Brain glioma diagnosis system based on resting state brain function contralateral positioning

Abstract

The invention relates to the technical field of resting-state brain function diagnosis, in particular to a brain glioma diagnosis system based on resting-state brain function contralateral positioning, comprising: an acquisition module for acquiring brain magnetic resonance image data of a user; the conversion module is used for calling the user brain magnetic resonance image data acquired in the acquisition module and converting the user brain magnetic resonance image data into a waveform image; the segmentation module is used for receiving the waveform image obtained by converting the brain magnetic resonance image data of the user in the conversion module and segmenting the waveform image; according to the method, the brain glioma risk of the user is pre-estimated through analysis of the brain magnetic resonance image data of the user, manual work is replaced to a large extent, the diagnosis precision is not influenced by manual professional knowledge accumulation and experience any more, and the brain glioma risk of the patient can be pre-estimated accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of resting-state brain function diagnosis, and in particular to a brain glioma diagnosis system based on contralateral positioning of resting-state brain function. Background Art

[0002] Gliomas are the most common primary intracranial tumors that originate from glial cells. They have the characteristics of invasive growth, unclear boundaries, and are difficult to remove surgically. They can cause symptoms such as headaches, vomiting, epilepsy, and limb dysfunction, seriously affecting brain function. They can be divided into many types according to cell types, which pose a great threat to the life and health of patients.

[0003] The invention patent with application number 200310109069.X discloses a method for implementing a computer-aided diagnosis system for brain glioma based on data mining, which is characterized by comprising the following specific steps: digital processing of attributes: for each attribute in the case database, sort them according to different description forms and replace them with integers in corresponding order, that is, map the attribute value into an integer sequence, and indicate the non-existent attribute values; establish fuzzy membership for each digitized attribute: after the digitization process is completed, define a sensitivity factor r, and establish a membership relationship between different values ​​of the same attribute, that is, for a certain attribute A, its kth value v k and the jth value v j The membership value between Here, max A is the maximum value of attribute A, min A is the minimum value of attribute A, and r generally takes a value between 0 and 10.

[0004] The application aims to solve the following problems: "The decision tree method is used to discover diagnostic rules, but the generated decision tree has many nodes, which is difficult to understand and prone to overfitting problems; the rough set theory is not easy to handle real data and uncertain descriptions, and the number of rules generated is large, which is not concise and difficult to understand; the multilayer perceptron network based on the BP training learning algorithm can achieve high diagnostic accuracy but has very poor comprehensibility."

[0005] However, currently, glioma disease can be predicted and prevented by obtaining brain magnetic resonance images through regular physical examinations. However, this process requires doctors to observe the magnetic resonance images based on their own experience and then make a predictive diagnosis. This requires a high level of professional knowledge and experience from the doctor, which makes this method impossible to promote and use.

[0006] Therefore, a brain glioma diagnosis system based on contralateral positioning of resting brain function was proposed. Summary of the invention

[0007] In view of the above-mentioned shortcomings of the prior art, the present invention provides a brain glioma diagnosis system based on contralateral positioning of resting brain function, which solves the technical problems raised in the above-mentioned background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] The brain glioma diagnosis system based on the contralateral positioning of resting brain function includes:

[0010] An acquisition module is used to acquire the user's brain magnetic resonance image data; a conversion module is used to retrieve the user's brain magnetic resonance image data acquired in the acquisition module and convert the user's brain magnetic resonance image data into a waveform image; a segmentation module is used to receive the waveform image converted from the user's brain magnetic resonance image data in the conversion module and segment the waveform image; an evaluation module is used to obtain the waveform image that has completed the segmentation processing in the segmentation module and evaluate the risk of glioma based on the waveform image; a calibration module is used to receive the glioma risk evaluation result in the evaluation module and further calibrate the evaluation result; a diagnosis module is used to obtain the evaluation result calibrated by the calibration module and diagnose whether the user has a risk of glioma based on the evaluation result;

[0011] The acquisition module is interactively connected to the preprocessing and database at the lower level through a wireless network, the acquisition module is interactively connected to the conversion module at the lower level through a wireless network, the conversion module is interactively connected to the segmentation unit through a wireless network, the conversion module is interactively connected to the segmentation module and the evaluation module through a wireless network, the evaluation module is interactively connected to the calculation unit at the lower level through a wireless network, the evaluation module is interactively connected to the calibration module and the diagnosis module through a wireless network, and the calibration module is interactively connected to the calculation unit through a wireless network.

