Brain cognitive function data processing method and device, storage medium and electronic equipment

Abstract

The invention relates to the field of data processing, in particular to a brain cognitive function data processing method and device, a storage medium and electronic equipment. The brain cognitive function data processing method comprises the following steps: collecting brain fingerprint data of a to-be-detected object based on a target brain fingerprint index, and collecting genome data of the to-be-detected object based on a target genome index; performing dimension reduction processing on the brain fingerprint data and the genome data to obtain core feature difference data of the to-be-detected object relative to a contrast group; and inputting the core feature difference data into a pre-trained difference value prediction model to obtain a brain cognitive function difference value output by the difference value prediction model. According to the brain cognitive function data processing method provided by the invention, the data accuracy of the brain cognitive function difference value of the plateau immigrant population relative to the plain population can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing, and particularly to a method, apparatus, storage medium, and electronic device for processing brain cognitive function data. Background Art

[0002] In the past decade, more and more people living on plains have migrated to plateau areas due to economic, social, and national defense needs. The health of the migrant population has gradually become an important public health issue. A series of studies have confirmed that short-term or long-term exposure to the hypoxic environment on the plateau can induce abnormal functions of the central nervous system of the body and cause brain cognitive function damage.

[0003] Previous studies have found that chronic hypoxia exposure induces damage to verbal and visual working memory, executive control function, and psychomotor function. At the same time, irreversible structural and functional changes occur at four levels of "decrease in the number of neurons - weakening of activity - weakening of connections - abnormal network" in brain regions centered on the striatum and hippocampus. Therefore, the abnormal brain structure, brain function, and brain network caused by the hypoxic environment on the plateau are the key pathological mechanisms affecting brain cognitive function. Currently, due to the poor accuracy of the data for evaluating the brain cognitive function of the population in the plateau environment, the issues of individualized prediction and how to screen susceptible populations in the field of special medicine have not been resolved.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present disclosure is to provide a method, apparatus, storage medium, and electronic device for processing brain cognitive function data, aiming to improve the data accuracy of the difference value of the brain cognitive function of plateau migrant populations relative to plain populations.

[0006] Other features and advantages of the present disclosure will become apparent through the following detailed description, or will be partially learned through the practice of the present disclosure.

[0007] According to one aspect of the embodiments of the present disclosure, a method for processing brain cognitive function data is provided, including:

[0008] Collecting brain fingerprint data of a to-be-detected object based on a target brain fingerprint index, and collecting genomic data of the to-be-detected object based on a target genomic index;

[0009] Performing dimensionality reduction processing on the brain fingerprint data and the genomic data to obtain core feature difference data of the to-be-detected object relative to a control population;

[0010] Input the core feature difference data into a pre-trained difference value prediction model to obtain the brain cognitive function difference value output by the difference value prediction model;

[0011] Among them, the difference value prediction model is trained based on the brain fingerprint data, genomic data, and brain cognitive function scores corresponding to the target population and the control population respectively; both the object to be detected and the target population are plateau immigrant populations, and the control population is a plain population.

[0012] According to some embodiments of the present disclosure, based on the foregoing solution, the method further includes: pre-training the difference value prediction model, and the pre-training of the difference value prediction model includes:

[0013] Obtain a data set; the data set includes a training set and a test set, and the data set includes core feature difference data and the brain cognitive function difference value corresponding to the core feature difference data;

[0014] Select core feature difference data from the training set and input it into the initial difference value prediction model to obtain the predicted value of the brain cognitive function difference value output by the initial difference value prediction model;

[0015] Compare the predicted value of the brain cognitive function difference value with the brain cognitive function difference value in the training set to calculate the loss function;

[0016] Adaptively adjust the model parameters of the initial difference value prediction model according to the loss function, and stop training when the training stop condition is met to obtain the trained difference value prediction model;

[0017] Use the test set to verify the accuracy of the trained difference value prediction model for the predicted value of the brain cognitive function difference value. If the verification is passed, end the training to obtain the verified difference value prediction model.

[0018] According to some embodiments of the present disclosure, based on the foregoing solution, the obtaining of the data set includes:

[0019] Collect the brain fingerprint data and genomic data of the target population and the control population based on the target brain fingerprint index and the target genomic index respectively;

[0020] Perform dimensionality reduction processing on the brain fingerprint data and genomic data of the target population and the control population to obtain the core feature difference data in the data set;

[0021] Obtain the first brain cognitive function score and the second brain cognitive function score obtained by performing neurobehavioral tests on the target population and the control population;

[0022] Use the difference between the first brain cognitive function score and the second brain cognitive function score as the brain cognitive function difference value in the dataset.

[0023] According to some embodiments of the present disclosure, based on the foregoing solution, the method further includes: determining the target brain fingerprint index; the determining the target brain fingerprint index includes:

[0024] Obtain the first brain fingerprint data of the target population and the second brain fingerprint data of the control population collected based on the initial brain fingerprint index respectively;

[0025] Perform differential analysis on the first brain fingerprint data and the second brain fingerprint data to determine the target brain fingerprint index; the differential analysis includes one or more of whole-brain functional connectivity differential analysis, dynamic functional connectivity differential analysis, and topological property differential analysis; the target brain fingerprint index is the brain fingerprint index with a significant difference in brain cognitive function impairment between the target population and the control population.

