CN113947577B - Method, system, device, processor and storage medium for realizing brain image feature normalization processing based on healthy population distribution - Google Patents
Method, system, device, processor and storage medium for realizing brain image feature normalization processing based on healthy population distribution Download PDFInfo
- Publication number
- CN113947577B CN113947577B CN202111208007.9A CN202111208007A CN113947577B CN 113947577 B CN113947577 B CN 113947577B CN 202111208007 A CN202111208007 A CN 202111208007A CN 113947577 B CN113947577 B CN 113947577B
- Authority
- CN
- China
- Prior art keywords
- distribution
- brain
- healthy
- characteristic
- covariate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 210000004556 brain Anatomy 0.000 title claims abstract description 84
- 238000010606 normalization Methods 0.000 title claims abstract description 55
- 238000012545 processing Methods 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000011218 segmentation Effects 0.000 claims abstract description 14
- 230000000877 morphologic effect Effects 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims description 45
- 238000012937 correction Methods 0.000 claims description 9
- 239000000126 substance Substances 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000007689 inspection Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 230000002452 interceptive effect Effects 0.000 claims description 4
- 201000010099 disease Diseases 0.000 abstract description 11
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 abstract description 11
- 238000004458 analytical method Methods 0.000 abstract description 4
- 208000020016 psychiatric disease Diseases 0.000 description 9
- 238000002610 neuroimaging Methods 0.000 description 6
- 238000012360 testing method Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 208000024827 Alzheimer disease Diseases 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000005821 brain abnormality Effects 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 206010015037 epilepsy Diseases 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G06T5/77—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20076—Probabilistic image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Abstract
The invention relates to a method for realizing brain image characteristic normalization processing based on healthy population distribution, which comprises the following steps: obtaining the morphological characteristics of the healthy tested brain through brain region segmentation and cortex reconstruction; obtaining the characteristics after the influence of the covariates is removed through a covariate regression model; obtaining a characteristic distribution probability density function through kernel density estimation; and obtaining quantiles in the distribution of healthy people under the characteristic dimension as a normalization result. The invention also relates to a device for realizing the brain image characteristic normalization processing based on the healthy population distribution, a processor and a computer readable storage medium thereof. The method, the system, the device, the processor and the computer readable storage medium for realizing the brain image feature normalization processing based on the healthy population distribution are suitable for the feature normalization processing in a real-time feature analysis system, do not depend on the distribution of data with the same sample label, and can realize the feature normalization of various diseases and unknown disease types.
Description
Technical Field
The invention relates to the field of magnetic resonance image data processing, in particular to the field of mental disorder diagnosis and treatment, and specifically relates to a method, a system, a device, a processor and a computer readable storage medium for realizing brain image feature normalization processing based on healthy population distribution.
Background
The relevance of mental disorders and brain abnormalities is well known for a long time, and neuroimaging is used as a noninvasive means for accurately measuring brain structures and functions, so that the neuroimaging can play a greater role in diagnosis and treatment of mental disorders, and provides quantitative objective indexes for clinical diagnosis and curative effect prediction of mental disorders, but a technology for objectively evaluating brain structures and functional abnormalities through neuroimaging is not available at present.
In evaluating the imaging characteristics extracted from neuroimaging data, there are differences in age, sex, and whole brain volume between individual patients, which all affect neuroimaging characteristics, but are not disease-related. Therefore, it is necessary to develop a technique for evaluating abnormality of neuroimaging features based on the control of the above variations.
The preprocessing aiming at the features in the existing method mainly comprises covariate regression, 0-1 standardization, z-fraction standardization and the like, and such operations need to utilize all tested image features included in the research, so that the method is not suitable for processing the unknown tested features in a real-time report generation system; meanwhile, the image feature distribution of different disease groups has great difference, so the feature processing operation can affect the quality of the features to a certain extent under the condition of mental diseases or unknown disease types.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a method, a system, a device, a processor and a computer readable storage medium for realizing brain image characteristic normalization processing based on healthy population distribution, which have the advantages of high accuracy, high precision and wider application range.
