Children PET brain template construction and quantitative analysis method

Abstract

The invention relates to a child PET brain template construction and quantitative analysis method. The method comprises the following steps: acquiring a preprocessed PET image of a normal participant of a child; performing spatial registration on the PET image to CHN-PD, performing intensity normalization processing and data screening, and performing average calculation to obtain a preliminary child PET brain template image; performing spatial registration on the screened PET images to the preliminary child PET brain template image, and performing intensity normalization processing, data screening, average calculation and smoothing processing to obtain a final child PET brain template image; after the PET image is subjected to space registration to a final child PET brain template image and a brain region is divided, a first quantitative parameter is calculated to be used for constructing a brain metabolism data database; and after performing spatial registration on the child PET image to be analyzed to the final child PET brain template image and dividing a brain region, calculating a second quantitative parameter and comparing the second quantitative parameter with the database for quantitative analysis. The method has the beneficial effects that a normal children brain metabolism data database is established, a matched quantitative metabolism analysis process is provided for clinical use, and the registration effect and the diagnosis and treatment efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of medical imaging technology, and in particular to a method for constructing and quantitatively analyzing a pediatric PET brain template. Background Technology

[0002] Research on pediatric brain diseases is a complex and evolving field, involving various disease types, primarily including neurodevelopmental disorders, congenital metabolic abnormalities, and acquired injuries. Currently, clinical research mainly relies on conventional imaging techniques, such as computed tomography (CT) and magnetic resonance imaging (MRI). CT only provides anatomical imaging capabilities and lacks metabolic imaging capabilities, while MRI offers both conventional structural and functional metabolic imaging, providing detailed information on brain structure and function. However, its effectiveness in diagnosing metabolic and developmental disorders is limited. Unlike MRI, positron emission tomography (PET) imaging can diagnose metabolic activity in the nervous system. PET images allow physicians to detect areas of metabolic abnormalities. Compared to MRI, PET metabolic detection has better sensitivity and specificity, and quantitative analysis of metabolism in PET images is more advantageous. Furthermore, PET offers advantages over MRI in areas such as metabolite detection, neurotransmitter assessment, molecular specificity, and dynamic metabolic tracking.

[0003] With the continuous advancement of brain imaging technology, brain templates play a crucial role in disease research. By registering patient images to a standard template space, a series of analyses can be conducted. Currently, most widely used brain templates are designed for adults. The lack of dedicated pediatric brain metabolic templates based on FDG-PET image data limits the research and diagnosis of pediatric neurometabolic diseases. Current brain template construction primarily relies on MRI and CT images, while research based on PET images is relatively limited. Clinically, patient PET images can only be registered with MRI templates, and multimodal registration is often less effective than same-modal registration and more complex, thus impacting diagnosis and treatment. While adult PET brain templates have been publicly released, no publicly available pediatric PET brain templates or related research have been found. Children's brain structures are significantly different, and their brains are not fully developed; using adult templates can easily lead to registration errors and biased analysis. Therefore, constructing pediatric PET brain templates is a critical clinical challenge that urgently needs to be addressed.

[0004] Quantitative analysis methods for the brain provide crucial information for disease diagnosis, treatment, and research by quantifying indicators related to brain structure, function, and metabolism. In the quantitative analysis of pediatric brain PET images, constructing specialized PET brain templates adapted to the developmental characteristics of children's brains is fundamental to improving image registration accuracy, while accumulating large-scale brain metabolic data from normal children is a key prerequisite for establishing reliable quantitative analysis models. However, current research faces two major challenges: First, existing quantitative brain analysis techniques are primarily developed for adults whose brain structures have stabilized. In contrast, children's brains exhibit significant morphological and metabolic dynamics with age (e.g., age-dependent differences in gray / white matter ratios and peak metabolic rates), requiring separate research. Second, a globally shared pediatric PET brain metabolism database has not yet been established. The lack of accurate and readily available metabolic data in clinical studies has long led to errors, affecting the accuracy of disease diagnosis and efficacy assessment. Therefore, establishing a standardized brain metabolism database and designing quantitative analysis methods based on the developmental patterns of children's brain metabolism are urgent tasks that need to be promoted. Summary of the Invention

[0005] Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method for constructing and quantitatively analyzing PET brain templates for children, which solves the technical problem of how to construct PET brain templates suitable for Chinese children and conduct quantitative analysis research based on the brain templates.

[0007] Technical solution

[0008] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0009] This invention provides a method for constructing and quantitatively analyzing a pediatric PET brain template, comprising:

[0010] Obtain preprocessed PET images of normal children participants;

[0011] The PET images were spatially registered to the pediatric MRI brain atlas CHN-PD to obtain the registered PET images.

[0012] Intensity normalization and data filtering were performed on the registered PET images to obtain the filtered normalized PET images;

[0013] The average of the selected normalized PET images was calculated to obtain preliminary PET brain template images of children.

[0014] The selected PET images were spatially registered to the preliminary pediatric PET brain template images, and intensity normalization, data filtering, averaging and smoothing were performed to obtain the final pediatric PET brain template images.

[0015] The PET images were spatially registered to the final pediatric PET brain template images and the brain regions were divided. The first quantitative parameter of each brain region of each normal participant was calculated, and the mean and standard deviation of each brain region and each parameter of all normal participants were calculated. By calculating the corresponding mean ± 2 × standard deviation, the normal metabolic range was obtained, and the construction of the brain metabolism database was completed.

[0016] After spatially registering the PET images of the child to be analyzed to the final PET brain template image of the child and dividing the brain regions, the second quantitative parameter is calculated and compared with the brain metabolism database to complete the quantitative analysis.

[0017] Optionally, the PET image space is spatially registered to the pediatric MRI brain atlas CHN-PD to obtain the registered PET image, including:

[0018] Define the initial deformation field, set the similarity metric function and regularization parameters;

[0019] Calculate the similarity between PET images and pediatric MRI brain atlas CHN-PD;

[0020] Based on the similarity measurement results, an optimization algorithm is used to update the transformation parameters so that the similarity measurement function reaches its maximum value;

[0021] Determine if the iteration has reached the convergence condition. If it has converged, stop the iteration and the spatial registration is complete; otherwise, continue the iterative optimization.

[0022] Optionally, the similarity between PET images and pediatric MRI brain atlas CHN-PD is calculated, including:

[0023] The similarity index (MI) between PET images and pediatric MRI brain atlases (CHN-PD) is calculated using the mutual information method, as follows:

[0024] Where p(x, y) represents the joint probability distribution at corresponding pixel (x, y) in the two images, and p(x) and p(y) represent the edge probability distributions at pixel x and y in the two images, respectively.

[0025] Optionally, the registered PET images are subjected to intensity normalization and data filtering to obtain filtered normalized PET images, including:

[0026] The intensity of the registered PET image is normalized according to the following formula:

[0027] I k (i)=G k (i) / G k_mean_ref Among them, G kFor the registered PET image of participant k, i is the pixel index, G k_mean_ref I represents the average pixel value of the cerebellar reference region for the participants. k PET images of participant k after normalization;

[0028] The sum of absolute errors between intensity-normalized PET images and children's MRI brain atlas CHN-PD is calculated using the following formula:

[0029] SAD k =∑ i |I k (i)-I ref (i)|, where, I ref For normalized pediatric MRI brain atlas CHN-PD, i is the pixel index, SAD k For I k with I ref The sum of the horizontal absolute errors between the images;

[0030] Anomaly detection is performed on intensity-normalized PET images using the following formula:

[0031] DIF k =|SAD k -SAD mean |-3×SAD std Among them, SAD mean and SAD std point

[0032] Let SAD be the mean and standard deviation of all participants; if DIF k If the value is greater than 0, the participant k data is considered to have potential metabolic abnormalities and needs to be removed. The remaining PET images are used as the normalized PET images after screening.

[0033] Alternatively, the average can be calculated using the following formula:

[0034] Where i represents the pixel index, n represents the number of images used for averaging, and I k denoted as the intensity-normalized image of the k-th participant, and I represents the preliminary pediatric PET brain template obtained by averaging.

[0035] Optionally, the method further includes:

[0036] The regional anatomical differences between the final pediatric PET brain template and the pediatric MRI brain atlas CHN-PD image were calculated based on the Mean Sum of Squared Errors (MSD) algorithm and the Normalized Product Correlation (NCC) algorithm. The calculation formula is as follows:

[0037]

[0038] Where I represents the final pediatric PET brain template obtained from the construction, I ref This represents the CHN-PD MRI brain atlas for children, where i represents the pixel index and n represents the pixel index. roi i_roi represents the number of pixels in the region of interest, and i_roi represents the pixel index within the region of interest. mean_roi and I ref_mean_roi Representing images I and I respectively ref The average pixel value within the region of interest; the smaller the MSD value, the smaller the anatomical difference between the two image regions; the larger the MSD value, the greater the anatomical difference between the two image regions; the closer the NCC value is to 1, the smaller the anatomical difference between the two image regions; the closer the NCC value is to -1, the greater the anatomical difference between the two image regions; the closer the NCC value is to 0, the two image regions are uncorrelated.

[0039] Optionally, the first quantitative parameter includes the average gray value (Gray). mean Average standard intake value for SUVs mean The second quantitative parameter includes the gray average value (Gray) compared to the standard intake value (SUVr). mean Average standard intake value for SUVs mean Standard intake value ratio (SUVr) and z-score (Z-score).

[0040] Optionally, the grayscale average value Gray is calculated according to the following formula. mean Average standard intake value for SUVs mean Standard intake value ratio (SUVr) and z-score:

[0041]

[0042]

[0043] Where R represents the individual image registered with the final pediatric PET brain template, S represents the individual image after converting voxel values ​​to SUV values, i represents the pixel index, roi represents the region of interest, n represents the number of pixels, ref represents the cerebellar region of the image itself, and database represents the brain metabolism data database; the formula for calculating the SUV value is:

[0044] When calculating the first quantitative parameter for constructing a brain metabolism database, the individual image is an image of a normal participant; when calculating the second quantitative parameter for quantitative analysis, the individual image is a PET image of a child to be analyzed.

[0045] Optionally, the SyN algorithm can be used for image registration.

[0046] Optionally, PET data of normal participants of different ages and genders can be obtained to construct a whole template for children aged 6-12 years and sub-templates with an age interval of 1 year. The whole template for children aged 6-12 years includes brain templates for boys, girls and children of undifferentiated genders, and a database of brain metabolism data corresponding to the templates can be obtained through quantitative analysis.

[0047] Beneficial effects

[0048] The beneficial effects of this invention are as follows: This invention provides a method for constructing and quantitatively analyzing pediatric PET brain templates, resulting in age- and sex-specific pediatric PET brain templates, corresponding brain metabolism data databases, and clinical application methods. The advantages are significant: First, it overcomes a series of problems in the template construction process, establishing ethically sound and accurate pediatric PET brain templates, filling a gap in current research on pediatric PET brain templates. Based on this template, better registration results can be obtained in clinical and research settings without necessarily using patient MRI images as an intermediary. Second, this invention establishes a pediatric PET brain metabolism data database and provides quantitative analysis methods for clinical applications, which can help doctors detect abnormalities early and improve diagnostic and treatment efficiency. Finally, this invention provides software tools to support and enhance the applicability of the templates. Attached Figure Description

[0049] Figure 1 A flowchart illustrating a method for constructing and quantitatively analyzing a pediatric PET brain template, provided in an embodiment of the present invention;

[0050] Figure 2 A schematic diagram of the axial plane of one of the templates established by the method for constructing and quantitatively analyzing a child PET brain template provided in an embodiment of the present invention—a template for children aged 6-12 years without distinguishing gender;

[0051] Figure 3 A schematic diagram of the coronal plane of one of the templates established by the method for constructing and quantitatively analyzing a child PET brain template provided in an embodiment of the present invention—a template for children aged 6-12 years without distinguishing gender;

[0052] Figure 4 This is a sagittal plane schematic diagram of one of the templates established by the method for constructing and quantitatively analyzing a child's PET brain template provided in an embodiment of the present invention—a template for children aged 6-12 years who are not gender-specific. Detailed Implementation

[0053] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] Constructing pediatric PET brain templates is a pressing clinical challenge. However, the process is extremely challenging. First, acquiring the data required for atlas construction is difficult. Compared to adults, pediatric subjects are more difficult to recruit and collect data from, requiring consideration of safety and cooperation issues, resulting in a scarcity of high-quality data suitable for atlas construction. Second, children's brain structure undergoes greater changes with age than adults, and brain morphology differs between sexes. Therefore, age and sex characteristics must be carefully considered when constructing pediatric PET brain templates. These intertwined factors make the construction of pediatric PET brain templates even more challenging, severely hindering the in-depth advancement of pediatric neuroscience research.

[0055] In quantitative analysis based on PET images, doctors primarily focus on quantitative indicators of brain regions of interest (such as the frontal lobe, temporal lobe, and parietal lobe). Commonly used brain region division methods include AAL (Anatomical Automatic Labeling), an automated brain region labeling template provided by the Montreal Neuroscience Institute in Canada, which divides the brain into 116 regions. Common quantitative indicators include the average grayscale value and the average standard uptake value (SUV). mean The standard ingress value ratio (SUVr) and Z-score are used to measure the average intensity of pixel values ​​in an image. mean The average uptake value can be represented by the PET image, indicating the average uptake value of a specific region or the entire brain. SUVr represents the ratio of the radioactive imaging agent concentration of the tissue of interest to that of corresponding muscle or other normal tissue in the same patient. The Z-score is a standardized value typically used to assess the deviation of a measurement from the mean; when the mean is based on a database of normal individuals, this value can be used to identify abnormalities in the patient's image. Clinically, the calculation and analysis of these quantitative parameters are of great significance for the diagnosis and treatment of diseases.

[0056] To construct a pediatric PET brain template to fill the current gap in clinical research, this invention is based on the pediatric MRI brain atlas CHN-PD and uses data from 400 normal pediatric subjects. 18Using F-FDG PET images and an image registration algorithm based on Symmetric Normalization (SyN), a method for constructing pediatric PET brain templates was designed and implemented, ultimately obtaining age-specific templates (overall templates for children aged 6-12 years and sub-templates at 1-year intervals) and sex-specific templates. For each template, based on AAL116 partitioning, quantitative analysis indicators were further calculated to establish a database of normal pediatric brain metabolic data. This template set provides a complete quantitative metabolic analysis workflow, including standardization, registration, partitioning, and statistics. Ultimately, the PET pediatric brain templates can be widely used for the diagnosis and analysis of pediatric neurological diseases such as neurodevelopmental disorders, congenital metabolic abnormalities, and acquired injuries, especially for the quantitative analysis of PET in diseases such as epilepsy, encephalitis, autism, and developmental delay.

[0057] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0058] Firstly, based on the CHN-PD pediatric MRI brain atlas from Beijing Normal University, this invention proposes a method using the SyN algorithm. 18 A method for constructing a pediatric PET brain template and a method for quantitative analysis of brain metabolism using F-FDG PET images. The brain template construction method first preprocesses the collected PET data from normal participants, including image cropping and position correction, to make the images more suitable for template construction. Then, a two-step process is used to construct the pediatric PET brain template, including SyN spatial registration, intensity normalization, data filtering, averaging, and smoothing. The brain metabolism quantitative analysis method first spatially registers the preprocessed normal participant data to the constructed template. Then, based on AAL116 partitioning, quantitative analysis indicators for each brain region are calculated, resulting in a brain metabolism database. This database can be used for clinical quantitative analysis and diagnosis.

[0059] Reference Figure 1 This embodiment provides a method for constructing and quantitatively analyzing a pediatric PET brain template, the specific steps of which include:

[0060] S1, Obtain preprocessed PET images of normal children.

[0061] A total of FDG-PET data from 460 healthy pediatric participants were collected for template construction and the creation of a brain metabolism database. Data screening and exclusion criteria:

[0062] (1) Age: 6-12 years old;

[0063] (2) Gender: Half male and half female;

[0064] (3) Image requirements: DICOM images of head PET only;

[0065] (4) Health status of enrolled children:

[0066] 1) No head diseases, and the head is not affected by any disease;

[0067] 2) The lesions throughout the body are relatively mild and have not caused systemic symptoms;

[0068] 3) The child has no cognitive or developmental abnormalities or mental illnesses;

[0069] 4) The child had never received radiotherapy, chemotherapy, or immunotherapy before treatment;

[0070] (5) Exclusion criteria:

[0071] 1) The disease or tumor affects multiple systems, for example, lymphoma affects the whole body;

[0072] 2) The child has cognitive or developmental abnormalities and mental illnesses;

[0073] 3) The child has already undergone radiotherapy, chemotherapy, and immunotherapy;

[0074] (6) Image quality: During the scan, the child's head should be kept as straight as possible without any displacement; the PET image should have clear edges without artifacts or blurring.

[0075] (7) Brief medical history requirements

[0076] 1) Basic information: Patient's name, gender, and age;

[0077] 2) Present illness: chief complaint, symptoms, and clinical diagnosis.

[0078] For the collected data, image preprocessing includes steps such as image cropping and position correction. First, the minimum bounding box size of the PET image is calculated, and then the surrounding blank areas are cropped. If the image is noisy, additional denoising processing is required. Next, using the Display function in the SPM analysis software, AC-PC (Anterior Commissure-Posterior Commissure) origin correction is manually performed to ensure image standardization and consistency, making it more conducive to subsequent registration operations. For template data, the image is multiplied by its brain label to obtain template data retaining only the brain image for subsequent registration.

[0079] S2, spatially register the PET image to the pediatric MRI brain atlas CHN-PD to obtain the registered PET image.

[0080] Optionally, the SyN algorithm can be used for image registration.

[0081] This invention first employs the SyN registration algorithm to spatially register the preprocessed PET image onto the Beijing Normal University Children's MRI Brain Atlas CHN-PD, thus mapping the PET image into a standard brain atlas space and eliminating morphological differences between subjects. The SyN registration algorithm is based on differential homeomorphic transformation and aims to align the floating image with the reference image through a smooth and reversible deformation field.

[0082] Optionally, the PET image space is spatially registered to the pediatric MRI brain atlas CHN-PD to obtain the registered PET image, including:

[0083] Define the initial deformation field, set the similarity metric function and regularization parameters;

[0084] Calculate the similarity between PET images and pediatric MRI brain atlas CHN-PD;

[0085] Based on the similarity measurement results, an optimization algorithm is used to update the transformation parameters so that the similarity measurement function reaches its maximum value;

[0086] Determine if the iteration has reached the convergence condition. If it has converged, stop the iteration and the spatial registration is complete; otherwise, continue the iterative optimization.

[0087] In SyN registration, the regularization parameter is a core parameter controlling the smoothness of the deformation field and preventing overfitting. Its setting needs to be optimized comprehensively based on image characteristics, registration objectives, and algorithm type. A typical regularization term is the L2 norm of the deformation field gradient: assuming the deformation field parameters are vectors. Then the L2 norm regularization term is Where λ is the regularization parameter. In the objective function of SyN registration, it typically takes the form:

[0088]

[0089] By minimizing this objective function, the complexity of the deformation field can be controlled by limiting the size of the deformation field parameters while optimizing the similarity metric.

[0090] Optimization algorithms include gradient descent, conjugate gradient, etc.

[0091] The updated transformation parameters are deformation field parameters, typically displacement vector field parameters and velocity field parameters. The entire update process is iterative. In each iteration, a similarity metric is calculated based on the current deformation field parameters, and then the deformation field parameters are updated based on the gradient of the objective function. This process is repeated until the convergence condition is met.

[0092] The convergence condition is that the change in the objective function value is less than a preset threshold, the change in the deformation field parameter is less than a preset threshold, or the maximum number of iterations is reached.

[0093] Optionally, the similarity between PET images and pediatric MRI brain atlas CHN-PD is calculated, including:

[0094] The similarity index (MI) between PET images and pediatric MRI brain atlases (CHN-PD) is calculated using the mutual information method, as follows:

[0095] Where p(x, y) represents the joint probability distribution at corresponding pixel (x, y) in the two images, and p(x) and p(y) represent the edge probability distributions at pixel x and y in the two images, respectively.

[0096] The performance of the SyN algorithm is highly dependent on the similarity measurement function, with mutual information (MI) being a common method. Mutual information, as a similarity measurement method, treats two images as two random variables and measures their similarity by calculating their mutual information value. A higher mutual information value indicates a higher correlation between the two images, resulting in better registration. This measurement method has the advantage of being insensitive to image grayscale variations and noise, making it very suitable for image registration.

[0097] The formula for calculating mutual information is usually expressed as:

[0098]

[0099] In the case of continuous random variables, summation is replaced by double definite integrals:

[0100]

[0101] The output of this step is the participant's PET image after spatial registration with the Beijing Normal University Children's MRI Brain Atlas CHN-PD, which is in the standard brain atlas space, and can be used for the subsequent template construction operation.

[0102] S3. Perform intensity normalization and data filtering on the registered PET image to obtain the filtered normalized PET image.

[0103] Optionally, the registered PET images are subjected to intensity normalization and data filtering to obtain filtered normalized PET images, including:

[0104] The registered PET images are normalized according to the following formula:

[0105] I k (i)=G k (i) / G k_mean_ref Among them, G k For the registered PET image of participant k, i is the pixel index, G k_mean_ref I represents the average pixel value of the cerebellar reference region for the participants. k The image shows the normalized PET image of participant k.

[0106] The method for normalizing CHN-PD intensity in pediatric MRI brain atlases is the same as above.

[0107] The sum of absolute errors between intensity-normalized PET images and children's MRI brain atlas CHN-PD is calculated using the following formula:

[0108] SAD k =∑ i |I k (i)-I ref (i)|, where, I ref For normalized pediatric MRI brain atlas CHN-PD, i is the pixel index, SAD k For I k with I ref The sum of the horizontal absolute errors between the images;

[0109] Anomaly detection is performed on intensity-normalized PET images using the following formula:

[0110] DIF k =|SAD k -SAD mean |-3×SAD std Among them, SAD mean and SAD std point

[0111] Let SAD be the mean and standard deviation of all participants; if DIF k If the value is greater than 0, the participant k data is considered to have potential metabolic abnormalities and needs to be removed. The remaining PET images are used as the normalized PET images after screening.

[0112] Considering the potential for inconsistent pixel intensity ranges among participants' PET images due to variations in the image acquisition process, intensity normalization was performed in this step to ensure that all participants' images were within a comparable intensity range. The method employed was to divide all voxel values ​​of the registered participant's PET image by the average pixel value of a reference region. The reference region was the participant's own cerebellum (using the AAL116 partition for region division; alternatively, the alairach partition, Brodmann partition, and Harvard-Oxford Atlas could also be selected).

[0113] To ensure the health and normality of the participant dataset, a data screening step was performed after data spatial registration and intensity normalization. The method used was the Sum of Absolute Differences (SAD) algorithm.

[0114] After calculating the SAD of all participants, a statistical method is used to detect anomalies. A relatively simple parametric anomaly detection model assumes that the samples follow a univariate normal distribution, and a data point is considered an anomaly if its difference from the mean is greater than three standard deviations.

[0115] S4. The average value of the selected normalized PET images is calculated to obtain preliminary PET brain template images of children.

[0116] Alternatively, the average can be calculated using the following formula:

[0117] Where i represents the pixel index, n represents the number of images used for averaging, and I k denoted as the intensity-normalized image of the k-th participant, and I represents the preliminary pediatric PET brain template obtained by averaging.

[0118] S5. The selected PET images are spatially registered to the preliminary pediatric PET brain template images, and intensity normalization, data filtering, averaging and smoothing are performed to obtain the final pediatric PET brain template images.

[0119] The specific steps for spatially registering the selected PET images to the initial pediatric PET brain template images, and then performing intensity normalization, data filtering, averaging, and smoothing to obtain the final pediatric PET brain template images are similar to those in the above method and will not be repeated here. The final PET brain template construction process is largely the same as the initial PET brain template construction process, both involving four steps: image-template spatial registration, intensity normalization, data filtering, and averaging. However, there are some differences, as follows:

[0120] In the image space registration process of the final brain template construction, the preliminary PET brain template obtained in the above steps is used as the registration reference template image, and the SyN registration algorithm is still used.

[0121] The template used is a preliminary PET brain template. Therefore, when performing image screening for deviation calculation, the degree of deviation between the intensity-normalized image and the preliminary PET brain template should be calculated.

[0122] The final template image obtained by averaging needs to undergo additional smoothing to reduce the noise caused by individual anatomical differences and to conform to the Gaussian distribution assumption, providing a basis for subsequent use. The smoothing parameter is determined based on parameters such as the spatial resolution of the collected participant PET images, and is full width at half height FWHM = [4 4 4].

[0123] After this step, the final PET brain template for children is constructed.

[0124] S6. Spatial registration of PET images to the final pediatric PET brain template image and division of brain regions. Calculation of the first quantitative parameter of each brain region for each normal participant and calculation of the mean and standard deviation of each brain region and each parameter for all normal participants. By calculating the corresponding mean ± 2 × standard deviation, the normal metabolic range is obtained, and the construction of the brain metabolism database is completed.

[0125] The normal metabolic range was obtained by calculating the mean ± 2 × standard deviation for each brain region and each parameter in all normal participants.

[0126] S7. After spatially registering the PET image of the child to be analyzed to the final PET brain template image of the child and dividing the brain regions, calculate the second quantitative parameter and compare it with the brain metabolism database to complete the quantitative analysis.

[0127] Optionally, the first quantitative parameter includes the average gray value (Gray). mean Average standard intake value for SUVs mean The second quantitative parameter includes the gray average value (Gray) compared to the standard intake value (SUVr). mean Average standard intake value for SUVs mean Standard intake value ratio (SUVr) and z-score (Z-score).

[0128] Optionally, the grayscale average value Gray is calculated according to the following formula. mean Average standard intake value for SUVs mean Standard intake value ratio (SUVr) and z-score:

[0129]

[0130] Where R represents the individual image registered with the final pediatric PET brain template, S represents the individual image after converting voxel values ​​to SUV values, i represents the pixel index, roi represents the region of interest, n represents the number of pixels, ref represents the cerebellar region of the image itself, and database represents the brain metabolism data database; the formula for calculating the SUV value is:

[0131] When calculating the first quantitative parameter for constructing a brain metabolism database, the individual image is an image of a normal participant; when calculating the second quantitative parameter for quantitative analysis, the individual image is a PET image of a child to be analyzed.

[0132] Optionally, the method further includes:

[0133] The regional anatomical differences between the final pediatric PET brain template and the pediatric MRI brain atlas CHN-PD image were calculated based on the Mean Sum of Squared Errors (MSD) algorithm and the Normalized Product Correlation (NCC) algorithm. The calculation formula is as follows:

[0134]

[0135] Where I represents the final pediatric PET brain template obtained from the construction, I ref This represents the CHN-PD MRI brain atlas for children, where i represents the pixel index and n represents the pixel index. roi i_roi represents the number of pixels in the region of interest, and i_roi represents the pixel index within the region of interest. mean_roi and I ref_mean_roi Representing images I and I respectively ref The average pixel value within the region of interest; the smaller the MSD value, the smaller the anatomical difference between the two image regions; the larger the MSD value, the greater the anatomical difference between the two image regions; the closer the NCC value is to 1, the smaller the anatomical difference between the two image regions; the closer the NCC value is to -1, the greater the anatomical difference between the two image regions; the closer the NCC value is to 0, the two image regions are uncorrelated.

[0136] This invention provides an evaluation method for the constructed template to ensure its accuracy and reliability. The template is compared with the Beijing Normal University MRI template CHN-PD, which serves as the research baseline, to calculate the regional anatomical differences between the two template images. The region segmentation label uses AAL116 partitions (in addition to AAL partitions, Talairach, Brodmann, and Harvard-Oxford Atlas partitions can also be selected). The calculation metrics employ the Mean Square Differences (MSD) algorithm and the Normalized Cross Correlation (NCC) algorithm.

[0137] Optionally, PET data of normal participants of different ages and genders can be obtained to construct a whole template for children aged 6-12 years and sub-templates with an age interval of 1 year. The whole template for children aged 6-12 years includes brain templates for boys, girls and children of undifferentiated genders, and a database of brain metabolism data corresponding to the templates can be obtained through quantitative analysis.

[0138] This invention constructs a set of age- and sex-specific brain templates for children. The age templates are divided into templates for 6-7 years, 7-8 years, 8-9 years, 9-10 years, 10-11 years, and 11-12 years, with 1-year intervals, as well as a general template for 6-12 years. The general template for 6-12 years further includes templates for boys, girls, and gender-neutral templates. The construction method has been detailed in the above steps. The difference is that during the template construction process, the CHN-PD pediatric MRI template from Beijing Normal University should also be selected as the baseline corresponding to the age and sex.

[0139] In clinical applications, patient images are first spatially registered to pre-constructed age and gender templates, using the SyN registration algorithm. Then, the registered images are segmented into brain regions based on AAL partitioning (in addition to AAL partitioning, Talairach, Brodmann, and Harvard-Oxford Atlas partitioning can also be used, and the partitioning method must be consistent with that used to construct the database), and the quantitative parameter value Gray is calculated. mean SUV mean SUVr and Z-score; finally, by comparing the calculated values ​​with the corresponding template metabolic values ​​in the brain metabolism database, the degree of abnormality in the patient's brain region can be determined. Gray mean SUV mean If the SUVr value falls outside the normal metabolic range in the brain metabolism database, it indicates an abnormality in the brain region. The Z-score is analyzed using a threshold: if 1.96 ≤ |Z-score| < 2.58, it indicates a mild abnormality; if |Z-score| ≥ 2.58, it indicates a moderate to severe abnormality. Through these analyses, the abnormalities in the brain regions of patients can be identified, aiding in clinical diagnosis.

[0140] This invention can be implemented as a software module, including functions such as image uploading, template registration, quantitative report generation, and result analysis. It can also be installed as a plugin into third-party software to improve the applicability of the template.

[0141] This invention ultimately yields age-specific (overall templates for children aged 6-12 years and sub-templates at 1-year intervals) and sex-specific (overall templates for children aged 6-12 years include boys, girls, and gender-neutral templates) child brain templates, and quantitatively analyzes them to obtain a database of brain metabolism data corresponding to the templates. This invention also provides quantitative analysis methods and software tools for clinical applications. Figure 2 , Figure 3 and Figure 4 The image shown is a schematic diagram of three directions for one of the established templates—a template for children aged 6-12 years who are not gender-specific.

[0142] In a second aspect, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed, implements the method for constructing and quantitatively analyzing a pediatric PET brain template as described in any of the first aspects above.

[0143] Thirdly, embodiments of the present invention provide a storage device, including a storage medium and a processor, wherein the storage medium stores a computer program, and when the program is executed by the processor, it implements the method for constructing and quantitatively analyzing a pediatric PET brain template as described in any of the first aspects above.

[0144] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0145] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, then this invention should also include these modifications and variations.

[0146] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make modifications, alterations, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for constructing and quantitatively analyzing a pediatric PET brain template, characterized in that, include: Obtain preprocessed PET images of normal children participants; The PET images were spatially registered to the pediatric MRI brain atlas CHN-PD to obtain the registered PET images. Intensity normalization and data filtering were performed on the registered PET images to obtain the filtered normalized PET images; The average of the selected normalized PET images was calculated to obtain preliminary PET brain template images of children. The selected PET images were spatially registered to the preliminary pediatric PET brain template images, and intensity normalization, data filtering, averaging and smoothing were performed to obtain the final pediatric PET brain template images. The PET images were spatially registered to the final pediatric PET brain template images and the brain regions were divided. The first quantitative parameter of each brain region of each normal participant was calculated, and the mean and standard deviation of each brain region and each parameter of all normal participants were calculated. By calculating the corresponding mean ± 2 × standard deviation, the normal metabolic range was obtained, and the construction of the brain metabolism database was completed. After spatially registering the PET images of the children to be analyzed to the final PET brain template image of the children and dividing the brain regions, the second quantitative parameter is calculated and compared with the brain metabolism database to complete the quantitative analysis.

2. The method for constructing and quantitatively analyzing a pediatric PET brain template according to claim 1, characterized in that, The PET images were spatially registered to the pediatric MRI brain atlas CHN-PD to obtain the registered PET images, including: Define the initial deformation field, set the similarity metric function and regularization parameters; Calculate the similarity between PET images and pediatric MRI brain atlas CHN-PD; Based on the similarity measurement results, an optimization algorithm is used to update the transformation parameters so that the similarity measurement function reaches its maximum value; Determine if the iteration has reached the convergence condition. If it has converged, stop the iteration and the spatial registration is complete; otherwise, continue the iterative optimization.