[0026] According to some embodiments of the present disclosure, based on the foregoing solution, the method further includes: determining the target genome index; the determining the target genome index includes:

[0027] Perform spatial correlation analysis on the brain fingerprint data and the whole-brain gene expression data of the target population to obtain gene expression data related to the brain fingerprint data;

[0028] Based on the gene expression data and the cognitive data of the target population and the control population, screen out the risk alleles with a significant difference in brain cognitive function impairment as the target genome index.

[0029] According to some embodiments of the present disclosure, based on the foregoing solution, after obtaining the brain cognitive function difference value, the method further includes:

[0030] Draw a predicted weight brain region map and / or a brain network group map based on the brain cognitive function difference value.

[0031] According to some embodiments of the present disclosure, based on the foregoing solution, the neurobehavioral test includes any one or more of a working memory test, a sustained attention test, and a cognitive flexibility test.

[0032] According to a second aspect of the embodiments of the present disclosure, there is provided a brain cognitive function data processing device, including:

[0033] An acquisition module, configured to acquire the brain fingerprint data of the object to be detected based on the target brain fingerprint index, and acquire the genome data of the object to be detected based on the target genome index;

[0034] An analysis module for performing dimensionality reduction processing on the brain fingerprint data and the genomic data to obtain core feature difference data of the object to be detected relative to the control population;

[0035] A prediction module for inputting the core feature difference data into a pre-trained difference value prediction model to obtain a brain cognitive function difference value output by the difference value prediction model; wherein, the difference value prediction model is trained according to the brain fingerprint data, genomic data and brain cognitive function scores corresponding to the target population and the control population respectively; the object to be detected and the target population are both plateau immigrant populations, and the control population is a plain population.

[0036] According to a third aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, it implements the brain cognitive function data processing method as in the above embodiments.

[0037] According to a fourth aspect of the embodiments of the present disclosure, there is provided an electronic device, comprising: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the brain cognitive function data processing method as in the above embodiments.

[0038] The exemplary embodiments of the present disclosure may have some or all of the following beneficial effects:

[0039] In the technical solutions provided by some embodiments of the present disclosure, data collection is first performed through pre-determined target brain fingerprint indicators and target genomic indicators, and then the radiomics data of the brain fingerprint data and the genomics data of the genomic data are combined to predict brain cognitive function impairment, and finally the impairment prediction result of the object to be detected is obtained. On the one hand, compared with single-modal prediction, the multi-modal features of radiomics and genomics in the present disclosure make the calculated brain cognitive function difference value more accurate and have higher data accuracy; on the other hand, the brain cognitive function data used to calculate the brain cognitive function difference value is collected according to the pre-determined target brain fingerprint indicators and target genomic indicators, and they are all core data related to brain cognitive function assessment. While reducing the amount of data in the data calculation process, it also further improves the data accuracy.

[0040] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts. In the drawings:

[0042] Figure 1 Schematically shows a flowchart of a method for processing brain cognitive function data in an exemplary embodiment of the present disclosure.

[0043] Figure 2 Schematically shows a flowchart of a process for training a difference value prediction model in an exemplary embodiment of the present disclosure.

[0044] Figure 3 Schematically shows a predicted weighted brain region atlas and a brain network group atlas in an exemplary embodiment of the present disclosure.

[0045] Figure 4 Schematically shows a flowchart of a process for determining a difference value prediction model in an exemplary embodiment of the present disclosure.

[0046] Figure 5 Schematically shows a schematic diagram of the composition of a brain cognitive function data processing device in an exemplary embodiment of the present disclosure.

[0047] Figure 6 Schematically shows a schematic diagram of a computer-readable storage medium in an exemplary embodiment of the present disclosure.

[0048] Figure 7 Schematically shows a schematic diagram of the structure of a computer system of an electronic device in an exemplary embodiment of the present disclosure. Detailed implementation manners

[0049] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more complete and comprehensive, and will fully convey the concept of the example embodiments to those skilled in the art.

[0050] In addition, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present disclosure. However, those skilled in the art will realize that the technical solutions of the present disclosure can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, well-known methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of the present disclosure.

[0051] The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0052] The flowcharts shown in the drawings are only illustrative and do not necessarily include all content and operations / steps, nor do they have to be executed in the described order. For example, some operations / steps can be decomposed, while some operations / steps can be combined or partially combined, so the actual execution order may change according to the actual situation.

[0053] The plateau area and the plateau-residing population in our country are the largest in the world. Moreover, the plateau area is located at the border and is rich in resources, having a very important strategic position. In recent years, with the strengthening and development of social, economic and military activities in the plateau areas of our country, the number of people migrating from plain areas to high-altitude areas has been continuously increasing. The most prominent factor affecting the human body in the plateau area is hypoxia. As an organ with high oxygen consumption, the human brain consumes about 25% of the total oxygen consumption of the human body and is the most sensitive to low oxygen.

[0054] The special environment of low oxygen in the plateau area will have a significant impact on the cognitive function of the human brain. Rapid ascent to the plateau in a short period of time can cause damage to various types of brain cognitive functions such as attention, memory, perception ability, and thinking and judgment ability. The cognitive damage caused by long-term low oxygen exposure is more prominent, mainly manifested as delayed cognitive reaction time, decreased attention, and reduced executive ability and working memory.

