A method, system, device and storage medium for predicting ADHD pathogenic subcutaneous nuclei

Abstract

The present application discloses a method, system, device, and storage medium for predicting ADHD pathogenic subcutaneous nuclei, wherein the method includes the following steps: obtaining a first magnetic resonance image dataset of each subtype of ADHD case, a second magnetic resonance image dataset of subcutaneous nuclei of normal children, and first scale data of each subtype case; performing structural covariation analysis of the subcutaneous nuclei on the first magnetic resonance image dataset and the second magnetic resonance image dataset to obtain a first abnormal subcutaneous nucleus; obtaining a first contribution score based on the volumes of two subcutaneous brain regions of the first abnormal subcutaneous nucleus; obtaining a trained prediction regression model based on the first contribution score and the first scale data; inputting the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain a prediction result for the subcutaneous nucleus. This method can provide a scientific basis for the effectiveness of clinical treatment and rehabilitation decisions. This application can be widely used in the field of medical technology.

Description

Technical Field

[0001] The present application relates to the field of medical technology, and in particular to a method, system, device and storage medium for predicting ADHD pathogenic subcutaneous nuclei. Background Art

[0002] Attention Deficit Hyperactivity Disorder (ADHD) is a common neuropsychiatric disorder. The incidence of ADHD is approximately 5-7% in children and 2.5% in adults, and has been on a continuous upward trend in recent years. ADHD is clinically divided into three subtypes: the first is the rare subtype characterized by hyperactivity and impulsivity (ADHD-Hyperactive); the second is the more common subtype characterized by inattention (ADHD-Inattentive); and the third is the most common subtype characterized by a mixture of hyperactivity and inattention (ADHD-Combined). Increasing evidence indicates that the inattentive subtype of ADHD is a distinct disorder from the other two subtypes, with distinct cognitive and behavioral characteristics and underlying neurobiological underpinnings. In particular, its clinical symptoms are completely opposite to those of the other two subtypes, and are therefore considered "heterogeneous." In existing technologies, ADHD symptom assessments mostly use EEG, eye movement, and behavioral scales to evaluate the cognitive level and symptoms of ADHD children. However, insufficient attention and research has been paid to abnormal subcutaneous nuclei, and there is no method to predict the pathogenic subcutaneous nuclei based on the patient's brain imaging data. Therefore, a new method for predicting the pathogenic subcutaneous nuclei of ADHD is urgently needed. Summary of the Invention

[0003] The purpose of this application is to solve one of the technical problems existing in the prior art to at least a certain extent.

[0004] To this end, one purpose of the embodiments of the present application is to provide a method, system, device and storage medium for predicting ADHD pathogenic subcutaneous nuclei. This method can provide a scientific basis for the effectiveness of clinical treatment and rehabilitation decisions, which will greatly improve the efficiency of disease diagnosis and treatment of patients and avoid unnecessary consumption of manpower, material resources and time.

[0005] In order to achieve the above-mentioned technical objectives, the technical solutions adopted in the embodiments of the present application include: a method for predicting pathogenic subcutaneous nuclei of ADHD, comprising the following steps: obtaining a first magnetic resonance image data set of each subtype of ADHD cases, a second magnetic resonance image data set of subcutaneous nuclei of normal children, and a first scale data of each subtype case; performing structural covariation analysis of the subcutaneous nuclei on the first magnetic resonance image data set and the second magnetic resonance image data set to obtain a first abnormal subcutaneous nucleus; obtaining a first contribution score based on the volumes of the two subcutaneous brain regions of the first abnormal subcutaneous nucleus; obtaining a trained prediction regression model based on the first contribution score and the first scale data; inputting the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain a prediction result of the subcutaneous nucleus.

[0006] In addition, the method for predicting ADHD pathogenic subcutaneous nuclei according to the above embodiment of the present invention may also have the following additional technical features:

[0007] Furthermore, in an embodiment of the present application, the step of performing structural covariation analysis of the subcutaneous nuclei on the first magnetic resonance image data set and the second magnetic resonance image data set to obtain the first abnormal subcutaneous nucleus specifically includes: performing structural covariation analysis on the first magnetic resonance image data to obtain a first partial correlation coefficient of the 14 subcutaneous brain regions of the case; performing structural covariation analysis on the second magnetic resonance image data to obtain a second partial correlation coefficient of the 14 subcutaneous brain regions of the normal children; determining a first inter-group difference coefficient between normal children and each subtype case based on the first partial correlation coefficient and the second partial correlation coefficient; and determining the first abnormal subcutaneous nucleus based on the first inter-group difference coefficient.