In order to achieve the above object, the method, system, apparatus, processor and computer readable storage medium for implementing brain image feature normalization processing based on healthy people distribution of the present invention are as follows:
the method for realizing the brain image characteristic normalization processing based on the healthy population distribution is mainly characterized by comprising the following steps of:
(1) collecting magnetic resonance brain images of healthy people, and acquiring the morphological characteristics of the brain of a healthy subject in a brain region segmentation and cortex reconstruction mode;
(2) carrying out covariate correction on the characteristics of the inspection image through a covariate regression model to obtain the characteristics without the influence of the covariate;
(3) obtaining a characteristic distribution probability density function of healthy people on characteristic dimensions through kernel density estimation;
(4) and obtaining quantiles in the distribution of healthy people under the characteristic dimension through a characteristic distribution probability density function, and taking the quantiles as the value results after the normalization of the characteristic dimension.
Preferably, the step (2) is specifically:
(2.1) for each characteristic dimension, establishing a covariate regression model by taking the age, the sex, the whole brain volume and the interactive items among indexes as covariates;
and (2.2) obtaining the characteristics after the influence of the covariates is removed through a covariate regression model.
Preferably, the characteristics obtained in step (2) after the influence of the covariate is removed are specifically:
the characteristic after removing the influence of the covariates is obtained according to the following formula:
wherein the content of the first and second substances,
for all the characteristics of healthy people on a certain characteristic dimension obtained after brain region segmentation and cortex reconstruction,
the dimensional feature prediction value is the value influenced by covariates in the original feature.
Preferably, the predicted value of the dimension characteristic is obtained by a covariate regression model, specifically:
obtaining a dimension characteristic predicted value according to the following formula:
wherein the content of the first and second substances,
respectively the age, sex and whole brain volume of the healthy subject population, k1～k8Each represents a regression coefficient.
Preferably, the obtaining of the feature distribution probability density function in step (3) is specifically:
a feature distribution probability density function is obtained according to the following formula:
wherein, K () is a kernel function adopting standard normal distribution, and h is a bandwidth.
Preferably, the bandwidth is obtained by optimizing an average integral squared error, or is set based on an adaptive bandwidth estimated from sample point to sample point.
Preferably, the quantile in the distribution of the healthy population under the characteristic dimension obtained in the step (4) is specifically:
the quantile in the distribution of healthy people in the characteristic dimension is found according to the following formula:
wherein rho is in a range of 0 to 1,
are features corrected by covariates.
The system for realizing brain image feature normalization processing based on healthy crowd distribution is mainly characterized by comprising the following components:
the brain morphological characteristic acquisition functional module is used for collecting magnetic resonance brain images of healthy people and acquiring the brain morphological characteristics of the healthy people to be tested in a brain region segmentation and cortex reconstruction mode;
the covariate correction function module is used for carrying out covariate correction on the characteristics of the inspection image through the covariate regression model to obtain the characteristics without the influence of the covariate;
the kernel density estimation function module is used for obtaining a feature distribution probability density function of the healthy population on the feature dimension through kernel density estimation;
and the quantile obtaining functional module is used for obtaining the quantile in the distribution of the healthy population under the characteristic dimension through a characteristic distribution probability density function and taking the quantile as a value result after the characteristic dimension is normalized.
The device for realizing brain image feature normalization processing based on healthy crowd distribution is mainly characterized by comprising the following steps:
a processor configured to execute computer-executable instructions;
and the memory stores one or more computer-executable instructions, and when the computer-executable instructions are executed by the processor, the memory realizes the steps of the method for realizing the brain image characteristic normalization processing based on the healthy population distribution.
The processor for implementing the brain image feature normalization processing based on the healthy crowd distribution is mainly characterized in that the processor is configured to execute computer-executable instructions, and when the computer-executable instructions are executed by the processor, the steps of the method for implementing the brain image feature normalization processing based on the healthy crowd distribution are implemented.
The computer-readable storage medium is mainly characterized by storing a computer program thereon, wherein the computer program can be executed by a processor to realize the steps of the method for realizing the brain image feature normalization processing based on the healthy people distribution.
The method, the system, the device, the processor and the computer readable storage medium for realizing the brain image feature normalization processing based on the healthy crowd distribution are suitable for the feature normalization processing in a real-time feature analysis system, do not depend on the distribution of data with the same label of a sample needing the normalization processing, and can realize the feature normalization under the conditions of various diseases and unknown disease types.
Drawings
Fig. 1 is a flowchart of a method for implementing brain image feature normalization processing based on healthy population distribution according to the present invention.
