Repeated transcranial magnetic stimulation intervention system

Abstract

The invention relates to the technical field of repeated transcranial magnetic stimulation, and particularly discloses a repeated transcranial magnetic stimulation intervention system, which comprises a brain region initial positioning data acquisition module, a basic near infrared spectrum data acquisition module, an updated infrared spectrum data acquisition module and a target brain region positioning detection module, the positioning module is used for determining whether the positioning accuracy of the target brain region meets the requirement or not based on comparison between the updated infrared spectrum data and the basic near infrared spectrum data; and a positioning adjustment module. Therefore, the positioning accuracy of the target brain region can be automatically judged and whether deviation occurs or not can be detected on the basis of infrared spectrum semantics and brain region hemodynamic changes before and after rTMS stimulation, and a positioning adjustment prompt signal is sent out when deviation occurs; in this way, it can be ensured that the positioning accuracy of the target brain area can be ensured in each time of stimulation in an automatic mode in the repeated transcranial magnetic stimulation process, and then the intelligent level of the repeated transcranial magnetic stimulation intervention system is improved.

Description

[0001] Repetitive transcranial magnetic stimulation intervention system Technical Field

[0002] The present application relates to the technical field of repetitive transcranial magnetic stimulation, and more specifically, to a repetitive transcranial magnetic stimulation intervention system. Background Art

[0003] Repetitive Transcranial Magnetic Stimulation (rTMS) is a non-invasive brain stimulation method that affects the activity of neurons in the cerebral cortex by changing the magnetic field above the scalp. rTMS has shown potential in treating neurological diseases such as depression and Parkinson's disease, and has been used to study the relationship between brain function and structure. However, the effect of rTMS is highly dependent on whether the target brain area can be accurately located. If the stimulation location is inaccurate, the desired therapeutic purpose may not be achieved.

[0004] Traditional methods of locating the target brain area are usually based on previously acquired imaging data (such as MRI) to set the position of the coil before treatment. However, during the actual treatment process, the patient's head posture may change, which will cause the predetermined stimulation position to deviate from the target brain area. In addition, the brain is a dynamic organ, and its internal structure and activity patterns will also change over time, and static positioning cannot respond to this. In other words, most rTMS positioning technologies lack a real-time monitoring mechanism, which means that even if the initial positioning is correct, any changes during the treatment process (such as patient movement or changes in brain state) may affect the final stimulation effect. Summary of the invention

[0005] The present application provides a repetitive transcranial magnetic stimulation intervention system, which can automatically determine the positioning accuracy of the target brain area and detect whether a deviation occurs based on the infrared spectrum semantics and brain area hemodynamic changes before and after rTMS stimulation, and send a positioning adjustment prompt signal when a deviation occurs. In this way, it can ensure the positioning accuracy of the target brain area in each stimulation in an automated manner during the repetitive transcranial magnetic stimulation process, thereby improving the intelligence level of the repetitive transcranial magnetic stimulation intervention system.

[0006] In a first aspect, a repetitive transcranial magnetic stimulation intervention system is provided, comprising: A brain region initial positioning data acquisition module is used to record the initial positioning data of the target brain region; A basic near-infrared spectral data acquisition module, used to acquire basic near-infrared spectral data of the target brain area without any stimulation using a NIRS monitoring device; An updated infrared spectrum data acquisition module is used to control the rTMS device to stimulate the target brain area, and at the same time, continuously monitor the near-infrared spectrum changes of the target brain area using the NIRS device to obtain updated infrared spectrum data; A target brain region positioning detection module, used to determine whether the positioning accuracy of the target brain region meets the requirements based on the comparison between the updated infrared spectrum data and the basic near-infrared spectrum data; The positioning adjustment module is used to generate a target brain area positioning adjustment prompt in response to the positioning accuracy of the target brain area not meeting the requirements.

