Intelligent intervention system for preventing and controlling myopia in teenagers based on dynamic axial monitoring

Abstract

This invention relates to the field of axial length data processing technology, specifically to an intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring. The invention uses normally labeled samples of the same sex as reference samples for the samples to be analyzed, obtaining multiple axial length time periods. Based on the axial length distribution of the samples to be analyzed and the reference samples within each axial length time period, the degree of visual impairment for each sample within each axial length time period, and the critical value for the degree of impairment of non-impaired samples, are obtained. Based on the distribution of visual impairment degree and axial length of the samples to be analyzed and impaired samples within each axial length time period, overall similar samples are obtained. Based on the distribution of visual impairment degree of overall similar samples within different axial length time periods, and the critical value for the degree of impairment, dynamic axial length monitoring is performed on the samples to be analyzed. This invention improves the accuracy of dynamic axial length monitoring by accurately analyzing the degree of visual impairment and the critical value for the degree of impairment of the samples.

Description

Technical Field

[0001] This invention relates to the field of axial length data processing technology, specifically to an intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring. Background Technology

[0002] Because adolescents frequently experience eye strain, and their axial length is still developing, they are more prone to myopia. Therefore, it is necessary to monitor the axial length of adolescents' eyes to allow for timely myopia control and prevent further progression. Dynamic axial length monitoring, which relies on historical changes in axial length, provides more accurate myopia control compared to single-sample monitoring. Therefore, dynamic axial length monitoring is a crucial element in myopia control for adolescents.

[0003] In existing technologies, axial length monitoring is achieved based on age curves that conform to the standard of axial length development. The degree to which the dynamic axial length of the sample exceeds the standard curve is used to determine myopia in adolescents. However, it is possible that vision may deteriorate but still remain within the normal range, resulting in untimely prevention and control of myopia. Alternatively, if the adolescent already has myopia, the axial length monitoring effect may be poor. Summary of the Invention

[0004] To address the technical problem that relying solely on analyzing axial length data to determine if it falls within the normal range leads to poor axial length monitoring results, this invention aims to provide an intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring. The specific technical solution adopted is as follows: This invention proposes an intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps: The axial length of the adolescent sample at each moment, including a preset myopia label, and the detection label, which is a normal label or an abnormal label; For the sample to be analyzed, the same-sex sample with normal labeling is used as the reference sample. Based on the myopia label distribution of axial length of the sample to be analyzed and the reference sample at different times, multiple axial length time periods are obtained for each sample to be analyzed. Based on the axial length distribution of the sample to be analyzed and the reference sample in each axial length time period, the axial growth rate of each sample at each time is obtained. For each axial length time interval, non-deteriorating samples are selected based on the axial growth rate of the sample to be analyzed and the reference sample at different times; based on the distribution of the axial growth rate of the sample to be analyzed and the non-deteriorating samples at different times, the degree of visual deterioration of each sample in each axial length time interval is obtained, as well as the deterioration threshold of the non-deteriorating samples; based on the distribution of visual deterioration of the sample to be analyzed and the non-deteriorating samples in each axial length time interval, and the axial length distribution, overall similar samples are obtained. Based on the distribution of visual deterioration in similar samples across different time periods and the critical value of deterioration, dynamic axial length monitoring was performed on the samples to be analyzed.

[0005] Furthermore, the method for obtaining the axial length time period includes: The maximum and minimum values ​​of axial length for all reference samples at each time step are obtained to form the baseline range of axial length at each time step. If the axial length of the sample to be analyzed is within the baseline range of axial length at each time point, and the preset myopia label is the preset first label value, the corresponding time point will be regarded as a normal time point; otherwise, the corresponding time point will be regarded as an abnormal time point. A series of normal moments constitutes a normal time period for the axial length, while a series of abnormal moments constitutes an abnormal time period for the axial length, thus forming multiple time periods for the axial length.