[0008] Furthermore, in an embodiment of the present application, the step of obtaining a first contribution score based on the two subcutaneous brain area volumes of the first abnormal subcutaneous nucleus specifically includes: obtaining the first subcutaneous brain area volume and the second subcutaneous brain area volume of the first abnormal subcutaneous nucleus; inputting the first subcutaneous brain area volume and the second subcutaneous brain area volume into a calculation formula to obtain a first contribution score; wherein the formula is

[0009]

[0010] Where p is the first contribution score, X i is the volume of the first subcutaneous brain area of ​​the case, Y i为 The volume of the second subcutaneous brain area, X represents the average volume of the first subcutaneous brain area of ​​the subtype data set to which the case belongs, Y represents the average volume of the second subcutaneous brain area of ​​the subtype data set to which the case belongs, S x represents the standard deviation of the volume of the first subcutaneous brain area, S y represents the standard deviation of the volume of the second subcortical brain area.

[0011] Furthermore, in an embodiment of the present application, the step of obtaining a trained predictive regression model based on the first contribution score and the first scale data specifically includes: constructing a support vector regression model and setting first model parameters; standardizing the first contribution score to obtain standardized data; inputting the standardized data into the support vector regression model to obtain first predictive scale data; determining the correlation probability based on the first predictive scale data and the first scale data; if the correlation probability is less than or equal to a preset first threshold, the support vector regression model is a predictive regression model; if the correlation probability is greater than the first threshold, adjusting the first model parameters and performing model training to make the correlation probability less than or equal to the first threshold.

[0012] Furthermore, in an embodiment of the present application, the step of inputting the first magnetic resonance image of the ADHD case to be predicted into the predictive regression model to obtain a prediction result of the subcutaneous nuclei specifically includes: inputting the first magnetic resonance image of the ADHD case to be predicted into the predictive regression model to obtain second prediction scale data of the subcutaneous nuclei; obtaining a scale data prediction interval based on the second prediction scale data; judging whether the first scale data belongs to the scale data prediction interval, and if so, the prediction result is that the first abnormal subcutaneous nucleus belongs to the pathogenic nucleus; if not, the prediction result is that the first abnormal subcutaneous nucleus does not belong to the pathogenic nucleus.

[0013] Furthermore, in an embodiment of the present application, the step of performing structural covariation analysis on the first magnetic resonance image dataset to obtain the first partial correlation coefficients of the 14 subcutaneous brain regions of the case specifically includes: segmenting the first magnetic resonance image through a toolkit to obtain 14 subcutaneous brain regions; performing structural covariation analysis on the 14 subcutaneous brain regions, and calculating the partial correlation coefficients of the 14 subcutaneous brain regions of each subtype of ADHD cases and healthy children.

[0014] Furthermore, in an embodiment of the present application, the step of determining the first abnormal subcutaneous nucleus based on the first inter-group difference coefficient specifically includes: obtaining the first inter-group difference coefficient; if the absolute value of the first inter-group difference coefficient is greater than a preset second threshold, then the subcutaneous nucleus is the first abnormal subcutaneous nucleus.

[0015] On the other hand, the present invention also provides an ADHD pathogenic subcutaneous nucleus prediction system, comprising:

[0016] an acquisition unit, configured to acquire a first magnetic resonance image dataset of each subtype ADHD case, a second magnetic resonance image dataset of subcutaneous nuclei of normal children, and first scale data of each subtype case;

[0017] an analyzing unit, configured to perform structural covariation analysis of subcutaneous nuclei on the first magnetic resonance image dataset and the second magnetic resonance image dataset to obtain a first abnormal subcutaneous nucleus;

[0018] a first processing unit, configured to obtain a first contribution score according to the volumes of the two subcutaneous brain regions of the first abnormal subcutaneous nucleus;

[0019] a second processing unit, configured to obtain a trained prediction regression model according to the first contribution score and the first scale data;

[0020] The third processing unit is configured to input the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain a prediction result of the subcutaneous nuclei.

[0021] On the other hand, the present application also provides an ADHD pathogenic subcutaneous nucleus prediction device, comprising:

[0022] at least one processor;

[0023] at least one memory for storing at least one program;

[0024] When the at least one program is executed by the at least one processor, the at least one processor implements a method for predicting ADHD pathogenic subcutaneous nuclei as described in any one of the invention contents.

[0025] In addition, the present application also provides a storage medium storing processor-executable instructions, which, when executed by the processor, are used to execute a method for predicting ADHD-causing subcutaneous nuclei as described in any one of the above items.