[0012] Furthermore, the user's brain magnetic resonance image data collected by the acquisition module is derived from a magnetic resonance imaging device, and the acquisition module is provided with submodules at a lower level, including:

[0013] A preprocessing unit, used for receiving the user's brain magnetic resonance image data acquired by the acquisition module, and performing smoothing processing on the user's brain magnetic resonance image data;

[0014] A database, used to obtain the user's brain magnetic resonance image data output after the smoothing process is completed in the preprocessing unit, and store the user's brain magnetic resonance image data;

[0015] Among them, the preprocessing unit is set with a user's brain magnetic resonance image data smoothing processing logic, and the user's brain magnetic resonance image data is preprocessed based on the user's brain magnetic resonance image data smoothing processing logic. Each group of user's brain magnetic resonance image data is marked with an acquisition timestamp. When the database stores the user's brain magnetic resonance image data, a storage sequence is generated based on the acquisition timestamp corresponding to the marked brain magnetic resonance image data of each user, and the user's brain magnetic resonance image data is sorted and stored based on the storage sequence.

[0016] Furthermore, the smoothing logic of the user's brain magnetic resonance image data in the preprocessing unit is expressed as:

[0017]

[0018] Where: S(x, y, z) is the gray value of the pixel in the brain magnetic resonance image data after smoothing; ω(i, j, k) is the gray difference weight; I(i, j, k) is the gray value of the pixel in the brain magnetic resonance image data;

[0019] in, σ is a sensitivity parameter; I(x, y, z) is the grayscale value of the pixel at the center of the brain magnetic resonance image data; the value of the sensitivity parameter σ is customized by the system user, and the value range of the sensitivity parameter σ is 0.1 to 1. Based on the above formula, each pixel in the user's brain magnetic resonance image data is smoothed to output the user's brain magnetic resonance image data after smoothing.

[0020] Furthermore, each time the conversion module is run, at least two or more sets of user brain magnetic resonance image data in the middle of the storage sequence are retrieved from the database. The conversion module is internally provided with submodules, including:

[0021] A segmentation unit, used to obtain the user's brain magnetic resonance image data retrieved by the conversion module, and symmetrically segment the user's brain magnetic resonance image data to obtain the user's left and right brain magnetic resonance image data;

[0022] After the segmentation unit performs segmentation processing on the user's brain magnetic resonance image data, the system end user customizes the ROI or performs waveform image conversion processing on the image data globally;

[0023] Among them, when the system user customizes ROI in the user's left and right brain magnetic resonance image data to perform waveform image conversion processing, the selected ROI in the user's left and right brain magnetic resonance image data is symmetrically distributed. After the user's brain magnetic resonance image data is segmented, the waveform images corresponding to the user's left and right brain magnetic resonance image data are converted separately based on Fourier transform.

[0024] Furthermore, the segmentation accuracy of the waveform image in the segmentation module is Segments represent the wave lines of a waveform;

[0025] L represents the total amount of wave lines that make up the waveform in the waveform image;

[0026] Among them, the segmentation accuracy of the waveform image in the segmentation module is customized by the system end user and obeys the segmentation logic that the higher the accuracy requirement for the risk assessment of glioma in the evaluation module, the smaller the segmentation accuracy, and vice versa.