Detailed Description
In order that the technical contents of the present invention can be more clearly described, the present invention will be further described with reference to specific embodiments. The invention discloses a method for realizing brain image characteristic normalization processing based on healthy crowd distribution, which comprises the following steps:
(1) collecting magnetic resonance brain images of healthy people, and acquiring the morphological characteristics of the brain of a healthy subject in a brain region segmentation and cortex reconstruction mode;
(2) carrying out covariate correction on the characteristics of the inspection image through a covariate regression model to obtain the characteristics without the influence of the covariate;
(3) obtaining a characteristic distribution probability density function of healthy people on characteristic dimensions through kernel density estimation;
(4) and obtaining quantiles in the distribution of healthy people under the characteristic dimension through a characteristic distribution probability density function, and taking the quantiles as the value results after the normalization of the characteristic dimension.
As a preferred embodiment of the present invention, the step (2) specifically comprises:
(2.1) for each characteristic dimension, establishing a covariate regression model by taking the age, the sex, the whole brain volume and the interactive items among indexes as covariates;
And (2.2) obtaining the characteristics after the influence of the covariates is removed through a covariate regression model.
As a preferred embodiment of the present invention, the characteristics obtained in step (2) after removing the influence of the covariate are specifically:
the characteristic after removing the influence of the covariates is obtained according to the following formula:
wherein, the first and the second end of the pipe are connected with each other,
for all the characteristics of healthy people on a certain characteristic dimension obtained after brain region segmentation and cortex reconstruction,
the dimensional feature prediction value is the value influenced by covariates in the original feature.
As a preferred embodiment of the present invention, the dimensional feature prediction value is obtained by a covariate regression model, specifically:
obtaining a dimension characteristic predicted value according to the following formula:
wherein the content of the first and second substances,
respectively the age, sex and whole brain volume of the healthy subject population, k1～k8Each represents a regression coefficient.
As a preferred embodiment of the present invention, the step (3) obtains a feature distribution probability density function, specifically:
a feature distribution probability density function is obtained according to the following formula:
wherein, K () is a kernel function adopting standard normal distribution, and h is a bandwidth.
Preferably, the bandwidth is obtained by optimizing the mean integral squared error, or is set based on an adaptive bandwidth estimated from sample point to sample point.
As a preferred embodiment of the present invention, the quantile in the distribution of healthy people in the characteristic dimension obtained in step (4) is specifically:
the quantile in the distribution of healthy people in the characteristic dimension is found according to the following formula:
wherein rho is in a range of 0 to 1,
are features corrected by covariates.
The invention relates to a system for realizing brain image characteristic normalization processing based on healthy crowd distribution, which comprises:
the brain morphological characteristic acquisition functional module is used for collecting magnetic resonance brain images of healthy people and acquiring the brain morphological characteristics of the healthy people to be tested in a brain region segmentation and cortex reconstruction mode;
the covariate correction function module is used for correcting the covariate of the characteristics of the image to be inspected through the covariate regression model to obtain the characteristics without the influence of the covariate;
the kernel density estimation function module is used for obtaining a feature distribution probability density function of healthy people on feature dimensions through kernel density estimation;
and the quantile obtaining functional module is used for obtaining quantiles in the distribution of healthy people under the characteristic dimension through a characteristic distribution probability density function and taking the quantiles as the value results after the characteristic dimension is normalized.
The invention relates to a device for realizing brain image characteristic normalization processing based on healthy crowd distribution, wherein the device comprises:
a processor configured to execute computer-executable instructions;
and the memory stores one or more computer-executable instructions, and when the computer-executable instructions are executed by the processor, the method for realizing the brain image characteristic normalization processing based on the healthy population distribution realizes the steps.
The processor for implementing brain image feature normalization based on healthy population distribution of the invention is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the steps of the method for implementing brain image feature normalization based on healthy population distribution are implemented.
The computer readable storage medium of the present invention has a computer program stored thereon, and the computer program can be executed by a processor to implement the steps of the above method for implementing the brain image feature normalization processing based on the healthy population distribution.
In a specific embodiment of the invention, a feature distribution probability density function of healthy people on the feature dimension is established through a kernel density estimation algorithm based on a feature data set of healthy people, and the quantile of the value of the feature of the inspected object on the healthy people probability density function is taken as the normalized result of the feature of the inspected object.
