Transcranial magnetic stimulation medical image processing system and method

Abstract

The invention relates to the technical field of transcranial magnetic stimulation medical image processing, and discloses a transcranial magnetic stimulation medical image processing system and method.The system comprises an image obtaining module used for collecting multi-modal medical image data and defining image partitions according to clinical diagnosis requirements; the feature analysis module is used for extracting stimulation associated features and constructing a feature mapping relation between the image subareas and the image positioning stimulation areas; the region fusion module is used for screening a core region and analyzing a parameter fusion matching degree to obtain parameter focusing fusion information; the mapping relation updating module is used for updating the feature mapping relation based on the parameter focusing fusion information; and the processing execution module is used for determining a processing target and a process according to the updated mapping relation and generating a processing execution library. The method corresponds to module steps of the system. According to the scheme, the precision and efficiency of transcranial magnetic stimulation medical image processing can be improved, and clinical diagnosis requirements are met.

Description

Technical Field

[0001] The present invention relates to the technical field of transcranial magnetic stimulation medical image processing, and in particular to a transcranial magnetic stimulation medical image processing system and method. Background Art

[0002] Transcranial magnetic stimulation (TMS), a non-invasive neuromodulation technique, is increasingly being used in the diagnosis and treatment of neuropsychiatric disorders. With the advancement of this technology, the acquisition and analysis of multimodal medical imaging data during TMS has become crucial. However, current medical image processing for TMS remains challenging.

[0003] In terms of feature parsing and mapping relationship construction, existing technologies, when extracting stimulus-related features from medical imaging data, struggle to efficiently call the parameters and states corresponding to the data, generate unrelated parameter parsing results, and accurately determine whether they are the target parameter parsing results, thereby determining the image-localized stimulation area. Furthermore, when constructing the feature mapping relationship between image partitions and image-localized stimulation areas, the information in the temporal feature domain, spatial feature location domain, and associated feature association domain of the image partitions, as well as the descriptive information and categories of the image-localized stimulation areas, may not be fully utilized. This results in an incomplete mapping relationship, which in turn affects the subsequent analysis and processing of the image-localized stimulation areas.

[0004] The implementation of the regional fusion module also has shortcomings. Traditional methods, when performing cluster analysis on image partitioning and image localization stimulation areas to screen core regions, may not accurately cluster them according to stimulation type, functional type, and action function, resulting in inaccurate identification of core regions. The calculation of feature similarity between core features and the fusion matching degree between parameters may also be based on unscientific methods, resulting in the calculation results of the fusion matching degree not accurately reflecting the actual mapping relationship between parameters, thereby affecting the acquisition of parameter focus fusion information.

[0005] When processing the update of the feature mapping relationship between the parameter focus fusion information and the image partition and the image localization stimulation area, the mapping relationship update module may not be able to accurately extract the time distribution probability of each parameter, set the target path and perform fitting, resulting in inaccurate calculation of the fusion probability. Therefore, when comparing with the reference of the feature mapping relationship between the image partition and the image localization stimulation area, it is impossible to accurately mark the difference value and classify and update the mapping relationship, affecting the timely update and optimization of the feature mapping relationship.

[0006] When determining processing targets and processes and generating a processing execution library, the processing execution module may be unable to accurately extract processing targets and sort them by probability based on the updated feature mapping relationship. This leads to an irrational formulation of the processing flow and the generated processing execution library failing to meet practical application requirements. These issues severely restrict the accuracy and efficiency of transcranial magnetic stimulation technology in medical image processing, and a more advanced and comprehensive medical image processing system and method are urgently needed to address them. Summary of the Invention

[0007] The object of the present invention is to provide a medical image processing system and method for transcranial magnetic stimulation to solve the problems raised in the above background technology.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a medical image processing system for transcranial magnetic stimulation, the system comprising: An image acquisition module is used to collect multimodal medical image data generated during transcranial magnetic stimulation and define image partitions of the medical image data relative to the neural stimulation area based on clinical diagnostic requirements; Feature parsing module, used to extract stimulus-related features from medical imaging data and construct feature mapping relationships between image partitions and image localization stimulus areas; The regional fusion module is used to select the core area from the image partition and the image localization stimulation area, and analyze the fusion matching degree between multiple parameters in the image localization stimulation area based on the core area to obtain parameter focus fusion information; a mapping relationship updating module, configured to verify information of the image localization stimulation area based on the parameter focus fusion information, identify the update status of the feature mapping relationship between the image partition and the image localization stimulation area, and update the feature mapping relationship between the image partition and the image localization stimulation area according to the update status of the feature mapping relationship; The processing execution module is used to determine the processing target and processing flow of the medical image data based on the updated feature mapping relationship between the image partitions and the image positioning stimulation areas, and generate a processing execution library.

[0009] Preferably, the implementation of the image acquisition module includes: For medical imaging data at any moment during the transcranial magnetic stimulation process, obtain a modality recognition model corresponding to the medical imaging data; Using the modality recognition model to classify the medical imaging data into modalities, at least one modality category is obtained; Identify the temporal features, spatial features, and correlation features in the medical image data under the corresponding modality category to form an image feature set. Medical image data of one modality corresponds to an image feature set containing temporal features, spatial features, and correlation features. The time domain features, the spatial features and the correlation features in the image feature set are analyzed respectively for each individual modality category, and the time domain feature domain, the spatial feature positioning domain and the correlation feature correlation domain corresponding to the image feature set are sequentially obtained and taken as the image partition of the medical image data relative to the neural stimulation region.

[0010] Preferably, the implementation manner of obtaining the time domain feature domain, the spatial feature positioning domain and the correlation feature correlation domain corresponding to the image feature set further comprises: The time domain features, the spatial features and the correlation features in the image feature set are integrated according to the modality categories to obtain a plurality of modality integration results; The time domain feature groups in the modality integration results are extracted, and the time domain feature groups are compared with the feature library to obtain the time domain feature domain; The positioning deviation rate of the spatial features and the correlation tightness of the correlation features in the modality integration results are extracted, and the modality integration results are divided according to the positioning deviation rate of the spatial features and the correlation tightness of the correlation features to obtain the spatial feature positioning domain and the correlation feature correlation domain.

[0011] Preferably, the implementation manner of defining the image partition of the medical image data relative to the neural stimulation region further comprises: The image partition is defined, the image frequency and the image duration of the neural stimulation in the image partition are analyzed, the image partition is fitted according to the image frequency and the image duration, and a mapping relationship from the image partition to the clinical actual demand is constructed.

[0012] Preferably, the implementation manner of the feature analysis module comprises: The parameters and the states corresponding to the medical image data are called to generate a plurality of unassociated parameter analysis results, and the unassociated parameter analysis results represent the parameters and the states that are not associated with the features in the image partition; It is judged whether the plurality of unassociated parameter analysis results are target parameter analysis results, and if so, the target parameter analysis results are taken as the image positioning stimulation area of the medical image data.

