Method and system for identifying abnormal brain development trajectory

By constructing a generalized additive location-scale-shape model and using offset correction techniques, the problems of offset in multi-center data fusion and limited evaluation dimensions were solved, enabling accurate identification and individualized quantification of abnormal brain development trajectories and providing intuitive identification reports.

CN121527007APending Publication Date: 2026-02-13ZHEJIANG XINGYU BRAIN TECHNOLOGY CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202511615674.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-11-06
Publication Date
2026-02-13

AI Technical Summary

Technical Problem

Existing technologies suffer from systematic bias in multi-center data fusion, limited dimensions and accuracy in brain development trajectory model assessment, lack of standardized scoring systems, and difficulty in accurately identifying abnormal brain development trajectories.

Method used

By acquiring T1-weighted magnetic resonance images of healthy individuals, performing offset correction processing, constructing a generalized additive location-scale-shape model of the healthy population, generating normative developmental trajectory curves and their normal variation ranges, and based on this, performing structural index correction and abnormality quantification on the individuals to be evaluated, generating an identification report.

Benefits of technology

It achieves accurate correction of image structure indicators from different imaging centers, can comprehensively identify abnormal brain development trajectories of individuals to be evaluated, provides individualized abnormality quantification and intuitive identification reports, and supports clinical applications and multi-center collaboration.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121527007A_ABST
    Figure CN121527007A_ABST
Patent Text Reader

Abstract

The invention discloses a brain abnormal development trajectory identification method and system, and belongs to the technical field of magnetic resonance image analysis, and the method comprises the following steps: obtaining T1 weighted magnetic resonance images of a plurality of healthy individuals, and carrying out the offset correction of the structure index of each brain region in the images, a generalized additive position-scale-shape model of the healthy crowd is constructed; generating a norm development trajectory curve and a normal change range thereof; obtaining a T1 weighted magnetic resonance image of a to-be-evaluated individual, and selecting a target brain region of the to-be-evaluated individual; obtaining various offset corrected structure indexes of the target brain region of the individual to be evaluated; obtaining a mean value and a standard deviation of each structure index at the position of the brain region corresponding to the same-age healthy population of the individual to be evaluated; and generating a recognition report of the abnormal brain development trajectory of the to-be-evaluated individual. The problem that it is difficult to accurately, clearly and visually reflect the abnormal recognition condition of the brain trajectory of the individual to be evaluated and the position of the corresponding abnormal brain region is solved.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of magnetic resonance image analysis, and particularly relates to a method and system for identifying abnormal brain development trajectories. BACKGROUND

[0002] With the progress of neuroimaging and artificial intelligence technology, brain development trajectory models have become a core tool for studying the dynamic changes of human brain structure. By analyzing the changes of key structural indicators such as brain cortical thickness, surface area, and gray matter volume with age, the model reveals that brain development has significant nonlinear characteristics and regional-specific differences.

[0003] However, there are still many bottlenecks in the actual application of brain development trajectory models. First, there are fundamental obstacles in multi-center data fusion: differences in magnetic resonance imaging equipment models, scanning protocols, and image preprocessing processes between different medical institutions result in systematic bias in brain structure indicators. Second, the evaluation dimension and accuracy of brain development trajectory models are limited, and most of them only focus on a single indicator such as cortical thickness, making it difficult to integrate multiple key structural characteristics such as cortical curvature and thickness variability, and unable to fully depict the development state of brain regions; at the same time, the modeling process is usually static and cannot be dynamically updated with new samples. Third, existing researches mostly stop at scientific chart output, lack of standardized scoring system for clinical interpretation, and do not establish automatic personalized parameter report mechanism, making it difficult to support efficient identification of abnormal brain regions and understanding of evaluation results. SUMMARY

[0004] In view of the above deficiencies in the prior art, the present application provides a method and system for identifying abnormal brain development trajectories, which solves the problem of accurately, clearly and intuitively reflecting the abnormal identification of the brain trajectory of the individual to be evaluated and the corresponding abnormal brain region position through multi-center data offset correction, multi-dimensional norm development trajectory construction and individualized abnormal quantification.

[0005] In order to achieve the above-mentioned application purposes, the technical scheme adopted by the present application is as follows: On the one hand, the present application provides a method for identifying abnormal brain development trajectories, comprising the following steps: S1, acquiring T1-weighted magnetic resonance images of a plurality of healthy individuals, and performing offset correction processing on the structural indicators of each brain region in the images to construct a generalized additive position-scale-shape model of healthy population; S2, generating norm development trajectory curves and their normal variation ranges based on the generalized additive position-scale-shape model of healthy population; S3, acquiring T1-weighted magnetic resonance images of the individual to be evaluated, and selecting a target brain region of the individual to be evaluated; S4, offset correction is performed on each structural index of the target brain region of the individual to be evaluated, to obtain each offset-corrected structural index of the target brain region of the individual to be evaluated; S5, based on the norm development trajectory curve and its normal variation range, the mean and standard deviation of each structural index at the corresponding brain region position of the individual to be evaluated are obtained; S6, based on the offset-corrected structural index of the target brain region of the individual to be evaluated, the mean and standard deviation of each structural index at the corresponding brain region position of the individual to be evaluated, the identification report of the abnormal development trajectory of the brain of the individual to be evaluated is generated.

