Data processing method and device and electronic equipment

Abstract

The invention relates to a data processing method and device and electronic equipment, and relates to the technical field of data processing.The method comprises the steps that when structure sequence data are obtained, image processing is conducted on the structure sequence data, and individual brain structure data are obtained; reading a target file in the individual brain structure data, and generating three-dimensional surface model results of different brain regions according to the target file; reading individual brain region volume data in the individual brain region data, and mapping the individual brain region volume data to preset crowd brain region volume distribution to obtain brain region volume distribution; performing logistic regression operation on the individual brain area volume data to obtain individual Alzheimer's disease risk level assessment; and performing individual brain tumor prediction processing on the structure sequence data to obtain an individual brain tumor risk prediction result. The application can promote the application of the 3D printing technology in the field of brain science, meet diversified analysis requirements, convert scientific achievements into actual analysis tools, and solve the defects in the aspect of disease prediction in the prior art.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a data processing method, apparatus and electronic device. Background Technology

[0002] In the fields of neuroscience and medical imaging, the need for analyzing and processing brain region data is increasing. However, existing technologies for analyzing and processing brain region data still have limitations and shortcomings.

[0003] Specifically, firstly, regarding brain region volume, existing technologies such as BrainKey.ai only provide a set of population norms, which cannot meet diverse analytical needs. This results in many shortcomings of existing technologies in terms of the breadth, adaptability, accuracy, and automation of data analysis.

[0004] Secondly, although the existing FreeSurfer software can distinguish brain regions, there is currently no platform on the market that can automatically generate personal brain region model STL files from data analyzed by FreeSurfer, thus limiting the application of 3D printing technology in the field of brain science.

[0005] Finally, in terms of disease prediction, although there are some research results on the prediction of Alzheimer's disease and brain tumors, most of these models remain at the level of academic papers and have not yet been transformed into practical analytical tools. Summary of the Invention

[0006] In view of this, this application provides a data processing method, apparatus and electronic device, the main purpose of which is to solve the problem of the lack of fully automated NMR data production platform in the market.

[0007] According to a first aspect of this disclosure, a data processing method is provided, the method comprising:

[0008] Acquire multiple raw NMR data;

[0009] When the original MRI data is structural sequence data, image processing is performed on the structural sequence data to obtain individual brain structure data, wherein the individual brain structure data includes individual brain region data;

[0010] Read the target file from the individual brain structure data, generate three-dimensional surface model results for different brain regions based on the target file, and convert the three-dimensional surface model results into the target file format;

[0011] Read the individual brain region volume data from the individual brain region data, and map the individual brain region volume data to a preset population brain region volume distribution to obtain the brain region volume distribution;

[0012] Logistic regression analysis was performed on the individual brain region volume data to obtain an individual disease risk level assessment, which included an individual Alzheimer's disease risk level assessment.

[0013] The structural sequence data is processed for individual brain tumor prediction to obtain individual brain tumor risk prediction results.

[0014] According to a second aspect of this disclosure, a data processing apparatus is provided, the apparatus comprising:

[0015] The acquisition module is used to acquire multiple raw NMR data.

[0016] The first processing module is used to perform image processing on the original MRI data when the original MRI data is structural sequence data to obtain individual brain structure data, wherein the individual brain structure data includes individual brain region data.

[0017] The generation module is used to read the target file from the individual brain structure data, generate three-dimensional surface model results of different brain regions based on the target file, and convert the three-dimensional surface model results into the target file format;

[0018] The mapping module is used to read the individual brain region volume data from the individual brain region data and map the individual brain region volume data to a preset population brain region volume distribution to obtain the brain region volume distribution.

[0019] The computation module is used to perform logistic regression on the individual brain region volume data to obtain an individual disease risk level assessment, which includes an individual Alzheimer's disease risk level assessment.

[0020] The prediction module is used to perform individual brain tumor prediction processing on the structural sequence data to obtain individual brain tumor risk prediction results.

[0021] According to a third aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of the first aspect described above.

[0022] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause a computer to perform the method of the first aspect described above.

[0023] According to a fifth aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method described in the first aspect above.

[0024] Compared with existing technologies, the data processing method, apparatus, and electronic equipment disclosed herein acquire multiple raw MRI data sets. When the raw MRI data are structural sequence data, image processing is performed on the structural sequence data to obtain individual brain structure data, which includes individual brain region data. A target file is read from the individual brain structure data, and three-dimensional surface model results for different brain regions are generated based on the target file. The three-dimensional surface model results are then converted into a target file format. Individual brain region volume data is read from the individual brain region data and mapped to a preset population brain region volume distribution to obtain a brain region volume distribution. Logistic regression is performed on the individual brain region volume data to obtain an individual disease risk assessment, which includes an individual Alzheimer's disease risk assessment. Individual brain tumor prediction processing is performed on the structural sequence data to obtain an individual brain tumor risk prediction result. In this embodiment, a personal brain region model (i.e., a three-dimensional surface model result) can be automatically generated from the raw MRI data and converted into a target file format suitable for 3D printing, thereby promoting the application of 3D printing technology in the field of brain science. The pre-defined population brain region volume distribution is a diverse database containing brain region volume data from different age groups and genders. This application maps individual brain region volume data to the pre-defined population brain region volume distribution to obtain the distribution of an individual's brain region volume within the population, thereby assessing whether an individual's brain region volume is above, below, or at the population average. Through more flexible and personalized brain region volume analysis, it meets diverse analytical needs and addresses the shortcomings of existing technologies in terms of the breadth, adaptability, accuracy, and automation of data analysis. This application's solution uses logistic regression analysis on individual brain region volume data to obtain an individual disease risk assessment, and performs individual brain tumor prediction processing on structural sequence data to obtain individual brain tumor risk prediction results, thus transforming research findings into practical analytical tools and addressing the shortcomings of existing technologies in disease prediction.