Taking the normalization of brain image characteristics as an example, in order to realize the normalization method based on the healthy population characteristic distribution, the invention collects a large number of age and sex ratio balanced magnetic resonance brain images of healthy subjects based on a public data set and data collected in a hospital, and obtains all brain morphological characteristics of the healthy subjects, such as the volume, the cortex thickness and the like of each brain area by utilizing brain area segmentation and cortex reconstruction algorithms. Aiming at each feature dimension, firstly, establishing a regression equation by taking the age, the sex, the whole brain volume and the interaction terms among the three indexes as covariates, and removing the influence of the covariates on the original features. Specifically, the characteristic δ after removing the influence of the covariates on a certain characteristic dimension is;
wherein, the first and the second end of the pipe are connected with each other,
for all the characteristics of healthy tested population on a certain characteristic dimension obtained after brain region segmentation and cortex reconstruction,
the predicted value of the dimension feature is obtained by the following covariate regression model:
wherein, the first and the second end of the pipe are connected with each other,
respectively the age, sex and whole brain volume of the healthy subject population, k1-k8Each represents a regression coefficient.
Establishing regression prediction model of covariates (including age, sex and whole brain volume) of the healthy tested population to image characteristics, and obtaining image characteristic prediction value by the model 
I.e. to characterize the covariate-affected part of the original feature, so that the original feature is used
Feature prediction value obtained by subtracting regression equation
The obtained delta is the removal covariateThe affected features.
And then, obtaining a feature distribution probability density function of the healthy population on the feature dimension by using a kernel density estimation algorithm. Specifically, for all n healthy subjects acquired, the feature in the current feature dimension is δ1、δ2……δnI.e. delta1、δ2……δnAre all independent and identically distributed sample points. And if the probability density function of the characteristic distribution is recorded as f, probability density estimation is carried out by utilizing the kernel function, and the obtained estimation of f is as follows:
wherein, K (-) is a non-negative function with an integral of 1 and according with the property of the probability density function, called as a kernel function, and generally adopts standard normal distribution; h > 0 is a smoothing parameter, called bandwidth, which can be obtained by optimizing the mean integrated squared error, and can also be set as an adaptive bandwidth setting based on a sample-point-by-sample-point estimation.
For the new test, the characteristics after regression covariates were obtained by the same processing method as described above
For each feature dimension, utilizing a feature distribution probability density function of healthy people under the feature dimension 
Obtaining the quantile rho of the tested person in the distribution of the healthy people under the characteristic dimension:
according to the definition of the probability density function, the value range of rho is 0-1, and the rho is used as the normalized value of the tested object under the characteristic dimension.
Here, the
The image characteristics of the examined patient which are not seen by the system are characteristics obtained by a covariate influence elimination operation, the delta in the formula is the characteristics obtained by the covariate influence elimination operation aiming at the existing healthy tested population, the two objects are different, and the two parameters are different.
Features targeted by the present invention include, but are not limited to, magnetic resonance features, and may also include CT image features, PET image features, and other types of features, such as blood routine, genomic, quantitative pathology analysis data, and the like.
Diseases to which the present invention is directed include, but are not limited to, psychiatric disorders, but also other diseases such as alzheimer's disease, epilepsy and the like.
In the method, a kernel function is used for probability density estimation, and the selection mode of the bandwidth h comprises but is not limited to obtaining through optimized mean integral square error and self-adaptive bandwidth setting based on sample point-by-sample point estimation, and also comprises the bandwidth setting modes such as setting according to experience and the like.
In the specific embodiment of the invention, 3000 mental disease data and 300 healthy people data collected by the mental health center in Shanghai are used for respectively extracting image features from original magnetic resonance data, healthy people distribution is used for carrying out normalization processing on the mental disease population features, the features obtained after the normalization processing are used for training and testing a support vector machine, and cross validation test results show that the support vector machine model obtained by training the features after the normalization processing is higher in classification accuracy of mental diseases than the model obtained by feature training by using other normalization processing (including z-fraction normalization and 0-1 normalization).
For a specific implementation scheme of this embodiment, reference may be made to relevant descriptions in the foregoing embodiments, which are not described herein again.