[0006] Further, the S1 comprises the following steps: S11, selecting T1-weighted magnetic resonance images of a plurality of healthy individuals from a plurality of imaging centers as control samples; S12, center labeling, structural index extraction, modeling and offset correction are performed on each control sample to obtain offset-corrected structural index of the control sample; S13, taking the age of the control sample as the independent variable, the offset-corrected structural index of the control sample as the dependent variable, and fitting the modeling based on the age distribution trend for each offset-corrected structural index of each brain region, to obtain a generalized additive position-scale-shape model for constructing a healthy population.

[0007] Further, the S12 comprises the following steps: S121, converting each control sample into a preset image format as an initial sample; S122, adding the modeling covariates corresponding to each initial sample to the initial sample to obtain a plurality of labeled samples, wherein the modeling covariates include age identifier, gender identifier and imaging center identifier; S123, extracting structural indexes of each labeled sample according to the brain region map, wherein the structural indexes include cortical surface area, average curvature, thickness standard deviation, average thickness and gray matter volume; S124, according to the generalized additive position-scale-shape model, each structural index of the labeled sample conforms to a four-parameter distribution model; S125, based on the four-parameter distribution model, the structural indexes of each labeled sample from different imaging centers are mapped to the same distribution using the generalized additive position-scale-shape model, to obtain the offset-corrected structural index of the control sample.

[0008] Further, the calculation expression of the four-parameter distribution model in S124 is as follows: , , , , , wherein, denotes the th structure indicator of the th annotated sample, denotes subject to, denotes a distribution function, denotes the th structure indicator of the th annotated sample, denotes the th structure indicator of the th annotated sample, denotes the th structure indicator of the th annotated sample, denotes the th structure indicator of the th annotated sample, denotes the random effect value of the imaging center on the mean of the th structure indicator of the th annotated sample, denotes the th structure indicator of the th annotated sample, denotes the th structure indicator of the th annotated sample, denotes the random effect value of the imaging center on the standard deviation of the th structure indicator of the th annotated sample, denotes a normal distribution function satisfying the mean variance parameter estimation, denotes a normal distribution function satisfying the standard deviation parameter estimation, wherein, .

[0009] Further, the calculation expression of the structure indicator after offset correction in the S125 is as follows: , wherein, denotes the th structure indicator after offset correction of the th annotated sample, denotes the the imaging center offset of the structure index of the i-th labeled sample after the item offset correction, the imaging center offset standard deviation of the structure index of the i-th labeled sample after the item offset correction, the imaging center offset of the structure index of the i-th labeled sample after the item offset correction, the imaging center offset standard deviation of the structure index of the i-th labeled sample after the item offset correction, the imaging center offset of the structure index of the i-th labeled sample after the item offset correction, the imaging center offset standard deviation of the structure index of the i-th labeled sample after the item offset correction, the imaging center offset standard deviation of the structure index of the i-th labeled sample after the item offset correction, the imaging center offset standard deviation of the structure index of the i-th labeled sample after the item offset correction. Further, the S2 comprises the following steps:

[0010] S21, obtaining the mean function and the standard deviation function of each structure index with respect to age based on the generalized additive location-scale-shape model of the healthy population; S22, generating the normal development trajectory curve and its normal variation range based on the mean function and the standard deviation function of each structure index with respect to age. Further, the normal development curve in the S22 is composed of the mean of each structure index with respect to age;

[0011] The calculation expression of the normal variation range of the normal development curve is as follows: wherein, the normal variation range of the normal development curve, the structure index belonging to, the mean function of the structure index with respect to age, the standard deviation function of the structure index with respect to age.