[0055] Previous studies have found that chronic low oxygen exposure induces damage to verbal and visual working memory, executive control function, and psychomotor function. At the same time, irreversible structural and functional changes occur at four levels of "decrease in the number of neurons - weakening of activity - weakening of connections - abnormal network" in brain regions centered on the striatum and hippocampus. Therefore, the abnormal brain structure, brain function and brain network caused by the plateau low oxygen environment are the key pathological mechanisms affecting brain cognitive function. Currently, in the field of special medicine, the early diagnosis of brain cognitive function damage, individualized prediction and how to screen susceptible populations in the plateau environment have not been solved.

[0056] Based on this, the present disclosure provides a method for processing brain cognitive function data, which performs correlation analysis on brain imaging data and genomic data. It can not only explore how gene mutations affect the brain structure and function of individuals, but also determine the correlation between specific genotypes and brain connection patterns and functional characteristics, thereby revealing the role of genes in regulating brain structure and function. This correlation helps to understand the regulatory role of genes in brain development, function and cognitive behavior, providing important clues for cognitive neuroscience research.

[0057] The implementation details of the technical solution of the embodiments of the present disclosure will be elaborated in detail below.

[0058] Figure 1 A schematic flowchart of a method for processing brain cognitive function data in an exemplary embodiment of the present disclosure is schematically shown. As Figure 1 shown, the method for processing brain cognitive function data includes steps S101 to S103:

[0059] Step S101, collecting brain fingerprint data of the object to be detected based on the target brain fingerprint index, and collecting genomic data of the object to be detected based on the target genomic index;

[0060] Step S102, performing dimensionality reduction processing on the brain fingerprint data and the genomic data to obtain core feature difference data of the object to be detected relative to the control population;

[0061] Step S103, inputting the core feature difference data into a pre-trained difference value prediction model to obtain a brain cognitive function difference value output by the difference value prediction model.

[0062] Based on the above method, data is first collected through the pre-determined target brain fingerprint index and target genomic index, and then the radiomics data of the brain fingerprint data and the genomics data of the genomic data are combined to predict brain cognitive function impairment, and finally the impairment prediction result of the object to be detected is obtained. On the one hand, compared with single-modal prediction, the multi-modal features of radiomics and genomics in the present disclosure make the calculated brain cognitive function difference value more accurate and have higher data accuracy; on the other hand, the brain cognitive function data used to calculate the brain cognitive function difference value is collected according to the pre-determined target brain fingerprint index and target genomic index, and they are all core data related to brain cognitive function assessment. While reducing the amount of data in the data calculation process, the data accuracy is further improved.

[0063] Next, each step of the method for processing brain cognitive function data in this exemplary embodiment will be described in more detail with reference to the accompanying drawings and embodiments.

[0064] In step S101, brain fingerprint data of the object to be detected is collected based on the target brain fingerprint index, and genomic data of the object to be detected is collected based on the target genomic index.

[0065] In an embodiment of the present disclosure, the object to be detected is a population of migrants to the plateau. A population of migrants to the plateau, that is, a population that has migrated from a plain area to a plateau area, can also be called an exposed population.

[0066] Before collecting data, it is first necessary to determine the metrics related to prediction, including the target brain fingerprint metrics and the target genomic metrics.

[0067] In one embodiment of the present disclosure, determining the target brain fingerprint metrics includes: obtaining the first brain fingerprint data of the target population and the second brain fingerprint data of the control population collected based on the initial brain fingerprint metrics respectively; performing a difference analysis on the first brain fingerprint data and the second brain fingerprint data to determine the target brain fingerprint metrics; the difference analysis includes one or more of whole-brain functional connectivity difference analysis, dynamic functional connectivity difference analysis, and topological property difference analysis; the target brain fingerprint metrics are the brain fingerprint metrics with significant differences in brain cognitive function impairment between the target population and the control population.

[0068] In one embodiment of the present disclosure, determining the target genomic metrics includes: performing a spatial correlation analysis on the brain fingerprint data and the whole-brain gene expression data of the target population to obtain gene expression data related to the brain fingerprint data; based on the gene expression data and the cognitive data of the target population and the control population, screening out the risk alleles with significant differences in brain cognitive function impairment as the target genomic metrics.

[0069] When collecting brain fingerprint data, the rs-fMRI data of the object to be detected can be collected using a GE Electric Discovery MR750 3.0T system (General Electric Company, USA). It is collected through an Echo Planar Imaging (EPI) sequence. Among them, the repetition time (TR) is set to 2000 ms; the echo time (TE) is set to 30 ms; the flip angle is set to 90°; the field of view is set to 220×220 mm 2 ; the acquisition / reconstruction matrix is set to 128×128; the slice thickness is set to 4 mm; the section gap is set to 0.6 mm; the number of slices is set to 30. The total scanning time is 6 minutes, the total number of samples (Total volumes) is 180, and the coverage range is the whole brain.

[0070] To collect genomic data, 10 ml of fasting venous blood can be drawn from the antecubital vein and placed into non-anticoagulant and anticoagulant blood collection tubes respectively, and then stored in a -80°C refrigerator for future use. Blood routine tests are used to detect blood cell count, lymphocyte count, red blood cell count, hemoglobin and other indicators through a Beckman Coulter LH750 hematology analyzer. Blood biochemistry tests are used to detect blood urea nitrogen, creatinine, total bilirubin and other indicators through a Beckman Coulter AU5800 clinical chemistry analyzer.

[0071] Although ordinary neuroimaging and genomics methods can provide information about brain structure and function as well as genetic information, there are certain precision limitations in the early diagnosis of brain cognitive function impairment due to the lack of individualized indicators. By pre-determining the indicators related to brain cognitive function impairment for data collection, the effectiveness of the collected data can be improved, and thus the accuracy of brain cognitive function data processing can be enhanced.

