Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. Based on the embodiments of the present application, every other embodiment obtained by a person skilled in the art without making any inventive effort falls within the scope of protection of the present application.
First, an application scenario to which the present application is applicable will be described. The application can be applied to the technical field of image processing.
It has been found that with the accumulation of neuroimage data and the advancement of technology of normalized modeling methods, there is an increasing demand for fitting dynamic changes of the brain.
Currently, common normalized modeling methods mainly include a parametric method, a non-parametric method, a semi-parametric method, and a bayesian statistical method. However, the existing brain charts still have some defects, most of the brain charts are built based on European and American population, and the sample size of the brain charts is small and the representativeness of the brain charts is insufficient (the sample size is less than 3%), so that the existing brain charts cannot fully reflect the uniqueness of the Chinese population and lack the representativeness of the brain charts aiming at the Chinese population, in addition, the existing brain charts are mostly concentrated in modeling of brain anatomical structures, and the modeling is not perfect for important functional indexes of the brain, such as blood flow perfusion, blood brain barrier permeability, neural activity pattern and the like of the brain region.
Based on the basic information of the object to be tested, the equipment model of the brain detection instrument, the brain function indexes and the brain structure indexes, a plurality of brain health deviation images are generated for representing the brain health state of the object to be tested, wherein each brain health deviation image comprises a plurality of brain chart curves and individuation data points, the plurality of brain chart curves are generated based on standard brain chart data under the corresponding brain indexes, the individuation data points are used for representing the ordering condition in the standard brain chart data under the corresponding brain indexes detected for the object to be tested, and the accuracy of brain health assessment is improved by simultaneously considering the plurality of brain function indexes and the plurality of brain structure indexes.
Referring to fig. 1, fig. 1 is a flowchart of an image processing method according to an embodiment of the application. As shown in fig. 1, the image processing method provided in the embodiment of the present application includes:
s101, acquiring multi-mode nerve image data detected by a to-be-detected object from a brain detection instrument in real time.
Here, all the multi-modal neuroimage data is in digital imaging and communication standard (DIGITAL IMAGING AND Communications IN MEDICINE, DICOM) data format, converted to nifi format using the dcm2niix tool for ease of reading and writing.
The multi-modal neuroimaging data includes a plurality of brain structural indicators and a plurality of brain functional indicators for characterizing brain dynamics of the subject to be tested.
Preferably, the plurality of brain structural indices comprise structural magnetic resonance imaging data and diffusion tensor weighted imaging data, and the plurality of brain functional indices comprise functional magnetic resonance imaging data, arterial spin labeling data, and vessel-water exchange imaging data.
The method comprises the following steps of extracting brain structural indexes aiming at multi-mode nerve image data to obtain the following data:
Structural magnetic resonance imaging data MRI (3D T1 WI) is mainly used for acquiring structural information of the brain, such as brain tissue distribution, brain region division, white matter and gray matter boundaries and the like, and is started from data preprocessing through a segmentation process of a free-form surface (FreeSurfer), wherein the data preprocessing comprises image registration, offset field correction and denoising, brain region separation is performed through brain tissue identification (brain extraction), cortex reconstruction is performed, white matter and gray matter boundaries are identified, and inner and outer cortex surfaces are generated. Thereafter, subcortical structures such as basal ganglia, thalamus, ventricles, etc. are further segmented and normalized brain region segmentation is performed, often using templates such as Deschin-Kiril brain atlas (Desikan-KILLIANY ATLAS). The segmentation process encompasses not only cortical and subcortical structures, but also the scaling of the gyrus and sulcus. Finally, the generated segmentation results include cortical surface, brain region labeling, and anatomical feature data.