It is understood that the same or similar parts in the above embodiments may be mutually referred to, and the same or similar contents in other embodiments may be referred to for the contents which are not described in detail in some embodiments.
It should be noted that, in the description of the present invention, the terms "first", "second", etc. are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. In addition, in the description of the present invention, the meaning of "a plurality" means at least two unless otherwise specified.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps in the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution device. For example, if implemented in hardware, as in another embodiment, any one or combination of the following technologies, which are well known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, and the corresponding program may be stored in a computer readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a separate product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The method, the system, the device, the processor and the computer readable storage medium for realizing the brain image feature normalization processing based on the healthy crowd distribution are suitable for the feature normalization processing in a real-time feature analysis system, do not depend on the distribution of data with the same label of a sample needing the normalization processing, and can realize the feature normalization under the conditions of various diseases and unknown disease types.
In this specification, the invention has been described with reference to specific embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
Claims (8)
1. A method for realizing brain image feature normalization processing based on healthy population distribution is characterized by comprising the following steps:
(1) collecting magnetic resonance brain images of healthy people, and acquiring the morphological characteristics of the brain of a healthy subject in a brain region segmentation and cortex reconstruction mode;
(2) carrying out covariate correction on the characteristics of the inspection image through a covariate regression model to obtain the characteristics without the influence of the covariate;
(3) Obtaining a feature distribution probability density function of healthy people on feature dimensions through kernel density estimation;
(4) obtaining quantiles in the distribution of healthy people under the characteristic dimension through a characteristic distribution probability density function, and taking the quantiles as a value result after normalization of the characteristic dimension;
the step (2) is specifically as follows:
(2.1) establishing a covariate regression model for each characteristic dimension by taking the age, the sex, the whole brain volume and the interactive items among indexes as covariates;
(2.2) obtaining the characteristics without the influence of the covariates through a covariate regression model;
the characteristics obtained in the step (2) after the influence of the covariates is removed are specifically as follows:
the characteristic after removing the influence of the covariate is obtained according to the following formula:
wherein the content of the first and second substances,
for all the characteristics of healthy people on a certain characteristic dimension obtained after brain region segmentation and cortex reconstruction,
the dimensional feature prediction value is a value influenced by covariates in the original feature;
the dimension characteristic predicted value is obtained through a covariate regression model, and specifically comprises the following steps:
obtaining the feature prediction value of the covariate on each feature dimension according to the following formula:
wherein the content of the first and second substances,
epsilon is the age, sex and total brain volume of healthy subject population, k 1～k8Each represents a regression coefficient.
2. The method for realizing brain image feature normalization processing based on healthy people distribution according to claim 1, wherein the feature distribution probability density function obtained in the step (3) is specifically:
a feature distribution probability density function is obtained according to the following formula:
wherein K () is a kernel function adopting standard normal distribution, h is a bandwidth, delta is a characteristic dimension, n is the total number of healthy subjects, and deltaiThe feature tested on the current feature dimension for the ith health, including δ1、δ2……δn。
3. The method of claim 2, wherein the bandwidth is obtained by optimizing mean-integral-squared error or adaptive bandwidth setting based on sample-point-by-sample-point estimation.
4. The method for realizing brain image feature normalization processing based on healthy people distribution according to claim 1, wherein the quantile in the healthy people distribution under the feature dimension obtained in the step (4) specifically comprises:
the quantile in the distribution of healthy people in the characteristic dimension is found according to the following formula:
wherein rho is in a value range of 0-1, t is an integral variable and is a continuous value, and the value is 
In the middle of the above-mentioned period,
the image features of the examined patient that are not seen by the system.
