is to lock the phase of the gamma-band sound wave with the flashing frequency of the visual target to form a neural entrainment coupling effect. Compared with existing technologies that rely on single visual stimulation, this approach improves the efficiency of neural plasticity induction through multi-sensory synergistic modulation. Furthermore, by constructing a closed-loop feedback through real-time attenuation index monitoring and dynamic threshold adjustment, it significantly enhances the inhibitory effect on neural response attenuation and forms a continuously optimized data-driven treatment paradigm.

[0114] The communication module locks the phase of the gamma-band acoustic signal with the flicker frequency of the electronic visual target, creating a neural entrainment coupling effect. It records the change in the attenuation index of the neural response intensity under the influence of this entrainment coupling effect and updates the historical neural response dataset.

[0115] Acquire the phase information of the gamma-band acoustic signal. Acquire the flicker frequency of the electronic visual target image.

[0116] Specifically, waveform data of the γ-band acoustic signal is acquired in real time from the output of the audio synthesizer. The instantaneous phase value of the γ-band acoustic signal is extracted using a zero-crossing detection algorithm. The zero-crossing detection algorithm calculates the phase value by identifying the crossing points of the acoustic signal from negative to positive. Each crossing point corresponds to a zero-phase point, and the instantaneous phase between adjacent crossing points is calculated by linear interpolation. At the same time, the flicker frequency parameters of the electronic target picture are extracted from the reference dynamic target parameter set. The flicker frequency parameters are provided by the reference dynamic target parameter set and continuously drive the operation of the co-viewing unit.

[0117] The phase information of the gamma-band acoustic signal is synchronized with the flashing frequency of the electronic visual target image through a phase-locked loop, maintaining the phase-locked state and forming a neural entrainment coupling effect.

[0118] Specifically, a hybrid analog-digital phase-locked loop (PLL) is used for synchronization control. The PLL consists of a phase detector, a loop filter, and a voltage-controlled oscillator (VCO). The phase detector compares the phase difference between the gamma-band acoustic signal and the electronic target picture's flashing signal, outputting an error voltage signal. This error voltage signal is smoothed by the loop filter and then input to the VCO. The VCO adjusts the phase output of the gamma-band acoustic signal, ensuring that the gamma-band acoustic signal continuously tracks the phase of the electronic target picture's flashing signal. When the phase difference remains stable, a neural entrainment coupling effect is formed, at which point the gamma-band acoustic signal and the electronic target picture's flashing signal achieve phase synchronization.

[0119] Record the changes in the neural response intensity decay index under the neural entrainment coupling effect. Write the changes in the neural response intensity decay index into the historical neural response dataset and store them in chronological order.

[0120] Specifically, during phase-lock maintenance, the changes in the neural response intensity decay index are monitored in real time, and the difference between the neural response intensity decay index before and after the onset of the neural entrainment coupling effect is calculated as the change. The change is calculated using the difference between the current neural response intensity decay index and the neural response intensity decay index before the onset of the neural entrainment coupling effect.

[0121] The changes are associated with the current timestamp and patient identification code, and the historical neural response dataset in the pre-stored database is updated by appending to it. The storage format includes the timestamp, the change in the neural response intensity decay index, the corresponding stimulation frequency parameter, and the efficacy improvement rate identifier field. The historical neural response dataset adopts a time-series storage structure, with new data records appended to the end of the existing data to form a time-series set.

[0122] The present invention also provides a synoptophore electronic target device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of the aforementioned synoptophore electronic target system.

[0123] In summary, this invention generates personalized regulatory factors and a baseline dynamic target parameter set through a neuroplasticity model, achieving precise adaptation to the patient's neural characteristics; it calculates binocular fusion error and neural response intensity decay index based on frequency domain decomposition and spatial clustering algorithms, extracting rich binocular coordination information from time-frequency domain features; by locking the phase of the γ-band sound wave signal with the flicker frequency of the electronic target picture, a neural entrainment coupling effect is formed, which can effectively enhance the synchronous firing activity of visual cortex neurons; it improves the efficiency of neuroplasticity induction by utilizing a multi-sensory synergistic stimulation mechanism, and updates the historical dataset by recording the changes in the neural response intensity decay index in real time, constructing a data-driven closed-loop optimization mechanism to achieve continuous and stable regulation of neural excitability and iterative improvement of treatment effects.

[0124] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A synoptophore electronic visual target system, characterized in that: The application relates to a visual system for treating amblyopia, comprising, a storage module, which calls historical neural response data sets in a pre-stored database through a patient identification code, inputs the historical neural response data sets into a neural plasticity model, and outputs an individualized adjustment factor and a benchmark dynamic target parameter set; a high-precision display unit, which drives an optokinetic drum to generate an electronic target picture through the benchmark dynamic target parameter set and synchronously collects real-time pupil movement data; a data analysis module, which performs frequency domain decomposition and spatial clustering on the real-time pupil movement data based on the individualized adjustment factor, calculates a binocular fusion error value and a neural response intensity attenuation index; an adjustment mechanism, which dynamically sets a neural response entrainment threshold according to the binocular fusion error value and the individualized adjustment factor; a control module, which triggers an audio synthesizer to generate a gamma-band sound wave signal when the neural response intensity attenuation index exceeds the neural response entrainment threshold; otherwise, the control module maintains the benchmark dynamic target parameter set to continuously operate and continues to monitor the neural response intensity attenuation index; a communication module, which locks the phase of the gamma-band sound wave signal and the flicker frequency of the electronic target picture to form a neural entrainment coupling effect, records the change amount of the neural response intensity attenuation index under the action of the neural entrainment coupling effect, and updates the historical neural response data sets.

2. The synoptophore electronic target system of claim 1, wherein: The historical neural response data sets comprise neural response data recorded in previous treatment sessions; the neural response data comprises fatigue sensitivity, stimulation frequency and curative effect improvement rate.

3. The synoptophore electronic target system of claim 1, wherein: The output of the individualized adjustment factor and the benchmark dynamic target parameter set comprises the following specific steps, preprocessing the historical neural response data sets; training the neural plasticity model by using a machine learning algorithm to learn the mode of the neural response data; inputting the neural response data corresponding to the patient identification code into the trained neural plasticity model to output the individualized adjustment factor and the benchmark dynamic target parameter set.

4. The synoptophore electronic target system of claim 1, wherein: The benchmark dynamic target parameter set comprises the flicker frequency, spatial coordinates, dynamic change mode and duration parameters of the electronic target picture.

5. The synoptophore electronic target system of claim 1, wherein: The calculation of the binocular fusion error value and the neural response intensity attenuation index comprises the following specific steps, filtering the real-time pupil movement data by using the individualized adjustment factor; performing frequency domain decomposition on the filtered real-time pupil movement data to extract frequency components; performing spatial clustering on the frequency components to identify a binocular movement mode and calculate the binocular fusion error value; extracting the attenuation amplitude of the neural response intensity according to the continuous change trend of the binocular fusion error value in time and generating the neural response intensity attenuation index.

6. The synoptophore electronic target system of claim 1, wherein: The dynamic setting of the neural response entrainment threshold according to the binocular fusion error value and the individualized adjustment factor comprises mapping the binocular fusion error value to the neural response entrainment threshold range through a mapping function and adjusting the mapping function by using the individualized adjustment factor.

7. The synoptophore electronic target system of claim 1, wherein: The triggering of the audio synthesizer to generate the gamma-band sound wave signal refers to that a trigger instruction is sent out and the audio synthesizer is started when it is monitored that the neural response intensity attenuation index exceeds the neural response entrainment threshold, and the gamma-band sound wave signal is output.

8. The synoptophore electronic target system of claim 1, wherein: The formation of the neural entrainment coupling effect comprises the following specific steps, obtaining phase information of the gamma-band sound wave signal and obtaining the flicker frequency of the electronic target picture; The phase information of the gamma-band acoustic wave signal is synchronized with the flicker frequency of the electronic target picture through a phase-locked loop, a phase-locked state is maintained, and a neural entrainment coupling effect is formed.

9. The synoptophore electronic target system of claim 1, wherein: The updated historical neural response data set refers to a change amount of a neural response intensity decay index under the neural entrainment coupling effect; the change amount of the neural response intensity decay index is written into the historical neural response data set and is stored in a time sequence.

10. An electronic synoptophore vision target device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that, The computer program, when executed by a processor, implements the steps of the synoptophore electronic target system according to any one of claims 1-9.

Images