[0072] In step S102, dimensionality reduction processing is performed on the brain fingerprint data and the genomic data to obtain the core feature difference data of the object to be detected relative to the control population.

[0073] Specifically, the features of the brain fingerprint data (connectivity matrix, time fraction, mean residence time, state transition times, small-world property, degree centrality, nodal efficiency, global efficiency, etc.) and genomic data (risk alleles) of the object to be detected can be extracted, and dimensionality reduction processing can be performed through methods such as Principal Component Analysis (PCA) to remove redundant information, further select the core features of the data, and improve the computational efficiency and accuracy.

[0074] Principal Component Analysis is a multivariate statistical technique widely used in data analysis, machine learning, statistical modeling and other fields. Its main purpose is to use dimensionality reduction techniques to convert multiple variables (or features) in the original data into a new set of linearly independent composite variables (i.e., principal components). These new variables can retain as much information of the original data as possible while reducing the dimensionality of the data, facilitating subsequent analysis and processing.

[0075] Of course, it should be noted that the use of principal component analysis for data dimensionality reduction in this disclosure is only an exemplary illustration, and other methods such as factor analysis, independent component analysis (ICA), linear discriminant analysis (LDA), and singular value decomposition (SVD) can also be used. In addition to the above linear dimensionality reduction methods, there are also some non-linear dimensionality reduction methods such as Isomap, locally linear embedding (LLE), and t-distributed stochastic neighbor embedding (t-SNE). This disclosure does not make specific limitations.

[0076] In step S103, the core feature difference data is input into a pre-trained difference value prediction model to obtain the brain cognitive function difference value output by the difference value prediction model.

[0077] In recent years, the prediction models of machine learning algorithms have shown great potential in the prediction of brain cognitive functions and are a powerful tool for exploring the development of individual human brain cognitive functions. Therefore, data prediction can be achieved using a machine learning model, with the core feature difference data between brain fingerprint data and genomic data as the input of the model, and the difference value compared with the normal value of brain cognitive function as the output of the model.

[0078] Among them, the difference value prediction model is trained based on the brain fingerprint data, genomic data, and brain cognitive function scores corresponding to the target population and the control population respectively. In an embodiment of this disclosure, the method further includes: pre-training the difference value prediction model. Figure 2 Schematically shows a flowchart of training a difference value prediction model in an exemplary embodiment of this disclosure. Refer to Figure 2 As shown, the process of training the difference value prediction model specifically includes:

[0079] Step S201, obtaining a data set; the data set includes a training set and a test set, and the data set includes core feature difference data and the brain cognitive function difference value corresponding to the core feature difference data;

[0080] Step S202, selecting core feature difference data from the training set and inputting it into the initial difference value prediction model to obtain the predicted value of the brain cognitive function difference value output by the initial difference value prediction model;

[0081] Step S203, comparing the predicted value of the brain cognitive function difference value with the brain cognitive function difference value in the training set to calculate the loss function;

[0082] Step S204: Adaptively adjust the model parameters of the initial difference value prediction model according to the loss function, and stop training when the training stop condition is met to obtain the trained difference value prediction model.

[0083] Step S205: Use the test set to verify the accuracy of the trained difference value prediction model for the predicted values of brain cognitive function difference values. If the verification is passed, end the training to obtain the verified difference value prediction model.

[0084] Among them, when obtaining the data set in step S201, the specific method is as follows:

[0085] Step 1: Collect the brain fingerprint data and genomic data of the target population and the control population based on the target brain fingerprint index and the target genomic index respectively.

[0086] Step 2: Perform dimensionality reduction processing on the brain fingerprint data and genomic data of the target population and the control population to obtain the core feature difference data in the data set.

[0087] The process of collecting the brain fingerprint data and genomic data of the target population and the control population in the above steps 1 to 2 and then obtaining the core feature difference data of the object to be detected relative to the control population is the same as the processing of the object to be detected in steps S101 and S102, so it will not be elaborated here.

[0088] It should be noted that both the target population and the control population are the objects of data collection. Among them, the target population can be the plateau immigrant population, that is, the population who moves from the plain area to the plateau area, while the control population is the population who has long lived in the plain area, that is, the plain population.

[0089] Specifically, when establishing the plateau immigrant population cohort, the conditions for the population are restricted as follows: age 18 - 25 years old; permanent residence altitude below 900m; no high altitude exposure history (>2500m); no history of mental illness; no history of using drugs that affect brain cognitive function; no history of other major diseases; agree to participate in this study and sign the informed consent form. Follow - up for 2 years, once a year, collect the above data and track their high altitude exposure history.

[0090] When establishing the plain population cohort, the conditions for the population are restricted as follows: age 18 - 25 years old; permanent residence altitude below 900m; no high altitude exposure history (>2500m); no history of mental illness; no history of using drugs that affect brain cognitive function; no history of other major diseases; agree to participate in this study and sign the informed consent form.

[0091] Step 3: Conduct neurobehavioral tests on the target population and the control population respectively to obtain the first brain cognitive function score and the second brain cognitive function score.

[0092] Step 4: Use the difference between the first brain cognitive function score and the second brain cognitive function score as the brain cognitive function difference value in the dataset.

[0093] Specifically, neurobehavioral tests can be used to evaluate the brain cognitive function of a population. The neurobehavioral tests include any one or more of a working memory test, a sustained attention test, and a cognitive flexibility test.