[0012] Further, the S6 comprises the following steps: S61, constructing the target brain region structure index comparison histogram of the individual to be evaluated and the healthy population of the same age based on each structure index of the target brain region of the individual to be evaluated after the item offset correction and the mean of each structure index at the corresponding brain region position of the healthy population of the same age of the individual to be evaluated; S62, calculating the Z score of each structure index of the target brain region of the individual to be evaluated according to each structure index of the target brain region of the individual to be evaluated after the item offset correction and the mean and the standard deviation of each structure index at the corresponding brain region position of the healthy population of the same age of the individual to be evaluated; S63, the range between the difference and the sum of the mean of each structure index at the corresponding brain region position of the healthy population of the same age of the individual to be evaluated and 2 times the standard deviation is correspondingly taken as the normal variation range of the structure index of the healthy population of the same age of the individual to be evaluated; ​S64. If any structural index of the target brain region of the individual to be evaluated is not within the normal range of structural index variation of the same age healthy population, the target brain region is spatially located and highlighted to obtain the spatial coordinates of the abnormal brain region and the abnormal brain region annotation map. S65. By combining the target brain region structural index comparison bar chart, the Z-scores of various structural indices of the target brain region of the individual to be evaluated, the spatial coordinates of the abnormal brain region, and the abnormal brain region annotation map, a report on the identification of abnormal brain development trajectory of the individual to be evaluated is generated.

[0013] Furthermore, the calculation expressions for the Z scores of various structural indicators of the target brain region of the individual to be evaluated in S62 are as follows: , in, This indicates the first brain region of the individual to be evaluated. Z-score of the item structure index This indicates the first target brain region of the individual to be evaluated. Structural indices after offset correction This indicates the number of healthy individuals of the same age in the target brain region at the corresponding location. The mean of the item structure index, This indicates the number of healthy individuals of the same age in the target brain region at the corresponding location. The standard deviation of the item structure index, This indicates the age of the individual to be evaluated.

[0014] The beneficial effects of this invention are as follows: This invention provides a method for identifying abnormal brain development trajectories. By acquiring T1-weighted magnetic resonance images of several healthy individuals and performing offset correction processing on the structural indicators of each brain region in the images, it effectively and accurately corrects the systematic offsets of image structural indicators caused by differences in equipment models, scanning protocols, and preprocessing procedures among different imaging centers, providing a foundation for unified data analysis and fusion. Based on the structural indicators of different brain regions after multiple offset corrections, this invention constructs a generalized additive location-scale-shape model of healthy individuals, and based on this generalized additive location-scale-shape model of healthy individuals, it generates… This invention establishes a norm-based developmental trajectory curve and its normal range of variation, taking into account multiple key structural indicators of brain regions and covering norm-based developmental trajectories across a wide age range. This enables accurate and comprehensive identification of the degree of abnormality in the target brain region for the individual being evaluated. By calculating the Z-scores of various structural indicators in the target brain region of the individual being evaluated, this invention quantifies the degree of deviation of the individual being evaluated from the norm trajectory in the target brain region. By generating an identification report of the abnormal brain development trajectory of the individual being evaluated, this invention enables a direct comparison of the brain development trajectory of the individual being evaluated with that of healthy individuals of the same age, providing a foundation for improving effective communication in clinical applications and multi-center collaborations.

[0015] In another aspect, the present application also provides a system for identifying abnormal brain development trajectory, comprising: A first module for acquiring T1-weighted magnetic resonance images of a plurality of healthy individuals, and performing offset correction on structural indicators of each brain region in the images to construct a generalized additive position-scale-shape model of the healthy population; A second module for generating a normative development trajectory curve and its normal variation range based on the generalized additive position-scale-shape model of the healthy population; A third module for acquiring T1-weighted magnetic resonance images of the individual to be evaluated, and selecting a target brain region of the individual to be evaluated; A fourth module for performing offset correction on each structural indicator of the target brain region of the individual to be evaluated to obtain each offset-corrected structural indicator of the target brain region of the individual to be evaluated; A fifth module for obtaining the mean and standard deviation of each structural indicator at the corresponding brain region position of the healthy population of the same age as the individual to be evaluated based on the normative development trajectory curve and its normal variation range; A sixth module for generating an identification report of the abnormal brain development trajectory of the individual to be evaluated based on the offset-corrected structural indicators of the target brain region of the individual to be evaluated, and the mean and standard deviation of each structural indicator at the corresponding brain region position of the healthy population of the same age as the individual to be evaluated.

[0016] The system for identifying abnormal brain development trajectory provided by the present application is a system corresponding to the above-mentioned method for identifying abnormal brain development trajectory, and is used to implement the above-mentioned method for identifying abnormal brain development trajectory, so as to accurately, clearly and intuitively reflect the abnormal identification situation of the brain trajectory of the individual to be evaluated and the corresponding abnormal brain region position.

[0017] Other advantages of the present application will be analyzed in more detail in the subsequent embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0019] Figure 1 The flow chart of the steps of the method for identifying abnormal brain development trajectory in the embodiments of the present application.

[0020] Figure 2 The schematic diagram of the target brain region of the individual to be evaluated in the embodiments of the present application.

[0021] Figure 3 This is a bar chart comparing the cortical surface area of ​​the target brain region of the individual to be evaluated with the average cortical surface area of ​​the corresponding region in the target brain region of healthy individuals of the same age, as shown in this embodiment of the invention.