Diffusion tensor weighted imaging Data (DKI) provides information about the trend, density, integrity and the like of white matter fiber bundles of the brain, is helpful for understanding the microstructure of the brain, fits the data by using a diffusion kurtosis model through denoising, correction and spatial registration of images, calculates parameters such as diffusion tensor, kurtosis tensor and the like, and obtains diffusion characteristics in different directions. The anisotropy score value (Fractional Anisotropy, FA) reflects the anisotropy of water molecules in a specific direction as an important index. The FA values are then used to analyze the microstructure of the tissue, particularly in areas such as the brain white matter fiber bundles.
Extracting brain function indexes aiming at the multi-mode nerve image data to obtain the following data:
Functional magnetic resonance imaging data (fMRI), reflecting the functional activity of the brain in the execution of specific tasks or resting states, reveal the activation patterns and functional connections of the different brain regions, first, data preprocessing, including artifact and noise removal, and head movement correction, ensuring data alignment at all time points. Each brain image tested was mapped to a standard brain template by spatial normalization to facilitate comparison across individuals. Then, the signal-to-noise ratio is enhanced by smoothing, time series analysis is performed, activation signals of different areas of the brain are extracted, and brain region activities associated with specific tasks or stimuli are identified.
Arterial spin labeling data (ASL) for quantifying cerebral blood perfusion conditions, i.e. Cerebral Blood Flow (CBF), reflecting blood supply changes in different areas of the brain, is first subjected to data preprocessing, including artifact, noise removal, and then image alignment at all time points is ensured by head movement correction. Next, spatial registration is performed, aligning images of different time points or different subjects onto a standard brain template. The baseline signal is then removed and the difference between the arterial spin labeling signal and the control signal is calculated, resulting in a quantitative value for cerebral blood flow (Cerebral Blood Flow, CBF). Through these steps, regional cerebral blood flow information can be extracted from ASL data, revealing blood supply changes in different areas of the brain.
Vessel-water exchange imaging data (VEXI), providing information about blood brain barrier permeability, aid in understanding the mass exchange between cerebral vessels and brain tissue, and linearly register b= s/mm2 diffusion imaging data and bf= s/mm2, bd= s/mm2 VEXI data to T1WI using normalization tools (ANTs). T1WI is then registered from the local space to the study-specific template by a symmetric image normalization algorithm (ANTs (Advanced Normalization Tools) algorithm). The parametric map (e.g., apparent moisture exchange APPARENT WATER exchange across the blood-brain barrier, AXRBBB, etc.) is then iteratively aligned to the template by applying a linear registration transformation matrix and a previously generated deformation field. A gaussian filter with standard deviation of 1.5 voxel size was applied to the pre-processed VEXI data. Finally, MNI152 standard brain T1 weighted images (MNI 152T 1 WI) are non-linearly registered to the corresponding templates to transform some MNI152 template-based atlas into study specific template space to derive water-vessel imaging permeability parameters for specific Areas (AXRBBB).
S102, generating a plurality of brain health deviation reports according to basic information of the to-be-tested object, equipment model of the brain detection instrument, a plurality of brain function indexes and a plurality of brain structure indexes, and representing brain health states of the to-be-tested object.
The brain health deviation reports comprise a plurality of first brain health deviation images and a plurality of second brain health deviation images, each first brain health deviation image corresponds to a brain function index, each second brain health deviation image corresponds to a brain structure index, each brain health deviation image comprises a plurality of brain chart curves and individuation data points, the plurality of brain chart curves are generated based on standard brain chart data under the corresponding brain indexes, and the individuation data points are used for representing ordering conditions in the standard brain chart data under the corresponding brain indexes, which are detected for a to-be-tested object.
The application acquires the multi-mode nerve image data from the brain detection instrument in real time and generates a plurality of brain health deviation images, which can obviously improve the accuracy, efficiency and individuation degree of brain health assessment, provide powerful support for clinical diagnosis and treatment, and simultaneously provide important data resources for brain science research.