[0094] Specifically, the spatial span test, visual memory test, continuous performance test, Wisconsin Card Sorting Test, and maze test in the BCT test can be used to measure the brain cognitive function of the research object. The aim is to measure the working memory, sustained attention, cognitive flexibility, reasoning, and problem-solving abilities of the research object to identify the effects of chronic hypoxic exposure on word-, graph-, and space-related working memory, attention, cognitive flexibility, reasoning, problem-solving abilities, and reaction time, and finally obtain a comprehensive brain cognitive function score.

[0095] It should be noted that the brain cognitive function score can be the sum of the scores of each test, or the scores of each test can be weighted and summed, etc. The present disclosure does not make specific limitations. Then, subtract the brain cognitive function scores corresponding to the target population and the control population respectively to obtain the brain cognitive function difference value, which represents the difference in brain cognitive function between the high-altitude migrants and the plain population.

[0096] Finally, 80% of the obtained dataset is used as the training set, and 20% is used as the test set. In the training set, the above core feature difference data collected during the baseline survey of the collection object is used as the predictive variable, and the difference in the brain cognitive function scores of the collection object during the baseline and follow-up surveys is used as the response variable.

[0097] It should be noted that training can be performed based on different machine models.

[0098] For example, a prediction model based on RVR can be constructed. Use the Python sklearning-rvm package to establish an RVR model with expectation maximization. The model uses a linear kernel and an alpha selection criterion threshold with an associated vector number of 1e9. All other parameters use default values, such as the stopping criterion tolerance set to 1e5, the beta value is not fixed, there is no pre-specified initial alpha value, no bias is added to the decision function, and the maximum number of iterations is set to 5000 times.

[0099] A prediction model based on Lasso regression can also be constructed. The Lasso regression function in the Python scikit-learn package is used to perform predictive analysis on the dataset. The regularization method of Lasso regression is L1, and the penalty term coefficient of the regression is automatically selected by the model. The penalty regularization parameter alpha in Lasso is responsible for adjusting the severity of the penalty. The higher the value, the stronger the penalty for each parameter, which in turn leads to a greater shrinkage of the coefficient magnitude. The grid search space for the parameter alpha is specified as 0.001, 0.01, 0.1, 1, 10, 100. All other parameters use the default values.

[0100] A prediction model based on SVR can also be constructed. The LinearSVR tool with an insensitive loss function in the Python scikit-learn package is used to establish the SVR model. During the establishment of the SVR model, the hyperparameter C is tuned. In each iteration of cross-validation, the sklearn grid search method is used to perform a systematic hyperparameter search for C in the search space -7 2 -5 2 -3 2 -1 1, 2, 2 3 2 5 and 2 7 The scoring parameter is specified as negative MAE. Before applying the model with the optimal hyperparameter value to the test set, the entire training set is retrained from this cross-validation iteration. All other parameters use the default values.

[0101] The data of the test set is incorporated into the different prediction models trained above. The prediction efficiency is optimized through the combination of different data types and machine learning algorithms. The prediction model with the optimal prediction efficiency is initially screened through ten-fold cross-validation, and then the Pearson correlation coefficient, MAE, MSE, RMSE, and R 2 between the predicted brain cognitive function score and the actual brain cognitive function score are calculated to finally verify and evaluate the prediction efficiency of the model, and the optimal model is selected through comparison.

[0102] For the high-efficiency prediction model obtained after screening, a predicted weight brain region map or a brain network group map can be drawn to reveal the key brain regions or core brain networks highly correlated with the susceptibility to brain cognitive function impairment in plateau migrants.

[0103] Specifically, both the predicted weight brain region map and the brain network group map are important tools in cognitive neuroscience research, which can help understand the relationship between brain structure and function, and how the brain works in a hypoxic environment.

[0104] A predictive weight brain region atlas generally refers to using machine learning algorithms and brain imaging data to predict an individual's behavior or cognitive function and identifying the brain regions that have weights for these predictions. This method can help researchers understand which brain regions are more important for specific cognitive tasks or behavioral performances. For example, by analyzing the activity patterns of the brain when performing a specific task, a model can be established to predict an individual's performance on a similar task, and then the key brain regions affecting task performance can be identified. This method has extensive applications in cognitive neuroscience, psychology, and clinical neuroscience. Especially when studying brain diseases and cognitive disorders, it can provide a basis for early diagnosis and intervention.

[0105] The brain network group atlas, on the other hand, is a more macroscopic perspective. It not only focuses on individual brain regions but views the brain as a complex network composed of multiple interacting brain regions. This atlas reveals the network structure and dynamic characteristics of the brain by analyzing the functional or structural connections between brain regions. The study of the brain network group atlas can help scientists understand how the brain supports complex cognitive and behavioral functions through the collaborative work of different brain regions. For example, by analyzing the brain network group atlas of patients with depression, researchers may find that the connection patterns between certain brain regions are significantly different from those of healthy people, thus providing clues for understanding the neural mechanisms of depression.

[0106] In practical applications, the brain network group atlas can be used to reveal the network reconstruction of the brain in healthy and diseased states, as well as how individual differences affect the organization of the brain network. These atlases can also be integrated with other data such as genetic information and environmental factors to provide a more comprehensive understanding of brain function and diseases. With the progress of technology, the resolution and accuracy of these atlases are continuously improving, providing powerful tools for brain science research and clinical applications.

[0107] The predictive weight brain region atlas and the brain network group atlas are important tools in cognitive neuroscience and brain disease research, which provide detailed views of brain structure and functional connections. The following are some key information included in these atlases:

[0108] (1) Fine division of brain regions: The brain network group atlas usually divides the brain into more fine-grained sub-regions.

[0109] (2) Multimodal connection patterns: The atlas not only includes the fine structures of the cerebral cortex and subcortical nuclei but also quantitatively describes the anatomical and functional connection patterns of different regions of the brain.

[0110] (3) Network topological structure: The brain network group atlas reveals the topological structure of the brain network, including the degree of nodes, clustering coefficient, path length, etc. These topological features help to understand the organizational principles and information processing capabilities of the brain network.

[0111] (4)Dynamic change patterns: The dynamic change patterns of brain networks are the key to understanding brain functions. These maps can show how brain networks dynamically adjust their connection patterns in a hypoxic environment.

[0112] (5)Networks related to specific brain functions and diseases: The maps can also reveal the network characteristics under specific brain functions or disease states. For example, networks related to intelligence, attention, etc., and whether these networks have a genetic basis.

[0113] (6)Simulation and modeling: By simulating and modeling brain networks, researchers can better understand the working mechanism of the brain and provide new perspectives for the treatment of brain diseases.

[0114] (7)Individualization and clinical applications: The individualized mapping method of the connectome map and its application in precision diagnosis and treatment provide new research means for understanding the mechanism of brain injury in a hypoxic environment, discovering biological markers for early diagnosis and efficacy evaluation of brain injury.

[0115] (8)Predictive weights of brain regions: In the map of brain regions with predictive weights, it may contain the contribution weights of specific brain regions to the prediction of behavior or cognitive functions, which helps to identify brain regions, circuits, and potential neurobiological bases that are predictive of specific behavioral representations.

[0116] The applications of these maps are extensive and play an important role in basic scientific research, clinical diagnosis, and treatment. With the progress of technology, the resolution and accuracy of these maps are constantly improving, providing powerful tools for brain science research and clinical applications.

[0117] The map of brain regions with predictive weights and / or the connectome map can be drawn using the BrainNet toolbox in Matlab software, and the drawing results can screen out the brain regions and brain network connection patterns that contribute the most to the prediction.

[0118] Based on the above methods, the individualized mapping method of the connectome map and its application in precision diagnosis and treatment can provide new research means for understanding the mechanism of brain injury in a hypoxic environment, discovering biological markers for early diagnosis and efficacy evaluation of brain injury. In the map of brain regions with predictive weights, it may contain the contribution weights of specific brain regions to the prediction of cognitive function impairment, which helps to identify brain regions, circuits, and potential neurobiological bases that are predictive of cognitive function impairment manifestations in a hypoxic environment.

[0119] Figure 3Schematically shows a predicted weight brain region atlas and a brain network group atlas in an exemplary embodiment of the present disclosure. Among them, on the left is a voxel-based predicted weight brain region atlas, and the color bar represents the relative importance of the voxel in the decision function. On the right is a region-based brain network group atlas, and the color bar represents the normalized contribution of the region. Among them, (A) represents the atlas corresponding to the visual simple reaction time; (B) represents the atlas corresponding to the auditory simple reaction time; (C) represents the atlas corresponding to the visual discrimination reaction time; (D) represents the atlas corresponding to the auditory discrimination reaction time.

[0120] Subsequently, the difference value prediction model is trained using the training set, and then the difference value prediction model is tested using the test set. After the test is completed, the trained difference value prediction model is obtained. Its input is the core feature difference data between the brain fingerprint data and the genomic data, and the output is the difference value of the brain cognitive function compared with the normal value.

[0121] Figure 4 Schematically shows a flowchart of determining a difference value prediction model in an exemplary embodiment of the present disclosure. Refer to Figure 4 As shown, when determining the difference value prediction model, first obtain the brain fingerprint data, genomic data, and the difference in cognitive function scores of the target population and the control population. Then, based on the brain fingerprint data and genomic data, obtain the core feature difference data of the object to be detected relative to the control population. Then, use the core feature difference data and the difference in cognitive function scores for model training. Ten-fold cross-validation can be used, and model training can be carried out in ways such as support vector regression, least absolute shrinkage and selection operator regression, and relevance vector regression. Finally, calculate parameters such as the Pearson correlation coefficient, squared absolute error, root mean square error, and determination coefficient corresponding to the three models to evaluate the prediction results of these three models, and then select the optimal machine learning model to obtain the final brain cognitive function difference value prediction model. At the same time, for the high-performance prediction model obtained after screening, a predicted weight brain region atlas or a brain network group atlas can be drawn.

[0122] After obtaining the core feature difference data after dimensionality reduction processing of the brain fingerprint data and genomic data of the object to be detected, input the core feature difference data into the trained difference value prediction model, and finally obtain the brain cognitive function difference value output by the difference value prediction model. This brain cognitive function difference value can reflect the difference in brain function cognition between the object to be detected and the ordinary plain population.

[0123] Based on the above method, compared with single-modal prediction, the difference value prediction model of the multi-modal features combining radiomics and genomics provided by the present disclosure usually has higher prediction accuracy, which helps to promote the development of precision medicine and provides an important basis for deeply understanding the mechanism of brain cognitive function damage and predicting its development.

[0124] Figure 5 Schematically showing a schematic diagram of the composition of a brain cognitive function data processing device in an exemplary embodiment of the present disclosure, as Figure 5 shown, the brain cognitive function data processing device 500 may include an acquisition module 501, an analysis module 502, and a prediction module 503. Among them:

[0125] The acquisition module 501 is used to acquire the brain fingerprint data of the object to be detected based on the target brain fingerprint index, and acquire the genomic data of the object to be detected based on the target genomic index;

[0126] The analysis module 502 is used to perform dimensionality reduction processing on the brain fingerprint data and the genomic data to obtain the core feature difference data of the object to be detected relative to the control population;

[0127] The prediction module 503 is used to input the core feature difference data into a pre-trained difference value prediction model to obtain the brain cognitive function difference value output by the difference value prediction model; wherein, the difference value prediction model is trained according to the brain fingerprint data, genomic data, and brain cognitive function scores corresponding to the target population and the control population respectively; the object to be detected and the target population are both plateau immigrant populations, and the control population is a plain population.

[0128] According to an exemplary embodiment of the present disclosure, the brain cognitive function data processing device 500 may further include a model training unit, which is used to obtain a data set; the data set includes a training set and a test set, and the data set includes core feature difference data and the brain cognitive function difference value corresponding to the core feature difference data; select the core feature difference data in the training set and input it into the initial difference value prediction model to obtain the predicted value of the brain cognitive function difference value output by the initial difference value prediction model; compare the predicted value of the brain cognitive function difference value with the brain cognitive function difference value in the training set to calculate the loss function; adaptively adjust the model parameters of the initial difference value prediction model according to the loss function, and stop training when the training stop condition is met to obtain the trained difference value prediction model; use the test set to verify the accuracy of the trained difference value prediction model for the predicted value of the brain cognitive function difference value, and if the verification is passed, end the training to obtain the verified difference value prediction model.

[0129] According to an exemplary embodiment of the present disclosure, the model training unit is further configured to collect brain fingerprint data and genomic data of a target population and a control population based on the target brain fingerprint index and the target genomic index respectively; perform dimensionality reduction processing on the brain fingerprint data and genomic data of the target population and the control population to obtain core feature difference data in the dataset; obtain a first brain cognitive function score and a second brain cognitive function score obtained by performing neurobehavioral tests on the target population and the control population; and use the difference between the first brain cognitive function score and the second brain cognitive function score as the brain cognitive function difference value in the dataset.

[0130] According to an exemplary embodiment of the present disclosure, the brain cognitive function data processing device 500 may further include a first index determination unit for determining the target brain fingerprint index; including: obtaining first brain fingerprint data of a target population and second brain fingerprint data of a control population collected based on initial brain fingerprint indexes respectively; performing difference analysis on the first brain fingerprint data and the second brain fingerprint data to determine the target brain fingerprint index; the difference analysis includes one or more of whole-brain functional connectivity difference analysis, dynamic functional connectivity difference analysis, and topological property difference analysis; the target brain fingerprint index is a brain fingerprint index with a significant difference in brain cognitive function impairment between the target population and the control population.

[0131] According to an exemplary embodiment of the present disclosure, the brain cognitive function data processing device 500 may further include a second index determination unit for determining a target genomic index; including: performing spatial correlation analysis on the brain fingerprint data and the whole-brain gene expression data of the target population to obtain gene expression data related to the brain fingerprint data; and screening out risk alleles with a significant difference in brain cognitive function impairment based on the gene expression data and the cognitive data of the target population and the control population as the target genomic index.

[0132] According to an exemplary embodiment of the present disclosure, the model training unit is further configured to, after obtaining the brain cognitive function difference value, draw a predicted weight brain region map and / or a brain network group map based on the brain cognitive function difference value.

[0133] According to an exemplary embodiment of the present disclosure, the neurobehavioral test includes any one or more of a working memory test, a sustained attention test, and a cognitive flexibility test.

[0134] The specific details of each module in the above-mentioned brain cognitive function data processing device 500 have been described in detail in the corresponding brain cognitive function data processing method, and thus will not be elaborated here.

[0135] It should be noted that although several modules or units of a device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0136] In an exemplary embodiment of the present disclosure, a storage medium capable of implementing the above method is also provided. Figure 6 A schematic diagram schematically showing a computer-readable storage medium in an exemplary embodiment of the present disclosure is as Figure 6 shown, which describes a program product 600 for implementing the above method according to an embodiment of the present disclosure. It can be a portable compact disc read-only memory (CD-ROM) and includes program code, and can run on a terminal device, such as a mobile phone. However, the program product of the present disclosure is not limited thereto. In this document, the readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device.

[0137] In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided. Figure 7 A schematic diagram showing the structure of a computer system of an electronic device in an exemplary embodiment of the present disclosure.

[0138] It should be noted that Figure 7 the computer system 700 of the shown electronic device is only an example and should not bring any limitation to the functions and usage scope of the embodiments of the present disclosure.

[0139] As Figure 7 shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 702 or the program loaded from the storage section 708 into the random access memory (RAM) 703. In the RAM 703, various programs and data required for system operation are also stored. The CPU 701, ROM 702, and RAM 703 are connected to each other through a bus 704. The input / output (I / O) interface 705 is also connected to the bus 704.

[0140] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is mounted on the drive 710 as needed so that a computer program read therefrom is installed into the storage section 708 as needed.

[0141] Specifically, according to an embodiment of the present disclosure, the processes described below with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present disclosure includes a computer program product including a computer program carried on a computer-readable medium, the computer program including program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 709 and / or installed from the removable medium 711. When the computer program is executed by a central processing unit (CPU) 701, various functions defined in the system of the present disclosure are executed.