5. A system for realizing brain image feature normalization processing based on healthy crowd distribution is characterized by comprising:
the brain morphological characteristic acquisition functional module is used for collecting magnetic resonance brain images of healthy people and acquiring the morphological characteristics of the healthy tested brain through brain region segmentation and cortex reconstruction;
the covariate correction function module is used for carrying out covariate correction on the characteristics of the inspection image through the covariate regression model to obtain the characteristics without the influence of the covariate;
the kernel density estimation function module is used for obtaining a feature distribution probability density function of the healthy population on the feature dimension through kernel density estimation;
the quantile obtaining function module is used for obtaining quantiles in the distribution of healthy people under the characteristic dimension through a characteristic distribution probability density function and taking the quantiles as a value result after the characteristic dimension is normalized;
the system also comprises the following processing procedures:
(1) collecting magnetic resonance brain images of healthy people, and acquiring the morphological characteristics of the brain of a healthy subject in a brain region segmentation and cortex reconstruction mode;
(2) Carrying out covariate correction on the characteristics of the inspection image through a covariate regression model to obtain the characteristics without the influence of the covariate;
(3) obtaining a characteristic distribution probability density function of healthy people on characteristic dimensions through kernel density estimation;
(4) obtaining quantiles in the distribution of healthy people under the characteristic dimension through a characteristic distribution probability density function, and taking the quantiles as the value results after the normalization of the characteristic dimension;
the treatment process (2) is specifically as follows:
(2.1) for each characteristic dimension, establishing a covariate regression model by taking the age, the sex, the whole brain volume and the interactive items among indexes as covariates;
(2.2) obtaining the characteristics without the influence of the covariates through a covariate regression model;
the characteristic of removing the influence of the covariates is obtained in the processing process (2), and specifically comprises the following steps:
the characteristic after removing the influence of the covariate is obtained according to the following formula:
wherein the content of the first and second substances,
for all the characteristics of healthy people on a certain characteristic dimension obtained after brain region segmentation and cortex reconstruction,
the dimensional feature prediction value is a value influenced by covariates in the original feature;
the dimension characteristic predicted value is obtained through a covariate regression model, and specifically comprises the following steps:
Obtaining the characteristic predicted value of the covariate on each characteristic dimension according to the following formula:
wherein the content of the first and second substances,
epsilon is the age, sex and total brain volume of healthy subject population, k1～k8Each represents a regression coefficient.
6. An apparatus for implementing a brain image feature normalization process based on healthy population distribution, the apparatus comprising:
a processor configured to execute computer-executable instructions;
a memory storing one or more computer-executable instructions that, when executed by the processor, perform the steps of the method of performing brain image feature normalization based on healthy population distribution of any one of claims 1 to 4.
7. A processor for implementing brain image feature normalization based on healthy population distribution, wherein the processor is configured to execute computer-executable instructions, and when the computer-executable instructions are executed by the processor, the steps of the method for implementing brain image feature normalization based on healthy population distribution according to any one of claims 1 to 4 are implemented.
8. A computer-readable storage medium, having a computer program stored thereon, the computer program being executable by a processor to perform the steps of the method for performing the brain image feature normalization process based on the distribution of healthy people according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111208007.9A CN113947577B (en) | 2021-10-18 | 2021-10-18 | Method, system, device, processor and storage medium for realizing brain image feature normalization processing based on healthy population distribution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111208007.9A CN113947577B (en) | 2021-10-18 | 2021-10-18 | Method, system, device, processor and storage medium for realizing brain image feature normalization processing based on healthy population distribution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113947577A CN113947577A (en) | 2022-01-18 |
| CN113947577B true CN113947577B (en) | 2022-06-28 |
Family
ID=79331099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111208007.9A Active CN113947577B (en) | 2021-10-18 | 2021-10-18 | Method, system, device, processor and storage medium for realizing brain image feature normalization processing based on healthy population distribution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113947577B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108288070A (en) * | 2018-01-12 | 2018-07-17 | 迈格生命科技（深圳）有限公司 | A kind of nerve fingerprint extraction sorting technique and system |