[0027] Furthermore, after the evaluation module runs and obtains the waveform image that has completed the segmentation processing, the evaluation logic for the risk of glioma is expressed as follows:

[0028]

[0029] Where: r is the Pearson correlation coefficient between the waveform B and the reference waveform C corresponding to the left and right brain magnetic resonance image data; n is the set of waveform midpoints; y Bg is the value of the gth sampling point in waveform B; is the mean value of the sampling points of waveform B; y Cg is the value of the gth sampling point in the reference waveform C; is the mean value of the sampling points of the reference waveform C; r s is the comprehensive Pearson correlation coefficient of the waveform obtained by segmentation module; u is the total number of waveform segments; r′ v is the Pearson correlation coefficient between the waveform B corresponding to the left and right brain magnetic resonance image data and the vth waveform in the reference waveform C; ω v is the weight;

[0030] The waveform corresponding to the right brain magnetic resonance image data is denoted as A, and the reference waveform C is obtained by mirroring the waveform A. In formula (2), r′ v The calculation logic is the same as that of formula (1), and the weight ω v The value is subject to r′ v The longer the corresponding waveform length, the greater the weight ω v The larger the value, the more r and r s The larger the value, the higher the waveform symmetry, which means the lower the risk of brain glioma. Conversely, the lower the waveform symmetry, which means the greater the risk of brain glioma.

[0031] Furthermore, the evaluation module is provided with submodules at the lower level, including:

[0032] Nucleation unit, used to combine r and r s Comprehensively assess the risk of glioma;

[0033] The logic of comprehensively evaluating the risk of glioma in the calculation unit is:

[0034] R=r×a1+r s ×a2;

[0035] Where: R is the comprehensive assessment of the risk of glioma; a1 and a2 are weight factors;

[0036] Among them, weight factors a1 and a2 are customized by system users, and the sum of weight factors a1 and a2 is 1, a1 is always smaller than a2, and the value of weight factor a2 follows that the more waveform segments there are in the waveform image, the larger the value of weight factor a2, and vice versa.

[0037] Furthermore, the brain glioma risk assessment result received in the calibration module is the comprehensive assessment of the brain glioma risk R obtained by the operation of the nuclear calculation unit;

[0038] Each group of user brain magnetic resonance image data collected by the acquisition module obtains a group of R, and the corresponding results of each group of user brain magnetic resonance image data are recorded as R1, R2, R3, ...;

[0039] The logic of further calibration of the evaluation results by the calibration module is expressed as follows:

[0040]

[0041] Where: R NORR is the calibrated risk of glioma; max(R1, R2, ...) is the maximum value in the brackets; ε is the adjustment index; R1, R2, R3, ..., R(n-1), R(n) are the corresponding results of the brain magnetic resonance image data of each group of users;

[0042] Among them, the adjustment index ε takes the value of 1 or -1. If the numerator of the fraction where the adjustment index ε is located is greater than the denominator, the adjustment index ε takes the value of -1. Otherwise, the adjustment index ε takes the value of 1.

[0043] Furthermore, the diagnostic module operates the R calibrated in the calibration module. NORR Recording, synchronously set the judgment threshold, based on the judgment threshold and the recorded R NORR Compare and diagnose whether the user has the risk of developing glioma;

[0044] Among them, any two groups of R recorded continuously in the diagnosis module NORR , the previous group R NORR The next group R NORR When the value is large, the user is diagnosed with the risk of brain glioma.

[0045] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0046] The present invention provides a brain glioma diagnosis system based on resting-state brain function contralateral positioning. During operation, the system estimates the risk of a user suffering from brain glioma by analyzing the user's brain magnetic resonance image data, replacing manual work to a large extent. The diagnostic accuracy is no longer affected by the accumulation of manual professional knowledge and experience, and the patient's brain glioma risk can be estimated more accurately.

[0047] In the process of estimating the user's risk of glioma based on the user's brain MRI image data, the user's risk of glioma is comprehensively assessed through overall analysis of the user's brain MRI image and segmented analysis of the corresponding waveform images, so that the assessment results can more truly correspond to the actual situation shown by the user's brain MRI image data, providing doctors with more favorable data references and providing support for the diagnosis of the user's condition. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0049] Figure 1 This is a schematic diagram of the structure of a brain glioma diagnostic system based on contralateral positioning of resting brain function. DETAILED DESCRIPTION

[0050] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0051] The present invention will be further described below in conjunction with the embodiments.