3. The method for constructing and quantitatively analyzing a pediatric PET brain template according to claim 2, characterized in that, Calculate the similarity between PET images and pediatric MRI brain atlas CHN-PD, including: The similarity index (MI) between PET images and pediatric MRI brain atlases (CHN-PD) is calculated using the mutual information method, as follows: Where p(x, y) represents the joint probability distribution at corresponding pixel (x, y) in the two images, and p(x) and p(y) represent the edge probability distributions at pixel x and y in the two images, respectively.

4. The method for constructing and quantitatively analyzing a pediatric PET brain template according to claim 3, characterized in that, The registered PET images are subjected to intensity normalization and data filtering to obtain the filtered normalized PET images, including: The intensity of the registered PET image is normalized according to the following formula: I k (i)=G k (i) / G k_mean_ref Among them, G k For the registered PET image of participant k, i is the pixel index, G k_mean_ref I represents the average pixel value of the cerebellar reference region for the participants. k PET images of participant k after normalization; The sum of absolute errors between intensity-normalized PET images and children's MRI brain atlas CHN-PD is calculated using the following formula: SAD k =∑ i |I k (i)-I ref (i)|, where, I ref For normalized pediatric MRI brain atlas CHN-PD, i is the pixel index, SAD k For I k with I ref The sum of the horizontal absolute errors between the images; Anomaly detection is performed on intensity-normalized PET images using the following formula: DIF k =|SAD k -SAD mean |-3×SAD std Among them, SAD mean and SAD std point Let SAD be the mean and standard deviation of all participants; if DIF k If the value is greater than 0, the participant k data is considered to have potential metabolic abnormalities and needs to be removed. The remaining PET images are used as the normalized PET images after screening.

5. The method for constructing and quantitatively analyzing a pediatric PET brain template according to claim 4, characterized in that, The average is calculated using the following formula: Where i represents the pixel index, n represents the number of images used for averaging, and I k denoted as the intensity-normalized image of the k-th participant, and I represents the preliminary pediatric PET brain template obtained by averaging.

6. The method for constructing and quantitatively analyzing a pediatric PET brain template according to claim 5, characterized in that, The method further includes: The regional anatomical differences between the final pediatric PET brain template and the pediatric MRI brain atlas CHN-PD image were calculated based on the Mean Sum of Squared Errors (MSD) algorithm and the Normalized Product Correlation (NCC) algorithm. The calculation formula is as follows: Where I represents the final pediatric PET brain template obtained from the construction, I ref This represents the CHN-PD MRI brain atlas for children, where i represents the pixel index and n represents the pixel index. roi i_roi represents the number of pixels in the region of interest, and i_roi represents the pixel index within the region of interest. mean_roi and I ref_mean_roi Representing images I and I respectively ref The average pixel value within the region of interest; the smaller the MSD value, the smaller the anatomical difference between the two image regions; the larger the MSD value, the greater the anatomical difference between the two image regions; the closer the NCC value is to 1, the smaller the anatomical difference between the two image regions; the closer the NCC value is to -1, the greater the anatomical difference between the two image regions; the closer the NCC value is to 0, the two image regions are uncorrelated.

7. The method for constructing and quantitatively analyzing a pediatric PET brain template according to claim 6, characterized in that, The first quantitative parameter includes the average gray value (Gray). mean Average standard intake value for SUVs mean The second quantitative parameter includes the gray average value (Gray) compared to the standard intake value (SUVr). mean Average standard intake value for SUVs mean Standard intake value ratio (SUVr) and z-score (Z-score).

8. The method for constructing and quantitatively analyzing a pediatric PET brain template according to claim 7, characterized in that, The average grayscale value (Gray) is calculated using the following formula. mean Average standard intake value for SUVs mean Standard intake value ratio (SUVr) and z-score: Where R represents the individual image registered with the final pediatric PET brain template, S represents the individual image after converting voxel values ​​to SUV values, i represents the pixel index, roi represents the region of interest, n represents the number of pixels, ref represents the cerebellar region of the image itself, and database represents the brain metabolism data database; the formula for calculating the SUV value is: When calculating the first quantitative parameter for constructing a brain metabolism database, the individual image is an image of a normal participant; when calculating the second quantitative parameter for quantitative analysis, the individual image is a PET image of a child to be analyzed.

9. The method for constructing and quantitatively analyzing a pediatric PET brain template according to claim 8, characterized in that, The SyN algorithm was used for image registration.

10. The method for constructing and quantitatively analyzing a pediatric PET brain template according to claim 9, characterized in that, PET data of normal participants of different ages and genders were obtained to construct a global template for children aged 6-12 years and sub-templates with 1-year intervals. The global template for children aged 6-12 years included brain templates for boys, girls and children of undifferentiated genders, and a database of brain metabolism data corresponding to the templates was obtained through quantitative analysis.

Images