Here, the basic information of the subject to be tested includes age and sex, the abscissa of each brain health deviation image is age, the ordinate of each brain health deviation image is a unit of a corresponding brain index, and the plurality of brain graph curves of each brain health deviation image include a first brain index development curve for characterizing the sex of the subject to be tested and a change curve of an average value of all detection values of a corresponding brain index of an entire age range under a device model of a brain detection instrument in which the subject to be tested performs brain detection, and at least one second brain index development curve having a preset gap between each data point of the second brain index development curve and each data point of the first brain index development curve.
In an alternative embodiment, the first brain index development curve (black solid line) is a variation curve representing the average value of all detection values of the brain index corresponding to the whole age segment under the sex and the detection instrument of the subject to be detected, based on standard brain chart data (which may be the average value or range derived from the detection results of a large number of normal persons) corresponding to the brain index. This can be seen as an average human level over the index, a second brain index development curve (black dashed line) is a preset gap from the first brain index development curve, indicating that the threshold level is exceeded or fallen below, a preset interval is present, for example, the lower limit of the preset interval may be 60% of the average human brain health level, the upper limit of the preset interval may be 100% of the average human brain health level, and if a personalized data point is present above the second brain index development curve representing 100% of the average human brain health level, it indicates that the brain health of the subject to be tested exceeds 100% of the average human level, and is in a healthy state, and similarly, if a personalized data point is present below the second brain index development curve representing 60% of the average human brain health level, it indicates that the brain health of the subject to be tested is below 60% of the average human level, and is in a sub-healthy state, requiring further diagnosis. These limits are used to identify whether the brain index of the subject to be tested deviates from the normal range.
Here, a range of reference standards for normal population across life cycles, which varies with age, is created using a plurality of brain structures and brain function indexes of a large number of healthy population, and is referred to as normal population brain chart data (standard brain chart data).
The personalized data points represent the detection results of the object to be tested under the corresponding brain indexes, the ordering condition in the standard brain chart data, the positions of the personalized data points are relative to the positions of the brain chart curves, and whether the brain indexes of the object to be tested are higher or lower than the average level and the deviation degree can be intuitively displayed.
The abscissa of the brain health deviation image is age, the condition of different age groups is represented, the ordinate of the brain health deviation image is a unit corresponding to brain indexes such as fraction, percentage, volume and the like, and the health state of a to-be-tested object on different brain indexes and the change trend with age can be intuitively known by observing the image.
In particular, the brain health deviation images further comprise a percentile, wherein for each brain health deviation image, a corresponding percentile is displayed in the associated region of each data point in the brain health deviation image, the percentile being used to characterize the ratio of the individualized data point associated with the percentile to standard brain chart data at the corresponding age, the corresponding brain index.
Here, a percentile is a statistic that indicates how many proportions of data points are less than or equal to a particular value in a set of data. For example, the 50 th percentile (median) indicates that half of the data points are less than or equal to the value, while the other half are greater than the value.
For each brain health deviation image, a corresponding percentile is calculated and displayed for each personalized data point, reflecting the relative position between the data point of the subject to be tested and standard brain chart data (typically an average or range based on the detection results of a large number of normal people) at the same age and brain index.
In the brain health deviation image, the associated region of each personalized data point (possibly some small region around the data point or the data point itself) will display the corresponding percentile, which region may highlight the percentile information by color coding, numerical labeling, or other visual elements.
If the percentile of a data point is higher (e.g., near 100%), it is indicated that the data point is at a higher level for the same age and brain index than for standard brain chart data, whereas if the percentile is lower (e.g., near 0%), it is indicated that the data point is at a lower level, and by comparing the percentiles of different data points, it is intuitively known whether the health status of the subject under test on different brain indexes deviates from the normal range, and the degree of deviation.
Doctors and researchers can use these percentile-containing brain health deviation images to more accurately assess the brain health status of a subject to be tested, and these images can serve as important basis for developing personalized treatment plans, monitoring disease progression, and assessing treatment efficacy.