| CN110428396A (en) * | 2019-06-25 | 2019-11-08 | 郑州大学第一附属医院 | A kind of feature vectors dimensional down method and system based on CT images |
| CN111000571A (en) * | 2019-12-26 | 2020-04-14 | 上海市精神卫生中心（上海市心理咨询培训中心） | Method for predicting risk of post-traumatic psychological disease, related device and storage device |
| CN111488934A (en) * | 2020-04-13 | 2020-08-04 | 脑结构(南京)科技有限公司 | Brain image data processing method, storage medium, computer device and apparatus |
| CN113096127A (en) * | 2021-06-04 | 2021-07-09 | 壹药网科技(上海)股份有限公司 | System and method for generating brain network evolution model |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9471978B2 (en) * | 2004-10-04 | 2016-10-18 | Banner Health | Methodologies linking patterns from multi-modality datasets |
| US10586618B2 (en) * | 2014-05-07 | 2020-03-10 | Lifetrack Medical Systems Private Ltd. | Characterizing states of subject |
| EP3836157A1 (en) * | 2019-12-12 | 2021-06-16 | Siemens Healthcare GmbH | Method for obtaining disease-related clinical information |
-
2021
- 2021-10-18 CN CN202111208007.9A patent/CN113947577B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108288070A (en) * | 2018-01-12 | 2018-07-17 | 迈格生命科技（深圳）有限公司 | A kind of nerve fingerprint extraction sorting technique and system |
| CN110428396A (en) * | 2019-06-25 | 2019-11-08 | 郑州大学第一附属医院 | A kind of feature vectors dimensional down method and system based on CT images |
| CN111000571A (en) * | 2019-12-26 | 2020-04-14 | 上海市精神卫生中心（上海市心理咨询培训中心） | Method for predicting risk of post-traumatic psychological disease, related device and storage device |
| CN111488934A (en) * | 2020-04-13 | 2020-08-04 | 脑结构(南京)科技有限公司 | Brain image data processing method, storage medium, computer device and apparatus |
| CN113096127A (en) * | 2021-06-04 | 2021-07-09 | 壹药网科技(上海)股份有限公司 | System and method for generating brain network evolution model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113947577A (en) | 2022-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Oh et al. | Identifying schizophrenia using structural MRI with a deep learning algorithm | |
| CN109447183B (en) | Prediction model training method, device, equipment and medium | |
| Wee et al. | Sparse temporally dynamic resting-state functional connectivity networks for early MCI identification | |
| O’Donoghue et al. | Anatomical dysconnectivity in bipolar disorder compared with schizophrenia: A selective review of structural network analyses using diffusion MRI | |
| Misra et al. | Baseline and longitudinal patterns of brain atrophy in MCI patients, and their use in prediction of short-term conversion to AD: results from ADNI | |
| Dickinson et al. | Multivariate neural connectivity patterns in early infancy predict later autism symptoms | |
| Mahanand et al. | Identification of brain regions responsible for Alzheimer’s disease using a Self-adaptive Resource Allocation Network | |
| CN102665565B (en) | Medical image-processing apparatus | |
| US20110077503A1 (en) | Automatic MRI Quantification of Structural Body Abnormalities | |
| CN110840468B (en) | Autism risk assessment method and device, terminal device and storage medium | |
| Kim et al. | Multi-resolution statistical analysis of brain connectivity graphs in preclinical Alzheimer's disease | |
| Segal et al. | Regional, circuit and network heterogeneity of brain abnormalities in psychiatric disorders | |
| WO2020075737A1 (en) | Brain functional connectivity correlation value adjustment method, brain functional connectivity correlation value adjustment system, brain activity classifier harmonization method, brain activity classifier harmonization system, and brain activity biomarker system | |
| Cai et al. | Dynamic graph theoretical analysis of functional connectivity in Parkinson's disease: The importance of Fiedler value | |
| JP2020062369A (en) | Brain functional connectivity correlation value adjustment method, brain functional connectivity correlation value adjustment system, brain activity classifier harmonization method, brain activity classifier harmonization system, and brain activity biomarker system | |
| Dunås et al. | Multimodal image analysis of apparent brain age identifies physical fitness as predictor of brain maintenance | |
| CN112348785A (en) | Epileptic focus positioning method and system | |
| Lopez et al. | HAPPILEE: HAPPE In Low Electrode Electroencephalography, a standardized pre-processing software for lower density recordings | |
| CN112561848A (en) | Method, non-transitory computer-readable medium, and apparatus for regional brain age prediction | |
| Kruggel et al. | Analysis of longitudinal diffusion-weighted images in healthy and pathological aging: An ADNI study | |
| Joshi et al. | A novel technique for olfactory bulb measurements | |
| CN113362944A (en) | Machine learning-based prediction of functional dyspepsia and acupuncture curative effect | |
| CN113947577B (en) | Method, system, device, processor and storage medium for realizing brain image feature normalization processing based on healthy population distribution | |
| Erdaş et al. | A fully automated approach involving neuroimaging and deep learning for Parkinson’s disease detection and severity prediction | |
| US20210125332A1 (en) | Diagnosis assisting system, diagnosis assisting method and diagnosis assisting program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |