A permanent magnet magnetic resonance imaging processing system and an electronic device

Abstract

The present invention provides a permanent magnet magnetic resonance imaging processing system and an electronic device, belonging to the field of magnetic resonance technology, comprising: a magnetic resonance machine and an imaging processing platform; the magnetic resonance machine is provided with an imaging body and a data processing device, a magnet is installed on the upper part of the imaging body, a workbench is slidably connected to the imaging body, an imaging area is arranged inside the workbench, and a radio frequency coil is arranged in the imaging area; the data processing device is connected to the radio frequency coil to obtain radio frequency signals and convert them into magnetic resonance images; the imaging processing platform is communicatively connected to the data processing device to obtain the magnetic resonance images as the magnetic resonance images to be analyzed, and perform identification, storage and display. By performing multi-dimensional index analysis on the magnetic resonance images, the characteristics of the magnetic resonance images can be accurately described. The system displays the index analysis results to the user in the form of intuitive charts, images, etc., enabling the user to understand and interpret the analysis results.

Description

Technical Field

[0001] The present invention belongs to the technical field of magnetic resonance, and particularly relates to a permanent magnetic resonance imaging processing system and an electronic device. Background Art

[0002] A magnetic resonance imaging (MRI, full English name: Magnetic Resonance Imaging) system is a commonly used medical imaging device, that is, an imaging technology that uses the hydrogen atomic nuclei (protons) in human tissues to undergo nuclear magnetic resonance phenomena after being excited by radio frequency pulses in a magnetic field, generates magnetic resonance signals, and reconstructs an image of a certain layer of the human body through computer processing.

[0003] In the analysis of magnetic resonance images in related technologies, users mainly rely on experience to measure magnetic resonance imaging. The judgment criteria and measurement methods of different users may vary, resulting in the accuracy and consistency of diagnostic results being affected. It is impossible to select corresponding analysis methods according to different magnetic resonance images, resulting in low accuracy in the analysis of different magnetic resonance images. It is impossible to select appropriate analysis methods according to different types of target magnetic resonance images, resulting in inaccurate analysis of different magnetic resonance images. Summary of the Invention

[0004] The present invention provides a permanent magnetic resonance imaging processing system, which can accurately describe the characteristics of magnetic resonance images by performing multi-dimensional index analysis on magnetic resonance images.

[0005] The system includes: a magnetic resonance machine and an imaging processing platform;

[0006] The magnetic resonance machine is provided with an imaging body and a data processing device. A magnet is installed on the upper part of the imaging body. The imaging body is slidably connected with a workbench. An imaging area is arranged inside the workbench, and a radio frequency coil is arranged in the imaging area;

[0007] The data processing device is connected to the radio frequency coil to obtain radio frequency signals and convert them into magnetic resonance images;

[0008] The imaging processing platform is communicatively connected to the data processing device to obtain magnetic resonance images as the magnetic resonance images to be analyzed, and perform identification, storage and display.

[0009] Furthermore, it should be noted that the imaging processing platform obtains the imaging sequence corresponding to the magnetic resonance image to be analyzed, and obtains the target lesion type of the lesion index corresponding to the magnetic resonance image from the preset magnetic resonance examination data;

[0010] According to the imaging sequence, the magnetic resonance examination data and the target lesion type, construct the lesion index of the magnetic resonance image;

[0011] Perform index analysis on magnetic resonance images according to lesion indexes and imaging sequences to obtain the index analysis results.

[0012] Furthermore, it should be noted that the imaging processing platform is used to determine the lesion analysis method corresponding to the lesion index according to the target lesion type of the lesion index;

[0013] Perform index analysis on magnetic resonance images according to the lesion analysis method and imaging sequences to obtain the index analysis results.

[0014] Furthermore, it should be noted that extract tumor contour data at different time points from the imaging sequence; assume that the tumor is accurately segmented in each imaging, and obtain the three-dimensional coordinate information of the tumor;

[0015] For the tumor contour at each time point, use the tumor volume V calculation formula: Calculate the tumor volume value at each time point, where V i is the volume of each tiny voxel after the tumor contour is discretized.

[0016] Furthermore, it should be noted that the imaging processing platform arranges the tumor volume values at different time points in chronological order and draws the tumor volume-time change curve;

[0017] Define the time as t, and the fitting equation is V = at + b, where a is the regression coefficient and b is the intercept; the calculation methods of a and b are:

[0018]

[0019] where, t i is the i-th time point, n is the number of time points, is the average value of time, is the average value of the tumor volume; judge the change trend of the tumor volume according to the regression coefficient a. If a > 0, it means the tumor volume shows an increasing trend; if a < 0, it means the tumor volume shows a decreasing trend; if a ≈ 0, it means the tumor volume is relatively stable.

[0020] Furthermore, it should be noted that the imaging processing platform determines the target data to be analyzed from the imaging sequence according to the lesion analysis method; use the lesion analysis method to analyze and process the target data to obtain the index analysis results.

[0021] Furthermore, it should be noted that the imaging processing platform determines the target state data corresponding to the target lesion type from the imaging sequence according to the target lesion type; construct the lesion index according to the target state data and magnetic resonance examination data.

[0022] It should be further noted that the imaging processing platform is also used to obtain the imaging sequence corresponding to the magnetic resonance image to be analyzed, and obtain the target lesion type of the lesion index corresponding to the magnetic resonance image from the preset magnetic resonance examination data;

[0023] According to the target lesion type, determine the target state data corresponding to the target lesion type from the imaging sequence;

[0024] Combine the target state data and the magnetic resonance examination data to construct the lesion index of the magnetic resonance image;

[0025] Based on the target lesion type of the lesion index, determine the lesion analysis method corresponding to the lesion index;

[0026] Based on the lesion analysis method, determine the target data to be analyzed from the imaging sequence;

[0027] Use the lesion analysis method to analyze and process the target data to obtain the index analysis result;

[0028] Display the index analysis result.

[0029] It should be further noted that the lesion index includes a contrast lesion index, a trend lesion index, a correlation lesion index, and a prediction lesion index;

[0030] The lesion analysis methods include a contrast lesion analysis method, a trend lesion analysis method, a depth analysis process strategy based on the lesion index, a correlation lesion analysis method, and a prediction lesion analysis method.

[0031] According to another embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the permanent magnet magnetic resonance imaging processing system described above is implemented.

[0032] From the above technical solutions, it can be seen that the present invention has the following advantages:

[0033] The permanent magnet magnetic resonance imaging processing system provided by the present invention can accurately describe the characteristics of magnetic resonance images through multi-dimensional index analysis of magnetic resonance images. For example, in tumor analysis, not only can basic information such as the volume of the tumor be obtained, but also the changes of the tumor at different time points can be understood through trend analysis. The system automatically completes a series of processes such as imaging sequence acquisition, lesion index construction, and index analysis, improving the imaging analysis efficiency and accuracy.

[0034] The present invention can select corresponding lesion analysis methods according to different types of target lesions, and conduct targeted analysis on different lesions. Different lesions may require different analysis focuses and methods, such as trend analysis of tumors and comparative analysis of inflammations, to meet the usage requirements. It also displays the results of index analysis, presenting the data in the form of charts, images, etc. to users, enabling users to understand and interpret the analysis results more quickly.

[0035] The method for integrating voxel integrals by discretizing tumor contours based on time series in the present invention calculates volume changes in combination with three-dimensional coordinate information, breaking through the error limitations of traditional two-dimensional slice superposition. Based on the closed-loop logic of lesion type → index construction → analysis method → target data selection, it realizes dynamic adaptation from the target to the analysis method. It integrates magnetic resonance examination data and real-time imaging sequences to construct dynamic lesion indicators, solving the problem of one-sidedness in single data source analysis. Brief Description of the Drawings

[0036] To more clearly illustrate the technical solutions of the present invention, the drawings required for description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0037] Figure 1 It is a schematic diagram of a magnetic resonance machine;

[0038] Figure 2 It is a schematic diagram of a permanent magnet magnetic resonance imaging processing system;

[0039] Figure 3 It is a schematic diagram of an electronic device. Detailed Embodiments

[0040] The following will describe in detail the specific content of the permanent magnet magnetic resonance imaging processing system. For illustration rather than limitation, specific details such as specific system structures and technologies are proposed to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details.

[0041] It should be understood that when used in the specification of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations. The terms "comprising", "including", "having", and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0042] It should be understood that the "one or more" mentioned in this application means one, two or more than two, and the "multiple" mentioned in this application means two or more than two. In the description of this application, unless otherwise specified, " / " means "or". For example, A / B can mean A or B. The "and / or" herein is merely a relationship describing the associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone.

[0043] For the convenience of clearly describing the technical solutions of this application, terms such as "first" and "second" are used to distinguish the same items or similar items with basically the same functions and roles. Those skilled in the art can understand that the terms such as "first" and "second" do not limit the quantity and execution order, and the terms such as "first" and "second" do not necessarily limit to be different.

[0044] The statements such as "an embodiment" or "some embodiments" described in this application mean that the specific features, structures or characteristics described in the embodiment are included in one or more embodiments of this application. Thus, the statements such as "in an embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments" and the like that appear in different parts of this application do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways.

[0045] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0046] Please refer to Figure 1 and Figure 2 The figure shows a schematic diagram of a permanent magnet magnetic resonance imaging processing system in a specific embodiment. The system includes: a magnetic resonance machine and an imaging processing platform.

[0047] The imaging processing platform uses a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The components, their connections and relationships, and their functions shown herein are only examples and are not intended to limit the implementation of the embodiments of the present application described and / or required herein.

[0048] The magnetic resonance machine in this embodiment is provided with an imaging body 1 and a data processing device. A magnet 2 is installed on the upper part of the imaging body 1. The imaging body is slidably connected with a workbench 3. An imaging area 4 is arranged inside the workbench 3, and a radio frequency coil 5 is arranged in the imaging area 4. The data processing device is connected to the radio frequency coil 5 to obtain radio frequency signals and convert them into magnetic resonance images.

[0049] A data processing device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0050] Specifically, place the required radio frequency receiving coil on the workbench. Place the patient in a lying position on the workbench and position the scanning part in the radio frequency coil. Adjust the position of the workbench to place the scanned part of the patient in the imaging area for magnetic resonance imaging.

[0051] The imaging processing platform communicates with the data processing device to obtain magnetic resonance images, which are used as the magnetic resonance images to be analyzed, and then performs identification, storage, and display.

[0052] Based on the above embodiments, in order to further improve the reliability of the permanent magnet magnetic resonance imaging processing system provided in the above embodiments, the following is a more specific and feasible method. In one embodiment, the imaging processing platform in the permanent magnet magnetic resonance imaging processing system can execute the following method steps:

[0053] S101, obtain the imaging sequence corresponding to the magnetic resonance image to be analyzed, and obtain the target lesion type of the lesion index corresponding to the magnetic resonance image from the pre-set magnetic resonance examination data according to the magnetic resonance image.

[0054] In some embodiments, the magnetic resonance image refers to obtaining the tissue structure and lesion conditions.

[0055] The lesion index is a set of rules based on which quantitative analysis of magnetic resonance images is performed. The lesion index includes data analysis objects, analysis methods, and result generation methods.

[0056] Among them, the data analysis object can involve specific imaging parameters such as tumor volume, signal intensity, and edge sharpness. The analysis method is a mathematical formula for calculating the tumor volume, and the formula is volume = [(length × width × height × π) / 6]. The result generation method is an algorithm for converting the volume change rate into a trend curve.

[0057] The lesion types in this embodiment are classification labels for lesion indicators, used to identify the analysis direction. In this embodiment, corresponding analysis strategies are matched according to the target lesion type. For example, the trend analysis strategy extracts image data at multiple time points. Historical image data of the same patient can also be extracted from the imaging sequence based on trend analysis to guide the generation of an analysis report. The analysis report can generate a line chart using trend analysis and a differential heat map using comparative analysis.

[0058] Exemplarily, analyze the change in tumor volume during the patient's treatment. The data analysis object is the three-dimensional size of the tumor. The analysis method is volume = length × width × height × π / 6. Calculate the volume change rate between two adjacent examinations. Based on the trend lesion type, extract three groups of image data before, during, and after treatment, calculate the volume value each time, generate a trend curve, and mark the time points with rapid volume growth.

[0059] The imaging sequence in this embodiment is a combination method of a series of specific radiofrequency pulses and gradient pulses used for data acquisition in magnetic resonance imaging. Different imaging sequences will highlight different characteristics of tissues, thereby obtaining images with different contrasts to meet the observation requirements of different tissues and lesions.

[0060] Among them, radiofrequency pulses are used to excite atomic nuclei to produce magnetic resonance phenomena. Gradient pulses include slice selection gradients, phase encoding gradients, and frequency encoding gradients. The slice selection gradient is used to select the imaging slice; the phase encoding gradient is used to perform phase encoding on atomic nuclei at different positions to determine their spatial positions; time parameters involve repetition time, echo time, etc. TR refers to the time interval between two adjacent radiofrequency pulse excitations, and TE refers to the time interval from radiofrequency pulse excitation to the generation of the echo signal. Different combinations of TR and TE can produce images with different contrasts, such as T1-weighted images, T2-weighted images, and proton density-weighted images, etc.

[0061] The preset magnetic resonance examination data in this embodiment are a large amount of reference data related to magnetic resonance imaging stored in the imaging processing platform in advance. These data are accumulated based on a large number of clinical cases and research results and are used to assist in the analysis and diagnosis of newly acquired magnetic resonance images.

[0062] Normal tissue image data includes typical magnetic resonance images of normal tissues of different age groups, genders, and body parts under various imaging sequences as normal reference standards.

[0063] Lesion image data can collect magnetic resonance images of various lesions at different development stages and under different imaging sequences, as well as corresponding lesion feature descriptions.

[0064] The lesion characteristic parameters involve quantitative or semi - quantitative parameters such as the size, shape, boundary, signal intensity, enhancement pattern, etc. of the lesion.

[0065] The diagnostic result and case information are the final diagnostic results corresponding to the lesion image data, as well as the relevant clinical information of the patient, such as symptoms, medical history, laboratory test results, etc., for comprehensive analysis of the lesion situation.

[0066] Among the target lesion types of the lesion indicators corresponding to the magnetic resonance images, the lesion indicators are various parameters and information used to describe and evaluate the lesion characteristics in magnetic resonance images. It can involve the size of the lesion, such as diameter, volume, etc., the shape such as round, oval, lobulated, etc., and the boundary.

[0067] The target lesion type is the possible lesion type inferred from the preset magnetic resonance examination data according to the lesion indicators, providing a reference for clinical diagnosis. For example: neoplastic lesions, including benign tumors and malignant tumors.

[0068] The imaging sequence in this embodiment is the way of MRI data acquisition, the preset magnetic resonance examination data is the reference basis, the lesion indicators are used to describe the lesion characteristics, and the target lesion type is the possible lesion type inferred based on the lesion indicators. These concepts are interrelated and jointly constitute the analysis and diagnosis basis of the permanent magnetic resonance imaging processing system.

[0069] In order to perform index analysis on magnetic resonance images, in some embodiments, the imaging processing platform first obtains the imaging sequence corresponding to the magnetic resonance image to be analyzed.

[0070] In some embodiments, during the process of the imaging processing platform obtaining the imaging sequence corresponding to the magnetic resonance image to be analyzed, it can first determine the data type of the corresponding imaging sequence according to the magnetic resonance image to be analyzed, and then obtain the imaging sequence corresponding to this data type, thus obtaining the imaging sequence corresponding to the magnetic resonance image to be analyzed.

[0071] After determining the imaging sequence corresponding to the magnetic resonance image to be analyzed, the imaging processing platform can obtain the target lesion type of the lesion indicators corresponding to the magnetic resonance image from the preset magnetic resonance examination data according to the magnetic resonance image.

[0072] The preset magnetic resonance examination data refers to the magnetic resonance examination data that pre - defines the association between multiple magnetic resonance images and automatic imaging and analysis configuration parameters. The preset magnetic resonance examination data is used for the imaging processing platform to analyze multiple magnetic resonance images according to the preset magnetic resonance examination data.

[0073] In this embodiment, when analyzing the magnetic resonance image for the lesion index, the parsing method of the lesion index is defined. This embodiment may be related to the analysis of the trend lesion index. The analysis of the trend lesion index is to analyze the dynamic change law of the lesion characteristics by tracking the magnetic resonance images of the same patient at different time points (such as before, during, and after treatment), so as to assist clinical decision-making.

[0074] The trend lesion index includes calculating the volume growth rate of the lesion at different time points (such as the tumor doubling time). The invasiveness change of the lesion is evaluated through parameters such as the lobulation index and surface smoothness. The change in the degree of boundary blurring is quantified. For example, the boundary tends to be clear when the inflammation subsides.

[0075] The analysis of the trend lesion index can also be based on the signal intensity trend index, comparing the relative change of the lesion signal intensity in the same sequence. For example, the range of high T2 signal in the edema area decreases. Analyze the change of the peak enhancement time and enhancement degree of the lesion in dynamic contrast-enhanced MRI. For example, the enhancement weakens due to the reduction of tumor neovascularization. The dynamic characteristics of cell density or water molecule movement are reflected by the change of the apparent diffusion coefficient, that is, the ADC value. For example, the ADC value increases when tumor cells die.

[0076] This embodiment can also be combined with the time-index curve analysis to calculate the change slope of the index over time.

[0077] The magnetic resonance image of this embodiment can correspond to multiple lesion indexes, that is, different index analyses can be performed on the magnetic resonance image. The magnetic resonance image is analyzed by comparison. The magnetic resonance image can correspond to the comparative lesion index and the in-depth analysis process based on the lesion index. Moreover, the multiple lesion indexes corresponding to different magnetic resonance images are not exactly the same.

[0078] In the magnetic resonance image analysis, the in-depth analysis process based on the lesion index refers to an operation behavior of deeply exploring data. When facing multiple lesion indexes corresponding to a magnetic resonance image, more detailed and in-depth data information can be further mined based on a specific index or a group of indexes.

[0079] Exemplarily speaking, through the preliminary analysis of the magnetic resonance image, it is found that the lesion in a certain area shows an abnormal signal intensity as a lesion index on the T1-weighted image.

[0080] The execution of the in-depth analysis process based on lesion indicators is the in-depth analysis of image-related lesion indicators. For example, originally, only the basic lesion indicator of abnormal signal intensity in the lesion area of the magnetic resonance image was observed. After entering the in-depth analysis process based on lesion indicators, further multi-dimensional expansion analysis is performed on this signal intensity indicator. Analyze the change trend of this signal intensity in different imaging sequences, the evolution of the contrast difference with the signal intensity of the surrounding normal tissues over time, or analyze the characteristic changes of this signal intensity indicator according to different patient groups, etc. Through such expression conversion, it is more directly linked to the indicator analysis, clarifying that this is an in-depth analysis process centered around lesion indicators and emphasizing the close association between the analysis behavior and the lesion indicators.

[0081] In this embodiment, the lesion types of the lesion indicators may include regular lesion indicator types and big data model lesion indicator types. When performing analysis in this embodiment, trend analysis, contrast analysis, the in-depth analysis process based on lesion indicators, and predictive analysis can be combined for use. Since different magnetic resonance images respectively correspond to multiple lesion indicators, and the lesion indicators can be identified by lesion types, the magnetic resonance examination data predefining the association between different magnetic resonance images and the lesion types of the lesion indicators can be preset.

[0082] In some embodiments, the preset magnetic resonance examination data includes the magnetic resonance examination data of the association between different magnetic resonance images and the lesion types of the lesion indicators. The imaging processing platform obtains the target lesion type of the lesion indicator corresponding to the magnetic resonance image from the preset magnetic resonance examination data according to the magnetic resonance image. Exemplarily, the imaging processing platform can determine, according to the magnetic resonance image, the lesion type of the lesion indicator associated with the magnetic resonance image from the magnetic resonance examination data, and determine the lesion type of the lesion indicator associated with the magnetic resonance image in the magnetic resonance examination data as the target lesion type, and the target lesion type is at least one of the lesion types.

[0083] S102. Construct the lesion indicator of the magnetic resonance image according to the imaging sequence, the magnetic resonance examination data, and the target lesion type.

[0084] In some embodiments, the indicator analysis is defined. Since different magnetic resonance images respectively correspond to multiple lesion indicators, the multiple different lesion indicators corresponding to one magnetic resonance image represent that different indicator analysis methods are used to analyze one magnetic resonance image. Exemplarily, the magnetic resonance image corresponds to a trend lesion indicator and a contrast lesion indicator, indicating that trend analysis and contrast analysis can be performed on the magnetic resonance image. Different magnetic resonance images will simultaneously correspond to the same lesion indicator.

[0085] Different magnetic resonance images can correspond to the same lesion index simultaneously. Moreover, the lesion index analysis methods defined by the lesion index are the same. However, when different magnetic resonance images perform index analysis using the same lesion index analysis method, the analysis objects and the generation methods of the analysis results may be different. For example, both the first magnetic resonance image and the second magnetic resonance image correspond to a contrast lesion index. When the first magnetic resonance image performs a contrast analysis, it compares the value corresponding to the first with other values, while the contrast analysis of the second magnetic resonance image compares the value corresponding to the second with other values. Therefore, after determining the target lesion type corresponding to the magnetic resonance image, a corresponding lesion index is constructed for the magnetic resonance image.

[0086] In some embodiments, the imaging processing platform constructs the lesion index of the magnetic resonance image based on the imaging sequence, magnetic resonance examination data, and the target lesion type. That is, the imaging processing platform determines the imaging sequence required for the lesion index corresponding to the target lesion type according to the magnetic resonance examination data, and constructs the lesion index of the magnetic resonance image based on the determined imaging sequence.