A plurality of brain health deviation images of a subject to be tested is determined by:
Inputting the age, the sex, the equipment model of the brain detection instrument, the brain function indexes and the brain structure indexes of the object to be tested into a brain health evaluation model to obtain a plurality of brain health deviation images, wherein the brain health evaluation model is used for representing the relation between target parameters and brain health states, and the target parameters are the age, the sex, the brain indexes obtained by the brain detection of the object to be tested and the equipment model of the brain detection instrument.
Firstly, collecting basic information and brain index data of a to-be-tested object, wherein the age is recorded; sex recording sex (male/female) of the subject to be tested; the brain detection instrument comprises a brain detection instrument, a plurality of brain function indexes, a plurality of brain structure indexes, a brain detection instrument and a brain detection instrument, wherein the brain detection instrument is used for recording the equipment number of a specific instrument for performing brain detection so as to ensure the traceability and accuracy of data;
Then, applied to a brain health assessment model that is capable of characterizing the relationship between target parameters (age, gender, brain index, device model) and brain health status, a plurality of brain health deviation images are obtained from the model, the model will generate a series of brain health deviation images based on the input data, which images may be presented in two-dimensional or three-dimensional form, and display the degree of deviation of brain health status, which deviation images are analyzed by visual inspection or automated analysis tools to determine specific deviation areas and degrees of brain health status. This helps identify potential brain functional or structural abnormalities, and finally, the offset images are interpreted by an expert in the field of neuroscience or medical imaging to determine specific brain health problems and possible causes, and a treatment plan is formulated, i.e., a personalized treatment or intervention plan is formulated for the subject to be tested according to the interpretation results to improve brain health, and brain detection is suggested to be performed periodically to track changes in brain health and adjust the treatment plan.
Referring to fig. 2, fig. 2 is a flowchart illustrating a method for determining a brain health evaluation model according to an embodiment of the application. As shown in fig. 2, the image processing method provided in the embodiment of the present application includes:
S201, determining response variables and covariates.
Here, the response variable includes a brain health deviation image representing a brain health state of the subject to be tested, and the covariates include an age, a sex, a plurality of brain function indicators, and a plurality of brain structure indicators.
The response variable here is not a single value in the conventional sense, but a series of images, each of which characterizes the degree of deviation of the subject to be tested from a particular brain health indicator. However, in conventional applications of the GAMLSS model, the response variables are typically numerical. To incorporate brain health deviation images into the GAMLSS model, we need to extract numerical features from these images. These features may be average deviation values, maximum deviation values, sums of deviation values for specific areas in the image, or other quantization indices derived by image analysis algorithms.
The covariates herein include the following parameters:
the age is the actual age of the object to be tested and is a numerical variable;
Sex, sex of the subject to be tested, is usually treated as a classification variable, and can be coded into 0 (female) and 1 (male), or other suitable coding modes are adopted;
A plurality of brain function indicators, wherein the indicators may comprise cognitive function scores, memory test scores and the like, which are all numerical variables;
a number of brain structural indices, which may relate to brain volume, grey matter/white matter ratio etc., are also numerical variables.
And S202, fitting the response variable and the covariate according to a generalized additive model to obtain the brain health assessment model.
Firstly, data preprocessing is carried out, numerical type characteristics are extracted from brain health deviation images and used as response variables of GAMLSS models, normalization processing is carried out on the numerical type covariates (such as age, brain function indexes and brain structure indexes) to eliminate the influence of dimension differences on the models, and classification variables such as gender and the like are properly encoded for use in the models.
Then, an appropriate probability distribution type is selected according to the characteristics of the response variables. Since the response variable is a numerical feature extracted from an image, it is possible to conform to a normal distribution, a lognormal distribution, a gamma distribution, or the like, and define a regression function gj (xi) for each parameter (θ1, θ2, θ3, θ4). These functions may be linear functions or more complex nonlinear functions such as smoothing functions in a Generalized Additive Model (GAM), and by way of example, the four parameters (θ1, θ2, θ3, θ4) typically represent a position parameter (e.g., mean or median), a scale parameter (e.g., standard deviation or variance), a shape parameter (e.g., skewness), and a super-shape parameter (e.g., kurtosis), respectively.
The model is fitted using statistical software (e.g., gamlss packages in the R language). In the fitting process, the distribution type of the response variable and the regression function of each parameter are required to be specified, the fitting effect of the model is checked through residual analysis and other means, the reasonability of model assumption is ensured, and the model structure is adjusted according to the diagnosis result, such as changing the distribution type, adjusting the regression function form and the like, so that the prediction performance of the model is improved.
Modeling is carried out according to the brain structure and function indexes and the ages, and corresponding percentiles are obtained. GAMLSS allows modeling of distribution parameters such as position, scale, shape etc. as a function of covariates, so that percentiles of brain volumes in different age periods can be predicted according to the relationship between brain structure and function index and age.
In an alternative embodiment, the GAMLSS model is in the basic form:
Yi ~ f (yi∣θ1(xi), θ2(xi), θ3(xi),θ4(xi))
Where yi is the response variable, and f (yi|θ1 (xi), θ2 (xi), θ3 (xi), θ4 (xi)) is the probability distribution given a set of position, scale, and shape parameters. Specifically, f represents a probability density function (or probability mass function) in which the parameters θ1, θ2, θ3, θ4 are modeled by covariates. These four parameters (θ1, θ2, θ3, θ4) generally represent a position parameter (e.g., mean or median), a scale parameter (e.g., standard deviation or variance), a shape parameter (e.g., skewness), and a super-shape parameter (e.g., kurtosis), respectively.
Each parameter can be represented by a regression model as a function of a covariate:
θj (xi) =gj (xi), where j=1, 2,3,4, gj (xi) is a regression function, typically a linear regression model or a more complex nonlinear function.
Modeling is performed according to brain structure and function indexes (as covariates) and ages (also as covariates), and percentiles of brain volumes under different age periods are obtained, and firstly, a proper probability distribution is needed to describe brain volume data. For example, a normal distribution, a lognormal distribution, a gamma distribution, or the like may be selected, and then a position parameter (mean or median), a scale parameter (standard deviation or variance), a shape parameter (skewness), and a super-shape parameter (kurtosis) are modeled using GAMLSS models. These parameters may be expressed as a function of brain structure and function index and age, and a model GAMLSS may be fitted using statistical software (e.g., gamlss package in R language), and finally, once the model fitting is complete, the model may be used to predict percentiles of brain volumes at different ages. This typically involves calculating a quantile function at a given probability p.
For normal distribution data, the percentile can be calculated by the following formula:
P(p)=μ+σ×Φ−1(p)
Wherein μ is an estimated mean value (position parameter), σ is an estimated standard deviation (scale parameter), and Φ -1 (p) is a quantile function of a standard normal distribution, namely, a Z value corresponding to a given probability p is obtained.
The GAMLSS model provides a flexible method for modeling and predicting distribution parameters of complex data, and can accurately describe and predict the relationship between brain structures and functional indexes and ages by selecting proper distribution and regression functions, and obtain percentiles of brain volumes in different age periods.
Specifically, the brain health assessment model may be trained by:
A training sample set is obtained.
The training sample set comprises a plurality of training samples, wherein each training sample comprises a sample age, a sample sex, a device model of a sample brain detection instrument, a plurality of sample brain function indexes, a plurality of sample brain structure indexes and a plurality of sample brain health deviation images;
first, a large amount of brain health data needs to be obtained from a reliable medical database or research institution. These data should contain individuals of different ages, sexes, brain health conditions, after the data is collected, the data needs to be cleaned to remove incomplete, abnormal or repeated data, ensure the accuracy and consistency of the data, and for each training sample, the brain health state needs to be noted, which is usually based on diagnosis of a doctor or professional brain health assessment results.