[0013] The target parameter analysis results need to meet two core conditions, one is to form a precise match with the time domain feature domain, the spatial feature positioning domain and the correlation feature correlation domain of the image partition, such as the parameter time sequence being consistent with the fluctuation rule of the time domain feature domain, the spatial parameter being coincided with the coordinate range of the positioning domain; the other is to meet the definition standard of the image positioning stimulation area in the clinical diagnosis, including that the parameter value is within the clinical effective range (such as the stimulation frequency and intensity meeting the treatment specification of the corresponding disease), and the anatomical position corresponding to the analysis result is consistent with the clinical target brain area.

[0014] Preferably, the implementation method of constructing the feature mapping relationship between the image partition and the image positioning stimulation area includes: using the information representation of the time domain feature domain, spatial feature positioning domain and associated feature association domain existing in the image partition, the descriptive information of the image positioning stimulation area and the category of the image positioning stimulation area to construct the feature mapping relationship between the image partition and the image positioning stimulation area.

[0015] Preferably, the implementation of the regional fusion module includes: Perform cluster analysis on the image partitions and image localization stimulation areas according to stimulation type, function type, and action function, and set the largest cluster center after cluster analysis as the core area; Extracting core features of the core area, calculating feature similarities between the core features, and setting a common sequence related to the feature similarities between the core features; Using the common sequence related to the feature similarity between the core features, the parameters existing in the common sequence are extracted, and the longest common subsequence between the parameters is set, and the length of the longest common subsequence is set as the fusion matching degree between the parameters; The fusion matching degree between the parameters is set according to the time distribution probability of each parameter, and the parameters are focused on the fusion information.

[0016] Preferably, the implementation of the mapping relationship update module includes: Extracting the time distribution probability of each parameter from the parameter focus fusion information; setting the target path of the parameter focus fusion information according to the time period corresponding to the time distribution probability of each parameter; Fitting the target path of each parameter in the parameter focus fusion information to obtain a fitted target path, and setting the probability value of the fitted target path in each time period as the fusion occurrence probability of the parameter focus fusion information; The fusion occurrence probability of the parameter focus fusion information is compared with the reference situation of the feature mapping relationship between the image partition and the image positioning stimulation area, and the difference value that appears is marked. The difference value is used to reflect the update status of the feature mapping relationship between the image partition and the image positioning stimulation area. The feature mapping relationship between the image partition and the image positioning stimulation area is classified according to the difference value that appears, and the update of the feature mapping relationship between the image partition and the image positioning stimulation area is completed.

[0017] Preferably, the processing execution module is implemented as follows: Based on the updated feature mapping relationship between the image partitions and the image localization stimulation areas, the processing targets in the medical image data are extracted, and the processing targets are sorted according to the occurrence probability of the processing targets to obtain the processing flow in the medical image data; the processing targets and processing flows are combined in a structured form to obtain a processing execution library.

[0018] Preferably, the present invention further includes a transcranial magnetic stimulation medical image processing method, which is applied to the above-mentioned transcranial magnetic stimulation medical image processing system, and the method comprises the following steps: Collect multimodal medical imaging data generated during transcranial magnetic stimulation and define the image partitions of the medical imaging data relative to the neural stimulation area based on clinical diagnostic needs; Extract stimulus-related features from medical imaging data and construct feature mapping relationships between image partitions and image-localized stimulus areas; The core area is selected from the image partition and the image localization stimulation area, and the fusion matching degree between multiple parameters in the image localization stimulation area is analyzed based on the core area to obtain parameter focus fusion information; Based on the parameter focus fusion information, the image localization stimulation area is verified, the update status of the feature mapping relationship between the image partition and the image localization stimulation area is identified, and the feature mapping relationship between the image partition and the image localization stimulation area is updated according to the update status of the feature mapping relationship; Based on the updated feature mapping relationship between the image partitions and the image localization stimulation areas, the processing objectives and processing flow of the medical image data are determined, and a processing execution library is generated.

[0019] Compared with the prior art, the present invention has the following beneficial effects: In the image acquisition link, the corresponding modal recognition model is obtained by targeting the medical imaging data at any moment during the transcranial magnetic stimulation process, and the data is modally classified. The time domain, space and correlation features are identified to form an image feature set, and further analysis is performed to obtain the time domain feature domain, spatial feature positioning domain and correlation feature correlation domain as image partitions. At the same time, the image frequency and duration of neural stimulation in the image partitions are analyzed, and a mapping relationship to actual clinical needs is constructed. This can accurately define image partitions according to clinical diagnostic needs, make the partitions more in line with the actual neural stimulation areas, and provide accurate basic data for subsequent processing.

[0020] The feature parsing module calls the parameters and status of medical imaging data to generate uncorrelated parameter parsing results, determines whether it is the target result to determine the image localization stimulation area, and constructs a feature mapping relationship based on the time domain, space, and correlation feature domain information of the image partition and the descriptive information and category of the image localization stimulation area, thereby achieving efficient extraction and accurate mapping of stimulus-related features, ensuring that the determination of the image localization stimulation area and the construction of the mapping relationship are scientific and reasonable, and providing a reliable basis for subsequent analysis.

[0021] The regional fusion module clusters the image partition and the image positioning stimulation area according to stimulation types, function types and action functions, determines a core area, extracts core features, calculates similarity, sets a common sequence and a longest common subsequence to determine the fusion matching degree between parameters, and then obtains parameter focused fusion information. This way can accurately screen the core area, scientifically analyze the fusion matching degree between parameters, make the parameter focused fusion information more in line with the actual situation, and improve the accuracy of analyzing the parameter mapping relationship in the image positioning stimulation area.

[0022] The mapping relationship updating module extracts the time distribution probability of the parameters from the parameter focused fusion information, sets a target path and fitting, obtains a fusion occurrence probability, compares the fusion occurrence probability with the reference situation of the feature mapping relationship between the image partition and the image positioning stimulation area, marks a difference value and classifies and updates the mapping relationship, realizes dynamic updating and optimization of the feature mapping relationship between the image partition and the image positioning stimulation area, ensures the timeliness and accuracy of the feature mapping relationship, and makes the system better adapt to changes in the transcranial magnetic stimulation process.

[0023] The processing execution module extracts processing targets based on the updated feature mapping relationship, sorts the processing targets according to occurrence probabilities to obtain a processing flow, and combines and generates a processing execution library in a structured form, so as to ensure that the processing targets and the flow are determined to be scientific and reasonable, the generated processing execution library can efficiently guide processing of medical image data, and the processing efficiency and accuracy are improved to meet actual needs of clinical diagnosis. In summary, through the cooperative work of the modules, the precision, efficiency and adaptability of the transcranial magnetic stimulation medical image processing are comprehensively improved, and stronger technical support is provided for diagnosis and treatment of related diseases. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 A working principle diagram of the transcranial magnetic stimulation medical image processing system and method is shown. Figure 2 A working principle diagram of the image acquisition module is shown. Figure 3 A working principle diagram of the regional fusion module is shown. Figure 4 A working principle diagram of the mapping relationship updating module is shown. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0026] Please refer to Figures 1-4The present invention relates to a transcranial magnetic stimulation medical image processing system, the core components of which include an image acquisition module, a feature analysis module, a region fusion module, a mapping relationship update module, and a processing execution module. Specific implementation methods are as follows: An image acquisition module is used to collect multimodal medical image data generated during transcranial magnetic stimulation and define image partitions of the medical image data relative to the neural stimulation area based on clinical diagnostic requirements; Feature parsing module, used to extract stimulus-related features from medical imaging data and construct feature mapping relationships between image partitions and image localization stimulus areas; The regional fusion module is used to select the core area from the image partition and the image localization stimulation area, and analyze the fusion matching degree between multiple parameters in the image localization stimulation area based on the core area to obtain parameter focus fusion information; a mapping relationship updating module, configured to verify information of the image localization stimulation area based on the parameter focus fusion information, identify the update status of the feature mapping relationship between the image partition and the image localization stimulation area, and update the feature mapping relationship between the image partition and the image localization stimulation area according to the update status of the feature mapping relationship; The processing execution module is used to determine the processing target and processing flow of the medical image data based on the updated feature mapping relationship between the image partitions and the image positioning stimulation areas, and generate a processing execution library.

[0027] Example 1: The specific implementation of the image acquisition module is as follows: This module is used to collect multimodal medical imaging data generated during transcranial magnetic stimulation and define the image partitioning of the medical imaging data relative to the neural stimulation area based on clinical diagnostic needs. In practical applications, for medical imaging data at any time during transcranial magnetic stimulation, it is necessary to obtain the modality recognition model corresponding to the medical imaging data. The modality recognition model here can be trained by a large amount of labeled multimodal medical imaging data, which can accurately classify different types of medical imaging data.

[0028] The acquired modality recognition model is used to perform modality classification on the medical imaging data, obtaining at least one modality category. For example, common modality categories include magnetic resonance imaging (MRI), functional magnetic resonance imaging (fMRI), positron emission tomography (PET), and computed tomography (CT). Different modality categories can reflect the physiological and pathological characteristics of the neural stimulation area from different perspectives. After completing modality classification, it is necessary to identify the temporal, spatial, and correlation features present in the medical imaging data under the corresponding modality category to form an image feature set.

[0029] Identifying temporal features primarily involves analyzing the temporal patterns of changes in medical imaging data. For example, in fMRI images, the temporal characteristics of neuronal activity are captured by monitoring changes in blood oxygenation-dependent signals, including temporal fluctuations in signal intensity and oscillations at specific frequencies. Identifying spatial features involves determining geometric properties such as the location, shape, and size of tissues and structures in medical images. For example, image segmentation techniques can be used to separate the stimulated region from surrounding tissue, thereby acquiring features such as its spatial coordinates, volume, and shape descriptors. Identifying associative features focuses on the corresponding mapping relationships between different modalities and the strength of the connections between the stimulated region and surrounding tissue. For example, when fusing MRI and PET images, it is necessary to establish a spatial registration mapping relationship between the two, while also analyzing the functional connectivity between the stimulated region and other brain regions.

[0030] After forming the image feature set, the time domain features, spatial features, and correlation features in the image feature set need to be analyzed separately to obtain the time domain feature domain, spatial feature location domain, and correlation feature association domain corresponding to the image feature set in sequence, and use them as image partitions of the medical image data relative to the nerve stimulation area. When analyzing the time domain features, time series analysis methods such as Fourier transform and wavelet transform can be used to convert the time domain features into the frequency domain or time-frequency domain for analysis, thereby obtaining their distribution patterns in the time dimension and determining the time domain feature domain. For example, the frequency component of the time domain feature is obtained through Fourier transform, and then its main oscillation frequency range is determined as part of the time domain feature domain.

[0031] When analyzing spatial features, techniques from image recognition and computer vision, such as edge detection, contour extraction, and 3D reconstruction, are used to precisely determine the location and range of spatial features and obtain a spatial feature localization domain. For example, for a nerve stimulation region in an MRI image, edge detection algorithms are used to extract its boundaries, and 3D reconstruction techniques are then used to construct its specific position and shape in 3D space, thereby determining the spatial feature localization domain.

[0032] When analyzing correlation features, the degree of association and transmission paths are analyzed by calculating phase mapping coefficients and constructing graphical models, forming a correlation domain for the correlation features. For example, when analyzing functional connectivity, the time series phase mapping coefficients between the neural stimulation area and other brain regions are calculated to construct a functional connectivity network. This allows the strength of association and transmission paths between brain regions to be determined, which serves as the content of the correlation domain for the correlation features.

[0033] Furthermore, when defining the image partitions of medical imaging data relative to the neurostimulation area, further analysis of the image partitions is required. Specifically, the frequency and duration of neurostimulation within the image partitions should be analyzed. Frequency refers to the number of times neurostimulation occurs within a given timeframe within that image partition, while duration refers to the duration of each neurostimulation within that image partition.

[0034] Image partitions are fitted according to imaging frequency and duration, constructing a mapping relationship between image partitions and actual clinical needs. A variety of mathematical methods can be used in the fitting process, such as linear regression, nonlinear regression, and neural network fitting. This mapping relationship allows image partitions to be matched to specific clinical diagnostic needs. For example, it can determine which image partitions are associated with specific neurological dysfunction or treatment effects, thereby providing more targeted information for clinical diagnosis.

[0035] In practical applications, this image acquisition module can be integrated with various medical imaging devices, such as MRI scanners and PET scanners, to acquire multimodal medical imaging data in real time during transcranial magnetic stimulation. Furthermore, the module can flexibly adjust the definition criteria and methods of image partitions based on different clinical diagnostic needs to meet the diagnostic and treatment requirements of different diseases. For example, when diagnosing Parkinson's disease, the focus may be on imaging partitions related to the motor cortex; when diagnosing depression, the focus may be on imaging partitions related to the limbic system.

[0036] Through the above series of operations, the image acquisition module can accurately collect multimodal medical imaging data and define reasonable image partitions according to clinical diagnosis needs, providing high-quality data support for subsequent feature analysis, regional fusion and other modules, thereby ensuring that the entire transcranial magnetic stimulation medical image processing system can effectively assist clinical diagnosis and treatment.

[0037] Example 2: The implementation of the feature analysis module is as follows: this module is used to extract the stimulus-related features of medical image data and construct the feature mapping relationship between image partitions and image positioning stimulation areas. In actual operation, the feature analysis module first needs to call the corresponding parameters and states of the medical image data. These parameters and states cover multiple dimensions, such as the parameters of the medical image itself, including the intensity values of each pixel or voxel, the resolution of the image, the noise level, etc.; at the same time, it also includes related parameters in the transcranial magnetic stimulation process, such as the frequency, intensity, pulse width, and stimulation duration of the stimulation, as well as the working state parameters of the stimulation device, such as the position and angle of the coil.

[0038] After calling these parameters and states, the system will process them independently to generate multiple unassociated parameter analysis results. Here, "unassociated" means that these analysis results have not yet been associated with the time domain feature domain, spatial feature positioning domain, and associated feature correlation domain defined in the image partition. For example, when processing the stimulation intensity parameter, only the numerical range or change curve of the parameter may be obtained, without relating it to the spatial position or time sequence features of a specific region in the image partition. Each parameter analysis result is an independent analysis of a single parameter or state, without cross-association with other features.

[0039] The system needs to determine whether these unassociated parameter analysis results are target parameter analysis results. The judgment process is based on pre-set multi-dimensional conditions, which are set according to the clinical diagnosis requirements and the application scenarios of transcranial magnetic stimulation. For example, for the analysis result of the stimulation frequency parameter, the target conditions may include whether the frequency value is within the effective range for treating a certain disease (such as high-frequency stimulation above 10 Hz for treating depression), and whether the change trend conforms to the pre-set stimulation mode (such as continuous and stable frequency output). For the analysis result of the image intensity parameter, the target conditions may involve whether the intensity value exceeds the normal physiological range, whether there are abnormal local high-intensity areas, etc. If a parameter analysis result meets all the pre-set target conditions, it will be considered as an image positioning stimulation area of the medical image data. For example, when a parameter analysis result shows that the stimulation intensity is within a certain treatment range, and there is an abnormal neural activity signal in the corresponding image area, this area will be determined as an image positioning stimulation area.

[0040] When constructing the feature mapping relationship between image partitions and image localization stimulation areas, the information of the existing time domain feature domain, spatial feature localization domain, and association feature association domain in the image partition is required as the basic representation. The time domain feature domain contains the characteristic distribution of medical imaging data in the time dimension, such as the periodic fluctuation pattern of neural activity, the time delay of stimulus response, and other information. The spatial feature localization domain clarifies the specific position and range of each feature in three-dimensional space, such as the coordinate boundary and volume size of the neural stimulation area. The association feature association domain records the connection mapping relationship and mutual influence between different features, such as the functional connection strength between different brain regions and the conduction path of the stimulation signal.

[0041] The description of the image-localized stimulation area includes the anatomical name of the area (such as the dorsolateral prefrontal cortex) and the physiological response characteristics during stimulation (such as the amplitude of changes in blood oxygen levels). The categories of image-localized stimulation areas are divided according to their functional attributes, such as the motor cortex image-localized stimulation area, the sensory cortex image-localized stimulation area, and the limbic system image-localized stimulation area. The system establishes a multidimensional mapping mechanism to match the temporal feature domain information of the image partition with the time-related description information of the image-localized stimulation area. For example, the high-frequency oscillation characteristics of neural activity within a certain time period in the image partition are associated with the stimulation frequency parameters of the image-localized stimulation area during that time period. The coordinate range of the spatial feature localization domain is spatially aligned with the anatomical location of the image-localized stimulation area to ensure the consistency of their positions in three-dimensional space. The functional connectivity strength in the association feature association domain is mapped to the interaction mapping relationship between the image-localized stimulation area and other brain regions. For example, the high connectivity strength characteristics of the image partition with the language center are mapped to the description of the impact of the image-localized stimulation area on language function.

[0042] In practice, this mapping relationship can be achieved by constructing a data table or a graph structure model. In a data table, rows and columns correspond to the features of the image partition and the features of the image localization stimulation area, respectively, and the values ​​in the table indicate the degree of association between the two. In a graph structure model, nodes represent features of the image partition or the image localization stimulation area, and edge weights indicate the strength of the association between the features. For example, for a specific image partition, its spatial feature localization domain identifies a region located in the left dorsolateral prefrontal cortex, the temporal feature domain shows a significant increase in blood oxygenation levels 200ms after stimulation, and the association feature association domain indicates a strong functional connection between the region and the hippocampus. The image localization stimulation area is described as the left dorsolateral prefrontal cortex, the stimulation category is cognitive function modulation, and the stimulation parameters include a stimulation frequency of 10 Hz and a stimulation intensity of 1.5 Tesla. The system then matches the spatial location of the image partition with the anatomical location of the image localization stimulation area, aligns the time points of blood oxygenation elevation in the temporal features with the time series of stimulation parameters, and associates the hippocampal connectivity strength in the association features with the description of the effect of the image localization stimulation area on memory function, thereby establishing a complete feature mapping relationship.

[0043] In practical applications, the feature parsing module needs to work closely with the image acquisition module to ensure that the acquired image feature set is accurate. At the same time, the module also needs to dynamically adjust the judgment conditions of the target parameter parsing results and the construction rules of the feature mapping relationship according to different transcranial magnetic stimulation application scenarios and clinical diagnostic needs. For example, when treating obsessive-compulsive disorder, more attention may be paid to the imaging localization stimulation area related to the orbitofrontal cortex. At this time, the target conditions will focus on the matching of the imaging features of this area with specific stimulation parameters; when treating the sequelae of stroke, more attention may be paid to the remodeling of the motor cortex. The feature mapping relationship will emphasize the association between the imaging localization stimulation area and the imaging features related to motor function.

[0044] Example 3: The specific implementation method of the regional fusion module is as follows: This module is used to screen the core area from the image partition and the image positioning stimulation area, and analyze the fusion matching degree between multiple parameters in the image positioning stimulation area based on the core area to obtain parameter focus fusion information. In actual operation, the regional fusion module needs to cluster the image partition and the image positioning stimulation area according to multiple dimensions. The relevant data are classified according to the three major dimensions of stimulation type, functional type, and action function. Among them, the stimulation type includes different stimulation modes such as single pulse stimulation, repeated pulse stimulation, and burst stimulation. Each stimulation mode has different time parameters and energy release characteristics; the functional type involves the physiological function category affected by neural stimulation, such as motor function, cognitive function, sensory function, emotional regulation function, etc.; the action function refers to the specific effect of stimulation on neural tissue, such as excitatory effect, inhibitory effect, plasticity regulation effect, etc.

[0045] During cluster analysis, a clustering algorithm tailored to the characteristics of medical imaging data and stimulation parameters is used to process the feature vectors of image partitions and image localization stimulation areas. For example, for each image partition, its feature vector may include statistical parameters of the temporal feature domain, geometric parameters of the spatial feature localization domain, and connection strength values ​​of the association feature association domain. For each image localization stimulation area, the feature vector may include stimulation type parameters, functional type identifiers, action function indicators, and relevant anatomical location parameters. By calculating the similarity or distance between these feature vectors, data points with similar characteristics are grouped into the same cluster. After clustering is complete, the largest cluster center—the center of the cluster containing the largest number of data points—is determined and designated as the core region. This core region represents the most representative portion of the image partition and image localization stimulation area, concentrating on the common characteristics of the majority of data points.

[0046] After determining the core region, the core features of the core region need to be extracted. This extraction is based on the core region's feature vector and covers multiple aspects: morphological features, including geometric properties such as the core region's shape descriptor (e.g., circularity and complexity), volume, and surface area; physiological features, including physiological indicators such as the core region's signal intensity, metabolic level, and blood oxygen saturation in medical images; and stimulation parameter features, including parameters such as the stimulation frequency, intensity, pulse width, and duration corresponding to the core region. After extracting these core features, the feature similarity between each core feature is calculated. The method for calculating feature similarity varies depending on the feature type. For numerical features (e.g., stimulation intensity and volume), Euclidean distance or cosine similarity can be used; for categorical features (e.g., stimulation type and functional type), chi-square distance or mutual information can be used.

[0047] According to the feature similarity between each core feature, a common sequence related to the feature similarity between each core feature is set. The common sequence here can be understood as a feature sequence fragment shared by each core feature, or a feature combination that can reflect the common attributes of each core feature. For example, if the core features include a stimulation frequency of 10 Hz, a stimulation type of repeated pulse stimulation, and an excitatory effect, and another core feature includes a stimulation frequency of 12 Hz, a stimulation type of repeated pulse stimulation, and an excitatory effect, then "repeated pulse stimulation" and "excitatory effect" constitute the common sequence part of the two core features.

[0048] Utilizing common sequences related to feature similarity between core features, parameters present in these common sequences are extracted. These parameters may include stimulus type parameters, function type parameters, action function parameters, and other parameters, as well as ancillary parameters related to these parameters. A longest common subsequence is determined between each parameter. This longest common subsequence is defined as the longest subsequence that appears in two or more parameter sequences. The solution employs dynamic programming, constructing a two-dimensional array to record the solutions to the subproblems, ultimately determining the length of the longest common subsequence. This length is then used as the fusion matching degree between the parameters. The numerical value of the fusion matching degree reflects the degree of correlation and fusion potential between the parameters. A larger value indicates greater commonality between the parameters and greater ease of fusion.

[0049] The fusion matching degree between each parameter is used to set the parameter focus fusion information according to the time distribution probability of each parameter. The time distribution probability of each parameter is obtained through statistical analysis, that is, the frequency or probability density of each parameter at different time points during the transcranial magnetic stimulation process is statistically analyzed. For example, the probability of a certain stimulation intensity parameter appearing within 0-10 seconds after the start of stimulation is 0.3, and the probability of appearing within 10-20 seconds is 0.5, etc. According to the time distribution probability, the fusion matching degree is weighted so that the parameters that are more likely to appear in time and their fusion mapping relationship occupy a more important position in the parameter focus fusion information. The parameter focus fusion information obtained in this way takes into account both the intrinsic correlation between the parameters and their temporal distribution characteristics, and can more accurately reflect the actual fusion situation between multiple parameters in the image positioning stimulation area.

[0050] In practical applications, the regional fusion module needs to work closely with the feature analysis module to ensure that the features of the acquired image partitions and image localization stimulation areas are accurate. At the same time, the module also needs to adjust the dimensions of cluster analysis, the extraction method of core features, and the calculation rules of fusion matching degree according to different transcranial magnetic stimulation treatment plans and clinical diagnostic needs. For example, in transcranial magnetic stimulation treatment for depression, more attention may be paid to the emotion regulation function parameters of the image partitions and image localization stimulation areas related to the limbic system. At this time, cluster analysis will focus on the functional type and action function dimension, and the extraction of core features will pay more attention to physiological indicators and stimulation parameters related to emotion regulation; in the treatment of Parkinson's disease, more attention may be paid to the motor function parameters of the image partitions and image localization stimulation areas related to the motor cortex. Cluster analysis and core feature extraction will accordingly revolve around the motor function dimension.

[0051] Example 4: The specific implementation of the mapping relationship update module is as follows: This module is used to verify the information of the image localization stimulation area based on the parameter focus fusion information, update the feature mapping relationship between the image partition and the image localization stimulation area, and update the feature mapping relationship between the image partition and the image localization stimulation area according to the update status of the feature mapping relationship. During actual operation, the mapping relationship update module needs to extract the time distribution probability of each parameter from the parameter focus fusion information. The parameter focus fusion information contains multiple parameters in the image localization stimulation area and their fusion matching information, and the time distribution probability of each parameter reflects the probability of these parameters appearing in different time periods during the transcranial magnetic stimulation process.

[0052] The feature mapping relationship between the image partition and the image localization stimulation area is constructed by a series of operations by the feature analysis module. The feature analysis module calls the parameters and states corresponding to the medical image data, and generates multiple unrelated parameter analysis results that are not associated with the image partition features. These parameters cover the pixel intensity and resolution of the image itself, the frequency and intensity and other process parameters of transcranial magnetic stimulation, and the working status parameters of the equipment. The system determines whether these unrelated parameter analysis results are target parameter analysis results. The target parameter analysis results must be accurately matched with the time domain feature domain, spatial feature localization domain, and associated feature association domain of the image partition, and meet the clinical diagnosis definition standards for the image localization stimulation area, including parameter values ​​within the clinical effective range and anatomical positions consistent with the clinical target brain area. Those that meet the conditions are determined to be image localization stimulation areas. Taking the temporal feature domain, spatial feature positioning domain, and associated feature association domain information in the image partition as representation, combined with the descriptive information and category of the image positioning stimulation area, a multi-dimensional mapping mechanism is established to correspond the image partition characteristics with the image positioning stimulation area characteristics, thereby forming a mapping relationship in the image positioning stimulation area. These mapping relationships cover multiple dimensions such as spatial position correspondence, temporal sequence association, and functional causal relationship.

[0053] For example, in a specific transcranial magnetic stimulation scenario, parameter focus fusion information may involve three parameters: stimulation frequency, stimulation intensity, and blood oxygen level change values ​​in image partitions. The probability of the stimulation frequency parameter appearing at 10 Hz within 0-10 seconds is 0.6, and the probability of appearing at 15 Hz within 10-20 seconds is 0.7; the probability of the stimulation intensity parameter appearing at 1.2 T within 0-10 seconds is 0.5, and the probability of appearing at 1.5 T within 10-20 seconds is 0.8; the probability of the blood oxygen level change value increasing by 0.5% within 0-10 seconds is 0.4, and the probability of increasing by 0.8% within 10-20 seconds is 0.6. These temporal distribution probabilities are obtained through statistical analysis of historical data from multiple transcranial magnetic stimulation processes and can reflect the temporal distribution patterns of the parameters.

[0054] The parameter-focused fusion information is used to set the target path for the time period corresponding to the time distribution probability of each parameter. The target path setting requires comprehensive consideration of the probability distribution of each parameter in different time periods, with the parameter combination with the highest probability being selected as the target path state within that time period. For example, in the 0-10 second time period, if the stimulation frequency is 10 Hz (probability 0.6), the stimulation intensity is 1.2 T (probability 0.5), and the blood oxygen level change value increases by 0.5% (probability 0.4), the high-probability combination of stimulation frequency and stimulation intensity might be selected as the target path segment for that time period. In the 10-20 second time period, if the stimulation frequency is 15 Hz (probability 0.7), the stimulation intensity is 1.5 T (probability 0.8), and the blood oxygen level change value increases by 0.8% (probability 0.6), the high-probability combination of these three parameters would be selected as the target path segment for that time period.

[0055] The target paths of each parameter in the parameter focus fusion information are fitted to obtain the fitted target path. The fitting process can adopt a variety of methods, such as spline curve fitting and polynomial fitting, and the appropriate fitting method is selected according to the temporal distribution characteristics of the parameters. For example, for the stimulation frequency parameter, its target values ​​for 0-10 seconds and 10-20 seconds are 10Hz and 15Hz, respectively. A linear fitting method can be used to obtain a smooth transition curve from 10Hz to 15Hz; for the stimulation intensity parameter, a similar fitting method can be used for the change from 1.2T to 1.5T. The fitted target path can more smoothly reflect the temporal variation trend of the parameter, facilitating subsequent analysis.

[0056] The probability value of the fitted target path in each time period is set as the fusion occurrence probability of the parameter-focused fusion information. The fusion occurrence probability represents the likelihood of the parameter-focused fusion information occurring within that time period and is derived by integrating the temporal distribution probabilities of each parameter within that time period. For example, in the 0-10 second time period, the fusion occurrence probability can be calculated by calculating the weighted average of the probabilities of the three parameters, stimulation frequency, stimulation intensity, and blood oxygen level change, within that time period. The weights are determined based on the importance of each parameter in the parameter-focused fusion information. In the 10-20 second time period, the weighted average of the probabilities of each parameter is similarly calculated as the fusion occurrence probability.

[0057] The fusion occurrence probability of the parameter focus fusion information is compared with the citation of the feature mapping relationship between the image partition and the image localization stimulation area. The citation of the feature mapping relationship between the image partition and the image localization stimulation area can be recorded by establishing a table, which contains information such as the mapping relationship type, mapping relationship description, historical citation frequency, and current association strength. The following is a specific example table: Table 1 Comparison and judgment table of unrelated parameter analysis results and target parameter standards in the feature analysis module

[0058] In this example, the historical citation frequency indicates how often the mapping relationship has been used in past transcranial magnetic stimulation cases, ranging from 0 to 1; the current association strength indicates the degree of association between the mapping relationship and other mapping relationships in the current parameter focus fusion information, also ranging from 0 to 1.

[0059] Compare the fusion probability with the historical citation frequency and current association strength in the table, and mark the difference value that appears. The difference value is used to reflect the updated status of the feature mapping relationship between the image partition and the image localization stimulus area. For example, if the fusion probability is 0.75 in the 0-10 second time period, while the historical citation frequency of "temporal association" is 0.8 and the current association strength is 0.78, the difference value may be small; while the historical citation frequency of "structural association" is 0.6 and the current association strength is 0.55, the difference between the fusion probability and the historical citation frequency may be larger. The calculation method of the difference value can be set according to the specific situation, such as absolute difference, relative difference, etc.

[0060] The feature mapping relationship between the image partition and the image localization stimulation area is classified according to the difference value, and the feature mapping relationship between the image partition and the image localization stimulation area is updated. The classification standard can be set as follows: when the difference value is less than a certain threshold (such as 0.1), the mapping relationship is in a "stable" state and does not need to be updated; when the difference value is between 0.1 and 0.2, the mapping relationship is in a "needs adjustment" state and the association strength needs to be fine-tuned; when the difference value is greater than 0.2, the mapping relationship is in a "needs reconstruction" state and the mapping relationship needs to be re-established.

[0061] For example, in the above example, the difference value of "structural association" may be greater than 0.2, and it is classified as a "requires reconstruction" state. At this time, it is necessary to re-analyze the boundary overlap mapping relationship between the image partition and the image positioning stimulation area, and the mapping relationship may be updated by re-image segmentation, adjusting spatial registration parameters, etc.; the difference value of "spatial association" may be between 0.1 and 0.2, and it is classified as a "requires adjustment" state, and its association strength needs to be fine-tuned; while the difference values ​​of "temporal association" and "functional association" are smaller, and are classified as a "stable" state, and remain unchanged.

[0062] In practical applications, the mapping relationship update module needs to work closely with the regional fusion module to ensure that the obtained parameter focus fusion information results are accurate. At the same time, the module also needs to adjust the calculation method of the fusion probability, the calculation standard of the difference value, and the classification threshold of the mapping relationship according to different transcranial magnetic stimulation treatment plans and clinical diagnosis needs. For example, when treating depression, more attention may be paid to "functional association" and "temporal association", and the difference value thresholds of these mapping relationships may be set more strictly; when treating stroke sequelae, more attention may be paid to "spatial association" and "structural association", and the thresholds and classification standards may be adjusted accordingly.

[0063] Through these operations, the mapping relationship update module can verify the information of the image localization stimulation area based on the parameter focus fusion information, identify the mapping relationship that needs to be updated, and update the feature mapping relationship between the image partition and the image localization stimulation area according to the updated status. These updated mapping relationships can more accurately reflect the true relationship between the image data and the image localization stimulation area during the transcranial magnetic stimulation process, providing a more reliable basis for subsequent processing execution modules, thereby improving the adaptability and accuracy of the entire medical image processing system and better assisting clinical diagnosis and treatment.

[0064] During implementation, additional parameters and mapping relationship types can be combined for analysis, such as the location parameters of the stimulation coil and the connection mapping relationships of nerve fiber bundles, making the mapping relationship updates more comprehensive and accurate. Furthermore, as transcranial magnetic stimulation case studies continue to accumulate, historical citation frequency and association strength data will be continuously updated, allowing the mapping relationship update module to dynamically adapt to new situations and continuously optimize feature mapping relationships.

[0065] Example 5: The specific implementation of the processing execution module is as follows: This module is used to determine the processing objectives and processing flow of medical image data based on the updated feature mapping relationship between the image partitions and the image localization stimulation areas, and to generate a processing execution library. In practical applications, the operation of the processing execution module must be based on the updated feature mapping relationship, which includes the corresponding relationship between the temporal, spatial, and correlation characteristics of the image partitions and the parameters, functions, and anatomical information of the image localization stimulation areas.

[0066] Extract processing targets from medical imaging data based on the updated feature mapping relationship. The types of processing targets cover multiple dimensions, such as image preprocessing targets (such as noise reduction and artifact removal), feature extraction targets (such as lesion identification and functional area positioning), stimulation effect evaluation targets (such as neural response analysis and treatment effect prediction), etc. Taking a specific transcranial magnetic stimulation scenario as an example, suppose that the updated feature mapping relationship shows that the blood oxygen level in the left dorsolateral prefrontal region in the image partition is significantly increased under 10Hz stimulation, and this region has a strong functional connection with the default network. At this time, the processing targets may include: denoising the fMRI images of this region, extracting the blood oxygen change characteristics of this region, and evaluating the regulatory effect of 10Hz stimulation on the default network.

[0067] After extracting the processing targets, they need to be sorted according to their probability of occurrence to determine the processing flow of the medical imaging data. The probability of occurrence of the processing targets is determined by a combination of multiple factors, including the urgency of the clinical diagnosis needs, the execution frequency of the target in historical cases, and the feature matching degree of the current medical imaging data. For example, in the treatment scenario of depression, clinical diagnosis pays more attention to the impact of stimulation on brain areas related to emotion regulation. If the current imaging data shows that the image-localized stimulation area is strongly associated with the amygdala, the probability of occurrence of the target "assessing the impact of stimulation on amygdala activity" will be higher. The following is an example table of processing targets and their probability of occurrence: Table 2. Difference value calculation and classification update strategy of feature mapping relationship of image localization stimulation area

[0068] In this example, the clinical need weight is set based on the standard treatment plan for depression. The historical execution frequency is based on statistics from 100 similar cases. The feature match is determined by the degree of correlation between the current imaging data and the target features. The overall occurrence probability is calculated through a weighted calculation, for example: Overall occurrence probability = clinical need weight × 0.4 + historical execution frequency × 0.3 + feature match × 0.3.

[0069] The processing objectives are sorted according to their combined probability of occurrence to form a processing flow. In the above example, the processing flow is as follows: fMRI image motion correction (probability 0.75) → feature extraction of blood oxygen level changes in the image-localized stimulation area (probability 0.82) → analysis of the impact of stimulation on default network functional connectivity (probability 0.81) → visualization of three-dimensional connections between the image-localized stimulation area and related brain regions (probability 0.58). It should be noted that the ordering of the processing flow must take into account the dependency mapping between objectives. For example, "feature extraction" should be performed after "image preprocessing." If an objective with a high probability of occurrence depends on an objective with a low probability, the execution order should be adjusted.

[0070] After determining the processing objectives and process flow, combine them in a structured format to generate a processing execution library. This structured format uses a standardized data format, such as JSON or XML, to ensure data storability and callability. The following is an example of the JSON format for a processing execution library: {"Processing Execution Library Identifier":"TMS_Processing_20250622","Processing Target List":[{"Target ID":"T001","Target Description":"Motion Correction of fMRI Images","Target Category":"Image Preprocessing","Execution Probability":0.75,"Dependent Targets":[]},{"Target ID":"T002","Target Description":"Feature Extraction of Blood Oxygen Level Changes in Image-Based Stimulation Areas","Target Category":"Feature Extraction","Execution Probability":0.82,"Dependent Targets":["T001"]},{"Target ID":"T003","Target Description":"Analysis of the Effect of Stimulation on Default Network Functional Connectivity"},{"Target ID":"T004","Target Description":"Visualization of the three-dimensional connection between the image-localized stimulation area and related brain areas","Target Category":"Visualization","Execution Probability":0.58,"Dependent Targets":["T002","T003"]}],"Processing Procedure":["T001","T002","T003","T004"],"Associated Parameters":{"Stimulation Frequency":"10Hz","Stimulation Intensity":"1.5T","Imaging Modality":"fMRI","Imaging Localized Stimulation Area":"Left Dorsolateral Prefrontal Cortex"}} In the above JSON structure, each processing target contains a unique identifier, description, category, execution probability, and dependency mapping. The processing flow lists the target IDs in order, and the associated parameters record key information about the current transcranial magnetic stimulation session. This structured format facilitates system calls for processing targets and processes, and also facilitates subsequent maintenance and updates.

[0071] In practical applications, the processing execution module needs to work closely with the mapping relationship update module to ensure that processing targets and processes are generated based on the latest feature mapping relationships. At the same time, this module must also dynamically adjust the extraction rules for processing targets, the calculation method for occurrence probability, and the structured parameter configuration based on different clinical application scenarios (such as depression, Parkinson's disease, and stroke treatment) and medical imaging modalities (such as fMRI, PET, and MRI). For example, in Parkinson's disease treatment, the processing objectives may focus more on feature extraction of the motor cortex and assessment of the impact of stimulation on motor function; in PET image processing, the processing objectives may include analysis of the distribution of radioactive tracers.

[0072] Furthermore, the generated processing library is not static; it can be iteratively optimized based on feedback from actual processing. For example, if the current algorithm for correcting fMRI image head motion is found to be ineffective, the execution parameters for that target can be manually adjusted or the algorithm can be replaced, and the processing library can be updated. This dynamic optimization mechanism improves the adaptability and effectiveness of the processing library.

[0073] Through the above steps, the processing execution module transforms abstract feature mapping relationships into specific processing objectives and executable processes, generating a structured processing execution library. This library provides clear operational guidelines for transcranial magnetic stimulation medical image processing, making the entire processing process operational and traceable, thereby assisting clinicians in more efficient diagnosis and treatment planning. During implementation, the processing execution library can also be integrated with the hospital's picture archiving and communication system (PACS) and electronic medical record system (EMR) to achieve full-process management of medical imaging data and processing results.

[0074] It should be noted that, in this document, mapping terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual mapping relationship or order between these entities or operations. Moreover, the terms "include," "comprise," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0075] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A medical image processing system for transcranial magnetic stimulation, characterized in that: include: An image acquisition module is used to collect multimodal medical image data generated during transcranial magnetic stimulation and define image partitions of the medical image data relative to the neural stimulation area based on clinical diagnostic requirements; Feature parsing module, used to extract stimulus-related features from medical imaging data and construct feature mapping relationships between image partitions and image localization stimulus areas; The regional fusion module is used to select the core area from the image partition and the image localization stimulation area, and analyze the fusion matching degree between multiple parameters in the image localization stimulation area based on the core area to obtain parameter focus fusion information; a mapping relationship updating module, configured to verify information of the image localization stimulation area based on the parameter focus fusion information, identify the update status of the feature mapping relationship between the image partition and the image localization stimulation area, and update the feature mapping relationship between the image partition and the image localization stimulation area according to the update status of the feature mapping relationship; The processing execution module is used to determine the processing target and processing flow of the medical image data based on the updated feature mapping relationship between the image partitions and the image positioning stimulation areas, and generate a processing execution library.

2. The medical image processing system for transcranial magnetic stimulation according to claim 1, characterized in that: The implementation methods of the image acquisition module include: For medical imaging data at any moment during the transcranial magnetic stimulation process, obtain a modality recognition model corresponding to the medical imaging data; Using the modality recognition model to classify the medical imaging data into modalities, at least one modality category is obtained; Identify the temporal features, spatial features, and correlation features in the medical image data under the corresponding modality category to form an image feature set. Medical image data of one modality corresponds to an image feature set containing temporal features, spatial features, and correlation features. Each individual modality category is processed separately, and the time domain features, spatial features and correlation features in the image feature set are analyzed respectively. The time domain feature domain, spatial feature positioning domain and correlation feature correlation domain corresponding to the image feature set are obtained in turn, and used as the image partition of the medical image data relative to the nerve stimulation area.

3. The medical image processing system for transcranial magnetic stimulation according to claim 2, characterized in that: The implementation methods for obtaining the temporal feature domain, spatial feature location domain, and correlation feature association domain corresponding to the image feature set include: The temporal features, spatial features and correlation features in the image feature set are integrated according to the modality category to obtain multiple modality integration results; Extract the time domain feature group from the modal integration result, compare the time domain feature group with the feature library, and obtain the time domain feature domain; The positioning deviation rate of spatial features and the correlation density of associated features in the modal integration results are extracted. The modal integration results are divided according to the positioning deviation rate of spatial features and the correlation density of associated features to obtain the spatial feature positioning domain and the correlation feature correlation domain.

4. The medical image processing system for transcranial magnetic stimulation according to claim 1, characterized in that: The image acquisition module is also used to: Define the image partitions, analyze the image frequency and image duration of nerve stimulation in the image partitions, fit the image partitions according to the image frequency and image duration, and construct a mapping relationship between the image partitions and actual clinical needs.

5. The transcranial magnetic stimulation medical image processing system according to claim 1, characterized in that: The implementation of the feature parsing module includes: Calling parameters and states corresponding to medical image data to generate multiple unrelated parameter analysis results, where the unrelated parameter analysis results represent parameters and states that are not associated with features in image partitions; It is determined whether the multiple unrelated parameter analysis results are consistent with the target parameter analysis result. If they are consistent with the target parameter analysis result, the target parameter analysis result is regarded as the image positioning stimulation area of ​​the medical image data.

6. The transcranial magnetic stimulation medical image processing system according to claim 3, characterized in that: The implementation method of constructing the feature mapping relationship between the image partition and the image positioning stimulation area includes: using the information of the time domain feature domain, the spatial feature positioning domain and the associated feature association domain existing in the image partition as representation, combining the description information of the image positioning stimulation area and the category of the image positioning stimulation area, and constructing the feature mapping relationship between the image partition and the image positioning stimulation area.

7. The medical image processing system for transcranial magnetic stimulation according to claim 1, characterized in that: The implementation methods of the regional fusion module include: Perform cluster analysis on the image partitions and image localization stimulation areas according to stimulation type, function type, and action function, and set the largest cluster center after cluster analysis as the core area; Extracting core features of the core area, calculating feature similarities between the core features, and setting a common sequence related to the feature similarities between the core features; Using the common sequence related to the feature similarity between the core features, the parameters existing in the common sequence are extracted, and the longest common subsequence between the parameters is set, and the length of the longest common subsequence is set as the fusion matching degree between the parameters; The fusion matching degree between each parameter is set according to the time distribution probability of each parameter, and the parameters are focused on the fusion information.

8. The medical image processing system for transcranial magnetic stimulation according to claim 7, characterized in that: The implementation of the mapping relationship update module includes: Extracting the time distribution probability of each parameter from the parameter focus fusion information; setting the target path of the parameter focus fusion information according to the time period corresponding to the time distribution probability of each parameter; Fitting the target path of each parameter in the parameter focus fusion information to obtain a fitted target path, and setting the probability value of the fitted target path in each time period as the fusion occurrence probability of the parameter focus fusion information; The fusion occurrence probability of the parameter focus fusion information is compared with the reference situation of the feature mapping relationship between the image partition and the image positioning stimulation area, and the difference value that appears is marked. The difference value is used to reflect the update status of the feature mapping relationship between the image partition and the image positioning stimulation area. The feature mapping relationship between the image partition and the image positioning stimulation area is classified according to the difference value that appears, and the update of the feature mapping relationship between the image partition and the image positioning stimulation area is completed.

9. The transcranial magnetic stimulation medical image processing system according to claim 1, characterized in that: The implementation of the processing execution module is as follows: Based on the updated feature mapping relationship between the image partitions and the image localization stimulation areas, the processing targets in the medical image data are extracted, and the processing targets are sorted according to the occurrence probability of the processing targets to obtain the processing flow in the medical image data; The processing targets and processing flows are combined in a structured form to obtain a processing execution library.

10. A transcranial magnetic stimulation medical image processing method, applied to the transcranial magnetic stimulation medical image processing system according to any one of claims 1 to 9, characterized in that: The following steps are involved: Collect multimodal medical imaging data generated during transcranial magnetic stimulation and define the image partitions of the medical imaging data relative to the neural stimulation area based on clinical diagnostic needs; Extract stimulus-related features from medical imaging data and construct feature mapping relationships between image partitions and image-localized stimulus areas; The core area is selected from the image partition and the image localization stimulation area, and the fusion matching degree between multiple parameters in the image localization stimulation area is analyzed based on the core area to obtain parameter focus fusion information; Based on the parameter focus fusion information, the image localization stimulation area is verified, the update status of the feature mapping relationship between the image partition and the image localization stimulation area is identified, and the feature mapping relationship between the image partition and the image localization stimulation area is updated according to the update status of the feature mapping relationship; Based on the updated feature mapping relationship between the image partitions and the image localization stimulation areas, the processing objectives and processing flow of the medical image data are determined, and a processing execution library is generated.

Images