[0007] The present application provides a repetitive transcranial magnetic stimulation intervention system, which processes basic near-infrared spectral data and updated infrared spectral data together through a data processing and analysis algorithm based on artificial intelligence and signal processing, so as to capture the basic near-infrared spectral image semantics and updated near-infrared spectral image semantics, as well as the fine-grained contrast feature representation information between the two, so as to judge the positioning accuracy of the target brain area to determine whether the positioning accuracy of the target brain area meets the requirements. In this way, the positioning accuracy of the target brain area can be automatically judged based on the infrared spectral semantics and brain area hemodynamic changes before and after rTMS stimulation, and whether a deviation occurs, and a positioning adjustment prompt signal is issued when a deviation occurs, so that each stimulation can ensure the positioning accuracy of the target brain area in an automated manner during the repetitive transcranial magnetic stimulation process, thereby improving the intelligence level of the repetitive transcranial magnetic stimulation intervention system. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0009] Figure 1 This is a schematic block diagram of a repetitive transcranial magnetic stimulation intervention system according to an embodiment of the present application.

[0010] Figure 2 This is a schematic block diagram of a repetitive transcranial magnetic stimulation intervention system according to an embodiment of the present application.

[0011] Figure 3 This is a schematic diagram of the data flow of the target brain area positioning detection module in the repetitive transcranial magnetic stimulation intervention system of an embodiment of the present application.

[0012] Figure 4 This is a schematic block diagram of a near-infrared spectral dominant feature enhancement processing unit in a repetitive transcranial magnetic stimulation intervention system according to an embodiment of the present application.

[0013] Figure 5This is a schematic block diagram of a near-infrared spectroscopy interactive encoding unit in a repetitive transcranial magnetic stimulation intervention system according to an embodiment of the present application.

[0014] Figure 6 This is a schematic block diagram of a near-infrared spectroscopy fine-grained semantic shared compensation subunit in a repetitive transcranial magnetic stimulation intervention system according to an embodiment of the present application. DETAILED DESCRIPTION

[0015] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0016] Near-Infrared Spectroscopy (NIRS) is a technology that can measure changes in hemoglobin concentration in brain tissue, thereby reflecting the metabolic activity of brain regions. NIRS has high temporal resolution and good spatial resolution, and can monitor hemodynamic changes during brain activity in real time. This immediacy is crucial to ensure the accuracy of the stimulation location, so it has unique advantages in evaluating the effects of rTMS on specific brain regions and localizing the target brain regions.

[0017] Based on this, in the technical solution of the present application, a repetitive transcranial magnetic stimulation intervention system is proposed, which aims to improve the accuracy of positioning the target brain area during rTMS intervention. By combining NIRS monitoring with rTMS stimulation, the system can compare and analyze the physiological responses of the target brain area before and after the stimulation to ensure the correctness of the stimulation position. Specifically, the system records the initial positioning data of the target brain area, and uses the NIRS device to obtain the basic near-infrared spectral data in the non-stimulated state as a reference. Then, the updated NIRS data is continuously monitored and collected during the rTMS stimulation process. By comparing these two sets of data, it can be determined whether the brain area actually stimulated is consistent with the predetermined target area and whether the position of the stimulated target brain area deviates.

[0018] Specifically, Figure 1As shown, the repetitive transcranial magnetic stimulation intervention system 1 includes: a brain area initial positioning data acquisition module 10, which is used to record the initial positioning data of the target brain area; a basic near-infrared spectrum data acquisition module 20, which is used to use a NIRS monitoring device to collect basic near-infrared spectrum data of the target brain area without any stimulation; an updated infrared spectrum data acquisition module 30, which is used to control the rTMS device to stimulate the target brain area, and at the same time use the NIRS device to continuously monitor the near-infrared spectrum changes of the target brain area to obtain updated infrared spectrum data; a target brain area positioning detection module 40, which is used to determine whether the positioning accuracy of the target brain area meets the requirements based on the comparison between the updated infrared spectrum data and the basic near-infrared spectrum data; a positioning adjustment module 50, which is used to generate a target brain area positioning adjustment prompt in response to the positioning accuracy of the target brain area not meeting the requirements.

[0019] Exemplarily, in the brain region initial positioning data acquisition module 10, the initial positioning data of the target brain region is recorded. It should be understood that recording the initial positioning data of the target brain region is crucial for repetitive transcranial magnetic stimulation (rTMS) treatment because it ensures that the stimulation can accurately act on the predetermined target brain region, which is essential for achieving effective treatment results and reducing potential side effects. Since the brain is a dynamic organ, its internal structure and activity patterns change over time, and static positioning methods may not be able to respond to these changes. Therefore, recording the initial positioning data not only provides a precise starting point for treatment, but also allows real-time monitoring and adjustment of any position deviations during treatment, thereby maintaining the consistency and effectiveness of each stimulation.

[0020] In one embodiment, the initial positioning data of the target brain area is recorded, including: first, by using structural imaging technology such as MRI (magnetic resonance imaging), high-resolution images of the patient's brain can be obtained, providing detailed anatomical information to help accurately locate the brain area that needs to be stimulated. In order to ensure the consistency of positioning between different patients, a standard brain coordinate system (such as MNI space or Talairach space) is often used to map the individualized MRI scan results into a common spatial framework to achieve cross-individual comparison and positioning. In addition, the technician will set several reference points on the patient's scalp and associate these reference points with the corresponding positions on the MRI image through special equipment (such as laser pointers, robotic arms, etc.). This step ensures that the position of the rTMS coil can be quickly recalibrated even if the patient's head moves slightly. Sometimes electrophysiological signals such as EEG (electroencephalogram) are combined as an auxiliary means, especially when looking for cortical locations associated with specific functions. EEG activity patterns during certain cognitive tasks can help identify the brain area responsible for performing the task.

[0021] Exemplarily, in the basic near-infrared spectral data acquisition module 20, the basic near-infrared spectral data of the target brain area without any stimulation is collected using a NIRS monitoring device. It should be understood that using a NIRS (near-infrared spectral imaging) monitoring device to collect basic near-infrared spectral data of the target brain area without any stimulation is crucial to ensuring the accuracy and effectiveness of repetitive transcranial magnetic stimulation (rTMS) treatment. This process not only provides a reliable benchmark for the subsequent rTMS stimulation effect, but also helps to improve positioning accuracy, monitor brain area activity, and reduce the impact of interference factors. Specifically, the basic NIRS data collected without any external stimulation can be used as a benchmark to compare the changes after subsequent rTMS stimulation. This comparison helps to evaluate the impact of rTMS on specific brain areas and confirm whether the expected effect has been achieved. At the same time, by analyzing physiological parameters such as hemoglobin concentration and oxygenation level in the resting state, the location of the target brain area can be more accurately determined, providing an important reference for the placement of the rTMS coil, thereby improving the accuracy of the stimulation position.

[0022] In addition, NIRS can monitor hemodynamic changes during brain activity in real time, reflecting the metabolic activity of brain regions. This information can help identify which areas are more active in their natural state and which may require intervention, providing support for personalized treatment plans. Because the brain is a dynamic organ, its internal structure and activity patterns change over time. Basic NIRS data continuously tracks changes in brain regions throughout the course of treatment and provides a stable reference point, helping the system to respond flexibly to these changes and ensure that each stimulation acts on the correct brain area. Data obtained under non-stimulation conditions excludes the influence of external intervention, making subsequent comparisons more reliable and able to more clearly distinguish specific changes caused by rTMS rather than the results of other variables.

[0023] The process of collecting basic near-infrared spectroscopy data of the target brain area without any stimulation first requires ensuring that the patient is in a relaxed state and avoiding any factors that may cause abnormal brain activity, such as tension or thinking about complex problems. Patients are usually required to remain quiet and close their eyes for a few minutes to obtain data that is closest to the natural state. The NIRS monitoring device is then properly worn on the patient's head, ensuring that the sensor (light source and detector) is located above the target brain area and is properly calibrated to ensure the accuracy of the measurement results. Use a standardized coordinate system (such as MNI or Talairach space) to correspond the position of the NIRS probe to the position of the target brain area previously determined by MRI or other imaging techniques to achieve precise spatial positioning.