Features useful for brain health prediction, including sample age, sample gender, device model of the sample brain test instrument, multiple sample brain function indicators (e.g., memory, attention, etc.), multiple sample brain structure indicators (e.g., brain volume, sulcus depth, etc.), and possibly genetic information, are then extracted from the raw data, and multiple sample brain health deviation images are output as models. The images may be brain scan images such as MRI, CT, etc., and the health-related features extracted by image processing techniques.
The method comprises the steps of taking a sample age, a sample gender, a device model of a sample brain detection instrument, a plurality of sample brain function indexes and a plurality of sample brain structure indexes as inputs of an initial brain health assessment model, and taking a plurality of sample brain health deviation images as outputs of the initial brain health assessment model so as to train the initial brain health assessment model.
Depending on the complexity of the problem and the nature of the data, a suitable machine learning or deep learning model is selected. For example, convolutional Neural Networks (CNNs) perform well in processing image data, while cyclic neural networks (RNNs) are suitable for processing time-series data.
Taking sample age, sample gender, equipment model of a sample brain detection instrument, a plurality of sample brain function indexes and a plurality of sample brain structure indexes as input features of a model, taking a plurality of sample brain health deviation images as output targets of the model, training an initial model by using a training sample set, minimizing prediction errors by adjusting parameters of the model, and in the training process, evaluating the performance of the model by using cross-validation and other technologies, so as to prevent overfitting.
And (3) evaluating the performance of the model by using a test data set, wherein the indexes comprise accuracy, recall rate, F1 score and the like, fine tuning the parameters of the model according to the evaluation result so as to improve the performance of the model, and knowing which features have the greatest influence on the prediction result of the model through feature importance analysis, thereby being beneficial to further optimizing feature selection and model design.
The image processing method and the image processing device provided by the embodiment of the application are used for acquiring the multi-modal neural image data detected for the object to be tested from the brain detection instrument in real time, wherein the multi-modal neural image data comprises a plurality of brain structure indexes and a plurality of brain function indexes used for representing the dynamic change of the brain of the object to be tested, and generating a plurality of brain health deviation images used for representing the brain health state of the object to be tested according to the basic information of the object to be tested, the equipment model of the brain detection instrument, the plurality of brain function indexes and the plurality of brain structure indexes, wherein each brain health deviation image comprises a plurality of brain chart curves and personalized data points, the plurality of brain chart curves are generated based on standard brain chart data under the corresponding brain indexes, and the personalized data points are used for representing the ordering condition in the standard brain chart data under the corresponding brain indexes detected for the object to be tested. The application improves the accuracy of brain health assessment and provides powerful support for clinical diagnosis and treatment.
The application can flexibly cope with challenges of uneven data distribution and multi-mode fitting, and firstly, the application models according to the Chinese crowd god image data of a large sample. Secondly, the normalized modeling method adopted by the application can effectively improve the fitting accuracy when the data distribution is sparse, and can reasonably fuse the data of different modes, thereby overcoming the defect of the traditional method in the multi-mode data fitting. By means of the method, the change track of the brain structure and function can be accurately drawn, the change process of the whole life cycle of an individual is covered, and an accurate brain chart is formed.
Compared with the prior art, the application has remarkable advantages. Firstly, a brain structure and a functional brain chart specific to the normal population in China are established for the first time, and the method has innovation and representativeness. Secondly, through optimization of a standardized modeling method, the problem of data distribution non-uniformity can be well solved, and therefore accuracy and stability of a brain structure and function change model are improved. In addition, aiming at the fusion of multi-mode data, the application can effectively perform unified modeling on different nerve image modes (such as structural magnetic resonance imaging, functional magnetic resonance imaging and the like), and eliminates the difficulty of the traditional method in mode conversion. This allows for a more comprehensive and accurate construction of brain charts. Finally, the application has higher clinical application value and can provide reliable support for diagnosis and treatment of nervous system diseases.
The application establishes the normal reference value (namely brain chart) of the specific multi-mode fitting brain structure and function of Chinese crowd, has good self-adaptive capacity of the brain structure and function chart, has good transfer generalization capacity, and can be applied to new data.
