WO2019136745A1 - Magnetic resonance image based brain age test method and apparatus - Google Patents

Abstract

Disclosed are a magnetic resonance image based brain age test method, a magnetic resonance image based brain age test apparatus, an electronic device and a computer-readable storage medium. The magnetic resonance image based brain age test method comprises: acquiring a T1-weighted brain magnetic resonance image of an object to be tested (S101); determining, based on the T1-weighted brain magnetic resonance image, a brain structure, a brain lobe and a brain tissue segmentation map of the object to be tested (S102); calculating a normalized volume of the brain structure and a brain atrophy value of the brain lobe (S103); and inputting the normalized volume and the brain atrophy value into a brain age estimation model to acquire a brain age of the object to be tested (S104). The method can be used to estimate a brain age, and facilitates people in intervening with the health of the brain in advance and improves the health awareness of people for the brain.

Description

Brain age test method based on magnetic resonance image and brain age test device Technical field

The present application belongs to the field of image processing technologies, and in particular, to a brain age testing method based on magnetic resonance images, a brain age testing device based on magnetic resonance images, an electronic device, and a computer readable storage medium.

Background technique

From fetus to infant, to children, adolescents, middle-aged, old age, brain and body at the same time, the brain age (referred to as brain age) is used to measure the development of the human brain. In addition, people also have a physiological age, which is used to describe the time from the date of birth. Normally, the physiological age is not equal to the brain age. For example, when a person's brain develops a lesion, it causes a sharp aging of the brain, a memory loss and a slow response, which makes the brain older than the physiological age. When people exercise regularly and maintain physical and mental pleasure for a long time, it will delay the aging of the brain. That makes the brain age less than the physiological age.

The determination of brain age can raise people's awareness of brain health, intervene in brain health in advance, and delay brain aging.

technical problem

In view of this, the embodiments of the present application provide a brain age testing method based on magnetic resonance images, a brain age testing device based on magnetic resonance images, an electronic device, and a computer readable storage medium, which can implement testing of human brain age.

Technical solution

A first aspect of the present application provides a brain age testing method based on a magnetic resonance image, comprising:

Obtaining a T1-weighted brain magnetic resonance image of the individual to be tested;

Determining a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject based on the T1-weighted brain magnetic resonance image;

Calculating a normalized volume of the brain structure, and calculating a brain atrophy value of the brain lobe based on the brain tissue segmentation map;

Entering the normalized volume and the brain atrophy value into a brain age estimation model to obtain a brain age of the test subject, wherein the brain age estimation model is based on a normalized volume of the training sample individual and a brain atrophy value And brain age training is available.

A second aspect of the present application provides a brain age testing device based on a magnetic resonance image, comprising:

An image acquisition unit, configured to acquire a T1-weighted brain magnetic resonance image of the individual to be tested;

An image analyzing unit, configured to determine a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject based on the T1-weighted brain magnetic resonance image;

a parameter calculation unit, configured to calculate a normalized volume of the brain structure, and calculate a brain atrophy value of the brain lobe based on the brain tissue segmentation map;

a brain age acquisition unit, configured to input the normalized volume and the brain atrophy value into a brain age estimation model, and obtain a brain age of the test subject, wherein the brain age estimation model is based on the individuality of the training sample individual A volume, brain atrophy value and brain age training are obtained.

A third aspect of the present application provides an electronic device including a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor executes the computer program The steps of the method as described above are achieved.

A fourth aspect of the present application provides a computer readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the steps of the method as described above.

Beneficial effect

As can be seen from the above, the present application provides a brain age testing method based on magnetic resonance images, first obtaining a T1-weighted brain magnetic resonance image of an individual to be tested, and secondly, analyzing the T1-weighted brain magnetic resonance image to obtain a brain structure, Brain lobe and brain tissue segmentation map, and calculating a normalized volume of the test subject and a brain atrophy value according to the obtained brain structure, brain lobe and brain tissue segmentation map, the normalized volume and the brain The atrophy value is input into the brain age estimation model, and the brain age of the individual to be tested is obtained, thereby realizing the test of the brain age of the individual to be tested. For the first time, the brain size and brain atrophy value are used as brain age test parameters. Through the application of the present application, the user can estimate his or her brain age, so as to understand the state of his brain health, and facilitate the intervention of his brain health in advance to delay. Brain aging, through the technical solutions provided by this application can estimate brain degeneration and improve people's awareness of brain health.

DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings used in the embodiments or the prior art description will be briefly described below. Obviously, the drawings in the following description are only the present application. For some embodiments, other drawings may be obtained from those skilled in the art without departing from the drawings.

1 is a schematic flowchart showing an implementation process of a brain age testing method based on magnetic resonance images according to Embodiment 1 of the present application;

2 is a schematic diagram showing an implementation flow of a brain age testing method based on magnetic resonance images provided in Embodiment 2 of the present application;

3 is a schematic structural diagram of a brain age testing device based on magnetic resonance images provided in Embodiment 3 of the present application;

4 is a schematic structural diagram of a brain age testing device based on magnetic resonance images provided in Embodiment 4 of the present application;

FIG. 5 is a schematic diagram of an electronic device according to Embodiment 5 of the present application.

Embodiments of the invention

In the following description, for purposes of illustration and description However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the application.

The brain age testing method based on the magnetic resonance image provided by the embodiment of the present application is applicable to an electronic device. Illustratively, the above electronic device includes, but is not limited to, a desktop computer, a tablet computer, a cloud server, a mobile phone terminal, and the like.

In order to explain the above technical solutions of the present application, the following description will be made by way of specific embodiments.

Example 1

The method for testing the brain age based on the magnetic resonance image provided in the first embodiment of the present application is described below. Referring to FIG. 1, the brain age testing method in the embodiment of the present application includes:

Step S101, acquiring a T1-weighted brain magnetic resonance image of the individual to be tested;

In the embodiment of the present application, the T1 weighted brain magnetic resonance image of the individual to be tested is first acquired, so that the related parameters about brain development of the test subject can be obtained from the T1 weighted brain magnetic resonance image, thereby utilizing the The parameter estimates the brain age of the individual to be tested, wherein the T1-weighted brain magnetic resonance image of the individual to be tested is a three-dimensional image.

Step S102, determining a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject according to the T1 weighted brain magnetic resonance image;

In the embodiment of the present application, the brain tissue segmentation map includes a gray matter segmentation map, a white matter segmentation map, and a cerebrospinal fluid segmentation map. In order to ensure more accurate determination of the brain structure, brain lobe, and brain tissue segmentation map of the individual to be tested, The above T1-weighted brain magnetic resonance image can be registered with a preset brain template library. Wherein, the brain template library is pre-stored, and the brain template library includes T1 weighted brain magnetic resonance images of two or more different brains, that is, including two or more different templates, preferably, to ensure the brain structure of the individual to be tested, The accuracy of the brain leaf and brain tissue segmentation map, which may include T1 weighted brain magnetic resonance images of people with different brain health, gender, age and age between 10 and 90 years old, and the number of images is more than 5 One.

In the embodiment of the present application, the brain structure and the brain leaf can be manually segmented in each T1 weighted brain magnetic resonance image (each template) in the brain template library, and the brain structure and brain corresponding to each template in the brain template library are obtained. Leaf, the above brain structure may include brain parenchyma, cerebellum, hippocampus, amygdala, ventricle, lateral ventricle, thalamus, caudate nucleus, putamen, globus pallidus, nucleus accumbens, midbrain, pons, cerebral ventricle, etc. An important brain structure associated with development and aging, including the left frontal lobe, left parietal lobe, left occipital lobe, left temporal lobe, left cingulate gyrus and left insular lobe, and right frontal lobe, right parietal lobe, right occipital lobe, Right temporal lobe, right cingulate and right island leaves. At the same time, the brain tissue is automatically segmented in each T1 weighted brain magnetic resonance image in the brain template library, and manual manual correction is performed on the basis of automatic computer segmentation, so that the brain tissue probability map corresponding to each template in the brain template library is obtained. The tissue probability map includes a white matter probability map, a gray matter probability map, and a cerebrospinal fluid probability map.

Specifically, the brain structure, the brain lobe, and the brain tissue segmentation map of the individual to be tested based on the brain template library may be:

S1021, using a non-linear registration to map each template in the brain template library to a T1-weighted brain magnetic resonance image of the individual to be tested, to obtain a T1-weighted brain magnetic resonance image between each template in the brain template library and the individual to be tested. Spatial mapping relationship;

S1022, by using the spatial mapping relationship, mapping each brain structure partition, brain leaf partition, and brain tissue probability map corresponding to each template in the brain template library to a T1-weighted brain magnetic resonance image of the individual to be tested, and obtaining each template Corresponding individual brain tissue division and brain leaf division, and brain tissue probability map of the individual to be tested;

Wherein, the above nonlinear registration can adopt a symmetric nonlinear registration algorithm based on a differential homeomorphic model. The brain tissue probability map of the above-mentioned test subject is: the brain tissue probability map of the template with the highest similarity to the T1-weighted brain magnetic resonance image of the individual to be tested is obtained by spatial mapping.

S1023, using a label fusion algorithm to fuse the brain structure partition and the brain leaf partition of the final test object;

The label fusion algorithm can be used to fuse the brain structure partitions of the individuals to be tested according to each template to obtain the brain structure partition of the final test subject. Similarly, using the label fusion algorithm, the test samples obtained according to each template are to be tested. The individual's brain lobe is fused to obtain the brain lobe of the final individual to be tested. The label fusion algorithm is an algorithm that combines the segmentation results corresponding to each template by using a desired maximum algorithm. The label fusion algorithm can solve the deviation caused by a single template, and the result is more accurate. Commonly used tag fusion algorithms include a reliability-based tag fusion algorithm, a majority consent rule tag fusion algorithm, and a weighted tag fusion algorithm.

S1024, using the brain tissue probability map of the above-mentioned test subject as a priori knowledge, using a brain tissue segmentation method of the Bayesian network, and performing brain tissue segmentation on the T1-weighted brain magnetic resonance image of the individual to be measured, and obtaining brain tissue of the individual to be tested Split the map.

Step S103, calculating a normalized volume of the brain structure, and calculating a brain atrophy value of the brain lobe based on the brain tissue segmentation map;

In the embodiment of the present application, the volume of the brain structure of the individual to be tested obtained in step S102 can be estimated, and the total volume of the brain is divided to calculate the normalized volume of the brain structure of the individual to be tested.

The brain gray matter volume, the white matter volume, and the cerebrospinal fluid volume in the brain lobe may be calculated according to the brain lobe partition and the brain tissue segmentation map obtained in step S102, and the brain atrophy value of the brain lobe is calculated according to the brain atrophy calculation formula, wherein The formula for brain atrophy is:

The brain atrophy value obtained in the above formula can directly reflect the degree of atrophy of the brain lobe. The larger the value, the greater the degree of brain atrophy on the surface of the brain.

Step S104, input the normalized volume and the brain atrophy value into a brain age estimation model, and obtain the brain age of the test subject;

In the embodiment of the present application, the brain age estimation model may be established in advance, firstly, the normalized volume of the training sample individual and the brain atrophy value are collected, and the normalized volume of the training sample individual and the brain atrophy value are used as independent variables, and the training will be performed. The brain age of the sample individuals was used as the dependent variable, and a linear support vector machine was used to establish a brain age estimation model. In addition, the establishment of the brain age estimation model is not limited to the linear support vector machine, but also the hidden Markov model, the neural network, etc. can be used to establish the brain age estimation model.

The following is a detailed discussion of how to apply a linear support vector machine to establish a brain age estimation model.

Assuming that there are N training samples, the training data are: (X ₁ , y ₁ ), (X ₂ , y ₂ ), ... (X _N , y _N ), where X _{i =} (x _i,1 ,x _{i , 2} ... x _{i, K} ) ^Τ , i = 1, 2 ... N, X _i includes K parameters related to brain development, and in the embodiment of the present application, may be a normalized volume of multiple brain structures and The brain atrophy value of the brain lobe, y _i is the brain age corresponding to the i-th training sample individual.

Assume that the expression of the brain age estimation model is: f(X)=W·X+b, W∈R ^1×K , b∈R

The optimization goal is:

The specific values of W and b when the optimization target value is minimum can be obtained. The data fitting algorithm such as gradient descent algorithm or genetic algorithm can be used to solve the values of W and b, so as to obtain the brain age estimation model. In addition, the optimization target can also be

The formula for optimizing the target is not limited here.

Preferably, in order to more accurately estimate the brain age of the individual to be tested, the T1-weighted brain magnetic resonance image may also be pre-processed after acquiring the T1-weighted brain magnetic resonance image of the individual to be tested, the pre-processing including the following Or more than two: noise reduction, de-fielding, and pixel range normalization. The present invention estimates the Rice noise variance based on the gray-scale distribution skewness noise estimation method, and then uses the non-local mean algorithm to reduce noise according to the estimated noise variance; the present application de-biasing field is used to remove the uneven magnetic field. The phenomenon that the gray level of the brain tissue is inconsistent; this application uses the method of histogram matching to normalize the intensity ranges of different images to a common range.

A total of 1936 healthy training samples were collected. The ratio of male to female was 1:1, and the age span ranged from 40 to 90 years. The T1 weighted brain magnetic resonance images of the training samples were from different models of 16 different hospitals in China, including SIEMENS. GE, PHILIPS NMR machine. Firstly, the T1 weighted brain magnetic resonance images of the collected training sample individuals are preprocessed; the 38 normalized volumes corresponding to each training sample individual and 12 brain atrophy values are obtained; since the healthy individual's brain age and physiological age are close Equal, so when training the brain age estimation model, the physiological age of the training sample individual is used as the brain age of the training sample individual to establish a brain age estimation model. After training, the difference between the established brain age estimation model and its physiological age is 5.44 years old, that is, according to the technical solution provided by the first embodiment of the present application, the error of the established brain age estimation model is 5.44 years old. The individuals to be tested were 12 healthy individuals (age range 67.3±9.3 years) and 14 individuals with Alzheimer's disease (age range 61.3±15.6 years). The T1-weighted brain magnetic resonance images of the individuals to be tested were from PHILIPS NMR. The resonance machine first preprocesses the T1-weighted brain magnetic resonance image of the individual to be measured, and extracts 38 normalized brain volumes and 12 brain atrophy values of the individuals to be tested, and estimates according to the brain age estimation model obtained above. The age of the individual to be tested. The test results showed that the average age difference between the predicted brain age and the physiological age of the above 12 healthy individuals was 6.4 years, and the predicted brain age and the physiological age difference of the individuals with Alzheimer's disease were 19.7 years old. It can be seen from the above test results that the brain of the Alzheimer's disease patient has a higher degree of brain aging than the healthy individual.

In the embodiment of the present application, the brain structure size and the brain atrophy value are used as the brain age test parameters for the first time. Through the application scheme, the user can estimate the brain age of the user to understand the state of the brain health and facilitate the advancement of the brain health. Intervention to delay brain aging, through the technical solutions provided in this application can estimate brain degeneration and improve people's awareness of brain health.

Example 2

Another method for testing a brain age based on magnetic resonance images provided in the second embodiment of the present application is described below. Please refer to FIG. 2 . The brain age testing method in the embodiment of the present application includes:

Step S201, acquiring a T1-weighted brain magnetic resonance image of the training sample individual;

In the embodiment of the present application, a method for establishing a brain age estimation model is specifically given. In order to determine a more accurate estimation acquisition model, a large number of parameters related to brain development are often required. However, many parameters related to brain development are not necessary, even if this parameter is added to the brain age estimation model, it cannot be To increase the accuracy of the brain age estimation model too much, it will occupy a large amount of computing resources in the process of calculating the brain age of the individual to be tested. Therefore, this embodiment provides a method for establishing a brain age estimation model, which can eliminate parameters that are not needed when establishing a brain age estimation model, and release certain computing resources.

In order to establish a brain age estimation model, it is first necessary to obtain a T1-weighted brain magnetic resonance image of the training sample individual, so as to subsequently extract relevant parameters of brain development of the training sample individual, thereby using the parameter to establish a brain age estimation model.

In order to more accurately extract the relevant parameters of brain development of the training sample individual, the T1-weighted brain magnetic resonance image of each training sample individual may be pre-processed, and the above pre-processing includes one or more of the following: noise reduction, de-biasing The field and pixel range are normalized.

Step S202, acquiring a brain lobe and a brain tissue segmentation map of each training sample individual, and calculating a brain atrophy value of the brain lobe based on the brain tissue segmentation map;

In the embodiment of the present application, according to the T1-weighted brain magnetic resonance image of each training sample individual, the brain lobe and the brain tissue probability map of each training sample individual are obtained, and the brain atrophy value of the brain lobe is calculated, and the specific implementation method can be referred to The first embodiment S102 and S103 are as described above, and are not described herein again.

In this embodiment, it is preferred to obtain 12 brain atrophy values corresponding to each training sample individual, including left frontal lobe, left parietal lobe, left occipital lobe, left temporal lobe, left cingulate gyrus and left insular lobe, and right frontal lobe. Brain atrophy values of right parietal lobe, right occipital lobe, right temporal lobe, right cingulate gyrus, and right island lobes.

Step S203, acquiring an i-th brain structure of each training sample individual, calculating a normalized volume of the i-th brain structure, a normalized volume based on the i-th brain structure, and brain atrophy of the brain of the training sample individual Value, establish an i-th brain age estimation model, and obtain the error of the above i-th brain age estimation model;

In the embodiment of the present application, the brain structure in the brain is very large, but the size of some brain structures changes little with the development of the brain, and some changes in the size of the brain structure do not affect the degree of aging of the brain. Obviously, the structure of these brains The normalized volume is not required in brain age calculations. Therefore, in the embodiment of the present application, we can first select a certain brain structure, calculate the normalized volume of the selected brain structure, and use the normalized volume and the above 12 brain leaf atrophy values as the model for establishing the brain age. The 13 dependent variables can be used to establish the brain age estimation model corresponding to the brain structure by using the linear support vector machine, and the minimum value of the optimization target is used as the error of the brain age estimation model.

Step S204, determining whether to traverse all brain structures of each training sample individual;

In the embodiment of the present application, after step S203, it is determined whether all brain structures have been traversed. If not, step S205 is performed, and if yes, step S206 is performed.

Step S205, increasing the value of i by one;

In this step, if all the brain structures have not been traversed, the normalized volume of the next brain structure and the above 12 brain leaf atrophy values are used as the 13 dependent variables for establishing the brain age estimation model, and the next brain structure is established. The corresponding brain age estimation model is obtained, and the error of the brain age estimation model is obtained.

Step S206, selecting a normalized volume of the brain structure corresponding to the brain age estimation model with a small error, and establishing a final brain age estimation model according to the normalized volume of the selected brain structure and the brain blade atrophy value;

In this step, if all brain structures are traversed, the normalized volume of the brain structure corresponding to the brain age estimation model with less error is selected. The threshold may be preset, for example, 10 years old, and the error of the brain age estimation model corresponding to each brain structure obtained above is compared with 10 years old, and if it is less than the preset 10 years old, the normalized volume of the brain structure is selected. For example, the normalized volume of the brain structure that may be selected is only the normalized volume of the cerebellum and the normalized volume of the hippocampus. The normalized volume of the selected brain structure, ie, the normalized volume of the cerebellum and the hippocampus normalized. The volume, and 12 brain leaf atrophy values were used as the 14 dependent variables of the final brain age estimation model to establish the final brain age estimation model.

S207. Acquire a T1-weighted brain magnetic resonance image of the individual to be tested;

S208, determining a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject according to the T1-weighted brain magnetic resonance image of the test subject;

S209, calculating a normalized volume of the brain structure of the test subject, and calculating a brain atrophy value of the brain of the test subject according to the brain tissue segmentation map of the test subject;

S210, inputting the normalized volume of the individual to be tested and the brain atrophy value of the test subject to the final brain age estimation model, and obtaining the brain age of the test subject;

In the embodiment of the present application, the determined brain structure of the test subject is corresponding to the brain structure in the final brain age estimation model, for example, if the final brain age estimation model has only the cerebellar normalized volume and the hippocampus If the volume is normalized, then in step S208, only the cerebellum and hippocampus of the individual to be tested need to be determined. In step S209, only the normalized volume of the cerebellum and the normalized volume of the brain structure need to be calculated. In addition, the foregoing steps S207-S209 are the same as the implementations of the steps S101-S103 in the first embodiment. For details, refer to the description of the first embodiment, and details are not described herein again. In addition, in the embodiment of the present application, not every brain age test needs to perform steps S201-S206. After the final brain age estimation model is established, steps S201-S206 need not be performed in subsequent brain age test.

In the embodiment of the present application, a method for establishing a brain age estimation model is specifically provided, which can eliminate parameters that are not needed when establishing a brain age estimation model, release certain computing resources, and firstly determine brain structure size and brain atrophy value. As a brain age test parameter, the user can estimate the age of the brain by using the solution of the present application, in order to understand the state of the brain health, and facilitate the intervention of the brain health in advance to delay the aging of the brain, through the technical solution provided by the application. Brain degeneration can be estimated and people's awareness of brain health can be improved.

It should be understood that the size of the sequence of the steps in the above embodiments does not mean that the order of execution is performed. The order of execution of each process should be determined by its function and internal logic, and should not be construed as limiting the implementation process of the embodiments of the present application.

Example 3

A third embodiment of the present invention provides a brain age testing device based on a magnetic resonance image. For convenience of description, only parts related to the present application are shown. As shown in FIG. 3, the brain age testing device 300 includes:

The image obtaining unit 301 is configured to acquire a T1-weighted brain magnetic resonance image of the individual to be tested;

The image analyzing unit 302 is configured to determine a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject according to the T1 weighted brain magnetic resonance image, where the brain tissue segmentation map includes a gray matter segmentation map, a white matter segmentation map, and a cerebrospinal fluid Split the map;

The parameter calculation unit 303 is configured to calculate a normalized volume of the brain structure, and calculate a brain atrophy value of the brain lobe based on the brain tissue segmentation map;

The brain age obtaining unit 304 is configured to input the normalized volume and the brain atrophy value into a brain age estimation model to obtain a brain age of the test subject, wherein the brain age estimation model is based on a normalized volume of the training sample individual Brain atrophy and brain age training are obtained.

Preferably, the image analysis unit 302 is specifically configured to:

The T1 weighted brain magnetic resonance image is registered with a preset brain template library to obtain a brain structure, a brain lobe, and a brain tissue probability map corresponding to the T1 weighted brain magnetic resonance image; and the T1 weighted brain magnetic resonance image is obtained. Corresponding brain tissue probability map is used as a priori knowledge, and brain tissue segmentation is performed on the T1-weighted brain magnetic resonance image to obtain a brain tissue segmentation map corresponding to the T1-weighted brain magnetic resonance image;

The brain template library includes: a T1-weighted brain magnetic resonance image of two or more different brains, and a brain structure, a brain lobe, and a brain tissue probability map corresponding to each T1-weighted brain magnetic resonance image, and the brain tissue probability map. Including the gray matter probability map, the white matter probability map and the cerebrospinal fluid probability map.

Preferably, the parameter calculation unit 303 is specifically configured to:

Calculating a normalized volume of the brain structure, and determining a gray matter volume in the brain lobe of the T1-weighted brain magnetic resonance image of the individual to be tested based on a brain tissue segmentation map of the T1-weighted brain magnetic resonance image of the individual to be tested , white matter volume and cerebrospinal fluid volume;

The brain atrophy value of the brain lobe is calculated according to the brain atrophy calculation formula, wherein the above brain atrophy calculation formula is:

Preferably, the brain age testing device 300 further includes:

The model building unit is configured to use the normalized volume of the training sample individual and the brain atrophy value as independent variables, and use the brain support age of the training individual as a dependent variable, and apply a linear support vector machine to establish a brain age estimation model.

Preferably, the brain age testing device 300 further includes:

The individual preprocessing unit to be tested is configured to preprocess the T1-weighted brain magnetic resonance image of the test subject to be tested before determining the brain structure, the brain lobe and the brain tissue segmentation map of the test subject according to the T1 weighted brain magnetic resonance image. The above preprocessing includes one or more of the following: noise reduction, de-biasing, and pixel range normalization.

It should be noted that the information interaction, the execution process, and the like between the modules/units of the foregoing device are based on the same concept, the specific functions and the technical effects brought about by the method embodiment of the present application. Part of it will not be described here.

Example 4

A fourth embodiment of the present invention provides a brain age testing device based on a magnetic resonance image. For convenience of explanation, only parts related to the present application are shown. As shown in FIG. 4, the brain age testing device 400 includes:

a training sample individual image acquiring unit 401, configured to acquire a T1-weighted brain magnetic resonance image of the training sample individual;

The training sample individual brain atrophy calculation unit 402 is configured to acquire a brain lobe and a brain tissue segmentation map of each training sample individual, and calculate a brain atrophy value of the brain lobe based on the brain tissue segmentation map;

The i-th brain age model acquisition unit 403 is configured to acquire an i-th brain structure of each training sample individual, calculate a normalized volume of the i-th brain structure, based on a normalized volume of the i-th brain structure, and Brain atrophy value of the brain, establish an i-th brain age estimation model, and obtain the error of the above i-th brain age estimation model;

The determining unit 404 is configured to determine whether to traverse all brain structures of each training sample individual;

Calculating a sequence number increasing unit 405 for increasing the value of i by 1 without traversing all brain structures;

The final model establishing unit 406 is configured to select a normalized volume of the brain structure corresponding to the brain age estimation model with less error, and establish a final brain according to the normalized volume of the selected brain structure and the brain blade atrophy value. Age estimation model;

The individual image acquisition unit 407 to be tested is configured to acquire a T1-weighted brain magnetic resonance image of the individual to be tested;

The individual image analysis unit 408 to be tested is configured to determine a brain structure, a brain lobe, and a brain tissue segmentation map of the individual to be tested based on the T1 weighted brain magnetic resonance image of the individual to be tested;

The individual parameter calculation unit 409 to be tested is configured to calculate a normalized volume of the brain structure of the individual to be tested, and calculate a brain atrophy value of the brain of the individual to be tested based on the brain tissue segmentation map of the individual to be tested;

The individual brain age obtaining unit 410 is configured to input the normalized volume of the test subject and the brain atrophy value of the test subject into a final brain age estimation model, and obtain the brain age of the test subject;

In the embodiment of the present application, the determined brain structure of the test subject is corresponding to the brain structure in the final brain age estimation model, for example, if the final brain age estimation model has only the cerebellar normalized volume and the hippocampus For the normalized volume, the individual image analysis unit 408 to be tested only needs to determine the cerebellum and hippocampus of the individual to be tested, and the individual parameter calculation unit 409 to be tested only needs to calculate the normalized volume of the cerebellum and the normalization of the brain structure. volume. In addition, the foregoing units 407-409 are the same as the implementations of the units 301-303 in the third embodiment. For details, refer to the description of the third embodiment, and details are not described herein again.

Preferably, the brain age testing device 400 further includes:

a pre-processing unit, configured to pre-process the T1-weighted brain magnetic resonance image of the individual to be tested, and to obtain each training sample individual, before determining the brain structure, the brain lobe, and the brain tissue segmentation map of the individual to be tested Before the brain structure, brain lobe and brain tissue segmentation map, the T1-weighted brain magnetic resonance images of each training sample individual are preprocessed, and the above preprocessing includes one or more of the following: noise reduction, de-biasing field, pixel range Normalized.

It should be noted that the content of the information exchange, the execution process, and the like between the modules/units of the foregoing device are based on the same concept, the specific functions and the technical effects brought by the second embodiment of the present application. The second part is not repeated here.

Example 5

FIG. 5 is a schematic diagram of an electronic device according to Embodiment 5 of the present application. As shown in FIG. 5, the electronic device 5 of this embodiment includes a processor 50, a memory 51, and a computer program 52 stored in the above-described memory 51 and operable on the processor 50 described above. When the processor 50 executes the computer program 52 described above, the steps in the above various method embodiments are implemented, such as steps S101 to S104 shown in FIG. Alternatively, when the processor 50 executes the computer program 52, the functions of the modules/units in the above-described respective device embodiments, such as the functions of the modules 301 to 304 shown in FIG. 3, are implemented.

The above electronic device 5 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server. The above electronic device may include, but is not limited to, the processor 50 and the memory 51. It will be understood by those skilled in the art that FIG. 5 is only an example of the electronic device 5, and does not constitute a limitation on the electronic device 5, and may include more or less components than those illustrated, or combine some components, or different components. For example, the electronic device 5 described above may further include an input/output device, a network access device, a bus, and the like.

The processor 50 may be a central processing unit (CPU), or may be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor or any conventional processor or the like.

The memory 51 may be an internal storage unit of the electronic device 5, such as a hard disk or an internal memory of the electronic device 5. The memory 51 may be an external storage device of the electronic device 5, such as a plug-in hard disk equipped with the above-mentioned electronic device 5, a smart memory card (SMC), a Secure Digital (SD) card, and a flash memory. Flash card, etc. Further, the above-mentioned memory 51 may also include both an internal storage unit of the above-described electronic device 5 and an external storage device. The above memory 51 is used to store the above computer program and other programs and data required for the above electronic device. The above-described memory 51 can also be used to temporarily store data that has been output or is about to be output.

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the division of each functional unit and module described above is exemplified. In practical applications, the above functions may be assigned to different functional units as needed. The module is completed by dividing the internal structure of the above device into different functional units or modules to perform all or part of the functions described above. Each functional unit and module in the embodiment may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit, and the integrated unit may be hardware. Formal implementation can also be implemented in the form of software functional units. In addition, the specific names of the respective functional units and modules are only for the purpose of facilitating mutual differentiation, and are not intended to limit the scope of protection of the present application. For the specific working process of the unit and the module in the foregoing system, reference may be made to the corresponding process in the foregoing method embodiment, and details are not described herein again.

In the above embodiments, the descriptions of the various embodiments are different, and the parts that are not detailed or described in a certain embodiment can be referred to the related descriptions of other embodiments.

Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the solution. A person skilled in the art can use different methods to implement the described functions for each particular application, but such implementation should not be considered to be beyond the scope of the present application.

In the embodiments provided by the present application, it should be understood that the disclosed apparatus/electronic device and method may be implemented in other manners. For example, the device/electronic device embodiments described above are merely illustrative. For example, the division of the above modules or units is only a logical functional division, and may be implemented in another manner, such as multiple units or Components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in electrical, mechanical or other form.

The units described above as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

The above-described integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application implements all or part of the processes in the above embodiments, and may also be completed by a computer program to instruct related hardware. The computer program may be stored in a computer readable storage medium. The steps of the various method embodiments described above may be implemented when executed by a processor. Wherein, the above computer program comprises computer program code, and the computer program code may be in the form of source code, object code form, executable file or some intermediate form. The computer readable medium may include any entity or device capable of carrying the above computer program code, a recording medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a Read-Only Memory (ROM), a random Access memory (RAM, Random Access Memory), electrical carrier signals, telecommunications signals, and software distribution media. It should be noted that the contents of the above computer readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in jurisdictions. For example, in some jurisdictions, according to legislation and patent practice, computer readable media are not Includes electrical carrier signals and telecommunications signals.

The above embodiments are only used to explain the technical solutions of the present application, and are not limited thereto; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the foregoing embodiments can still be The technical solutions are modified, or some of the technical features are replaced by equivalents; and the modifications or substitutions do not deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the present disclosure. Within the scope of protection of the application.

Claims (10)

1. A brain age testing method based on magnetic resonance images, characterized in that it comprises:

   Obtaining a T1-weighted brain magnetic resonance image of the individual to be tested;

   Determining a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject based on the T1-weighted brain magnetic resonance image;

   Calculating a normalized volume of the brain structure, and calculating a brain atrophy value of the brain lobe based on the brain tissue segmentation map;

   Entering the normalized volume and the brain atrophy value into a brain age estimation model to obtain a brain age of the test subject, wherein the brain age estimation model is based on a normalized volume of the training sample individual and a brain atrophy value And brain age training is available.
2. The cerebral age testing method according to claim 1, wherein the determining a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject based on the T1-weighted brain magnetic resonance image comprises:

   Registering the T1-weighted brain magnetic resonance image with a preset brain template library to obtain a brain structure, a brain lobe, and a brain tissue probability map corresponding to the T1-weighted brain magnetic resonance image;

   Using the brain tissue probability map corresponding to the T1-weighted brain magnetic resonance image as a priori knowledge, the brain tissue segmentation is performed on the T1-weighted brain magnetic resonance image, and the brain tissue segmentation corresponding to the T1-weighted brain magnetic resonance image is obtained. Map

   The brain template library includes: a T1-weighted brain magnetic resonance image of two or more different brains, and a brain structure, a brain lobe, and a brain tissue probability map corresponding to each T1-weighted brain magnetic resonance image, respectively.
3. The brain age testing method according to claim 1, wherein the calculating a brain atrophy value of the brain lobe based on the brain tissue segmentation map comprises:

   Determining a gray matter volume, a white matter volume, and a cerebrospinal fluid volume in the brain lobe based on the brain tissue segmentation map;

   Calculating a brain atrophy value of the brain lobe according to a brain atrophy calculation formula, wherein the brain atrophy calculation formula is:

   
4. The brain age testing method according to claim 1, wherein the brain age testing method further comprises:

   The normalized volume of the training sample individual and the brain atrophy value are used as independent variables, and the brain age of the training sample individual is used as the dependent variable, and the brain age estimation model is established by applying a linear support vector machine.
5. The brain age testing method according to any one of claims 1 to 4, wherein the determining a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject based on the T1-weighted brain magnetic resonance image Previously, it also included:

   Performing preprocessing on the T1-weighted brain magnetic resonance image of the individual to be tested, the pre-processing including one or more of the following: noise reduction, de-biasing field, pixel range normalization;

   Determining, according to the T1-weighted brain magnetic resonance image, a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject:

   A brain structure, a brain lobe, and a brain tissue segmentation map of the test subject are determined based on the T1-weighted brain magnetic resonance image obtained after the pre-processing.
6. A brain age testing device based on magnetic resonance images, characterized in that it comprises:

   An image acquisition unit, configured to acquire a T1-weighted brain magnetic resonance image of the individual to be tested;

   An image analyzing unit, configured to determine a brain structure, a brain lobe, and a brain tissue segmentation map of the test subject based on the T1-weighted brain magnetic resonance image;

   a parameter calculation unit, configured to calculate a normalized volume of the brain structure, and calculate a brain atrophy value of the brain lobe based on the brain tissue segmentation map;

   a brain age acquisition unit, configured to input the normalized volume and the brain atrophy value into a brain age estimation model, and obtain a brain age of the test subject, wherein the brain age estimation model is based on the individuality of the training sample individual A volume, brain atrophy value and brain age training are obtained.
7. The cerebral age testing device according to claim 6, wherein the image analyzing unit is specifically configured to:

   Registering the T1-weighted brain magnetic resonance image with a preset brain template library to obtain a brain structure, a brain lobe, and a brain tissue probability map corresponding to the T1-weighted brain magnetic resonance image;

   Using the brain tissue probability map corresponding to the T1-weighted brain magnetic resonance image as a priori knowledge, the brain tissue segmentation is performed on the T1-weighted brain magnetic resonance image, and the brain tissue segmentation corresponding to the T1-weighted brain magnetic resonance image is obtained. Map

   The brain template library includes: a T1-weighted brain magnetic resonance image of two or more different brains, and a brain structure, a brain lobe, and a brain tissue probability map corresponding to each T1-weighted brain magnetic resonance image, respectively.
8. The cerebral age testing device according to claim 6, wherein the parameter calculating unit is specifically configured to:

   Calculating a normalized volume of the brain structure, and determining a gray matter volume, a white matter volume, and a cerebrospinal fluid volume in the brain lobe based on the brain tissue segmentation map;

   Calculating a brain atrophy value of the brain lobe according to a brain atrophy calculation formula, wherein the brain atrophy calculation formula is:

   
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor executes the computer program as claimed in claim 1 5 The steps of any of the methods described.
10. A computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the method of any one of claims 1 to 5.

Images