[0025] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0026] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0027] To more clearly illustrate the technical applications in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a schematic flowchart of a data processing method provided in an embodiment of the present disclosure;

[0029] Figure 2 A three-dimensional surface model of the brain provided in this embodiment of the present disclosure;

[0030] Figure 3 An Alzheimer's disease susceptibility index map provided in an embodiment of this disclosure;

[0031] Figure 4 This is a schematic flowchart of another data processing method provided in an embodiment of the present disclosure;

[0032] Figure 5 A brain structure diagram provided in an embodiment of this disclosure;

[0033] Figure 6-9 A trend distribution chart based on BGI norms and world norms is provided for embodiments of this disclosure;

[0034] Figure 10 This is a schematic diagram of the structure of a data processing apparatus provided in an embodiment of the present disclosure. Detailed Implementation

[0035] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description. It should be noted that, unless otherwise specified, the embodiments of this disclosure and the features described therein can be combined with each other.

[0036] The data processing methods, apparatus, and electronic devices of embodiments of this disclosure are described below with reference to the accompanying drawings.

[0037] This disclosure provides a data processing method, apparatus, and electronic device, primarily aimed at addressing the technical problems that existing technologies cannot meet diverse analytical needs, cannot automatically generate personal brain region model STL files, and cannot transform research results into practical analytical tools.

[0038] like Figure 1As shown, embodiments of this disclosure provide a data processing method, which may include:

[0039] Step 101: Obtain multiple raw NMR data.

[0040] The raw NMR data can be the raw signal data obtained through NMR scanning.

[0041] In the embodiments of this disclosure, the executing entity may be a data processing device or equipment, which can be used to acquire multiple raw NMR data. The Flowhub cloud computing platform is used as an example to illustrate the technical solutions in this disclosure, but this does not constitute a specific limitation on the technical solutions in this disclosure. Flowhub can be a workflow hosting platform.

[0042] Step 102: When the original MRI data is structural sequence data, perform image processing on the structural sequence data to obtain individual brain structure data, which includes individual brain region data.

[0043] Structural sequence data can be medical image data obtained through magnetic resonance imaging (MRI) technology, showing the anatomical structure of an individual's brain. In MRI scans, structural sequences can provide detailed information about brain morphology, such as gray matter regions and white matter regions.

[0044] Individual brain structure data refers to the individual brain structure and tissue obtained through image processing technology based on structural sequence data. It can be used for neuroscience research, clinical diagnosis, and surgical planning.

[0045] Brain regions refer to different functional areas of the brain, each with its specific function and role. Brain regions can be classified and named according to the brain's anatomical structure and functional characteristics.

[0046] Individual brain region data can include quantitative measurement indicators such as individual brain region volume data, individual brain region surface area data, and individual brain region thickness data.

[0047] In this embodiment of the disclosure, when the original MRI data is structural sequence data, image processing is performed on the structural sequence data to obtain individual brain structure data. The image processing process may include image segmentation, image registration, image correction, image reconstruction, etc., and is not specifically limited thereto.

[0048] Step 103: Read the target file from the individual brain structure data, generate three-dimensional surface model results for different brain regions based on the target file, and convert the three-dimensional surface model results into the target file format.

[0049] The target file can be an aseg.mgz file, and the target file format can be an STL file.

[0050] In this embodiment of the disclosure, the Freesurfer software in the MRI data processing platform can be used to analyze and process individual brain structure data. Freesurfer can be an open-source toolkit for processing and analyzing brain MRI images (such as individual brain structure data).

[0051] Before using Freesurfer software to analyze and process individual brain structure data, one possible approach is to first obtain the corresponding execution software (such as Freesurfer) for performing the MRI sequence analysis process. Docker containerization technology can then be used to deploy all the execution software in a containerized form, creating a complete MRI sequence analysis workflow. Docker containerization ensures that each software program operates in the same environment, avoiding software compatibility issues caused by environmental differences, and also simplifies the software deployment and maintenance process.

[0052] By using Docker containerization and version control, we ensured that the versions of dependent software used were consistent with the paradigm, reducing the possibility of parameter variables and errors. Furthermore, the NMR data processing platform allows for the simultaneous submission of multiple tasks for batch analysis, saving time and improving analysis efficiency.

[0053] Version control is a method for managing software versions, ensuring that the versions of dependent software used are consistent with the paradigm, avoiding compatibility issues caused by using different versions of software, and ensuring the consistency and accuracy of analysis.

[0054] In the embodiments of this disclosure, when using the NMR data processing platform to process data according to the analysis process corresponding to the original NMR data in the NMR sequence analysis process, the data can be processed quickly, and the number of tasks delivered at one time is unlimited. Users do not need to wait for the tasks to complete and can process multiple tasks at the same time. When a large amount of data needs to be analyzed (e.g., 500 bits of data), the analysis speed will not be affected, which means that the processing time is the same as the time required to analyze a small amount of data (e.g., 1 bit of data).

[0055] This disclosure reduces repetitive and tedious configuration tasks and simplifies software installation and deployment by integrating multiple sequence analyses into a single workflow. This improves the scalability and usability of the analysis platform, thereby increasing work efficiency and output quality.

[0056] Compared to existing technologies, this disclosure is the first standardized platform that integrates multiple sequence analyses and can be infinitely expanded.

[0057] The aseg.mgz file can contain labels for multiple brain regions, which correspond to different brain structures, such as gray matter, white matter, ventricles, amygdala, brainstem, hippocampus, etc., without specific limitations; the STL file is a standard file format used for 3D printing and manufacturing, which describes the geometry of a 3D object.

[0058] In this embodiment of the disclosure, the Freesurfer software can be used to read the aseg.mgz file from individual brain structure data, and based on original code, three-dimensional surface model results of different brain regions can be automatically generated using the aseg.mgz file. The three-dimensional surface model results can then be converted into STL file format, where the three-dimensional surface model results can contain more than 20 brain regions, such as... Figure 2 As shown, users can easily browse their own unique brain anatomy model on their mobile phones.

[0059] In this embodiment of the disclosure, the application of 3D printing technology in the field of brain science is promoted by automatically generating a personal brain region model (i.e., a three-dimensional surface model result) from raw MRI data and converting it into an STL file format that can be used for 3D printing.

[0060] Step 104: Read the individual brain region volume data from the individual brain region data, and map the individual brain region volume data to the preset population brain region volume distribution to obtain the brain region volume distribution.

[0061] In this embodiment of the disclosure, individual brain region volume data can be read from individual brain region data, wherein the individual brain region volume data may include data such as individual cortical gray matter volume, individual subcortical gray matter volume, individual cortical white matter volume, and individual ventricular volume.

[0062] Individual brain region volume data can be mapped to a preset population brain region volume distribution to obtain the brain region volume distribution. The preset population brain region volume distribution can be a database containing brain region volume data of people of different ages, genders, etc.

[0063] By mapping individual brain region volume data to a predefined population brain region volume distribution, the distribution of an individual's brain region volume within the same age and sex population can be obtained. This allows for the assessment of whether an individual's brain region volume is higher, lower, or at the average level of their age and sex population. Through more flexible and personalized brain region volume analysis, diverse analytical needs can be met, addressing the shortcomings of existing technologies in terms of data analysis breadth, adaptability, accuracy, and automation.

[0064] Among them, brain region volume distribution can be used to represent the position of an individual's brain region volume relative to the average level of people of the same age and sex.

[0065] Step 105: Perform logistic regression on the individual brain region volume data to obtain the individual disease risk level assessment, which includes the individual Alzheimer's disease risk level assessment.

[0066] In this embodiment of the disclosure, an individual risk assessment of disease incidence, such as an assessment of the risk of developing Alzheimer's disease, can be calculated based on individual brain region volume data. Since the volume of certain brain regions (such as the hippocampus and amygdala) is significantly reduced in Alzheimer's patients, as a possible method, a logistic regression model can be used to perform logistic regression calculations on the individual brain region volume data to obtain the individual risk assessment of disease incidence.

[0067] In medical research, logistic regression can be used to predict the probability of disease occurrence by analyzing various indicators of an individual (such as brain region volume) to predict their risk of developing the disease.

[0068] In specific application scenarios, such as Figure 3 As shown, based on individual brain region volume data, the logistic regression model calculated the current probability of the subject having Alzheimer's disease to be 1.87%. As a possible approach, the current probability of disease can be compared with a preset probability of disease. If the current probability of disease is greater than the preset probability of disease, it indicates that the individual has an extremely high risk of Alzheimer's disease; if the current probability of disease is equal to the preset probability of disease, it indicates that the individual has a relatively high risk of Alzheimer's disease; and if the current probability of disease is less than the preset probability of disease, it indicates that the individual has a relatively low risk of Alzheimer's disease.

[0069] Step 106: Perform individual brain tumor prediction processing on the structural sequence data to obtain individual brain tumor risk prediction results.

[0070] In this embodiment of the disclosure, an individual brain tumor prediction model can be used to process structural sequence data for individual brain tumor prediction, thereby obtaining individual brain tumor risk prediction results, thus transforming scientific research results into practical analytical tools and solving the shortcomings of existing technologies in disease prediction.

[0071] In summary, the data processing method provided in this disclosure, compared with the prior art, involves: acquiring multiple raw MRI data sets; when the raw MRI data are structural sequence data, performing image processing on the structural sequence data to obtain individual brain structure data, which includes individual brain region data; reading target files from the individual brain structure data, generating three-dimensional surface model results for different brain regions based on the target files, and converting the three-dimensional surface model results into a target file format; reading individual brain region volume data from the individual brain region data, mapping the individual brain region volume data to a preset population brain region volume distribution to obtain a brain region volume distribution; performing logistic regression on the individual brain region volume data to obtain an individual disease risk level assessment, which includes an individual Alzheimer's disease risk level assessment; and performing individual brain tumor prediction processing on the structural sequence data to obtain an individual brain tumor risk prediction result. In this embodiment, a personal brain region model (i.e., a three-dimensional surface model result) can be automatically generated from the raw MRI data and converted into a target file format suitable for 3D printing, thereby promoting the application of 3D printing technology in the field of brain science. The pre-defined population brain region volume distribution is a diverse database containing brain region volume data from different age groups and genders. This application maps individual brain region volume data to the pre-defined population brain region volume distribution to obtain the distribution of an individual's brain region volume within the population, thereby assessing whether an individual's brain region volume is above, below, or at the population average. Through more flexible and personalized brain region volume analysis, it meets diverse analytical needs and addresses the shortcomings of existing technologies in terms of the breadth, adaptability, accuracy, and automation of data analysis. This application's solution uses logistic regression analysis on individual brain region volume data to obtain an individual disease risk assessment, and performs individual brain tumor prediction processing on structural sequence data to obtain individual brain tumor risk prediction results, thus transforming research findings into practical analytical tools and addressing the shortcomings of existing technologies in disease prediction.

[0072] Furthermore, as a refinement and extension of the above embodiments, and in order to fully illustrate the specific implementation process of the method disclosed herein, this disclosure provides the following... Figure 4 The specific method shown includes:

[0073] Step 201: Perform image processing on the structural sequence data to obtain individual brain structure data.

[0074] In this embodiment of the disclosure, image processing is performed on structural sequence data to obtain individual brain structure data. Specifically, this may include using an image segmentation algorithm to identify the skull and other non-brain tissues in the structural sequence data, and removing the skull and other non-brain tissues.

[0075] Image registration was performed on the structural sequence data after removing the skull and other non-brain tissues, and intensity non-uniformity correction was performed on the registered structural sequence data to obtain the corrected structural sequence data.

[0076] The corrected structural sequence data is processed by image segmentation to obtain different brain tissue types, and the relevant brain tissue types are mapped to the cortical surface to obtain the reconstructed cerebral cortex surface. According to the preset brain region template or preset anatomical structure, the reconstructed cerebral cortex surface is divided into brain regions to obtain different individual brain structure data.

[0077] In specific application scenarios, raw MRI data (such as DCM format or Siemens proprietary IMA format) uploaded to the MRI data processing platform is automatically converted into the standard image processing formats NIFTI or BIDS. Traditional tools like dcm2niix can only convert DCM format data to NIFTI, while the workflow integrated into the MRI data processing platform can convert data from DCM to BIDS format, making data processing more flexible and better meeting the needs of neuroimaging data processing. BIDS format, in particular, has a superior structure and is more suitable for multi-center, multimodal neuroimaging data processing. The entire data processing flow is continuous, and the output of this part can directly connect to the data input of the next dimension without manual intervention, improving data processing efficiency.

[0078] When the original MRI data is structural sequence data, preprocessing is performed on the structural sequence data, such as removing skull images and other non-brain tissue images from the structural sequence data, performing image orientation correction on the structural sequence data, and performing intensity non-uniformity correction on individual brain structure image data (to make the pixel values ​​in the image more uniform), etc.

[0079] Secondly, Freesurfer's automated algorithms can be used to reconstruct the cortical surface of structural sequence data. This can include segmenting the structural sequence data into different tissue types (such as gray matter, white matter, and cerebrospinal fluid), and then mapping these tissue types onto the cortical surface to obtain the reconstructed cortical surface. Freesurfer can divide the cortical surface into different brain regions based on standard brain region templates (i.e., preset brain region templates, such as the MNI template (i.e., the Montreal Neuroscience Institute template)) and individual anatomical structures (i.e., preset anatomical structures). These brain regions can be named and labeled according to their function or anatomical characteristics to generate a text file (i.e., individual brain structure data) containing information such as the volume and thickness of different brain regions of the subject.

[0080] Step 202: Map individual brain region volume data to a preset population brain region volume distribution to obtain the brain region volume distribution.

[0081] Among them, such as Figure 5As shown, individual brain region volume data may include individual cortical gray matter volume, individual subcortical gray matter volume, individual cortical white matter volume, and individual ventricular volume; the preset population brain region volume distribution includes preset foreign population brain region volume distribution and preset domestic population brain region volume distribution.

[0082] In this embodiment of the disclosure, text files can be organized in a special way and integrated into the MRIfig tool. Relying on literature support and self-integrated code, it is possible to generate the brain region volume distribution of subjects at a specific age and gender (such as individual cortical gray matter volume, individual subcortical gray matter volume, individual cortical white matter volume, and individual ventricular volume) in preset foreign population brain region data and preset domestic population brain region data.

[0083] Among them, the preset brain region data of the domestic population is the BGI norm (e.g. Figure 6-9 (as shown); preset brain region data from foreign populations, i.e., world norms (such as...) Figure 6-9 (As shown).

[0084] The two sets of norms, BGI norms and world norms, can be trained on different populations. World norms can be trained using data from foreign populations, while BGI norms can be trained using data from domestic populations. The distribution of individual brain regions differs between the world norms and BGI norms, but the absolute volume of individual brain regions remains unchanged.

[0085] In this embodiment of the disclosure, individual brain region volume data is mapped to a preset population brain region volume distribution to obtain the brain region volume distribution, such as... Figure 6-9 As shown, it may specifically include:

[0086] The individual cortical gray matter volume, individual subcortical gray matter volume, individual cortical white matter volume, and / or individual ventricular volume are mapped to the corresponding preset brain region volume distribution of foreign populations, so as to obtain the distribution level of individual cortical gray matter volume, individual subcortical gray matter volume, individual ventricular volume, and individual white matter volume in foreign populations of the same age and sex.

[0087] The individual cortical gray matter volume, individual subcortical gray matter volume, individual cortical white matter volume, and / or individual ventricular volume are mapped to the corresponding preset brain region volume distribution in the domestic population. This yields the distribution levels of individual cortical gray matter volume, individual subcortical gray matter volume, individual ventricular volume, and individual white matter volume in the domestic population of the same age and sex.

[0088] Step 203: Convert the structural sequence data into JPEG format image data; input the converted structural sequence data into the trained individual brain tumor prediction model; use the individual brain tumor prediction model to perform individual brain tumor prediction processing on the structural sequence data to obtain the individual brain tumor risk prediction result.

[0089] In the embodiments of this disclosure, structural sequence data is typically stored in DICOM format or other formats, but these formats are not supported by all image processing software. For ease of processing and analysis, the image format of the structural sequence data can be converted to a more universal image format, such as JPEG.

[0090] The converted structural sequence data can be input into an individual brain tumor prediction model trained with a large amount of data. This individual brain tumor prediction model can be used to identify image features related to brain tumors.

[0091] The individual brain tumor prediction model can predict the risk of an individual having a brain tumor based on image features related to brain tumors, and output the individual brain tumor risk prediction result. The individual brain tumor risk prediction result can be the probability value of an individual having a brain tumor or a direct indication of whether an individual has a brain tumor.

[0092] Specifically, the training process for an individual tumor prediction model may include:

[0093] Obtain structural sequence data in JPEG format and the corresponding individual brain tumor information;

[0094] The structural sequence data is used as the input feature of the individual tumor prediction model, and the individual brain tumor status is used as the training label of the individual brain tumor prediction model. The individual brain tumor prediction model is trained iteratively until the loss function value of the individual brain tumor prediction model is less than a preset threshold, at which point the individual brain tumor prediction model is considered to have completed training.

[0095] Step 204: When the original MRI data is functional sequence data, the functional sequence data is processed using the fMRIPREP time series processing tool based on the structural sequence data to obtain time series data; brain region functional connectivity is calculated on the time series data to obtain the individual functional connectivity matrix.

[0096] Among them, functional sequence data can be raw data obtained by functional magnetic resonance imaging (fMRI) that records changes in functional activity between different brain regions when the brain is performing a specific task or in a resting state.

[0097] Time series data can be data on the signal intensity of one or more brain regions at each time point recorded within a certain time interval.

[0098] Brain region functional connectivity refers to the degree of functional connection between different brain regions.

[0099] In specific application scenarios, after structural phase data processing, functional phase data processing can be performed automatically. Functional phase data processing can rely on fMRIPREP software, which integrates image registration tools (Advanced Normalization Tools, ANT), FSL (fMRI data analysis), FreeSurfer tools, Nipype (neuroimaging data processing workflow), SPM (functional magnetic resonance imaging data analysis software package), AFNI (data analysis), etc.

[0100] The aforementioned tools can be used to process functional sequence data to obtain time series data. Furthermore, the time series data can be embedded in the Nilearn software package. Based on the original Python code in the Nilearn software package, brain region functional connectivity calculations can be performed on the time series data to obtain the individual functional connectivity matrix for each subject.

[0101] This matrix can be used to derive each subject's age index, IQ index, and so on through a weighted algorithm.

[0102] Step 205: When the original NMR data is diffuse sequence data, the diffuse sequence data is processed using the Mrtrix3 software tool based on the structural sequence data to obtain the diffusion coefficient and anisotropy index.

[0103] Among them, diffusion sequence data can be used to record the diffusion of water molecules in the brain;

[0104] The Apparent Diffusion Coefficient (ADC) is a parameter that quantifies the diffusion capacity of water molecules, reflecting the degrees of freedom of water molecule movement within tissues. Under certain pathological conditions, such as acute stroke, the ADC value in the damaged area decreases because the movement of water molecules within cells is restricted.

[0105] Anisotropy indices can be used to reflect the directionality of diffusion. The most commonly used are fractional anisotropy (FA) and relative anisotropy (RA). A high FA value usually indicates that diffusion mainly occurs in one direction, which is common in white matter fiber bundles; a low FA value indicates that diffusion is more uniform, which is more common in gray matter or damaged white matter.

[0106] In the embodiments of this disclosure, when the original NMR data is diffuse sequence data, the diffuse sequence data can be image-processed using the Mrtrix3 software tool based on the structural sequence data to obtain the diffusion coefficient and anisotropy index.

[0107] In summary, the data processing method provided in this disclosure, compared with the prior art, involves: acquiring multiple raw MRI data sets; when the raw MRI data are structural sequence data, performing image processing on the structural sequence data to obtain individual brain structure data, which includes individual brain region data; reading target files from the individual brain structure data, generating three-dimensional surface model results for different brain regions based on the target files, and converting the three-dimensional surface model results into a target file format; reading individual brain region volume data from the individual brain region data, mapping the individual brain region volume data to a preset population brain region volume distribution to obtain a brain region volume distribution; performing logistic regression on the individual brain region volume data to obtain an individual disease risk level assessment, which includes an individual Alzheimer's disease risk level assessment; and performing individual brain tumor prediction processing on the structural sequence data to obtain an individual brain tumor risk prediction result. In this embodiment, a personal brain region model (i.e., a three-dimensional surface model result) can be automatically generated from the raw MRI data and converted into a target file format suitable for 3D printing, thereby promoting the application of 3D printing technology in the field of brain science. The pre-defined population brain region volume distribution is a diverse database containing brain region volume data from different age groups and genders. This application maps individual brain region volume data to the pre-defined population brain region volume distribution to obtain the distribution of an individual's brain region volume within the population, thereby assessing whether an individual's brain region volume is above, below, or at the population average. Through more flexible and personalized brain region volume analysis, it meets diverse analytical needs and addresses the shortcomings of existing technologies in terms of the breadth, adaptability, accuracy, and automation of data analysis. This application's solution uses logistic regression analysis on individual brain region volume data to obtain an individual disease risk assessment, and performs individual brain tumor prediction processing on structural sequence data to obtain individual brain tumor risk prediction results, thus transforming research findings into practical analytical tools and addressing the shortcomings of existing technologies in disease prediction.

[0108] Based on the above Figure 1 and Figure 4 To provide a specific implementation of the method shown, this embodiment offers a data processing device, such as... Figure 10 As shown, the device includes: an acquisition module 31, a first processing module 32, a generation module 33, a mapping module 34, a calculation module 35, and a prediction module 36;

[0109] Acquisition module 31 is used to acquire multiple raw NMR data;

[0110] The first processing module 32 is used to perform image processing on the original MRI data when the original MRI data is structural sequence data to obtain individual brain structure data, wherein the individual brain structure data includes individual brain region data.

[0111] The generation module 33 is used to read the target file in the individual brain structure data, generate three-dimensional surface model results of different brain regions according to the target file, and convert the three-dimensional surface model results into the target file format.

[0112] Mapping module 34 is used to read individual brain region volume data from the individual brain region data, and map the individual brain region volume data to a preset population brain region volume distribution to obtain the brain region volume distribution.

[0113] The calculation module 35 is used to perform logistic regression calculation on the individual brain region volume data to obtain an individual disease risk level assessment, which includes an individual Alzheimer's disease risk level assessment.

[0114] The prediction module 36 is used to perform individual brain tumor prediction processing on the structural sequence data to obtain individual brain tumor risk prediction results.

[0115] In specific application scenarios, the first processing module 32 can be used to identify the skull and other non-brain tissues in the structural sequence data using an image segmentation algorithm, and remove the skull and other non-brain tissues.

[0116] Image registration processing is performed on the structural sequence data after removing the skull and other non-brain tissues, and intensity non-uniformity correction processing is performed on the registered structural sequence data to obtain the corrected structural sequence data.

[0117] The corrected structural sequence data is processed by image segmentation to obtain different brain tissue types, and the relevant brain tissue types are mapped onto the cortical surface to obtain a reconstructed cortical surface. According to a preset brain region template or preset anatomical structure, the reconstructed cortical surface is divided into brain regions to obtain different individual brain structure data.

[0118] In specific application scenarios, individual brain region volume data includes individual cortical gray matter volume, individual subcortical gray matter volume, individual cortical white matter volume, and individual ventricular volume; the preset population brain region volume distribution includes preset foreign population brain region volume distribution and preset domestic population brain region volume distribution. The mapping module 34 can be used to map the individual cortical gray matter volume, the individual subcortical gray matter volume, the individual cortical white matter volume, and / or the individual ventricular volume to the corresponding preset foreign population brain region volume distribution, to obtain the distribution level of the individual cortical gray matter volume in the foreign population of the same age and sex, the distribution level of the individual subcortical gray matter volume in the foreign population of the same age and sex, the distribution level of the individual ventricular volume in the foreign population of the same age and sex, and the distribution level of the individual white matter volume in the foreign population of the same age and sex.

[0119] The individual's cortical gray matter volume, subcortical gray matter volume, cortical white matter volume, and / or ventricular volume are mapped to corresponding preset brain region volume distributions in the domestic population. This yields the distribution levels of the individual's cortical gray matter volume, subcortical gray matter volume, ventricular volume, and white matter volume in the domestic population of the same age and sex.

[0120] In specific application scenarios, the prediction module 36 can be used to convert the structure sequence data into JPEG format image data;

[0121] The converted structural sequence data is input into the trained individual brain tumor prediction model;

[0122] The individual brain tumor prediction model is used to perform individual brain tumor prediction processing on the structural sequence data to obtain the individual brain tumor risk prediction result.

[0123] In specific application scenarios, such as Figure 10 As shown, the device also includes: a training module 37;

[0124] Training module 37 is used to acquire JPEG format structural sequence data and the individual brain tumor information corresponding to the structural sequence data;

[0125] The structural sequence data is used as the input features of the individual tumor prediction model, and the individual brain tumor condition is used as the training label of the individual brain tumor prediction model. The individual brain tumor prediction model is trained iteratively until the loss function value of the individual brain tumor prediction model is less than a preset threshold, at which point the training of the individual brain tumor prediction model is considered complete.

[0126] In specific application scenarios, such as Figure 10 As shown, the device also includes: a second processing module 38 and a computing module 39;

[0127] The second processing module 38 is used to process the functional sequence data based on the structural sequence data using the FMRIPREP time series processing tool to obtain time series data when the original NMR data is functional sequence data.

[0128] The calculation module 39 is used to perform brain region functional connectivity calculations on the time series data to obtain an individual functional connectivity matrix.

[0129] In specific application scenarios, such as Figure 10 As shown, the device also includes: a third processing module 40;

[0130] The third processing module 40 is used to perform image processing on the diffuse sequence data based on the structure sequence data using the Mrtrix3 software tool when the original NMR data is diffuse sequence data, to obtain the diffusion coefficient and anisotropy index.

[0131] It should be noted that other corresponding descriptions of the functional units involved in the data processing apparatus provided in this embodiment can be found in [reference needed]. Figure 1 and Figure 4 The corresponding descriptions of the Chinese methods will not be repeated here.

[0132] Based on the above, Figure 1 and Figure 4 Accordingly, this disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. Figure 1 and Figure 4 The method shown.

[0133] Based on this understanding, the technical solution of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, or portable hard drive) and includes several instructions to cause a computer device (such as a personal computer, server, or network device) to execute the methods of various implementation scenarios of this disclosure.

[0134] Based on the above, Figure 1 and Figure 4 The method shown, and Figure 10 To achieve the above objectives, this disclosure also provides an electronic device, configurable on a vehicle (e.g., an electric vehicle), in accordance with the illustrated virtual device embodiment. The device includes a storage medium and a processor; the storage medium stores a computer program; the processor executes the computer program to implement the above-described virtual device. Figure 1 and Figure 4 The method shown.

[0135] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0136] Those skilled in the art will understand that the physical device structure provided in this disclosure does not constitute a limitation on the physical device, and may include more or fewer components, or a combination of certain components, or different component arrangements.

[0137] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0138] Through the above description of the embodiments, those skilled in the art can clearly understand that this disclosure can be implemented by means of software plus necessary general-purpose hardware platforms, or it can be implemented by hardware. Compared with the prior art, the data processing method, apparatus and electronic device provided by this disclosure acquires multiple raw MRI data; when the raw MRI data is structural sequence data, image processing is performed on the structural sequence data to obtain individual brain structure data, wherein the individual brain structure data includes individual brain region data; target files are read from the individual brain structure data, three-dimensional surface model results of different brain regions are generated according to the target files, and the three-dimensional surface model results are converted into target file format; individual brain region volume data are read from the individual brain region data, and the individual brain region volume data is mapped to a preset population brain region volume distribution to obtain brain region volume distribution; logistic regression is performed on the individual brain region volume data to obtain an individual disease risk level assessment, which includes an individual Alzheimer's disease risk level assessment; and individual brain tumor prediction processing is performed on the structural sequence data to obtain an individual brain tumor risk prediction result. In this embodiment, a personal brain region model (i.e., a three-dimensional surface model result) can be automatically generated from raw MRI data and converted into a target file format suitable for 3D printing, thereby promoting the application of 3D printing technology in the field of brain science. The preset population brain region volume distribution is a diverse database containing brain region volume data from different age groups, genders, etc. This application maps individual brain region volume data to the preset population brain region volume distribution to obtain the distribution of an individual's brain region volume within the population, thus allowing assessment of whether an individual's brain region volume is higher, lower, or at the population average. Through more flexible and personalized brain region volume analysis, diverse analytical needs are met, addressing the shortcomings of existing technologies in terms of the breadth, adaptability, accuracy, and automation of data analysis. This application's solution performs logistic regression analysis on individual brain region volume data to obtain an individual disease risk assessment, and performs individual brain tumor prediction processing on structural sequence data to obtain individual brain tumor risk prediction results, thereby transforming research results into practical analytical tools and addressing the shortcomings of existing technologies in disease prediction.

[0139] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "comprising" or any other variations thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0140] The above are merely specific embodiments of this disclosure, enabling those skilled in the art to understand or implement this disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to these embodiments, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A data processing method, characterized in that, The method includes: Acquire multiple raw NMR data; When the original MRI data is structural sequence data, image processing is performed on the structural sequence data to obtain individual brain structure data, wherein the individual brain structure data includes individual brain region data; Read the target file from the individual brain structure data, generate three-dimensional surface model results for different brain regions based on the target file, and convert the three-dimensional surface model results into the target file format; Read the individual brain region volume data from the individual brain region data, and map the individual brain region volume data to the preset population brain region volume distribution to obtain the brain region volume distribution; Logistic regression analysis was performed on the individual brain region volume data to obtain an individual disease risk level assessment, which included an individual Alzheimer's disease risk level assessment. The structural sequence data is processed for individual brain tumor prediction to obtain individual brain tumor risk prediction results.

2. The method according to claim 1, characterized in that, The process of image processing of the structural sequence data to obtain individual brain structure data includes: Image segmentation algorithms are used to identify the skull and other non-brain tissues in the structural sequence data, and the skull and other non-brain tissues are removed. Image registration processing is performed on the structural sequence data after removing the skull and other non-brain tissues, and intensity non-uniformity correction processing is performed on the registered structural sequence data to obtain the corrected structural sequence data. The corrected structural sequence data is processed by image segmentation to obtain different brain tissue types, and the relevant brain tissue types are mapped onto the cortical surface to obtain a reconstructed cortical surface. According to a preset brain region template or preset anatomical structure, the reconstructed cortical surface is divided into brain regions to obtain different individual brain structure data.

3. The method according to claim 1, characterized in that, The individual brain region volume data includes individual cortical gray matter volume, individual subcortical gray matter volume, individual cortical white matter volume, and individual ventricular volume; the preset population brain region volume distribution includes preset foreign population brain region volume distribution and preset domestic population brain region volume distribution. The step of mapping the individual brain region volume data to a preset population brain region volume distribution to obtain the brain region volume distribution includes: The individual's cortical gray matter volume, subcortical gray matter volume, cortical white matter volume, and / or ventricular volume are mapped to corresponding preset brain region volume distributions in foreign populations to obtain the distribution levels of the individual's cortical gray matter volume, subcortical gray matter volume, ventricular volume, and white matter volume in foreign populations of the same age and sex. The individual's cortical gray matter volume, subcortical gray matter volume, cortical white matter volume, and / or ventricular volume are mapped to corresponding preset brain region volume distributions in the domestic population. This yields the distribution levels of the individual's cortical gray matter volume, subcortical gray matter volume, ventricular volume, and white matter volume in the domestic population of the same age and sex.

4. The method according to claim 1, characterized in that, The process of performing individual brain tumor prediction processing on the structural sequence data to obtain individual brain tumor risk prediction results includes: Convert the structure sequence data into JPEG format image data; The converted structural sequence data is input into the trained individual brain tumor prediction model; The individual brain tumor prediction model is used to perform individual brain tumor prediction processing on the structural sequence data to obtain the individual brain tumor risk prediction result.

5. The method according to claim 4, characterized in that, The training process of the individual tumor prediction model includes: Obtain structural sequence data in JPEG format and the corresponding individual brain tumor information; The structural sequence data is used as the input features of the individual tumor prediction model, and the individual brain tumor condition is used as the training label of the individual brain tumor prediction model. The individual brain tumor prediction model is trained iteratively until the loss function value of the individual brain tumor prediction model is less than a preset threshold, at which point the training of the individual brain tumor prediction model is considered complete.

6. The method according to claim 1, characterized in that, The method further includes: When the original NMR data is functional sequence data, the functional sequence data is processed using the FMRIPREP time series processing tool based on the structural sequence data to obtain time series data. Brain region functional connectivity is calculated from the time-series data to obtain an individual functional connectivity matrix.

7. The method according to claim 1, characterized in that, The method further includes: When the original NMR data is diffuse sequence data, the diffuse sequence data is processed using the Mrtrix3 software tool based on the structural sequence data to obtain the diffusion coefficient and anisotropy index.

8. A data processing apparatus, characterized in that, The device includes: The acquisition module is used to acquire multiple raw NMR data. The first processing module is used to perform image processing on the original MRI data when the original MRI data is structural sequence data to obtain individual brain structure data, wherein the individual brain structure data includes individual brain region data. The generation module is used to read the target file from the individual brain structure data, generate three-dimensional surface model results of different brain regions based on the target file, and convert the three-dimensional surface model results into the target file format; The mapping module is used to read the individual brain region volume data from the individual brain region data and map the individual brain region volume data to a preset population brain region volume distribution to obtain the brain region volume distribution. The computation module is used to perform logistic regression on the individual brain region volume data to obtain an individual disease risk level assessment, which includes an individual Alzheimer's disease risk level assessment. The prediction module is used to perform individual brain tumor prediction processing on the structural sequence data to obtain individual brain tumor risk prediction results.

9. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-7.

Images