Since the multi-modal neuroimaging data covers a plurality of brain structural indexes and brain functional indexes, the application can provide a more comprehensive and detailed brain health assessment, which is helpful for identifying various potential problems in brain health, including but not limited to structural abnormalities, dysfunctions and the like, and acquiring and processing data in real time means that brain health status feedback of a subject to be tested can be obtained immediately. The method is particularly important for the situation that medical decision needs to be made rapidly, such as emergency treatment when acute brain injury or disease occurs, the generated brain health deviation image can reflect the difference among individuals by combining the basic information of the to-be-tested object and the equipment model of the brain detection instrument, so that more personalized brain health assessment and treatment scheme formulation can be realized, the brain health deviation image intuitively displays the ratio condition of the to-be-tested object in standard brain chart data under different brain indexes through a plurality of brain chart curves and data points, and doctors and other medical professionals can understand and interpret the data more easily in a visual mode, so that more accurate diagnosis can be made.
The brain health deviation image provided by the application can be used as an important reference basis for clinical decision. The doctor can make more accurate and effective treatment scheme according to the information provided by the images and other clinical data, and the method is not only beneficial to clinical practice, but also provides new tools and visual angles for neuroscience research. By further analyzing these multimodal neuroimaging data, researchers can further understand the structure and function of the brain and their relationships with various neurological diseases.
Based on the same inventive concept, the embodiment of the present application further provides an image processing device corresponding to the image processing method, and since the principle of solving the problem by the device in the embodiment of the present application is similar to that of the image processing method in the embodiment of the present application, the implementation of the device may refer to the implementation of the method, and the repetition is omitted.
Referring to fig. 3, fig. 3 is a schematic structural diagram of an image processing apparatus according to an embodiment of the application. As shown in fig. 3, the image processing apparatus 300 includes:
An acquisition module 301, configured to acquire, in real time, multi-modal neuroimaging data detected for a subject to be tested from a brain detection apparatus, where the multi-modal neuroimaging data includes a plurality of brain structural indicators and a plurality of brain function indicators for characterizing brain dynamic changes of the subject to be tested;
The generating module 302 is configured to generate a plurality of brain health deviation reports according to the basic information of the subject to be tested, the equipment model of the brain detection instrument, the plurality of brain function indexes and the plurality of brain structure indexes, where the plurality of brain health deviation reports include a plurality of first brain health deviation images and a plurality of second brain health deviation images, each of the first brain health deviation images corresponds to a brain function index, each of the second brain health deviation images corresponds to a brain structure index, each of the brain health deviation images includes a plurality of brain chart curves and individualized data points, the plurality of brain chart curves are generated based on standard brain chart data under the corresponding brain indexes, and the individualized data points are used for characterizing the ordering condition in the standard brain chart data under the corresponding brain indexes detected for the subject to be tested.
Referring to fig. 4, fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the application. As shown in fig. 4, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.
The memory 520 stores machine-readable instructions executable by the processor 510, and when the electronic device 500 is running, the processor 510 communicates with the memory 520 through the bus 530, and when the machine-readable instructions are executed by the processor 510, the steps of the image processing method in the method embodiment shown in fig. 1 can be executed, and the specific implementation can be referred to the method embodiment and will not be described herein.
The embodiment of the present application further provides a computer readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the steps of the image processing method in the embodiment of the method shown in fig. 1 may be executed, and a specific implementation manner may refer to the embodiment of the method and will not be described herein.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. The storage medium includes a U disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.
It should be noted that the foregoing embodiments are merely illustrative embodiments of the present application, and not restrictive, and the scope of the application is not limited to the embodiments, and although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that any modification, variation or substitution of some of the technical features of the embodiments described in the foregoing embodiments may be easily contemplated within the scope of the present application, and the spirit and scope of the technical solutions of the embodiments do not depart from the spirit and scope of the embodiments of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.