[0026] The advantages and benefits of this application will be partially given in the following description, and partially become apparent from the following description, or learned through practice of this application:

[0027] The present application can perform structural covariation analysis of subcutaneous nuclei on magnetic resonance image datasets of ADHD cases and magnetic resonance image datasets of normal children to obtain a first abnormal subcutaneous nucleus, and a first contribution score can be obtained through the first subcutaneous nucleus. The first contribution score and the data scale of each subtype of ADHD cases are input into a support vector regression model and trained to obtain a predictive regression model; the first magnetic resonance image of the ADHD case to be predicted is input into the trained predictive regression model, and the pathogenic subcutaneous nuclei of ADHD can be predicted from the magnetic resonance images of the ADHD case, which can provide a scientific basis for the effectiveness of clinical treatment and rehabilitation decisions, which will greatly improve the efficiency of disease diagnosis and treatment for patients and avoid unnecessary consumption of manpower, material resources and time. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1This is a schematic diagram of the steps of a method for predicting ADHD pathogenic subcutaneous nuclei in a specific embodiment of the present invention;

[0029] Figure 2 This is a schematic diagram of the steps of obtaining a trained prediction regression model according to the first contribution score and the first scale data in a specific embodiment of the present invention;

[0030] Figure 3 This is a schematic structural diagram of a system for predicting ADHD pathogenic subcutaneous nuclei in a specific embodiment of the present invention;

[0031] Figure 4 Schematic diagram of the structure of a device for predicting ADHD pathogenic subcutaneous nuclei in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following describes the embodiments of the present invention in detail with reference to the accompanying drawings to illustrate the principles and processes of the method, system, device, and storage medium for predicting ADHD pathogenic subcutaneous nuclei in the embodiments of the present invention.

[0033] Reference Figure 1 The present invention provides a method for predicting ADHD pathogenic subcutaneous nuclei, which may include the following steps:

[0034] S1. Acquire a first magnetic resonance image dataset of each subtype of ADHD cases, a second magnetic resonance image dataset of subcutaneous nuclei of normal children, and first scale data of each subtype case;

[0035] In an embodiment of the present application, the first magnetic resonance image data set may be a collection of magnetic resonance imaging data of ADHD cases, and the first scale data refers to the scale data corresponding to each subtype of ADHD cases diagnosed by clinicians, the second magnetic resonance image data set may be a magnetic resonance image data set of normal children, the first scale data may be scale data used by clinicians to assess ADHD patients based on the ADHD rating scale or the Wechsler Intelligence Scale for Chinese Children, and the first scale data includes subtypes, behaviors, and scores of ADHD patients; the magnetic resonance image data may be obtained through network data transmission or USB transmission.

[0036] S2. performing structural covariation analysis of subcutaneous nuclei on the first magnetic resonance image dataset and the second magnetic resonance image dataset to obtain a first abnormal subcutaneous nucleus;

[0037] In an embodiment of the present application, the first magnetic resonance image data set can be preprocessed, and the image data of each subtype case in the first magnetic resonance image data set can be segmented into 14 subcutaneous brain regions of the left and right brain using toolkits such as Freesurfer, SPM, and FSL, namely, the thalamus, caudate nucleus, hippocampus, amygdala, putamen, globus pallidus, and nucleus accumbens; every two subcutaneous brain regions can form a pair of subcutaneous nuclei, and structural covariation analysis is performed on the 14 segmented subcutaneous brain regions, and the correlation coefficients between the 14 subcutaneous brain regions in the groups of different subtype cases are calculated respectively. and the correlation coefficient between the 14 subcutaneous brain regions in the normal children control group. According to the correlation coefficient of the subcutaneous brain regions, the inter-group difference coefficient between cases and normal children can be obtained. The inter-group difference coefficient can reflect the difference between the 14 subcutaneous brain regions of cases of different subtypes and the 14 subcutaneous brain regions of normal children. When the difference coefficient of the subcutaneous brain region is greater than 1.96, the subcutaneous nucleus composed of the corresponding two subcutaneous brain regions is the first abnormal subcutaneous nucleus. In this embodiment of the present application, the first abnormal subcutaneous nucleus can be a significantly abnormal subcutaneous nucleus, and the first abnormal subcutaneous nucleus can be one or more.

[0038] S3. Obtaining a first contribution score based on the volumes of the two subcutaneous brain regions of the first abnormal subcutaneous nucleus;

[0039] In the embodiment of the present application, the first abnormal subcutaneous nucleus is the most significant abnormal subcutaneous nucleus. There are differences between the two brain regions of the abnormal subcutaneous nucleus and the two corresponding subcutaneous brain regions of normal children. Since the subcutaneous nucleus includes two subcutaneous brain regions, the first contribution score can be obtained based on the volume of the two subcutaneous nuclei of the case, the volume of the subcutaneous brain region corresponding to the subtype of the case, and the standard deviation of the volume of the subcutaneous brain region corresponding to the subtype; the volume of the subcutaneous brain region corresponding to the subtype of the case can specifically be the volume of the subcutaneous brain region corresponding to the three typical subtypes of ADHD: the subtype with hyperactivity and impulsivity (ADHD-Hyperactive, ADHD-H); the subtype with attention deficit (ADHD-Inattentive, ADHD-I); the third is the volume of 14 subcutaneous brain regions of the left and right brain, including the thalamus, caudate nucleus, hippocampus, amygdala, putamen, globus pallidus and nucleus accumbens.

[0040] S4. Obtaining a trained prediction regression model based on the first contribution score and the first scale data;

[0041] In an embodiment of the present application, after obtaining the first contribution score through the volume of the first abnormal subcutaneous nucleus, it can be input into the support vector regression model for training to obtain a trained prediction regression model; subsequently, the first magnetic resonance image of the ADHD case to be predicted can be input into the trained prediction regression model to predict the scale data corresponding to the case, and based on the scale data corresponding to the case, it can be further predicted whether the corresponding first abnormal subcutaneous nucleus is the ADHD pathogenic subcutaneous nucleus.

[0042] S5. Inputting the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain a prediction result of the subcutaneous nuclei;

[0043] In an embodiment of the present application, the first magnetic resonance image of the ADHD case to be predicted is input into the predictive regression model to obtain predicted scale data; the predicted scale data is compared with the real data scales of patients of different subtypes assessed by doctors based on the ADHD rating scale or the Wechsler Intelligence Scale for Chinese Children-Revised. If the real data scale falls within the scale interval corresponding to the predicted scale data, the corresponding first abnormal subcutaneous nucleus is the pathogenic subcutaneous nucleus, and the obtained pathogenic subcutaneous nucleus can provide a certain reference for the patient's clinical treatment and rehabilitation decision-making.

[0044] Furthermore, the step of performing structural covariation analysis of subcutaneous nuclei on the first magnetic resonance image dataset and the second magnetic resonance image dataset to obtain a first abnormal subcutaneous nucleus may specifically include steps S21, S22, S23, and S24;

[0045] S21. Performing structural covariation analysis on the first magnetic resonance image dataset to obtain first partial correlation coefficients of 14 subcutaneous brain regions for each subtype case;

[0046] Specifically, in the embodiment of the present application, since the image data of each subtype case is divided into 14 subcutaneous brain regions of the left and right brain, the first partial correlation coefficient may refer to the partial correlation coefficient corresponding to the 14 subcutaneous brain regions of each subtype case, which may be a 14*14 data matrix, where each matrix element represents the correlation coefficient of one brain region with other brain regions, and two subcutaneous brain regions may be combined into a subcutaneous nucleus; the partial correlation coefficient of each subcutaneous brain region corresponding to each subtype may be calculated by using software such as R, SPSS or Matlab.

[0047] S22. Performing structural covariation analysis on the first magnetic resonance image data to obtain the second partial correlation coefficients of the 14 subcutaneous brain regions of the case;

[0048] Specifically, in the embodiment of the present application, similarly to the first correlation coefficient, the second partial correlation coefficient may refer to the partial correlation coefficient corresponding to the 14 subcutaneous brain regions of normal children, which may be a 14*14 data matrix, where each matrix element represents the correlation coefficient of one brain region with other brain regions, and two subcutaneous brain regions may be combined into one subcutaneous nucleus; the partial correlation coefficient of each subcutaneous brain region corresponding to normal children may be calculated by using software such as R, SPSS or Matlab.

[0049] S23. Determine a first intergroup difference coefficient between normal children and each subtype case based on the first partial correlation coefficient and the second partial correlation coefficient;

[0050] In the embodiment of the present application, the first inter-group difference coefficient may include the inter-group difference coefficient between normal children and the brain regions corresponding to each subtype. Specifically, the first inter-group difference coefficient may include the inter-group difference coefficient between the left thalamus of normal children and the left thalamus of each subtype case. It may also include the inter-group difference coefficient between normal children and the brain regions corresponding to each subtype, and the inter-group difference coefficient between the normal children group and the ADHD case group can be determined according to the first partial correlation coefficient and the second partial correlation coefficient of the ADHD case. Further, according to the formula

[0051]

[0052] Among them, Z is the inter-group difference coefficient, r1 is the second partial correlation coefficient of case children; r2 is the first partial correlation coefficient of normal children; n1 is the number of subtype data groups corresponding to case children; n2 is the number of data groups of normal children, and these two numbers can be set according to different situations. Usually, the data group of the subtype dominated by hyperactivity and impulsivity (ADHD-Hyperactive, ADHD-H) can be set to 1 sample; the data group of the subtype dominated by attention deficit (ADHD-Inattentive, ADHD-I) can be set to 66 samples; the third type is the data group of the mixed subtype of hyperactivity and attention deficit (ADHD-Combined, ADHD-C) can be set to 46 samples; and r1 and r2 correspond to the corresponding correlation coefficients when needed, such as r1 is the correlation coefficient between the left thalamus of the case child and the right thalamus, and r2 is the correlation coefficient between the left thalamus and the right thalamus of the normal child.

[0053] S24. Determine a first abnormal subcutaneous nucleus based on the first inter-group difference coefficient;

[0054] Specifically, in an embodiment of the present application, the inter-group difference coefficients of the corresponding brain regions of the normal children group and the different subtype groups of ADHD cases are obtained by the above formula. By setting the threshold of the inter-group difference, the subcutaneous nuclei corresponding to the two subcutaneous brain regions can be divided into significantly abnormal subcutaneous nuclei and non-significantly abnormal subcutaneous nuclei; when the inter-group difference coefficients of the two brain regions are both greater than the set threshold, the subcutaneous nuclei composed of the two brain regions are significantly abnormal subcutaneous nuclei.

[0055] Furthermore, performing structural covariation analysis on the first magnetic resonance image data set to obtain the first partial correlation coefficients of the 14 subcutaneous brain regions of the case may specifically include: step S211 and step S212

[0056] S211. Segment each image in the first magnetic resonance image dataset using a toolkit to obtain 14 subcutaneous brain regions; obtaining 14 subcutaneous brain regions;

[0057] S212, performing structural covariation analysis on the 14 subcutaneous brain regions, and calculating partial correlation coefficients of the 14 subcutaneous brain regions in ADHD cases of each subtype and healthy children;

[0058] Specifically, in an embodiment of the present application, each image in the first magnetic resonance image dataset can be segmented using toolkits such as Freesurfer, SPM, and FSL to obtain 14 subcutaneous brain regions for each image; structural covariation analysis is performed on the 14 subcutaneous brain regions, and the corresponding partial correlation coefficients of the 14 subcutaneous brain regions of each subtype ADHD case and healthy children in the first magnetic resonance image dataset can be calculated.

[0059] Furthermore, the step of determining the first abnormal subcutaneous nucleus according to the first inter-group difference coefficient may specifically include: step S241 and step S242

[0060] S241. Obtain the intergroup difference coefficient between the subcutaneous nuclei of cases and normal children;

[0061] S242: If the absolute value of the inter-group difference coefficient is greater than a preset second threshold, the subcutaneous nucleus is a first abnormal subcutaneous nucleus;

[0062] Specifically, in an embodiment of the present application, the second threshold value can be preset to 1.96. When the absolute value of the inter-group difference coefficient between the subcutaneous nuclei of case children and the subcutaneous nuclei of normal children is greater than 1.96, the corresponding subcutaneous nuclei are the first abnormal subcutaneous nuclei, that is, the significantly abnormal subcutaneous nuclei. If the absolute value is less than or equal to 1.96, the corresponding subcutaneous nuclei are non-significantly abnormal subcutaneous nuclei.

[0063] Furthermore, the step of obtaining the first contribution score according to the volumes of the two subcutaneous brain regions of the first abnormal subcutaneous nucleus specifically includes: step S31 and step S32

[0064] S31, obtaining the volume of the first subcutaneous brain region and the volume of the second subcutaneous brain region of the first abnormal subcutaneous nucleus;

[0065] S32. Input the first subcutaneous brain region volume and the second subcutaneous brain region volume into a calculation formula to obtain a first contribution score; wherein the formula is

[0066]

[0067] Where p is the first contribution score, X i is the volume of the first subcutaneous brain area of ​​the case, Y i is the volume of the second subcutaneous brain area, X represents the average volume of the first subcutaneous brain area of ​​the subtype data set to which the case belongs, Y represents the average volume of the second subcutaneous brain area of ​​the subtype data set to which the case belongs, S x represents the standard deviation of the volume of the first subcutaneous brain area, S y represents the standard deviation of the volume of the second subcortical brain area.

[0068] Specifically, the brain volume of 14 subcutaneous brain regions in the magnetic resonance image data is calculated by using toolkits such as Freesurfer, SPM, and FSL. According to the first abnormal subcutaneous nucleus determined in step S2 above, the volumes of the two subcutaneous brain regions corresponding to the first abnormal subcutaneous nucleus are determined, and the volumes of the two subcutaneous brain regions are input into the formula

[0069]

[0070] Where p is the first contribution score, X i is the volume of the first subcutaneous brain area of ​​the case, Y i is the volume of the second subcutaneous brain area, X represents the average volume of the first subcutaneous brain area of ​​the subtype data set to which the case belongs, Y represents the average volume of the second subcutaneous brain area of ​​the subtype data set to which the case belongs, S x represents the standard deviation of the volume of the first subcutaneous brain area, S y represents the standard deviation of the volume of the second subcortical brain area.