[0052] Embodiment 1:

[0053] The brain glioma diagnosis system based on resting state brain function contralateral positioning of this embodiment is as follows: Figure 1 As shown, including:

[0054] An acquisition module, used for acquiring the user's brain magnetic resonance image data;

[0055] The user's brain magnetic resonance image data collected by the acquisition module comes from the magnetic resonance imaging device. The acquisition module is provided with submodules at the lower level, including:

[0056] A preprocessing unit, used for receiving the user's brain magnetic resonance image data acquired by the acquisition module, and performing smoothing processing on the user's brain magnetic resonance image data;

[0057] A database, used to obtain the user's brain magnetic resonance image data output after the smoothing process is completed in the preprocessing unit, and store the user's brain magnetic resonance image data;

[0058] Wherein, the preprocessing unit is set with a user brain magnetic resonance image data smoothing processing logic, and the user brain magnetic resonance image data is preprocessed based on the user brain magnetic resonance image data smoothing processing logic, each group of user brain magnetic resonance image data is marked with an acquisition timestamp, and when the database stores the user brain magnetic resonance image data, a storage sequence is generated based on the acquisition timestamp corresponding to the marked brain magnetic resonance image data of each user, and the user brain magnetic resonance image data is sorted and stored based on the storage sequence;

[0059] The smoothing logic of the user's brain magnetic resonance image data in the preprocessing unit is expressed as:

[0060]

[0061] Where: S(x, y, z) is the gray value of the pixel in the brain magnetic resonance image data after smoothing; ω(i, j, k) is the gray difference weight; I(i, j, k) is the gray value of the pixel in the brain magnetic resonance image data;

[0062] in, σ is a sensitivity parameter; I(x, y, z) is the gray value of the pixel in the center of the brain magnetic resonance image data; the value of the sensitivity parameter σ is customized by the system end user, and the value range of the sensitivity parameter σ is 0.1 to 1. Based on the above formula, each pixel in the user's brain magnetic resonance image data is smoothed to output the user's brain magnetic resonance image data after smoothing;

[0063] A conversion module, used to retrieve the user's brain magnetic resonance image data collected by the acquisition module, and convert the user's brain magnetic resonance image data into a waveform image;

[0064] Each time the conversion module is run, at least two or more sets of user brain magnetic resonance image data in the center of the storage sequence are retrieved from the database. The conversion module is internally provided with submodules, including:

[0065] A segmentation unit, used to obtain the user's brain magnetic resonance image data retrieved by the conversion module, and symmetrically segment the user's brain magnetic resonance image data to obtain the user's left and right brain magnetic resonance image data;

[0066] After the segmentation unit performs segmentation processing on the user's brain magnetic resonance image data, the system end user customizes the ROI or performs waveform image conversion processing on the image data globally;

[0067] When the system end user customizes ROI in the user's left and right brain magnetic resonance image data to perform waveform image conversion processing, the selected ROI in the user's left and right brain magnetic resonance image data is symmetrically distributed. After the user's brain magnetic resonance image data is segmented, the waveform images corresponding to the user's left and right brain magnetic resonance image data are converted respectively based on Fourier transform;

[0068] A segmentation module is used to receive the waveform image obtained by converting the user's brain magnetic resonance image data in the conversion module, and segment the waveform image;

[0069] An evaluation module is used to obtain the waveform image that has completed the segmentation process in the segmentation module, and evaluate the risk of brain glioma based on the waveform image;

[0070] After the evaluation module runs and obtains the waveform image that has completed segmentation processing, the evaluation logic for the risk of glioma is expressed as follows:

[0071]