[0087] In this embodiment, the automatic imaging and analysis configuration parameters in the magnetic resonance examination data include the key lesion feature data associated with the lesion type, the lesion index analysis method, and the lesion result derivation algorithm. After determining the target lesion type, the key lesion feature data, the lesion index analysis method, and the lesion result derivation algorithm corresponding to the target lesion type are obtained from the magnetic resonance examination data. The data analysis object is obtained from the imaging sequence according to the key lesion feature data, and the lesion index is constructed based on the data analysis object, the lesion index analysis method, and the generation method of the analysis result.

[0088] In some specific embodiments, the magnetic resonance examination data further includes a lesion analysis process template for the lesion index associated with the lesion type. After determining the target lesion type as described above, the lesion analysis process template corresponding to the target lesion type is obtained from the magnetic resonance examination data. The lesion analysis process template includes fields corresponding to the data analysis object, the lesion index analysis method, and the lesion result derivation algorithm. The fields in the lesion analysis process template are parsed, the data information corresponding to the fields is obtained from the imaging sequence, the link relationship between the fields and the corresponding data information is established, and the lesion analysis process template is filled to construct the lesion index.

[0089] In a permanent magnetic resonance imaging processing system, the automatic imaging and analysis configuration parameters refer to a series of parameters that are preset and used to automatically control the imaging process and subsequent data processing. These parameters determine how the system automatically adjusts the imaging conditions and performs subsequent data analysis settings according to different examination requirements and lesion types. For example, the selection of imaging sequences for different parts (such as the brain, abdomen, etc.) and different diseases (such as tumors, inflammation, etc.), the setting of scanning parameters (such as magnetic field strength, radio frequency pulse frequency, pulse interval time, etc.), and some initial settings for data processing all fall within the scope of the automatic imaging and analysis configuration parameters.

[0090] The key data of lesion characteristics refers to the data that is closely related to a specific lesion type and plays a core role in determining the lesion index. These data can directly or indirectly reflect the characteristics of the lesion and are the key analysis objects selected from the imaging sequence data. For example, for tumor lesions, data such as the size, shape, edge characteristics, signal intensity changes under different imaging sequences, and the relationship with surrounding tissues; for multiple sclerosis lesions in the brain, data such as the location, number, size of demyelinating plaques in the white matter of the brain, and the signal characteristics under T1, T2 weighted images, and Flair sequences. These data are the basis for constructing the lesion index and can help the system accurately judge the nature and degree of the lesion.

[0091] The method of analyzing lesion indexes refers to a series of rules and standards followed when analyzing magnetic resonance imaging data to determine lesion indexes. These principles stipulate how to extract meaningful features from the original imaging data and key data information and how to judge the nature and degree of the lesion based on these features. For example, when judging whether a tumor is malignant, it may be analyzed according to rules such as the enhancement pattern, enhancement degree (the ratio of signal intensity compared with normal tissue), and the characteristics of the time-signal intensity curve during enhanced scanning; when analyzing brain lesions, the type of lesion is determined according to the change law of signal intensity under different imaging sequences. For example, low signal on T1 weighted image and high signal on T2 weighted image may indicate fluid or edema, etc.

[0092] The method of analyzing lesion indexes is to collect the typical manifestation data of various lesions under different imaging sequences and establish a database. Then, through statistical analysis and comparative research on these data, the characteristic rules that can effectively distinguish different lesion types and degrees are summarized, and these rules are transformed into specific analysis principles.

[0093] The lesion result derivation algorithm refers to the specific methods and algorithms for generating the final judgment result of the lesion based on the previous index analysis process and results. It comprehensively processes the lesion features extracted from the imaging sequence data and the analysis results obtained according to the lesion index analysis method to generate clear and clinically significant results. For example, by analyzing indicators such as the size, enhancement pattern, and signal intensity of a tumor, and using specific mathematical models or logical judgment rules, results such as whether the tumor is benign or malignant and at what stage can be obtained; for cardiovascular diseases, based on indicators such as the signal changes and thickness of the myocardium in magnetic resonance imaging, combined with corresponding algorithms, it is judged whether there are lesions such as myocardial ischemia and infarction, as well as the scope and degree of the lesions.

[0094] The lesion analysis process template is a pre-set framework for data processing and analysis for different types of lesions. It contains the key information required for analyzing specific types of lesions, such as which data needs to be extracted from the imaging sequence, according to what principles the index analysis is carried out, and how the analysis results are generated. It can be regarded as a standardized process template, and there is a corresponding template for each common type of lesion. When the system encounters a magnetic resonance image that needs to be analyzed, according to the determined target lesion type, the corresponding template is called, and data extraction, analysis, and result generation are carried out according to the regulations in the template, so as to efficiently construct lesion indicators. For example, for lung cancer lesions, the template will clearly stipulate that the size and shape data of the tumor under the lung window and mediastinal window, and the enhancement feature data in the enhanced scan need to be extracted, analyzed according to the lesion index analysis method related to lung cancer, and the analysis results such as lung cancer staging and type are generated according to specific algorithms.

[0095] In this way, through the above methods, automatic construction of lesion indicators can be achieved. In the process of analyzing magnetic resonance images, constructing lesion indicators can improve the automation degree of magnetic resonance image analysis, thereby improving the efficiency of magnetic resonance image analysis.

[0096] S103, perform index analysis on the magnetic resonance image according to the lesion indicators and the imaging sequence to obtain the index analysis result.

[0097] In some embodiments, the imaging processing platform performs index analysis on the magnetic resonance image according to the lesion indicators and the imaging sequence to obtain the index analysis result, which may be to process the imaging sequence according to the lesion indicators to achieve index analysis of the magnetic resonance image and obtain the index analysis result.

[0098] In this embodiment, the imaging processing platform obtains the value corresponding to the data analysis object from the imaging sequence according to the data analysis object, the lesion index parsing method, and the lesion result derivation algorithm included in the lesion index, analyzes and calculates the value according to the lesion index parsing method to obtain a calculation result, and processes the calculation result according to the lesion result derivation algorithm to obtain the lesion index analysis result.

[0099] It can be seen that the lesion parsing method corresponding to the lesion index determined by the imaging processing platform can be used for index analysis of magnetic resonance images, and by extracting information from the image data, it can assist medical diagnosis.

[0100] In an embodiment of the present invention, in combination with the above method, a more specific method will be given below.

[0101] S201, determine the lesion parsing method corresponding to the lesion index according to the target lesion type of the lesion index.

[0102] In some embodiments, tumor contour data at different time points are extracted from the imaging sequence. Assuming that the tumor is accurately segmented each time of imaging, the three-dimensional coordinate information of the tumor can be obtained.

[0103] For the tumor contour at each time point, using the tumor volume V calculation formula: , calculate the tumor volume value at each time point, where V i is the volume of each tiny voxel after the tumor contour is discretized.

[0104] The lesion index includes a comparison lesion index, a trend lesion index, a correlation lesion index, and a prediction lesion index. Correspondingly, the lesion parsing method includes a comparison lesion parsing method, a trend lesion parsing method, a depth analysis process strategy based on the lesion index, a correlation lesion parsing method, and a prediction lesion parsing method.

[0105] The comparison lesion parsing method refers to comparing and analyzing the value of the current institutional magnetic resonance image with other values, and giving a comparison analysis result. The value can be a reference value or an average value. The comparison analysis result refers to the difference result after comparison, and also includes suggestions for the difference and key data indicators affecting the difference in this magnetic resonance image.

[0106] The trend lesion parsing method refers to analyzing the change trend of the lesion index of the magnetic resonance image within a certain future time range, and giving a trend analysis result. The trend analysis result includes a change trend chart, and turning points are given in the trend to realize the longitudinal comparison of the magnetic resonance image. It also includes the analysis of the trend direction and key data indicators affecting the trend direction.

[0107] The correlation lesion analysis method refers to analyzing the indicators correlated with magnetic resonance images from the imaging sequence based on magnetic resonance examination data, then conducting research through experiments, analyzing the correlation relationships between the indicators, and finding the factors that can effectively drive the improvement of magnetic resonance images for continuous intervention.

[0108] The prediction lesion analysis method refers to reasonably constructing a machine learning prediction model according to the magnetic resonance image analysis target, training and analyzing the historical data of magnetic resonance images, and continuously adjusting the parameters of the prediction model so that the prediction model can accurately conduct prediction analysis on magnetic resonance images based on historical data.

[0109] S202, conduct index analysis on the magnetic resonance image according to the lesion analysis method and the imaging sequence to obtain the index analysis result.

[0110] In some embodiments, arrange the tumor volume values at different time points in chronological order to plot the tumor volume-time change curve. Define the time as t, and the fitting equation is V = at + b, where a is the regression coefficient and b is the intercept. Calculate the values of a and b through the following formula,

[0111]

[0112] where, is the average value of time, is the average value of tumor volume. Judge the change trend of the tumor volume according to the regression coefficient a. If a > 0, it means the tumor volume shows an increasing trend; if a < 0, it means the tumor volume shows a decreasing trend; if a ≈ 0, it means the tumor volume is relatively stable.

[0113] Analyze the key data indicators affecting the trend direction. If the angiogenesis around the tumor increases during a certain period and the tumor volume growth accelerates during this period, then the angiogenesis situation around the tumor can be used as the key data indicator affecting the tumor volume growth trend. Finally, organize the change trend chart, trend direction analysis, and key data indicators into a trend analysis result report for output.

[0114] Combined with the above method, as a more specific implementation method, this embodiment also involves the following method:

[0115] S301, determine the target data to be analyzed from the imaging sequence according to the lesion analysis method.

[0116] In some embodiments, the lesion analysis method corresponds to lesion indicators. Since the lesion indicators include data analysis objects, obtain the target data to be analyzed from the imaging sequence according to the relevant information of the data analysis objects. The target data refers to the entity data corresponding to the data analysis object. For example, if the data analysis object is the code submission volume, the target data is the specific value of the code submission volume.

[0117] Optionally, the imaging processing platform may determine corresponding lesion indicators according to the lesion analysis method, obtain relevant information of the data analysis object from the lesion indicators, use the relevant information of the data analysis object as a query field, and obtain the target data to be analyzed from the imaging sequence.

[0118] S302. Analyze and process the target data using the lesion analysis method to obtain an index analysis result.

[0119] In some embodiments, after the imaging processing platform obtains the target data required in the lesion analysis method, it analyzes and processes the target data according to the lesion index analysis method and the lesion result derivation algorithm in the lesion analysis method to obtain an index analysis result.

[0120] In this way, by analyzing and processing the target data obtained according to the lesion analysis method, the index analysis of the magnetic resonance image is realized, and an index analysis result is obtained.

[0121] In the above embodiments, it is mentioned that the lesion indicators of the magnetic resonance image are constructed according to the imaging sequence, the magnetic resonance examination data, and the target lesion type. Next, an implementation manner of the imaging processing platform for constructing the lesion indicators of the magnetic resonance image will be described.

[0122] In the system of this embodiment, the imaging processing platform also executes based on the following manner. Specifically, it includes the following steps:

[0123] S401. Determine the target state data corresponding to the target lesion type from the imaging sequence according to the target lesion type.

[0124] Among them, the imaging sequence includes the state data corresponding to each lesion type respectively, and the state data corresponding to each lesion type respectively is obtained by pre-dividing the imaging sequence according to each lesion type. The above imaging sequence is the imaging sequence classified according to the magnetic resonance image, and is the imaging sequence corresponding to the magnetic resonance image to be analyzed. This imaging sequence includes all data associated with the magnetic resonance image, and the target lesion type corresponding to the magnetic resonance image can be multiple lesion types. The imaging sequences associated with different lesion types are different. Therefore, the imaging sequence corresponding to the magnetic resonance image can be pre-divided according to the lesion type to obtain multiple state data. Exemplarily, the first data can be divided into trend lesion index data, contrast lesion index data, correlation lesion index data, and prediction lesion index data.

[0125] In some embodiments, determining the target state data corresponding to the target lesion type from the imaging sequence according to the target lesion type may be obtaining the target state data corresponding to the target lesion type from multiple state data corresponding to the imaging sequence according to the target lesion type. Since the target lesion type may be multiple lesion types, the target state data may be multiple state data.

[0126] S402. Construct a lesion index according to the target state data and the magnetic resonance examination data.

[0127] In the above embodiments, it is mentioned that index analysis is performed on the magnetic resonance image to obtain an index analysis result. Next, an implementation manner of the imaging processing platform processing the index analysis result will be described.

[0128] Furthermore, as a refinement and extension of the specific implementation manner of the above embodiments, in order to completely illustrate the specific implementation process in this embodiment, the imaging processing platform is also executed based on the following manner, including the following steps:

[0129] S511. Obtain the imaging sequence corresponding to the magnetic resonance image to be analyzed, and obtain the target lesion type of the lesion index corresponding to the magnetic resonance image from the preset magnetic resonance examination data according to the magnetic resonance image.

[0130] In this embodiment, the imaging processing platform first performs data interaction with the magnetic resonance machine to obtain the magnetic resonance image to be analyzed from the storage device or transmission interface of the magnetic resonance machine. At the same time, obtain the imaging sequence information used during the scanning of this magnetic resonance image, and these information are usually stored or transmitted together with the magnetic resonance image. Then, the imaging processing platform compares the characteristics of the magnetic resonance image with the preset magnetic resonance examination data. The preset magnetic resonance examination data contains a large number of magnetic resonance image feature samples of different lesion types and the corresponding lesion type annotations. Through image recognition technology and feature matching algorithms, find the lesion type that best matches the current magnetic resonance image, so as to determine the target lesion type.

[0131] S512. Determine the target state data corresponding to the target lesion type from the imaging sequence according to the target lesion type.

[0132] After determining the target lesion type, the imaging processing platform will refer to the relevant information about this target lesion type in the preset magnetic resonance examination data. These information will specify which data in the imaging sequence are key and lesion-related target state data for this lesion type. The platform filters out the corresponding target state data from various data parameters and image features of the imaging sequence according to these specified information. For example, if the target lesion type is a brain tumor, the target state data may include data such as the signal intensity of the tumor under different weighted images, the clarity of the tumor boundary, and the size of the edema area around the tumor.

[0133] S513. Combine the target status data and the magnetic resonance examination data to construct the lesion index of the magnetic resonance image.

[0134] The imaging processing platform combines the obtained target status data with the relevant standards and rules in the magnetic resonance examination data. The magnetic resonance examination data contains the definitions and calculation methods for how different target status data are transformed into lesion indexes. The platform processes and calculates the target status data according to these definitions and methods. For example, for the signal intensity data of a tumor, it may be determined whether it is a high signal, a low signal, or an isosignal according to the preset standard, and combined with other relevant data to calculate the lesion index reflecting the nature of the tumor, such as the contrast index of the tumor. In this way, a lesion index that can accurately describe the lesion characteristics in the magnetic resonance image is constructed.

[0135] S514. Based on the target lesion type of the lesion index, determine the lesion analysis method corresponding to the lesion index.

[0136] The imaging processing platform selects the lesion analysis method suitable for the lesion type from the preset lesion analysis method library according to the target lesion type. The lesion analysis method library stores a variety of lesion analysis methods for different lesion types, and each lesion analysis method contains specific analysis methods, rules, and processes. For example, for tumor lesions, there may be a contrast lesion analysis method, a trend lesion analysis method, etc. The platform selects the lesion analysis method that can most effectively analyze the lesion according to the characteristics of the target lesion type and the constructed lesion index.

[0137] S515. Based on the lesion analysis method, determine the target data to be analyzed from the imaging sequence.

[0138] After determining the lesion analysis method, the imaging processing platform screens out the target data that needs further analysis from the imaging sequence according to the requirements of this lesion analysis method. Different lesion analysis methods may focus on different aspects of the data in the imaging sequence. For example, the contrast lesion analysis method may need to obtain the numerical value of the current magnetic resonance image and the data related to the reference value or average value for comparison; the trend lesion analysis method may need to obtain the imaging sequence data at different time points. The platform extracts and determines these target data to be analyzed from the imaging sequence according to the specific requirements of the lesion analysis method.

[0139] S516. Use the lesion analysis method to analyze and process the target data to obtain the index analysis result.

[0140] The imaging processing platform performs corresponding analysis and processing on the determined target data according to the selected lesion analysis method. If it is a comparative lesion analysis method, the platform will compare the target data with the comparative data, calculate the differences, and evaluate and interpret the differences according to the preset rules to obtain the comparative analysis results; if it is a trend lesion analysis method, the platform will perform time series analysis on the target data at different time points, draw a change trend chart, find the turning points, and analyze the trend direction and influencing factors to obtain the trend analysis results. In this way, in-depth analysis of the target data is carried out to obtain the final index analysis results.

[0141] S517, display the index analysis results.

[0142] The imaging processing platform displays the obtained index analysis results in an intuitive and easy-to-understand manner. For example, for the trend analysis results, a line chart showing the change of tumor volume over time can be drawn, and the turning points and key data can be marked on the chart; for the comparative analysis results, the differences and evaluation conclusions of the comparison can be described in words. The platform presents these display contents to the user or other users to facilitate their viewing and understanding of the analysis results.

[0143] The system involved in this application can specifically construct lesion indicators by acquiring imaging sequences, determining the types of target lesions and target status data, thereby improving the accuracy of index analysis. Analyzing according to the preset process and strategy can quickly extract valuable information from magnetic resonance image data, improving the analysis efficiency. By selecting an appropriate lesion analysis method to deeply analyze the target data and presenting the results in an intuitive way.

[0144] As Figure 3 shown, this application also provides an electronic device, including a display module 103, a memory 102, a processor 101, and a computer program stored on the memory and executable on the processor 101. When the processor 101 executes the program, it implements a permanent magnetic resonance imaging processing system.

[0145] In an embodiment of the present invention, the electronic device may be an imaging processing platform. In an embodiment of the present application, the processor 101 may be implemented by using at least one of an application specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, a microcontroller, a microprocessor, and an electronic unit designed to execute the functions described herein. In some cases, such an implementation may be implemented in the controller. For a software implementation, an implementation such as a process or a function may be implemented with a separate software module that allows execution of at least one function or operation. The software code may be implemented by a software application (or program) written in any appropriate programming language. The software code may be stored in the memory and executed by the controller.

[0146] The display module 103 is configured to display information input by the user or information provided to the user. The display module 103 may include a display panel, and the display panel may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0147] The memory 102 may be used to store software programs and various data. The memory 102 may include a high-speed random access memory, and may further include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0148] The permanent magnet magnetic resonance imaging processing system combines the units and algorithm steps of the examples described in the embodiments disclosed herein, and can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such an implementation should not be considered to exceed the scope of the present invention.

[0149] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A permanent magnet magnetic resonance imaging processing system, characterized in that Including: A magnetic resonance machine and an imaging processing platform; The magnetic resonance machine is provided with an imaging body and a data processing device. A magnet is installed on the upper part of the imaging body. The imaging body is slidably connected to a workbench. An imaging area is arranged inside the workbench, and a radiofrequency coil is arranged in the imaging area; The data processing device obtains radiofrequency signals through connection with the radiofrequency coil and converts them into magnetic resonance images; The imaging processing platform obtains magnetic resonance images through communication connection with the data processing device, takes them as magnetic resonance images to be analyzed, and performs identification, storage and display; The imaging processing platform is further used to obtain the imaging sequence corresponding to the magnetic resonance image to be analyzed, and obtain the target lesion type of the lesion index corresponding to the magnetic resonance image from the preset magnetic resonance examination data; According to the target lesion type, determine the target state data corresponding to the target lesion type from the imaging sequence; Combine the target state data and the magnetic resonance examination data to construct the lesion index of the magnetic resonance image; Based on the target lesion type of the lesion index, determine the lesion analysis method corresponding to the lesion index; Based on the lesion analysis method, determine the target data to be analyzed from the imaging sequence; Use the lesion analysis method to analyze and process the target data to obtain an index analysis result; Display the index analysis result.

2. The permanent magnet magnetic resonance imaging processing system according to claim 1, wherein Extract tumor contour data at different time points from the imaging sequence; Assume that the tumor is accurately segmented for each imaging, and obtain the three-dimensional coordinate information of the tumor; For the tumor contour at each time point, use the tumor volume V calculation formula: , calculate the tumor volume value at each time point, where V i is the volume of each tiny voxel after the tumor contour is discretized.

3. The permanent magnet magnetic resonance imaging processing system according to claim 2, wherein The imaging processing platform arranges the tumor volume values at different time points in chronological order and draws a tumor volume-time change curve; Define the time as t, and the fitting equation is V = at + b, where a is the regression coefficient and b is the intercept; The calculation methods of a and b are: where t i is the i-th time point, n is the number of time points, is the average value of time, is the average value of tumor volume; the change trend of tumor volume is judged according to the regression coefficient a. If a > 0, it means the tumor volume shows an increasing trend; if a < 0, it means the tumor volume shows a decreasing trend; if a ≈ 0, it means the tumor volume is relatively stable.

4. The permanent magnet magnetic resonance imaging processing system according to claim 1, wherein The lesion index includes a contrast lesion index, a trend lesion index, a correlation lesion index, and a prediction lesion index; The lesion analysis methods include a contrast lesion analysis method, a trend lesion analysis method, a depth analysis process strategy based on the lesion index, a correlation lesion analysis method, and a prediction lesion analysis method.

5. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the permanent magnet magnetic resonance imaging processing system according to any one of claims 1 to 4.

Images