[0071] Further, refer to Figure 2 The step of obtaining a trained prediction regression model according to the first contribution score and the first scale data may specifically include steps S41, S42, S43, S44, S45, and S46;

[0072] S41, constructing a support vector regression model and setting first model parameters;

[0073] S42. Normalize the first contribution score to obtain standardized data;

[0074] S43, inputting the standardized data into the support vector regression model to obtain first prediction scale data;

[0075] S44. Determine a correlation probability based on the first prediction scale data and the first scale data;

[0076] S45. If the correlation probability is less than or equal to a preset first threshold, the support vector regression model is a predictive regression model;

[0077] S46. If the correlation probability is greater than the first threshold, adjust the first model parameters and perform model training to make the correlation probability less than or equal to the first threshold;

[0078] Specifically, in an embodiment of the present application, the first threshold value can be set to 0.05; the machine learning prediction model can select a support vector regression model for prediction; the support vector regression model is trained by a cross-validation method to find the first model parameter, and the first model parameter can be the optimal parameter before the start of training; wherein the optimal parameter is a parameter that can characterize the relationship between the contribution score and the scale indicator; for the training of the model, first, a set of magnetic resonance image data of each subtype to be trained can be obtained, which can be obtained from the existing publicly available magnetic resonance image data of ADHD children, which is a data set of all case imaging data in the ADHD children database, including different The imaging data of children with different ADHD subtypes were used to calculate the contribution scores of all different ADHD subtypes to the different subtype groups within the case imaging dataset. All contribution scores were then standardized. Then, the optimal parameters were substituted into a support vector regression model to predict each scale indicator. Finally, the correlation coefficient between the actual scale values ​​and the predicted values ​​from the support vector regression model was calculated to test whether the subcutaneous nuclei were significantly predictive of the scale indicators. The support vector regression model was tested for significance at P ≤ 0.05 using 5000 permutation tests. P represents the probability of occurrence, which reflects the likelihood of an event occurring. The P value obtained using statistical significance testing generally defines a statistically significant difference as P < 0.05, a statistically significant difference as P < 0.01, and an extremely significant difference as P < 0.001. A P < 0.05 is considered statistically significant. If significant, this indicates that the support vector regression model accurately characterizes the relationship between the subcutaneous nuclei data and the scale indicators in individual ADHD children. If it is not significant, it indicates that the support vector regression model cannot reliably predict the scale indicators from the subcutaneous nucleus data of individual ADHD children. In this case, it is necessary to repeatedly train the support vector regression model through cross-validation and adjust the first model parameters until P is less than 0.05. The final model can be simplified as follows:

[0079] Scale data = ADHD children's subcutaneous nucleus data * α + β

[0080] Among them, α and β represent model coefficients, "ADHD children's subcutaneous nucleus data" is the input of the model, and the model input is the contribution score of the subcutaneous nucleus of ADHD children, and "scale data" is the output of the model, and the data output by the model is the range of each score in the predicted scale data.

[0081] Furthermore, the step of inputting the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain the prediction result of the subcutaneous nuclei may specifically include: Step S51 to Step S53

[0082] S51, inputting the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain second prediction scale data of the subcutaneous nuclei;

[0083] S52. Obtaining a scale data prediction interval based on the second prediction scale data;

[0084] S53. Determine whether the first scale data belongs to the scale data prediction interval. If so, the prediction result is that the first abnormal subcutaneous nucleus belongs to the pathogenic nucleus; if not, the prediction result is that the first abnormal subcutaneous nucleus does not belong to the pathogenic nucleus.

[0085] Specifically, in an embodiment of the present application, the doctor uses a data scale of different subtypes of patients assessed according to the ADHD rating scale or the Wechsler Intelligence Scale for Chinese Children-Revised, and the second predicted scale data is the predicted scale data of the case obtained by the prediction model. The scale data prediction interval can be obtained based on the second scale data. Specifically, the upper limit of the predicted interval can be equal to the sum of the second predicted scale data and 1.96 times the standard deviation, and the upper and lower limits of the predicted interval are equal to the difference between the second predicted scale data and 1.96 times the standard deviation. The standard deviation can be obtained by calculation through the first scale data. After obtaining the scale data prediction interval, it is determined whether the first scale data belongs to the scale data prediction interval. If it does, the prediction result is that the first abnormal subcutaneous nucleus belongs to the pathogenic nucleus; if it does not, the prediction result is that the first abnormal subcutaneous nucleus does not belong to the pathogenic nucleus.

[0086] In addition, refer to Figure 3 ,and Figure 1 Corresponding to the method, an embodiment of the present application further provides an ADHD pathogenic subcutaneous nucleus prediction system, including: an acquisition unit, used to acquire a first magnetic resonance image data set of each subtype ADHD case, a second magnetic resonance image data set of subcutaneous nuclei of normal children, and first scale data of each subtype case; an analysis unit, used to perform structural covariation analysis of the subcutaneous nuclei on the first magnetic resonance image data set and the second magnetic resonance image data set to obtain a first abnormal subcutaneous nucleus; a first processing unit, used to obtain a first contribution score based on the volumes of two subcutaneous brain regions of the first abnormal subcutaneous nucleus; a second processing unit, used to obtain a trained prediction regression model based on the first contribution score and the first scale data; a third processing unit, used to input the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain a prediction result of the subcutaneous nucleus.