[0142] It should be noted that the computer-readable medium shown in the embodiments of the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program code is carried. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0143] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the above module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0144] The units involved in the embodiments described in this disclosure can be implemented in software or in hardware, and the described units can also be provided in a processor. Among them, the names of these units do not, in some cases, constitute a limitation on the unit itself.

[0145] On the other hand, the present disclosure also provides a computer-readable medium, which can be included in the electronic device described in the above embodiments; or can exist alone without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the methods described in the above embodiments.

[0146] It should be noted that although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0147] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present 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, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (such as a personal computer, a server, a touch terminal, or a network device, etc.) to execute the methods according to the embodiments of the present disclosure.

[0148] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present disclosure. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include known common general knowledge or conventional technical means in the technical field not disclosed by the present disclosure.

[0149] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A method for processing brain cognitive function data, characterized in that: The brain cognitive function data includes brain fingerprint data and genome data, and the method includes: Collecting brain fingerprint data of the subject to be detected based on the target brain fingerprint index, and collecting genome data of the subject to be detected based on the target genome index; Performing dimensionality reduction processing on the brain fingerprint data and the genome data to obtain core feature difference data of the subject to be detected relative to a control population; Inputting the core feature difference data into a pre-trained difference value prediction model to obtain a brain cognitive function difference value output by the difference value prediction model; Among them, the difference value prediction model is obtained by training based on the brain fingerprint data, genome data and brain cognitive function scores corresponding to the target population and the control population respectively; the subjects to be detected and the target population are both plateau immigrants, and the control population is a plain population.

2. The method for processing brain cognitive function data according to claim 1, characterized in that: The method further includes: pre-training the difference value prediction model, wherein the pre-training the difference value prediction model includes: Acquire a data set; the data set includes a training set and a test set, and the data set includes core feature difference data and brain cognitive function difference values ​​corresponding to the core feature difference data; Selecting core feature difference data from the training set and inputting them into the initial difference value prediction model to obtain a brain cognitive function difference value prediction value output by the initial difference value prediction model; Comparing the predicted value of the brain cognitive function difference value with the brain cognitive function difference value in the training set to calculate a loss function; Adaptively adjusting the model parameters of the initial difference value prediction model according to the loss function, and stopping the training when the training stop condition is met to obtain a trained difference value prediction model; The test set is used to verify the accuracy of the trained difference value prediction model in predicting the difference value of brain cognitive function. If the verification is passed, the training is terminated to obtain the verified difference value prediction model.

3. The method for processing brain cognitive function data according to claim 2, characterized in that: The acquiring of the data set comprises: Based on the target brain fingerprint index and the target genome index, brain fingerprint data and genome data of the target population and the control population are collected respectively; Performing dimensionality reduction processing on the brain fingerprint data and genome data of the target population and the control population to obtain core feature difference data in the data set; Obtaining a first brain cognitive function score and a second brain cognitive function score obtained by performing a neurobehavioral test on the target population and the control population; The difference between the first brain cognitive function score and the second brain cognitive function score is used as the brain cognitive function difference value in the data set.

4. The method for processing brain cognitive function data according to claim 1, characterized in that: The method further includes: determining the target brain fingerprint index; the determining the target brain fingerprint index includes: Acquire the first brain fingerprint data of the target population collected based on the initial brain fingerprint indicators, and the second brain fingerprint data of the control population; A difference analysis is performed on the first brain fingerprint data and the second brain fingerprint data to determine the target brain fingerprint index; the difference analysis includes one or more of whole-brain functional connectivity difference analysis, dynamic functional connectivity difference analysis and topological attribute difference analysis; the target brain fingerprint index is a brain fingerprint index with significant differences in brain cognitive function damage between the target population and the control population.

5. The method for processing brain cognitive function data according to claim 1, characterized in that: The method further includes: determining the target genome index; the determining the target genome index includes: Performing spatial correlation analysis on the brain fingerprint data and whole-brain gene expression data of the target population to obtain gene expression data related to the brain fingerprint data; Based on the gene expression data and the cognitive data of the target population and the control population, risk alleles with significant differences in brain cognitive function impairment are screened out as target genome indicators.

6. The method for processing brain cognitive function data according to claim 3, characterized in that: After obtaining the brain cognitive function difference value, the method further includes: Based on the brain cognitive function difference values, a predicted weighted brain region map and / or a brain network group map is drawn.

7. The method for processing brain cognitive function data according to claim 3, characterized in that: The neurobehavioral test includes any one or more of a working memory test, a sustained attention test, and a cognitive flexibility test.

8. A brain cognitive function data processing device, characterized in that: include: A collection module, used to collect brain fingerprint data of the subject to be detected based on the target brain fingerprint index, and to collect genome data of the subject to be detected based on the target genome index; An analysis module, used for performing dimensionality reduction processing on the brain fingerprint data and the genome data to obtain core feature difference data of the subject to be detected relative to the control population; A prediction module is used to input the core feature difference data into a pre-trained difference value prediction model to obtain a brain cognitive function difference value output by the difference value prediction model; wherein the difference value prediction model is trained based on the brain fingerprint data, genome data and brain cognitive function scores corresponding to the target population and the control population respectively; the subjects to be detected and the target population are both plateau immigrants, and the control population is a plain population.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the brain cognitive function data processing method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the brain cognitive function data processing method as described in any one of claims 1 to 7.

Images