[0022] Figure 4 This is a schematic diagram showing the relative position of the cortical surface area of ​​the target brain region of the individual to be evaluated and the norm developmental trajectory curve and its normal range of variation in an embodiment of the present invention.

[0023] Figure 5 This is a block diagram of a brain abnormal development trajectory recognition system according to an embodiment of the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0025] like Figure 1 As shown, in one embodiment of the present invention, the present invention provides a method for identifying abnormal brain development trajectories, comprising the following steps: S1. Obtain T1-weighted magnetic resonance images of several healthy individuals and perform offset correction on the structural indicators of each brain region in the images to construct a generalized additive location-scale-shape model of the healthy population. S1 includes the following steps: S11. Select T1-weighted magnetic resonance images of several healthy individuals from multiple imaging centers as control samples; in this embodiment, the age of each healthy individual is between 3 and 40 years old, covering the age-dependent developmental trajectory of the entire adolescence and early adulthood.

[0026] S12. Perform center labeling, structural index extraction, modeling, and offset correction on each control sample to obtain the offset-corrected structural index of the control sample. S12 includes the following steps: S121. Convert each control sample into a preset image format and use it as the initial sample; S122. Add the modeling covariates corresponding to each initial sample to the initial sample to obtain several labeled samples. The modeling covariates include age identifier, gender identifier and imaging center identifier. In this embodiment, the preset image format is a standard medical image format, either DICOM (Digital Imaging and Communications in Medicine) or NIfTI (Neuroimaging Information Science and Technology Initiative) format. DICOM format is an international standard format in the field of medical imaging, developed by the National Electrical Manufacturers Association (NEMA). It not only stores image data but also includes metadata such as individual information, scan parameters, and device model, ensuring image compatibility and traceability across different systems. NIfTI format is a format specifically designed for neuroimaging, supporting 3D and 4D (time-series) data. Its simple file structure facilitates efficient storage and processing.

[0027] S123. Based on the brain region map, extract the structural indicators of each labeled sample. The structural indicators include cortical surface area, mean curvature, thickness standard deviation, mean thickness, and gray matter volume. In this scheme, FreeSurfer software is used to automatically process the standard brain region maps of all labeled samples to extract structural indicators from the labeled samples. The extracted structural indicators from the labeled samples can form a sample feature matrix to facilitate batch processing of structural indicator values ​​extracted from different objects. Each row in the sample feature matrix corresponds to a structural indicator value extracted from a T1-weighted magnetic resonance image.

[0028] S124. According to the generalized additive position-scale-shape model, let all structural indices of the labeled samples follow a four-parameter distribution model. The calculation expression for the four-parameter distribution model in S124 is as follows: , , , , , in, Indicates the first The first labeled sample Item structure indicators, Indicates obedience, Represents the distribution function. Indicates the first The first labeled sample Conditional mean of the item structure index. Indicates the first The first labeled sample Conditional standard deviation of the item structure index Indicates the first The first labeled sample The mean function of the age component of the item structure indicator. Indicates the first The first labeled sample Mean function of the gender identifier component of the item structure index. Indicates the imaging center relative to the first The first labeled sample The random effects value of the mean of the item structure index, Indicates the first The first labeled sample The standard deviation function of the age component of the item structure index. Indicates the first The first labeled sample The standard deviation function of the gender component of the item structure index. Indicates the imaging center relative to the first The first labeled sample The random effects value of the standard deviation of the item structure index. Indicates that the mean and variance parameters are satisfied. The estimated normal distribution function, Indicates that the standard deviation parameter is satisfied. The estimated normal distribution function, where, .

[0029] In this scheme, the conditional mean, conditional standard deviation, conditional skewness, and conditional kurtosis of the labeled sample structure indicators are not simple sample statistics, but rather dynamic parametric functions based on covariates such as age, gender, and imaging center, modeled using a generalized additive location-scale-shape model. These functions reflect the systematic influence of the covariates on the distribution characteristics. The structural index is the average thickness. The time-structure index is the mean curvature. The structural index is the standard deviation of thickness. The structural index at that time is the cortical surface area. The time-structure index is the gray matter volume.

[0030] , , and These are smoothing functions for modeling the mean and standard deviation, respectively, used to capture the nonlinear effects and central differences of covariates such as age and gender on structural indicators. Age and gender, as basic demographic information, are essential synchronous data acquired during T1-weighted magnetic resonance imaging. and The specific imaging center identifier is the random effect term of the mean and standard deviation of the structural index of each imaging center. That is, the systematic bias caused by different imaging centers, which is an error source that is not directly concerned but must be modeled and corrected. They all follow a zero-mean normal distribution. and It is a variance parameter automatically estimated by a generalized additive position-scale-shape model, which can quantify the difficulty of systematic shifts in mean and standard deviation from different acquisition centers. Based on these two variance parameters, the random effect value of each imaging center can be calculated to... and Inclusion criteria parameters and Modeling was performed to achieve batch correction of data structure indicators for multiple imaging centers.

[0031] S125. Based on the four-parameter distribution model, the structural indices of each labeled sample from different imaging centers are mapped to the same distribution using the generalized additive position-scale-shape model, and the offset-corrected structural indices of the control sample are obtained.

[0032] The calculation expression for the offset-corrected structural parameters in S125 is as follows: , in, Indicates the first The first labeled sample Structural indices after offset correction Indicates the first The first labeled sample Mean image center offset of structural parameters after offset correction. Indicates the first The first labeled sample The standard deviation of the imaging center offset of the structural parameters after offset correction. Represents the first of all labeled samples The standard deviation of the imaging center reference offset of the structural parameters after offset correction. Represents the first of all labeled samples Mean of the imaging center reference offset of the structural parameters after offset correction.

[0033] In this plan, and The distribution parameters of the structural indicators of the brain regions are modeled by the generalized additive location-scale-shape model in compliance with the unified target distribution, and the mean difference and standard deviation of the subjects in different imaging centers can be adjusted respectively to realize the standardization processing of the structural indicators. The variance of the structural indicators after offset correction between different centers will be significantly reduced. For example, the mean square error of the average thickness indicators between two imaging centers is reduced from 0.092 to 0.021, thereby effectively eliminating the systematic offset caused by the differences in imaging equipment and processing flow.

[0034] S13, taking the age corresponding to the control sample as the independent variable, the offset corrected structural indicators of the control sample as the dependent variable, and each item of offset corrected structural indicators of each brain region based on the age distribution trend to fit and model, to obtain the generalized additive location-scale-shape model of the healthy population.

[0035] S2, generating the norm development trajectory curve and its normal variation range based on the generalized additive location-scale-shape model of the healthy population; The S2 comprises the following steps: S21, obtaining the mean function and standard deviation function of each structural indicator with age based on the generalized additive location-scale-shape model of the healthy population; S22, generating the norm development trajectory curve and its normal variation range based on the mean function and standard deviation function of each structural indicator with age.

[0036] The norm development curve in the S22 is composed of the mean value of each structural indicator with age; The calculation expression of the normal variation range of the norm development curve is as follows: , Wherein, represents the normal variation range of the norm development curve, represents the norm development curve, represents the mean function of the structural indicator with age, represents the standard deviation function of the structural indicator with age.

[0037] In the present scheme, the norm development trajectory curve and its normal variation range represent the development trajectory curve and the normal variation range of the healthy population corresponding to each brain region and structural indicator. In the present scheme, the range between the mean function with age plus or minus 2 times the standard deviation function with age is used as the normal variation range, which can represent the 95% confidence interval and is an important reference for measuring whether the brain development of an individual is abnormal.

[0038] S3, acquiring the T1 weighted magnetic resonance image of the to-be-evaluated individual, and selecting the target brain region of the to-be-evaluated individual; For example, Figure 2In the embodiment shown, the middle-posterior cingulate gyrus and sulcus of the individual to be evaluated is selected as the target brain region; S4. Offset correction is performed on each structural index of the target brain region of the individual to be evaluated to obtain each offset-corrected structural index of the target brain region of the individual to be evaluated. In this scheme, the offset correction method for each structural index of the target brain region of the individual to be evaluated is the same as the offset correction method for the brain region structural index of the T1-weighted magnetic resonance image of the healthy individual in S1.

[0039] In this scheme, the corrected structural index of the individual to be evaluated is a standardized value after the central effect is eliminated, which can ensure that the corrected structural index of the individual to be evaluated and the normative developmental trajectory curve are in a unified distribution scale, so that the individual data from different imaging centers are comparable.

[0040] S5. Based on the normative developmental trajectory curve and its normal variation range, the mean and standard deviation of each structural index at the corresponding brain region position of the healthy population of the same age as the individual to be evaluated are obtained. S6. Based on the offset-corrected structural index of the target brain region of the individual to be evaluated and the mean and standard deviation of each structural index at the corresponding brain region position of the healthy population of the same age as the individual to be evaluated, a recognition report of the abnormal brain development trajectory of the individual to be evaluated is generated.

[0041] The S6 includes the following steps: S61. Based on each offset-corrected structural index of the target brain region of the individual to be evaluated and the mean of each structural index at the corresponding brain region position of the healthy population of the same age as the individual to be evaluated, a target brain region structural index comparison bar chart of the individual to be evaluated and the healthy population of the same age is constructed. As Figure 3 In the embodiment shown, based on a large-scale healthy control sample of 3 to 40 years old, an age-dependent developmental trajectory covering the entire puberty and early adulthood is constructed for five structural indexes. For the cortical surface area of the middle-posterior cingulate gyrus and sulcus of the individual to be evaluated, a comparison bar chart is constructed with the mean of the healthy population of the same age, that is, the cortical surface area of the individual to be evaluated at the target brain region is 427 mm 2 , which is significantly lower than the average value 639.030 mm 2 of the healthy population of the same age, and the difference is obvious. This target brain region structural index comparison bar chart can be structured to derive and intuitively and accurately display the abnormal brain trajectory of the individual to be evaluated.

[0042] S62. Based on the offset-corrected structural indices of the target brain region of the individual to be evaluated, and the mean and standard deviation of the structural indices at the corresponding brain region locations of the healthy individuals of the same age, calculate the Z-scores of the structural indices of the target brain region of the individual to be evaluated. The Z-scores for the various structural indices of the target brain region of the individual to be evaluated in S62 are calculated as follows: , in, This indicates the first target brain region of the individual to be evaluated. Z-score of the item structure index This indicates the first target brain region of the individual to be evaluated. Structural indices after offset correction This indicates the number of healthy individuals of the same age in the target brain region at the corresponding location. The mean of the item structure index, This indicates the number of healthy individuals of the same age in the target brain region at the corresponding location. The standard deviation of the item structure index, This indicates the age of the individual to be evaluated.

[0043] In this embodiment, the mean and standard deviation of the structural indicators of the individual to be evaluated in the target brain region corresponding to healthy individuals of the same age refer to the mean and standard deviation of the cortical surface area, mean curvature, thickness standard deviation, mean thickness, and gray matter volume of the individual to be evaluated in the target brain region corresponding to healthy individuals of the same age.

[0044] In this scheme, the Z-score reflects the degree of deviation of the structural index of the individual being evaluated from that of healthy individuals of the same age. The larger the absolute value of the Z-score, the higher the degree of abnormality of the brain region structure of the individual being evaluated in that age group. In this embodiment, the Z-score of the cortical surface area of ​​the target brain region of the individual being evaluated was calculated to be -4.46, realizing the quantification and localization of structural abnormalities.

[0045] S63. The range between the difference between the mean and twice the standard deviation of the various structural indicators at the corresponding brain regions of the individual to be evaluated and the sum of these values ​​shall be taken as the normal range of variation of the structural indicators of the individual to be evaluated in the same healthy population of the same age. In this protocol, the normal range of variation for the individual being evaluated within a healthy population of the same age refers to the normal range of variation corresponding to the individual's age within the norm developmental trajectory curve and its normal range of variation. .

[0046] S64. If any structural index of the target brain region of the individual to be evaluated is not within the normal range of structural index variation of the same age healthy population, the target brain region is spatially located and highlighted to obtain the spatial coordinates of the abnormal brain region and the abnormal brain region annotation map. like Figure 4 As shown, in this embodiment, the individual to be evaluated is 6 years old, and the cortical surface area of ​​the target brain region is 427.00 mm². 2 Based on the norm-based developmental trajectory curves, it was found that the average cortical surface area at the corresponding location of the target brain region in healthy individuals of the same age was 1021.746445876289 mm². 2 The average cortical surface area at the corresponding location of the target brain region in a healthy group of the same age, plus two standard deviations, is used as the upper bound of the normal range of variation of the norm developmental trajectory curve: 1252.4853677491833 mm. 2 The lower bound of the normal range of variation in the developmental trajectory curve is calculated as follows: the average cortical surface area at the corresponding location of the target brain region in a healthy group of the same age, minus two standard deviations. (791.0075240033949 mm) 2 .based on Figure 4 It can detect the lower bound of the normal range of changes in the individual being evaluated that significantly deviate from the norm developmental trajectory.

[0047] S65. By combining the target brain region structural index comparison bar chart, the Z-scores of various structural indices of the target brain region of the individual to be evaluated, the spatial coordinates of the abnormal brain region, and the abnormal brain region annotation map, a report on the identification of abnormal brain development trajectory of the individual to be evaluated is generated.

[0048] like Figure 5 As shown, in another embodiment of the present invention, the present invention provides a system for recognizing abnormal brain development trajectories, comprising: The first module is used to acquire T1-weighted magnetic resonance images of several healthy individuals and perform offset correction processing on the structural indicators of each brain region in the images in order to construct a generalized additive location-scale-shape model of the healthy population. The second module is used to generate normative developmental trajectory curves and their normal range of variation based on a generalized additive location-scale-shape model of healthy individuals. The third module is used to acquire T1-weighted magnetic resonance images of the individual to be evaluated and to select the target brain region of the individual to be evaluated. The fourth module is used to perform offset correction on various structural indicators of the target brain region of the individual to be evaluated, so as to obtain the offset-corrected structural indicators of the target brain region of the individual to be evaluated. The fifth module is used to obtain the mean and standard deviation of various structural indicators at the corresponding brain regions of the individual to be evaluated in healthy peers based on the norm-model developmental trajectory curve and its normal range of variation. The sixth module is used to generate an identification report of abnormal brain development trajectories for the individual being evaluated, based on the offset-corrected structural indicators of the target brain region of the individual being evaluated, and the mean and standard deviation of various structural indicators at the corresponding brain region locations in healthy individuals of the same age as the individual being evaluated.

[0049] The brain abnormal development trajectory recognition system is a system corresponding to the brain abnormal development trajectory recognition method, and is used for implementing the brain abnormal development trajectory recognition method to accurately, clearly and intuitively reflect the abnormal recognition situation of the brain trajectory of the individual to be evaluated and the corresponding abnormal brain region position.

[0050] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for identifying abnormal brain development trajectories, characterized in that, Includes the following steps: S1. Obtain T1-weighted magnetic resonance images of several healthy individuals and perform offset correction on the structural indicators of each brain region in the images to construct a generalized additive location-scale-shape model of the healthy population. S2. Based on the generalized additive position-scale-shape model of healthy individuals, generate normative developmental trajectory curves and their normal range of variation; S3. Obtain T1-weighted magnetic resonance images of the individual to be evaluated and select the target brain region of the individual to be evaluated; S4. Perform offset correction on the various structural indicators of the target brain region of the individual to be evaluated to obtain the offset-corrected structural indicators of the target brain region of the individual to be evaluated. S5. Based on the norm-model developmental trajectory curve and its normal range of variation, obtain the mean and standard deviation of various structural indicators at the corresponding brain regions of the individual to be evaluated in the same age group of healthy people. S6. Based on the offset-corrected structural indicators of the target brain region of the individual to be evaluated, and the mean and standard deviation of various structural indicators at the corresponding brain region locations of the healthy individuals of the same age as the individual to be evaluated, a report on the identification of abnormal brain development trajectories of the individual to be evaluated is generated.

2. The method for identifying abnormal brain development trajectories according to claim 1, characterized in that, S1 includes the following steps: S11. Select T1-weighted magnetic resonance images of several healthy individuals from multiple imaging centers as control samples; S12. Perform center labeling, structural index extraction, modeling, and offset correction on each control sample to obtain the offset-corrected structural index of the control sample. S13. Using the age of the control sample as the independent variable and the offset-corrected structural index of the control sample as the dependent variable, and fitting and modeling each offset-corrected structural index of each brain region based on the age distribution trend, a generalized additive location-scale-shape model of healthy individuals is obtained.

3. The method for identifying abnormal brain development trajectories according to claim 2, characterized in that, S12 includes the following steps: S121. Convert each control sample into a preset image format and use it as the initial sample; S122. Add the modeling covariates corresponding to each initial sample to the initial sample to obtain several labeled samples. The modeling covariates include age identifier, gender identifier and imaging center identifier. S123. Based on the brain region map, extract the structural indicators of each labeled sample. The structural indicators include cortical surface area, mean curvature, thickness standard deviation, mean thickness, and gray matter volume. S124. According to the generalized additive position-scale-shape model, let all structural indices of the labeled samples follow a four-parameter distribution model. S125. Based on the four-parameter distribution model, the structural indices of each labeled sample from different imaging centers are mapped to the same distribution using the generalized additive position-scale-shape model, and the offset-corrected structural indices of the control sample are obtained.

4. The method for identifying abnormal brain development trajectories according to claim 3, characterized in that, The calculation expression for the four-parameter distribution model in S124 is as follows: , , , , , in, Indicates the first The first labeled sample Item structure indicators, Indicates obedience, Represents the distribution function. Indicates the first The first labeled sample Conditional mean of the item structure index. Indicates the first The first labeled sample Conditional standard deviation of the item structure index Indicates the first The first labeled sample The mean function of the age component of the item structure indicator. Indicates the first The first labeled sample Mean function of the gender identifier component of the item structure index. Indicates the imaging center relative to the first The first labeled sample The random effects value of the mean of the item structure index. Indicates the first The first labeled sample The standard deviation function of the age component of the item structure index. Indicates the first The first labeled sample The standard deviation function of the gender component of the item structure index. Indicates the imaging center relative to the first The first labeled sample The random effects value of the standard deviation of the item structure index. Indicates that the mean and variance parameters are satisfied. The estimated normal distribution function, Indicates that the standard deviation parameter is satisfied. The estimated normal distribution function, where, .

5. The method for identifying abnormal brain development trajectories according to claim 3, characterized in that, The calculation expression for the offset-corrected structural parameters in S125 is as follows: , in, Indicates the first The first labeled sample Structural indices after offset correction Indicates the first The first labeled sample The mean image center offset of structural parameters after offset correction. Indicates the first The first labeled sample The standard deviation of the imaging center offset of the structural parameters after offset correction. Represents the first of all labeled samples The standard deviation of the imaging center reference offset of the structural parameters after offset correction. Represents the first of all labeled samples Mean value of the imaging center reference offset of the structural parameters after offset correction.

6. The method for identifying abnormal brain development trajectories according to claim 3, characterized in that, S2 includes the following steps: S21. Based on the generalized additive location-scale-shape model of healthy individuals, the mean function and standard deviation function of each structural index as a function of age are obtained. S22. Based on the mean function and standard deviation function of the changes of each structural index with age, generate the norm developmental trajectory curve and its normal range of change.

7. The method for identifying abnormal brain development trajectories according to claim 6, characterized in that, The normative developmental curve in S22 is composed of the average values ​​of each structural index as a function of age. The formula for calculating the normal range of variation of the norm development curve is as follows: in, This indicates the normal range of variation in the normative developmental curve. Indicates belonging to, This represents the mean function of structural indicators as a function of age. The standard deviation function representing the change of structural indicators with age.

8. The method for identifying abnormal brain development trajectories according to claim 7, characterized in that, S6 includes the following steps: S61. Based on the structural indicators of the target brain region of the individual to be evaluated after offset correction, and the mean values ​​of the structural indicators of the corresponding brain regions of the healthy population of the same age as the individual to be evaluated, construct a comparative bar chart of the structural indicators of the target brain regions of the individual to be evaluated and the healthy population of the same age. S62. Based on the offset-corrected structural indices of the target brain region of the individual to be evaluated, and the mean and standard deviation of the structural indices at the corresponding brain region locations of the healthy individuals of the same age, calculate the Z-scores of the structural indices of the target brain region of the individual to be evaluated. S63. The range between the difference between the mean and twice the standard deviation of the various structural indicators at the corresponding brain regions of the individual to be evaluated and the sum of these values ​​shall be taken as the normal range of variation of the structural indicators of the individual to be evaluated in the same healthy population of the same age. S64. If any structural index of the target brain region of the individual to be evaluated is not within the normal range of structural index variation of the same age healthy population, the target brain region is spatially located and highlighted to obtain the spatial coordinates of the abnormal brain region and the abnormal brain region annotation map. S65. By combining the target brain region structural index comparison bar chart, the Z-scores of various structural indices of the target brain region of the individual to be evaluated, the spatial coordinates of the abnormal brain region, and the abnormal brain region annotation map, a report on the identification of abnormal brain development trajectory of the individual to be evaluated is generated.

9. The method for identifying abnormal brain development trajectories according to claim 8, characterized in that, The calculation expressions for the Z scores of various structural indicators of the target brain region of the individual to be evaluated in S62 are as follows: , in, This indicates the first brain region of the individual to be evaluated. Z-score of the item structure index This indicates the first brain region of the individual to be evaluated. Structural indices after offset correction This indicates the number of healthy individuals of the same age in the target brain region at the corresponding location. The mean of the item structure index, This indicates the number of healthy individuals of the same age in the target brain region at the corresponding location. The standard deviation of the item structure index, This indicates the age of the individual to be evaluated.

10. A system for identifying abnormal brain development trajectories, characterized in that, include: The first module is used to acquire T1-weighted magnetic resonance images of several healthy individuals and perform offset correction processing on the structural indicators of each brain region in the images in order to construct a generalized additive location-scale-shape model of the healthy population. The second module is used to generate normative developmental trajectory curves and their normal range of variation based on a generalized additive position-scale-shape model of healthy individuals. The third module is used to acquire T1-weighted magnetic resonance images of the individual to be evaluated and to select the target brain region of the individual to be evaluated. The fourth module is used to perform offset correction on various structural indicators of the target brain region of the individual to be evaluated, so as to obtain the offset-corrected structural indicators of the target brain region of the individual to be evaluated. The fifth module is used to obtain the mean and standard deviation of various structural indicators at the corresponding brain regions of the individual to be evaluated in healthy peers based on the norm-model developmental trajectory curve and its normal range of variation. The sixth module is used to generate an identification report of abnormal brain development trajectories for the individual being evaluated, based on the offset-corrected structural indicators of the target brain region of the individual being evaluated, and the mean and standard deviation of various structural indicators at the corresponding brain region locations in healthy individuals of the same age as the individual being evaluated.

Citation Information

Cited By

  • Brain development characteristic evolution analysis system and method based on multi-modal brain image data

    CN122025192A