[0087] and Figure 1 Corresponding to the method, the embodiment of the present application also provides an ADHD pathogenic subcutaneous nucleus prediction device, the specific structure of which can be referred to Figure 4 ,include:

[0088] at least one processor;

[0089] at least one memory for storing at least one program;

[0090] When the at least one program is executed by the at least one processor, the at least one processor implements the ADHD pathogenic subcutaneous nucleus prediction method.

[0091] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0092] and Figure 1 Corresponding to the method, an embodiment of the present application further provides a storage medium storing processor-executable instructions, which are used to execute the ADHD pathogenic subcutaneous nucleus prediction method when executed by the processor.

[0093] In some optional embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiments presented and described in the flow chart of the present application are provided in an exemplary manner for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operations and logical flows presented herein. Optional embodiments are contemplated in which the order of the various operations is changed and the sub-operations described as a part of a larger operation are performed independently.

[0094] In addition, although the present application is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present application. More specifically, given the properties, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the module will be understood within the routine skills of an engineer. Therefore, a person skilled in the art can implement the present application as set forth in the claims using ordinary techniques without undue experimentation. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present application, which is determined by the full scope of the appended claims and their equivalents.

[0095] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several programs for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0096] The logic and / or steps represented in a flowchart or otherwise described herein, for example, may be considered as an ordered list of executable programs for implementing the logical functions, and may be embodied in any computer-readable medium for use by, or in conjunction with, a program execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can retrieve and execute a program from a program execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, a program execution system, apparatus, or device.

[0097] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0098] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable program execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0099] In the above description of this specification, reference to the terms "one embodiment / example," "another embodiment / example," or "certain embodiments / examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples.

[0100] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and intent of the present application, and that the scope of the present application is defined by the claims and their equivalents.

[0101] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present application. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A method for predicting ADHD pathogenic subcutaneous nuclei, characterized in that: The following steps are involved: Acquire a first magnetic resonance image dataset of each subtype ADHD case, a second magnetic resonance image dataset of subcutaneous nuclei of normal children, and first scale data of each subtype case; performing structural covariation analysis of subcutaneous nuclei on the first magnetic resonance image dataset and the second magnetic resonance image dataset to obtain a first abnormal subcutaneous nucleus; the step of performing structural covariation analysis of subcutaneous nuclei on the first magnetic resonance image dataset and the second magnetic resonance image dataset to obtain the first abnormal subcutaneous nucleus specifically includes: Performing structural covariation analysis on the first magnetic resonance image dataset to obtain first partial correlation coefficients of 14 subcutaneous brain regions for each subtype case; performing structural covariation analysis on the second magnetic resonance image dataset to obtain second partial correlation coefficients of 14 subcortical brain regions in normal children; Determining a first intergroup difference coefficient between normal children and each subtype case based on the first partial correlation coefficient and the second partial correlation coefficient; determining a first abnormal subcutaneous nucleus according to the first inter-group difference coefficient; Obtaining a first contribution score according to the volumes of the two subcutaneous brain regions of the first abnormal subcutaneous nucleus; The step of obtaining a first contribution score based on the volumes of the two subcutaneous brain regions of the first abnormal subcutaneous nucleus specifically includes: Obtaining the volume of the first subcutaneous brain region and the volume of the second subcutaneous brain region of the first abnormal subcutaneous nucleus; The volume of the first subcutaneous brain region and the volume of the second subcutaneous brain region are input into a calculation formula to obtain a first contribution score, wherein the formula is: Where p is the first contribution score, X i is the volume of the first subcutaneous brain area of ​​the case, Y i is the volume of the second subcutaneous brain area, X represents the average volume of the first subcutaneous brain area of ​​the subtype data set to which the case belongs, Y represents the average volume of the second subcutaneous brain area of ​​the subtype data set to which the case belongs, S x represents the standard deviation of the volume of the first subcutaneous brain area, S y represents the standard deviation of the volume of the second subcortical brain area; Obtaining a trained prediction regression model according to the first contribution score and the first scale data; The step of obtaining a trained prediction regression model according to the first contribution score and the first scale data specifically includes: Constructing a support vector regression model and setting first model parameters; Normalizing the first contribution score to obtain standardized data; Inputting the standardized data into the support vector regression model to obtain first prediction scale data; determining a correlation probability based on the first prediction scale data and the first scale data; If the correlation probability is less than or equal to a preset first threshold, the support vector regression model is a predictive regression model; If the correlation probability is greater than the first threshold, adjusting the first model parameters and performing model training to make the correlation probability less than or equal to the first threshold; The first magnetic resonance image of the ADHD case to be predicted is input into the prediction regression model to obtain the prediction result of the subcutaneous nucleus.