[0072] Where: r is the Pearson correlation coefficient between the waveform B and the reference waveform C corresponding to the left and right brain magnetic resonance image data; n is the set of waveform midpoints; y Bg is the value of the gth sampling point in waveform B; is the mean value of the sampling points of waveform B; y Cg is the value of the gth sampling point in the reference waveform C; is the mean value of the sampling points of the reference waveform C; r s is the comprehensive Pearson correlation coefficient of the waveform obtained by segmentation module; u is the total number of waveform segments; r′ v is the Pearson correlation coefficient between the waveform B corresponding to the left and right brain magnetic resonance image data and the vth waveform in the reference waveform C; ω v is the weight;

[0073] The waveform corresponding to the right brain magnetic resonance image data is denoted as A, and the reference waveform C is obtained by mirroring the waveform A. In formula (2), r′ v The calculation logic is the same as that of formula (1), and the weight ω v The value is subject to r′ v The longer the corresponding waveform length, the greater the weight ω v The larger the value, the more r and r s The larger the value, the higher the waveform symmetry, which means the lower the risk of glioma. Conversely, the lower the waveform symmetry, which means the greater the risk of glioma.

[0074] The evaluation module is divided into submodules, including:

[0075] Nucleation unit, used to combine r and r s Comprehensively assess the risk of glioma;

[0076] The logic of comprehensive assessment of the risk of glioma in the nuclear calculation unit is:

[0077] R=r×a1+r s ×a2;

[0078] Where: R is the comprehensive assessment of the risk of glioma; a1 and a2 are weight factors;

[0079] Among them, the weight factors a1 and a2 are customized by the system end user, and the sum of the weight factors a1 and a2 is 1, a1 is always less than a2, and the value of the weight factor a2 follows that the more the number of waveform segments in the waveform image, the larger the value of the weight factor a2, and vice versa, the smaller the value of the weight factor a2;

[0080] A calibration module is used to receive the brain glioma risk assessment results from the assessment module and further calibrate the assessment results;

[0081] The brain glioma risk assessment result received in the calibration module is the comprehensive assessment of the brain glioma risk R obtained by the operation of the nuclear calculation unit;

[0082] Each group of user brain magnetic resonance image data collected by the acquisition module obtains a group of R, and the corresponding results of each group of user brain magnetic resonance image data are recorded as R1, R2, R3, ...;

[0083] The logic of the calibration module to further calibrate the evaluation results is expressed as:

[0084]

[0085] Where: R NORR is the calibrated risk of glioma; max(R1, R2, ...) is the maximum value in the brackets; ε is the adjustment index; R1, R2, R3, ..., R(n-1), R(n) are the corresponding results of the brain magnetic resonance image data of each group of users;

[0086] Among them, the adjustment index ε takes the value of 1 or -1. If the numerator of the fraction where the adjustment index ε is located is greater than the denominator, the adjustment index ε takes the value of -1. Otherwise, the adjustment index ε takes the value of 1.

[0087] The diagnosis module is used to obtain the evaluation results of the calibration module and diagnose whether the user has a risk of developing glioma based on the evaluation results;

[0088] The acquisition module is interactively connected to the preprocessing and database at its lower level through a wireless network, the acquisition module is interactively connected to the conversion module at its lower level through a wireless network, the conversion module is interactively connected to the segmentation unit through a wireless network, the conversion module is interactively connected to the segmentation module and the evaluation module through a wireless network, the evaluation module is interactively connected to the calculation unit at its lower level through a wireless network, the evaluation module is interactively connected to the calibration module and the diagnosis module through a wireless network, and the calibration module is interactively connected to the calculation unit through a wireless network.

[0089] In this embodiment, the acquisition module is operated to acquire the user's brain magnetic resonance image data, the preprocessing unit synchronously receives the user's brain magnetic resonance image data acquired by the acquisition module, and performs smoothing on the user's brain magnetic resonance image data. The database acquires the user's brain magnetic resonance image data output after the smoothing process in the preprocessing unit in real time, stores the user's brain magnetic resonance image data, and then the conversion module retrieves the user's brain magnetic resonance image data acquired in the acquisition module, and converts the user's brain magnetic resonance image data into a waveform image. The segmentation unit synchronously acquires the user's brain magnetic resonance image data retrieved by the conversion module, and symmetrically segments the user's brain magnetic resonance image data to obtain the user's left and right brain magnetic resonance image data. The segmentation module is post-operated to receive the waveform image obtained by converting the user's brain magnetic resonance image data in the conversion module, and segments the waveform image. The evaluation module further acquires the waveform image that has completed the segmentation process in the segmentation module, and evaluates the risk of glioma based on the waveform image. The nuclear calculation unit synchronously combines r and r s Comprehensively evaluate the risk of glioma. The calibration module further receives the glioma risk assessment results from the assessment module, further calibrates the assessment results, and finally obtains the assessment results calibrated by the calibration module through the diagnosis module. Based on the assessment results, it diagnoses whether the user has the risk of glioma.

[0090] Through the operation of the above system, predictive diagnostic opinions are provided for the diagnosis of brain glioma, which effectively realizes the prevention of such diseases and largely replaces the steps of manually observing brain magnetic resonance images, with lower demand for manual labor, making the diagnosis and predictive diagnosis of brain glioma faster and more accurate.

[0091] At the same time, through the operation of the brain magnetic resonance image processing module recorded by the system, it provides precision control and improvement for the predictive diagnosis of glioma, ensuring that the operation of the system can bring more effective reference information for the diagnosis of glioma.

[0092] Embodiment 2:

[0093] In terms of specific implementation, based on Example 1, this example refers to Figure 1The glioma diagnosis system based on the contralateral positioning of resting brain function in Example 1 is further specifically described as follows:

[0094] The segmentation accuracy of the waveform image in the segmentation module is The segments represent the wave lines of the waveform;

[0095] L represents the total amount of wave lines that make up the waveform in the waveform image;

[0096] Among them, the segmentation accuracy of the waveform image in the segmentation module is customized by the system end user and obeys the segmentation logic that the higher the accuracy requirement for the risk assessment of glioma in the evaluation module, the smaller the segmentation accuracy, and vice versa.

[0097] Through the above settings, the segmentation processing logic of the converted waveform image in the segmentation module is further limited, and this is used as a means of precision control of the system operation output results, which further constrains the system operation efficiency and provides convenience for different system usage scenarios.

[0098] like Figure 1 As shown, the diagnostic module runs the R calibrated in the calibration module. NORR Recording, synchronously set the judgment threshold, based on the judgment threshold and the recorded R NORR Compare and diagnose whether the user has the risk of developing glioma;

[0099] Among them, any two groups of R recorded continuously in the diagnosis module NORR , the previous group R NORR The next group R NORR When the value is large, the user is diagnosed with the risk of brain glioma.

[0100] Through the above settings, the operating logic and output content of the diagnostic module are further limited, so that the valuable data obtained from the system operation can be summarized and output, providing comprehensive data reference for system users, thereby making predictive diagnosis of brain glioma diseases more quickly.

[0101] In summary, in the above-mentioned embodiment, during operation, the system estimates the risk of the user suffering from glioma through analysis of the user's brain magnetic resonance image data, thereby replacing manual labor to a large extent. The accuracy of the diagnosis is no longer affected by the accumulation of manual professional knowledge and experience, and the patient's risk of glioma can be estimated more accurately. In the process of estimating the user's risk of glioma based on the user's brain magnetic resonance image data, the user's risk of glioma is comprehensively evaluated through the overall analysis of the user's brain magnetic resonance image and the segmented analysis of the corresponding waveform image, so that the evaluation result can more truly correspond to the actual situation shown by the user's brain magnetic resonance image data, provide more favorable data reference for doctors, and provide support for the diagnosis of the user's condition.

[0102] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A brain glioma diagnosis system based on contralateral positioning of resting brain function, characterized in that: include: An acquisition module, used for acquiring magnetic resonance imaging data of a user's brain; A conversion module, used to retrieve the user's brain magnetic resonance image data collected by the acquisition module, and convert the user's brain magnetic resonance image data into a waveform image; A segmentation module is used to receive the waveform image converted from the user's brain magnetic resonance image data in the conversion module, and segment the waveform image; An evaluation module is used to obtain the waveform image that has completed the segmentation process in the segmentation module, and evaluate the risk of brain glioma based on the waveform image; A calibration module is used to receive the risk assessment result of brain glioma in the assessment module and further calibrate the assessment result; The diagnosis module is used to obtain the evaluation results of the calibration module and diagnose whether the user has the risk of developing glioma based on the evaluation results.

2. The brain glioma diagnosis system based on resting-state brain function contralateral positioning according to claim 1 is characterized in that: The user's brain magnetic resonance image data collected by the acquisition module comes from a magnetic resonance imaging device. The acquisition module is provided with submodules at the lower level, including: A preprocessing unit, used for receiving the user's brain magnetic resonance image data acquired by the acquisition module, and performing smoothing processing on the user's brain magnetic resonance image data; A database, used to obtain the user's brain magnetic resonance image data output after the smoothing process is completed in the preprocessing unit, and store the user's brain magnetic resonance image data; Among them, the preprocessing unit is set with a user's brain magnetic resonance image data smoothing processing logic, and the user's brain magnetic resonance image data is preprocessed based on the user's brain magnetic resonance image data smoothing processing logic. Each group of user's brain magnetic resonance image data is marked with an acquisition timestamp. When the database stores the user's brain magnetic resonance image data, a storage sequence is generated based on the acquisition timestamp corresponding to the marked brain magnetic resonance image data of each user, and the user's brain magnetic resonance image data is sorted and stored based on the storage sequence.

3. The brain glioma diagnosis system based on resting state brain function contralateral positioning according to claim 2 is characterized in that: The smoothing logic of the user's brain magnetic resonance image data in the preprocessing unit is expressed as: Where: S(x, y, z) is the gray value of the pixel in the brain magnetic resonance image data after smoothing; ω(i, j, k) is the gray difference weight; I(i, j, k) is the gray value of the pixel in the brain magnetic resonance image data; in, σ is a sensitivity parameter; I(x, y, z) is the grayscale value of the pixel at the center of the brain magnetic resonance image data; the value of the sensitivity parameter σ is customized by the system user, and the value range of the sensitivity parameter σ is 0.1 to 1. Based on the above formula, each pixel in the user's brain magnetic resonance image data is smoothed to output the user's brain magnetic resonance image data after smoothing.

4. The brain glioma diagnosis system based on resting-state brain function contralateral positioning according to claim 1 is characterized in that: Each time the conversion module is run, at least two or more sets of user brain magnetic resonance image data stored in the middle of the sequence are retrieved from the database. The conversion module is internally provided with submodules, including: A segmentation unit, used to obtain the user's brain magnetic resonance image data retrieved by the conversion module, and symmetrically segment the user's brain magnetic resonance image data to obtain the user's left and right brain magnetic resonance image data; After the segmentation unit performs segmentation processing on the user's brain magnetic resonance image data, the system end user customizes the ROI or performs waveform image conversion processing on the image data globally; Among them, when the system user customizes ROI in the user's left and right brain magnetic resonance image data to perform waveform image conversion processing, the ROI selected in the user's left and right brain magnetic resonance image data is symmetrically distributed. After the user's brain magnetic resonance image data is segmented, the waveform images corresponding to the user's left and right brain magnetic resonance image data are converted separately based on Fourier transform.

5. The brain glioma diagnosis system based on resting-state brain function contralateral positioning according to claim 1 is characterized in that: The segmentation accuracy of the waveform image in the segmentation module is Segments represent the wave lines of a waveform; L represents the total amount of wave lines that make up the waveform in the waveform image; Among them, the segmentation accuracy of the waveform image in the segmentation module is customized by the system end user and obeys the segmentation logic that the higher the accuracy requirement for the risk assessment of glioma in the evaluation module, the smaller the segmentation accuracy, and vice versa.

6. The brain glioma diagnosis system based on resting-state brain function contralateral positioning according to claim 1 is characterized in that: After the evaluation module runs and obtains the waveform image that has completed the segmentation processing, the evaluation logic for the risk of glioma is expressed as follows: Where: r is the Pearson correlation coefficient between the waveform B and the reference waveform C corresponding to the left and right brain magnetic resonance image data; n is the set of waveform midpoints; y Bg is the value of the gth sampling point in waveform B; is the mean value of the sampling points of waveform B; y Cg is the value of the gth sampling point in the reference waveform C; is the mean value of the sampling points of the reference waveform C; r s is the comprehensive Pearson correlation coefficient of the waveform obtained by segmentation module; u is the total number of waveform segments; r′ v is the Pearson correlation coefficient between the waveform B corresponding to the left and right brain magnetic resonance image data and the vth waveform in the reference waveform C; ω v is the weight; The waveform corresponding to the right brain magnetic resonance image data is denoted as A, and the reference waveform C is obtained by mirroring the waveform A. In formula (2), r′ v The calculation logic is the same as that of formula (1), and the weight ω v The value is subject to r′ v The longer the corresponding waveform length, the greater the weight ω v The larger the value, the more r and r s The larger the value, the higher the waveform symmetry, which means the lower the risk of brain glioma. Conversely, the lower the waveform symmetry, which means the greater the risk of brain glioma.

7. The brain glioma diagnosis system based on resting-state brain function contralateral positioning according to claim 6 is characterized in that: The evaluation module is provided with submodules at the lower level, including: Nucleation unit, used to combine r and r s Comprehensively assess the risk of glioma; The logic of comprehensively evaluating the risk of glioma in the calculation unit is: R=r×a1+r s ×a2; Where: R is the comprehensive assessment of the risk of glioma; a1 and a2 are weight factors; Among them, weight factors a1 and a2 are customized by system users, and the sum of weight factors a1 and a2 is 1, a1 is always smaller than a2, and the value of weight factor a2 follows that the more waveform segments there are in the waveform image, the larger the value of weight factor a2, and vice versa.

8. The brain glioma diagnosis system based on resting-state brain function contralateral positioning according to claim 1 is characterized in that: The brain glioma risk assessment result received in the calibration module is the comprehensive assessment of the brain glioma risk R obtained by the operation of the nuclear calculation unit; Each group of user brain magnetic resonance image data collected by the acquisition module obtains a group of R, and the corresponding results of each group of user brain magnetic resonance image data are recorded as R1, R2, R3, ...; The logic of further calibration of the evaluation results by the calibration module is expressed as follows: Where: R NORR is the calibrated risk of glioma; max(R1, R2, ...) is the maximum value in the brackets; ε is the adjustment index; R1, R2, R3, ..., R(n-1), R(n) are the corresponding results of the brain magnetic resonance image data of each group of users; Among them, the adjustment index ε takes the value of 1 or -1. If the numerator of the fraction where the adjustment index ε is located is greater than the denominator, the adjustment index ε takes the value of -1. Otherwise, the adjustment index ε takes the value of 1.

9. The brain glioma diagnosis system based on resting-state brain function contralateral positioning according to claim 1, characterized in that: The diagnostic module operates on the R calibrated in the calibration module NORR Recording, synchronously set the judgment threshold, based on the judgment threshold and the recorded R NORR Compare and diagnose whether the user has the risk of developing glioma; Among them, any two groups of R recorded continuously in the diagnosis module NORR , the previous group R NORR The next group R NORR When the value is large, the user is diagnosed with the risk of brain glioma.

10. The brain glioma diagnosis system based on resting-state brain function contralateral positioning according to claim 1, characterized in that: The acquisition module is interactively connected to the preprocessing and database at the lower level through a wireless network, the acquisition module is interactively connected to the conversion module at the lower level through a wireless network, the conversion module is interactively connected to the segmentation unit through a wireless network, the conversion module is interactively connected to the segmentation module and the evaluation module through a wireless network, the evaluation module is interactively connected to the calculation unit at the lower level through a wireless network, the evaluation module is interactively connected to the calibration module and the diagnosis module through a wireless network, and the calibration module is interactively connected to the calculation unit through a wireless network.

Images