[0006] Furthermore, the method for obtaining the axial length growth rate includes: For the sample to be analyzed and the reference sample, obtain the axial length curve of each sample at all times within each axial length time period, and take the derivative of the axial length curve at each time point as the axial growth rate of each sample at each time point.

[0007] Furthermore, the method for obtaining the undeteriorated samples and deteriorated samples includes: For each time period of axial length, the reliability of visual deterioration for each sample is obtained based on the axial length growth rate of the sample to be analyzed and the reference sample at different times. If the confidence level of a sample’s vision deterioration is less than or equal to a preset probability threshold, the corresponding sample is considered as a non-deteriorated sample; other samples besides the non-deteriorated samples are considered as deteriorated samples.

[0008] Furthermore, the method for obtaining the reliability of the vision deterioration includes: For each time interval of the axial length, a box plot is used based on the axial growth rate of all samples at each time point. If the axial growth rate of a sample at each time point is not within the box plot, the minimum difference between the axial growth rate and the growth rate corresponding to the boundary in the box plot is calculated as the first deterioration coefficient of the corresponding sample at each time point; otherwise, the first deterioration coefficient corresponding to the axial growth rate at each time point is set to 0. Obtain the first deterioration coefficient sequence with consecutive non-zero values ​​in chronological order, and take the first deterioration coefficient sequence with the largest number of corresponding elements in all first deterioration coefficient sequences as the target sequence. The reliability of visual deterioration of the corresponding sample is obtained based on the number of elements in the target sequence and the first deterioration coefficient. Both the number of elements and the first deterioration coefficient are positively correlated with the reliability of visual deterioration.

[0009] Furthermore, the method for obtaining the degree of vision deterioration includes: For each axial length time period, obtain the first growth rate curve composed of the axial length growth rate of each sample at different times; obtain the second growth rate curve composed of the mean axial length growth rate of the non-deteriorated samples at different times; obtain the difference between the first growth rate curve and the second growth rate curve at each time point to form a difference curve. The difference curve is integrated within each axial length time interval, and the integration result is used as the degree of visual deterioration for each sample in each axial length time interval.

[0010] Furthermore, the method for obtaining the deterioration threshold includes: The fluctuation characteristics of visual acuity deterioration in different undeteriorated samples within each axial time period were obtained as the degree of visual acuity deterioration fluctuation; the mean of visual acuity deterioration in different undeteriorated samples within each axial time period was obtained as the level of visual acuity deterioration. The sum of the degree of visual impairment fluctuation and the level of visual impairment is used as the critical value for the degree of impairment.

[0011] Furthermore, the method for obtaining the overall similar samples includes: Based on the distribution of visual deterioration degree and axial length of different samples at each time interval, the optimized similarity of axial length between the sample to be analyzed and the deteriorated sample at each time interval is obtained. The relative distance between the sample to be analyzed and the deteriorated sample is obtained as a sequence of visual deterioration degree at different times within each axial length time period, and negative correlation mapping is performed as the deterioration similarity between the corresponding samples; if the deterioration similarity is greater than the preset similarity threshold, the corresponding sample is regarded as the deterioration similar sample. The mean value of the optimized similarity of the axial length between the sample to be analyzed and the deteriorated sample is obtained in each time period. If the mean value of the optimized similarity of the axial length is greater than the preset similarity threshold, the corresponding sample is regarded as a sample with similar axial length. For any given time period of axial length, obtain the same samples from all similar samples of deterioration and similar samples of axial length, and use them as the overall similar samples.

[0012] Furthermore, the method for obtaining the optimized similarity of the axial length includes: The relative distance between the sample to be analyzed and the deteriorated sample is obtained by the sequence of axial lengths at different times within each axial length time interval, and negative correlation mapping is performed as the axial similarity between the corresponding samples; the fluctuation characteristics of the degree of visual deterioration of the non-deteriorated sample within each axial length time interval are obtained as the normal fluctuation degree of deterioration within each axial length time interval. For any given time period of axial length, the ratio of axial length similarity to the degree of normal fluctuation in deterioration is obtained and normalized as the axial length optimization similarity.

[0013] Furthermore, the relative distance is obtained by calculating the Euclidean distance.

[0014] The present invention has the following beneficial effects: This invention uses normally labeled samples of the same sex as reference samples for the samples to be analyzed. Based on the myopia label distribution of axial length at different times for the samples to be analyzed and the reference samples, multiple axial length time periods are obtained for each sample. Instead of statically analyzing the entire period, the focus is on the time periods in which changes occur. Based on the axial length distribution of the samples to be analyzed and the reference samples within each axial length time period, the axial length growth rate for each sample at each time point is obtained to assess the rate of axial change. For any given axial length time period, based on the axial length growth rates of the samples to be analyzed and the reference samples at different times, samples that have not deteriorated and samples that have deteriorated are selected, and potential samples from the reference samples are identified. This invention employs samples that are present or about to deteriorate to ensure the normality of the control group. Based on the axial length growth rate distribution of the samples to be analyzed and those that have not deteriorated at different times, the degree of visual deterioration for each sample within each axial length time interval is obtained, along with the deterioration threshold for non-deteriorated samples, which helps quantify the normal deterioration boundary. Based on the distribution of visual deterioration degree and axial length distribution of different samples within each axial length time interval, overall similar samples are obtained, and core reference samples most comparable to the samples to be analyzed are selected. Based on the distribution of visual deterioration degree of overall similar samples within different axial length time intervals and the deterioration threshold, dynamic axial length monitoring is performed on the samples to be analyzed. This invention improves the accuracy of dynamic axial length monitoring by accurately analyzing the degree of visual deterioration and the deterioration threshold of the samples. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. 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.

[0016] Figure 1A flowchart illustrating the implementation method of an intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring, as provided in one embodiment of the present invention; Figure 2 A flowchart illustrating a method for obtaining the reliability of vision deterioration according to an embodiment of the present invention; Figure 3 This is a flowchart illustrating a method for obtaining overall similar samples according to an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring, as proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details a specific scheme for an intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring, provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a flowchart of an implementation method for an intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring, according to an embodiment of the present invention. The method specifically includes: Step S1: Obtain the axial length of the adolescent sample at each moment, including the preset myopia label, and the detection label, which is a normal label or an abnormal label.

[0021] In an embodiment of the present invention, considering that relying solely on whether the axial length data of adolescents is within the normal range may result in untimely prevention and control, and thus the results of monitoring the axial length of adolescents are inaccurate; therefore, information on the trend of axial length change is analyzed; firstly, adolescents are regularly organized to undergo physical examinations to obtain the axial length of the adolescents at each moment of the physical examination, and the axial length of the adolescent samples at each moment including the preset myopia label, as well as the detection label, are obtained.

[0022] It should be noted that, in the embodiments of the present invention, for adolescent samples, the obtained axial length is analyzed in a traditional manner based on a single axial length monitoring. If the adolescent's axial length is within the normal range of the standard axial length, the preset myopia label is set to 0; otherwise, the preset myopia label is set to 1. Furthermore, if the preset myopia label of the adolescent is 0 at all times, the adolescent is given a normal label; otherwise, the adolescent is given an abnormal label. The normal range of the standard axial length can be obtained in advance based on relevant professional data.

[0023] It should be noted that, in one embodiment of the present invention, the axial length is obtained and analyzed once a year, i.e., the time interval is 1 year; in other embodiments of the present invention, the time interval can be set according to specific circumstances, and will not be limited or described in detail here.

[0024] Step S2: For the sample to be analyzed, the corresponding normally labeled same-sex sample is used as the reference sample. Based on the myopia label distribution of the axial length of the sample to be analyzed and the reference sample at different times, multiple axial length time periods are obtained for each sample to be analyzed. Based on the axial length distribution of the sample to be analyzed and the reference sample in each axial length time period, the axial growth rate of each sample at each time is obtained.

[0025] To avoid the influence of gender on axial length and interference with the data processing, the analysis was conducted on adolescents of the same sex; for the samples to be analyzed, the corresponding normally labeled samples of the same sex were used as reference samples.

[0026] Normally, as people age, the axial length of the eye gradually increases, causing the focus to gradually shift forward. This results in normal vision followed by vision abnormalities over time. Therefore, normal and abnormal conditions do not alternate frequently. Based on the myopia label distribution of axial length at different times for the sample to be analyzed and the reference sample, multiple axial length time periods are obtained for each sample to be analyzed.

[0027] Preferably, in one embodiment of the present invention, the method for obtaining the axial length time period includes: The maximum and minimum values ​​of axial length for all reference samples at each time step are obtained to form the baseline range of axial length at each time step. If the axial length of the sample to be analyzed is within the baseline range of axial length at each time point, and the preset myopia label is the preset first label value, the corresponding time point will be regarded as a normal time point; otherwise, the corresponding time point will be regarded as an abnormal time point. A series of normal moments constitutes a normal time period for the axial length, while a series of abnormal moments constitutes an abnormal time period for the axial length, thus forming multiple time periods for the axial length.

[0028] It should be noted that, in one embodiment of the present invention, the size of the preset first tag value is set to 1.

[0029] Since the rate of increase of the axial length curve gradually decreases with time, the possibility of subsequent myopia risk is smaller. Therefore, the axial length growth rate is analyzed. Based on the axial length distribution of the sample to be analyzed and the reference sample in each axial length time period, the axial length growth rate of each sample at each time is obtained.

[0030] Preferably, in one embodiment of the present invention, a flowchart of the method for obtaining the axial length growth rate is provided; For the sample to be analyzed and the reference sample, obtain the axial length curve of each sample at all times within each axial length time period, and take the derivative of the axial length curve at each time point as the axial growth rate of each sample at each time point.

[0031] It should be noted that the derivative reflects the trend of changes in axial length. The larger the derivative, the greater the growth rate of axial length. The specific methods are well known to those skilled in the art and will not be elaborated here.

[0032] Step S3: For any axial length time period, based on the axial length growth rate of the sample to be analyzed and the reference sample at different times, filter out the non-deteriorating samples and the deteriorating samples; based on the axial length growth rate distribution of the sample to be analyzed and the non-deteriorating samples at different times, obtain the degree of visual deterioration of each sample in each axial length time period, and the deterioration threshold of the non-deteriorating samples; based on the distribution of visual deterioration degree of different deteriorating samples in each axial length time period, and the axial length distribution, obtain the overall similar samples.

[0033] As the samples aged, the axial length growth rate showed a decreasing trend. By analyzing the axial length growth rate at different times, the relationship between axial length growth and time can be reflected, allowing for a more accurate capture of the axial length growth trend and avoiding interference from abnormal growth. For any given axial length time period, based on the axial length growth rates of the sample to be analyzed and the reference sample at different times, samples that did not deteriorate and samples that deteriorated were selected.

[0034] Preferably, in one embodiment of the present invention, the method for obtaining undegraded samples and degraded samples includes: For each time period of axial length, the reliability of visual deterioration for each sample is obtained based on the axial length growth rate of the sample to be analyzed and the reference sample at different times. Preferably, in one embodiment of the present invention, the method for obtaining the reliability of vision deterioration is described in [reference needed]. Figure 2 It illustrates a flowchart of a method for obtaining the reliability of vision deterioration, including: Step S201: For any eye axis time period, use a box plot based on the eye axis growth rate of all samples at each time point. If the eye axis growth rate of a sample at each time point is not within the box plot, calculate the minimum difference between the eye axis growth rate and the growth rate corresponding to the boundary in the box plot, and use it as the first deterioration coefficient of the corresponding sample at each time point; otherwise, set the first deterioration coefficient of the eye axis growth rate at the corresponding time point to 0.

[0035] It should be noted that a box plot is a standard graphical representation method for displaying the distribution of a set of data. The upper and lower limits represent the reasonable range of the data distribution. The corresponding data can be obtained by calculating the upper and lower quartiles of the axial length growth rate. Data falling outside the upper and lower limits are considered to be abnormal. The specific methods are well known to those skilled in the art and will not be elaborated here.

[0036] Step S202: Obtain the first deterioration coefficient sequence with consecutive non-zero values ​​in chronological order, and obtain the first deterioration coefficient sequence with the largest number of corresponding elements in all first deterioration coefficient sequences as the target sequence.

[0037] The more consecutive non-zero first deterioration coefficients there are, the more anomalies there are, and the more they reflect the deterioration situation.

[0038] Step S203: Based on the number of elements in the target sequence and the first deterioration coefficient, obtain the credibility of vision deterioration of the corresponding sample. The number of elements and the first deterioration coefficient are both positively correlated with the credibility of vision deterioration.

[0039] It should be noted that the more elements there are, the larger the first deterioration coefficient is, the better it reflects the deterioration situation, and the greater the possibility of deterioration; therefore, both the number of elements and the first deterioration coefficient are positively correlated with the reliability of vision deterioration.

[0040] In one embodiment of the present invention, the ratio between the number of elements in the target sequence and the number of elements in the corresponding time period of the axial length is obtained, which reflects the proportion of time when the performance deteriorates. The higher the proportion, the greater the credibility of the deterioration. The mean of all elements in the target sequence is obtained, which reflects the overall deterioration level of the target sequence. The higher the overall deterioration level, the greater the credibility of the deterioration. Therefore, the product of the ratio result and the mean of the elements is obtained, which reflects the credibility of the visual deterioration of the corresponding sample.

[0041] The greater the credibility of vision deterioration, the better it indicates the extent of vision deterioration. The more obvious the degree of deterioration, the more likely it is to be a sample that has deteriorated. Clear, non-deteriorated samples can be selected for reference.

[0042] If the confidence level of a sample’s vision deterioration is less than or equal to a preset probability threshold, the corresponding sample is considered as a non-deteriorated sample; other samples besides the non-deteriorated samples are considered as deteriorated samples.

[0043] It should be noted that, in one embodiment of the present invention, the preset probability threshold is 0.9; in other embodiments of the present invention, the preset probability threshold may be set according to specific circumstances, which will not be elaborated here.

[0044] By excluding the influence of normal axial length growth through the axial growth rate of non-deteriorated samples, and combining the changes in axial length of samples with time, the range of deterioration fluctuations is quantified; based on the distribution of axial growth rate of the samples to be analyzed and non-deteriorated samples at different times, the degree of visual deterioration of each sample within each axial length time period is obtained, as well as the critical value of the degree of deterioration of non-deteriorated samples.

[0045] Preferably, in one embodiment of the present invention, the method for obtaining the degree of visual impairment includes: For each axial length time period, obtain the first growth rate curve composed of the axial length growth rate of the sample to be analyzed at different times; obtain the second growth rate curve composed of the mean axial length growth rate of the non-deteriorated sample at different times; obtain the difference between the first growth rate curve and the second growth rate curve at each time point to form a difference curve. The difference curve is integrated within each axial length time interval, and the integration result is used as the degree of visual deterioration for each sample in each axial length time interval.

[0046] Preferably, the degree of deterioration is affected by age; the younger the age, the greater the impact on the axial length and the greater the change in the rate of change of the axial length. As age increases, the axial length becomes relatively fixed, and the fluctuation of deterioration decreases. In one embodiment of the present invention, the method for obtaining the critical value of the degree of deterioration includes: The fluctuation characteristics of visual acuity deterioration in different undeteriorated samples within each axial time period were obtained as the degree of visual acuity deterioration fluctuation; the mean of visual acuity deterioration in different undeteriorated samples within each axial time period was obtained as the level of visual acuity deterioration. The sum of the degree of visual impairment fluctuation and the level of visual impairment is used as the critical value for the degree of impairment.