2. A method for predicting ADHD pathogenic subcutaneous nuclei according to claim 1, characterized in that: The step of inputting the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain the prediction result of the subcutaneous nuclei specifically includes: Inputting the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain second prediction scale data of the subcutaneous nuclei; Obtaining a scale data prediction interval based on the second prediction scale data; Determine whether the first scale data belongs to the scale data prediction interval. If so, the prediction result is that the first abnormal subcutaneous nucleus belongs to the pathogenic nucleus; if not, the prediction result is that the first abnormal subcutaneous nucleus does not belong to the pathogenic nucleus.

3. A method for predicting ADHD pathogenic subcutaneous nuclei according to claim 1, characterized in that: The step of performing structural covariation analysis on the first magnetic resonance image dataset to obtain first partial correlation coefficients of 14 subcutaneous brain regions of the case specifically includes: Each image in the first magnetic resonance image dataset was segmented using the toolkit to obtain 14 subcortical brain regions; Structural covariation analysis was performed on the 14 subcortical brain regions, and partial correlation coefficients of the 14 subcortical brain regions in ADHD cases of each subtype and healthy children were calculated.

4. A method for predicting ADHD pathogenic subcutaneous nuclei according to claim 1, characterized in that: The step of determining the first abnormal subcutaneous nucleus according to the first inter-group difference coefficient specifically includes: Get the first intergroup difference coefficient; If the absolute value of the first inter-group difference coefficient is greater than a preset second threshold, the subcutaneous nucleus is a first abnormal subcutaneous nucleus.

5. A prediction system for ADHD pathogenic subcutaneous nuclei, characterized by: include: an acquisition unit, configured to acquire a first magnetic resonance image dataset of each subtype ADHD case, a second magnetic resonance image dataset of subcutaneous nuclei of normal children, and first scale data of each subtype case; an analyzing unit configured to perform structural covariation analysis of subcutaneous nuclei on the first magnetic resonance image dataset and the second magnetic resonance image dataset to obtain a first abnormal subcutaneous nucleus; wherein the step of performing structural covariation analysis of subcutaneous nuclei on the first magnetic resonance image dataset and the second magnetic resonance image dataset to obtain the first abnormal subcutaneous nucleus specifically comprises: Performing structural covariation analysis on the first magnetic resonance image dataset to obtain first partial correlation coefficients of 14 subcutaneous brain regions for each subtype case; performing structural covariation analysis on the second magnetic resonance image dataset to obtain second partial correlation coefficients of 14 subcortical brain regions in normal children; Determining a first intergroup difference coefficient between normal children and each subtype case based on the first partial correlation coefficient and the second partial correlation coefficient; determining a first abnormal subcutaneous nucleus according to the first inter-group difference coefficient; a first processing unit, configured to obtain a first contribution score according to the volumes of the two subcutaneous brain regions of the first abnormal subcutaneous nucleus; The step of obtaining a first contribution score based on the volumes of the two subcutaneous brain regions of the first abnormal subcutaneous nucleus specifically includes: Obtaining the volume of the first subcutaneous brain region and the volume of the second subcutaneous brain region of the first abnormal subcutaneous nucleus; The volume of the first subcutaneous brain region and the volume of the second subcutaneous brain region are input into a calculation formula to obtain a first contribution score, wherein the formula is: Where p is the first contribution score, X i is the volume of the first subcutaneous brain area of ​​the case, Y i is the volume of the second subcutaneous brain area, X represents the average volume of the first subcutaneous brain area of ​​the subtype data set to which the case belongs, Y represents the average volume of the second subcutaneous brain area of ​​the subtype data set to which the case belongs, S x represents the standard deviation of the volume of the first subcutaneous brain area, S y represents the standard deviation of the volume of the second subcortical brain area; a second processing unit, configured to obtain a trained prediction regression model according to the first contribution score and the first scale data; The step of obtaining a trained prediction regression model according to the first contribution score and the first scale data specifically includes: Constructing a support vector regression model and setting first model parameters; Normalizing the first contribution score to obtain standardized data; Inputting the standardized data into the support vector regression model to obtain first prediction scale data; determining a correlation probability based on the first prediction scale data and the first scale data; If the correlation probability is less than or equal to a preset first threshold, the support vector regression model is a predictive regression model; If the correlation probability is greater than the first threshold, adjusting the first model parameters and performing model training to make the correlation probability less than or equal to the first threshold; The third processing unit is configured to input the first magnetic resonance image of the ADHD case to be predicted into the prediction regression model to obtain a prediction result of the subcutaneous nuclei.

6. A device for predicting ADHD pathogenic subcutaneous nuclei, characterized in that include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method for predicting ADHD pathogenic subcutaneous nuclei as described in any one of claims 1 to 4.

7. A storage medium storing instructions executable by a processor, characterized in that: The processor-executable instructions are used to execute the method for predicting ADHD pathogenic subcutaneous nuclei when executed by the processor according to any one of claims 1 to 4.

Images