[0047] It should be noted that, in one embodiment of the present invention, the fluctuation characteristics are reflected by calculating the standard deviation. The larger the standard deviation, the greater the fluctuation characteristics, and the smaller the standard deviation, the smaller the fluctuation characteristics. In other embodiments of the present invention, the fluctuation characteristics are reflected by calculating the variance or range. The specific means are well known to those skilled in the art and will not be described in detail here.

[0048] To reduce the error in the analysis, we selected adolescent samples from groups similar to the sample to be analyzed, who showed similar vision deterioration in the same time period. Such samples have greater reference value. Based on the distribution of vision deterioration degree and axial length of different samples in each time period, we obtained overall similar samples.

[0049] Preferably, in one embodiment of the present invention, the method for obtaining overall similar samples is described in [reference needed]. Figure 3 It illustrates a flowchart of a method for obtaining overall similar samples, including: Step S301: Based on the distribution of visual deterioration degree and axial length of different samples in each time period, obtain the optimized similarity of axial length between the sample to be analyzed and the deteriorated sample in each time period.

[0050] Preferably, in one embodiment of the present invention, the method for obtaining axial length optimization similarity includes: The relative distance between the sample to be analyzed and the deteriorated sample is obtained by the sequence of axial lengths at different times within each axial length time interval, and negative correlation mapping is performed as the axial similarity between the corresponding samples; the fluctuation characteristics of the degree of visual deterioration of the non-deteriorated sample within each axial length time interval are obtained as the normal fluctuation degree of deterioration within each axial length time interval. For any given time period of axial length, the ratio of axial length similarity to the degree of normal fluctuation in deterioration is obtained and normalized as the axial length optimization similarity.

[0051] It should be noted that, in the embodiments of the present invention, the relative distance is calculated using Euclidean distance or Manhattan distance, or by taking the reciprocal or an exponential function with the natural constant as the base. Negative correlation mapping is performed using techniques well-known to those skilled in the art, which will not be elaborated here.

[0052] It should be noted that, in the embodiments of the present invention, normalization is performed by linear normalization or a normalization function. The specific means are well known to those skilled in the art and will not be described in detail here.

[0053] Step S302: Obtain the relative distance between the sample to be analyzed and the deteriorated sample at different times within each axial length time interval, and perform negative correlation mapping as the deterioration similarity between corresponding samples; if the deterioration similarity is greater than the preset similarity threshold, the corresponding sample is regarded as the deterioration similar sample. The mean value of the optimized similarity of the axial length between the sample to be analyzed and the deteriorated sample is obtained at each time interval. If the mean value of the optimized similarity of the axial length is greater than the preset similarity threshold, the corresponding sample is regarded as a sample with similar axial length.

[0054] Based on this, the sample to be analyzed and each deteriorated sample were analyzed to screen out samples with similar deterioration and samples with similar axial length.

[0055] It should be noted that, in one embodiment of the present invention, the preset similarity threshold is set to 0.9; in other embodiments of the present invention, the size of the preset similarity threshold can be set according to specific circumstances, and will not be limited or elaborated here.

[0056] Step S303: For any time period of eye axis, obtain the same samples among all deterioration similar samples and eye axis length similar samples, and use them as the overall similar samples.

[0057] Based on this, the identical samples among the deterioration-similar samples and axial length-similar samples obtained can help analyze the changes in axial length of the sample to be analyzed.

[0058] Step S4: Based on the distribution of visual acuity deterioration in similar samples over different time periods and the critical value of deterioration, perform dynamic axial length monitoring on the samples to be analyzed.

[0059] It should be noted that dynamic monitoring of axial length includes: for any given axial length time period, obtaining the average degree of visual deterioration of all overall similar samples, obtaining the degree of visual deterioration fluctuation, and obtaining the sum of the average degree of visual deterioration and the degree of visual deterioration fluctuation of the overall similar samples as the overall degree of deterioration; By comparing the overall degree of deterioration and the critical value of deterioration across different axial length time periods, if the overall degree of deterioration across all axial length time periods is less than the critical value, the axial length growth of the sample under analysis is suppressed, which is a normal trend of axial length change. If the overall degree of deterioration across all axial length time periods is greater than the critical value, the axial length growth of the sample under analysis does not conform to the trend of decreasing with time, indicating an abnormality in the axial length growth of the corresponding sample, requiring myopia prevention and control reminders. For other cases with inconsistent values, the degree difference between the overall degree of deterioration and the critical value of deterioration is calculated. If the degree difference of the latest axial length time period is greater than that of the previous axial length time period, the overall degree of deterioration is greater, and the change in axial length is less consistent with the characteristic of slowing down with time, requiring myopia prevention and control. Conversely, if the degree difference of the latest axial length time period is not greater than that of the previous axial length time period, axial length growth is suppressed, and maintaining good eye habits is recommended.

[0060] Based on this, by analyzing similar samples as a whole, the accuracy of the analysis of adolescent samples to be analyzed can be improved, which helps to provide accurate and reliable myopia prevention and control.

[0061] In summary, this invention uses normally labeled samples of the same sex as reference samples for the samples to be analyzed, obtaining multiple axial length time periods. Based on the axial length distribution of the samples to be analyzed and the reference samples within each axial length time period, it obtains the degree of visual deterioration for each sample within each axial length time period, as well as the deterioration threshold for non-deteriorated samples. Based on the distribution of visual deterioration degree and axial length of the samples to be analyzed and deteriorated samples within each axial length time period, it obtains overall similar samples. Based on the distribution of visual deterioration degree of overall similar samples within different axial length time periods, and the deterioration threshold, it performs dynamic axial length monitoring on the samples to be analyzed. This invention improves the accuracy of dynamic axial length monitoring by accurately analyzing the degree of visual deterioration and the deterioration threshold of the samples.

[0062] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0063] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A smart intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it performs the following steps: The axial length of the adolescent sample at each moment, including a preset myopia label, and the detection label, which is a normal label or an abnormal label; For the sample to be analyzed, the same-sex sample with normal labeling is used as the reference sample. Based on the myopia label distribution of axial length of the sample to be analyzed and the reference sample at different times, multiple axial length time periods are obtained for each sample to be analyzed. Based on the axial length distribution of the sample to be analyzed and the reference sample in each axial length time period, the axial growth rate of each sample at each time is obtained. For any given time period of eye axis, non-deteriorating and deteriorating samples are selected based on the axial growth rate of the sample to be analyzed and the reference sample at different times; based on the distribution of the axial growth rate of the sample to be analyzed and the non-deteriorating samples at different times, the degree of visual deterioration of each sample in each time period of eye axis is obtained, as well as the critical value of the degree of deterioration of the non-deteriorating samples; based on the distribution of the degree of visual deterioration of different deteriorating samples in each time period of eye axis, as well as the distribution of axial length, the overall similar samples are obtained. Based on the distribution of visual deterioration in similar samples across different axial length time periods, and the critical value of deterioration, dynamic axial length monitoring was performed on the samples to be analyzed.

2. The intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring according to claim 1, characterized in that, The method for obtaining the axial length time period includes: The maximum and minimum values ​​of axial length for all reference samples at each time step are obtained to form the baseline range of axial length at each time step. If the axial length of the sample to be analyzed is within the baseline range of axial length at each time point, and the preset myopia label is the preset first label value, the corresponding time point will be regarded as a normal time point; otherwise, the corresponding time point will be regarded as an abnormal time point. A series of normal moments constitutes a normal time period for the axial length, while a series of abnormal moments constitutes an abnormal time period for the axial length, thus forming multiple time periods for the axial length.

3. The intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring according to claim 1, characterized in that, The method for obtaining the axial length growth rate includes: For the sample to be analyzed and the reference sample, obtain the axial length curve of each sample at all times within each axial length time period, and take the derivative of the axial length curve at each time point as the axial growth rate of each sample at each time point.

4. The intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring according to claim 1, characterized in that, The methods for obtaining the undeteriorated samples and deteriorated samples include: For each time period of axial length, the reliability of visual deterioration for each sample is obtained based on the axial length growth rate of the sample to be analyzed and the reference sample at different times. If the confidence level of a sample’s vision deterioration is less than or equal to a preset probability threshold, the corresponding sample is considered as a non-deteriorated sample; other samples besides the non-deteriorated samples are considered as deteriorated samples.

5. The intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring according to claim 4, characterized in that, The methods for obtaining the reliability of the vision deterioration include: For each time interval of the axial length, a box plot is used based on the axial growth rate of all samples at each time point. If the axial growth rate of a sample at each time point is not within the box plot, the minimum difference between the axial growth rate and the growth rate corresponding to the boundary in the box plot is calculated as the first deterioration coefficient of the corresponding sample at each time point; otherwise, the first deterioration coefficient corresponding to the axial growth rate at each time point is set to 0. Obtain the first deterioration coefficient sequence with consecutive non-zero values ​​in chronological order, and take the first deterioration coefficient sequence with the largest number of corresponding elements in all first deterioration coefficient sequences as the target sequence. The reliability of visual deterioration of the corresponding sample is obtained based on the number of elements in the target sequence and the first deterioration coefficient. Both the number of elements and the first deterioration coefficient are positively correlated with the reliability of visual deterioration.

6. The intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring according to claim 1, characterized in that, The methods for obtaining the degree of vision deterioration include: For each axial length time period, obtain the first growth rate curve composed of the axial length growth rate of each sample at different times; obtain the second growth rate curve composed of the mean axial length growth rate of the non-deteriorated samples at different times; obtain the difference between the first growth rate curve and the second growth rate curve at each time point to form a difference curve. The difference curve is integrated within each axial length time interval, and the integration result is used as the degree of visual deterioration for each sample in each axial length time interval.

7. The intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring according to claim 1, characterized in that, The method for obtaining the critical value of the degree of deterioration includes: The fluctuation characteristics of visual acuity deterioration in different undeteriorated samples within each axial time period were obtained as the degree of visual acuity deterioration fluctuation; the mean of visual acuity deterioration in different undeteriorated samples within each axial time period was obtained as the level of visual acuity deterioration. The sum of the degree of visual impairment fluctuation and the level of visual impairment is used as the critical value for the degree of impairment.

8. The intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring according to claim 1, characterized in that, The method for obtaining the overall similar samples includes: Based on the distribution of visual deterioration degree and axial length of different samples at each time interval, the optimized similarity of axial length between the sample to be analyzed and the deteriorated sample at each time interval is obtained. The relative distance between the sample to be analyzed and the deteriorated sample is obtained as a sequence of visual deterioration degree at different times within each axial length time period, and negative correlation mapping is performed as the deterioration similarity between the corresponding samples; if the deterioration similarity is greater than the preset similarity threshold, the corresponding sample is regarded as the deterioration similar sample. The mean value of the optimized similarity of the axial length between the sample to be analyzed and the deteriorated sample is obtained in each time period. If the mean value of the optimized similarity of the axial length is greater than the preset similarity threshold, the corresponding sample is regarded as a sample with similar axial length. For any given time period of axial length, obtain the same samples from all similar samples of deterioration and similar samples of axial length, and use them as the overall similar samples.

9. The intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring according to claim 8, characterized in that, The method for obtaining the optimized similarity of the axial length includes: The relative distance between the sample to be analyzed and the deteriorated sample is obtained by the sequence of axial lengths at different times within each axial length time interval, and negative correlation mapping is performed as the axial similarity between the corresponding samples; the fluctuation characteristics of the degree of visual deterioration of the non-deteriorated sample within each axial length time interval are obtained as the normal fluctuation degree of deterioration within each axial length time interval. For any given time period of axial length, the ratio of axial length similarity to the degree of normal fluctuation in deterioration is obtained and normalized as the axial length optimization similarity.

10. The intelligent intervention system for myopia prevention and control in adolescents based on dynamic axial length monitoring according to claim 8, characterized in that, The relative distance is obtained by calculating the Euclidean distance.

Images