[0024] Before starting the recording, allow the device to have a warm-up period to stabilize the signal, and then start the data acquisition program to continuously monitor the changes in hemoglobin concentration and other related parameters in the target brain area for a certain period of time. During the acquisition process, ensure that the environmental conditions are constant, such as light intensity, temperature, etc., to reduce the interference of external factors on the measurement results. After the data acquisition is completed, a specially designed software platform is used to process the raw data, remove noise and extract useful information, such as calculating the changes in light absorption at different wavelengths, and then inferring the changes in hemoglobin concentration. The basic NIRS data obtained from the analysis can be further used for feature extraction to generate image semantic encoding feature maps for subsequent comparison with the data collected after stimulation. All basic NIRS data collected should be properly stored and marked with relevant information, such as acquisition date, time, patient ID, etc., for future reference and analysis. These data will serve as an important reference during rTMS treatment to help determine whether the actual stimulated brain area is consistent with the predetermined target area and whether the location of the stimulated target brain area deviates.

[0025] Exemplarily, in the updated infrared spectrum data acquisition module 30, the rTMS device is controlled to stimulate the target brain area, and the NIRS device is used to continuously monitor the near-infrared spectrum changes of the target brain area to obtain updated infrared spectrum data. It should be understood that controlling the rTMS (repetitive transcranial magnetic stimulation) device to stimulate the target brain area and using the NIRS (near infrared spectral imaging) device to continuously monitor the near-infrared spectrum changes of the target brain area to obtain updated data is an important means to ensure the therapeutic effect and safety of rTMS. This process allows real-time evaluation of the effects of rTMS on the brain, ensures the accuracy of the stimulation position, and monitors potential physiological reactions. Specifically: the basic NIRS data collected without any external stimulation provides a reference point in the resting state. When rTMS begins to work, brain hemodynamics and metabolic activity will change, and these changes can be captured in real time by NIRS. By comparing the situation before and after the stimulation, it can be accurately judged whether rTMS has reached the expected target brain area, whether there is any deviation, and specific physiological responses can be observed. This real-time monitoring is crucial for dynamically adjusting rTMS parameters and optimizing treatment plans. In addition, it can help identify side effects or adverse reactions that may be caused by inappropriate stimulation, so that timely measures can be taken to correct them.

[0026] In one embodiment, the rTMS device is controlled to stimulate the target brain area, and the NIRS device is used to continuously monitor the near-infrared spectrum changes of the target brain area to obtain updated infrared spectrum data, including: first, ensuring that the patient is in a relaxed state to avoid interference from external factors, such as nervousness or complex thinking activities. The NIRS monitoring device is properly worn on the patient's head so that the sensor accurately covers the target brain area, and the necessary calibration steps are completed, including using a standardized coordinate system (such as MNI or Talairach space) to correspond the position of the NIRS probe to the previously determined target brain area position to ensure accurate spatial positioning. Next, the rTMS device is started and the target brain area is stimulated according to the predetermined parameter settings. At the same time, the NIRS device begins to continuously record the changes in hemoglobin concentration and other relevant physiological parameters in the target brain area. In this process, the environmental conditions are kept stable, such as light intensity and temperature, to reduce the influence of external factors on the measurement results. The NIRS device should have a sufficient sampling frequency to capture the rapid hemodynamic changes and ensure the integrity and accuracy of the data. As the rTMS stimulation proceeds, the NIRS device will continue to generate new data that reflects the immediate response of the target brain area after being stimulated. By analyzing these updated near-infrared spectral data, information about blood flow, oxygenation levels, etc. can be extracted to evaluate the effect of rTMS. All collected updated infrared spectral data should be properly stored and labeled with relevant information, such as acquisition timestamp, rTMS stimulation parameters, etc., for subsequent analysis. A specially designed software platform is used to process these data, remove noise and extract useful information, such as calculating changes in light absorption at different wavelengths and inferring changes in hemoglobin concentration. Furthermore, these data can be compared with the previously established basic NIRS data to generate image semantic coding feature maps to intuitively show the differences before and after stimulation.

[0027] Exemplarily, in the target brain area positioning detection module 40, based on the comparison between the updated infrared spectrum data and the basic near-infrared spectrum data, it is determined whether the positioning accuracy of the target brain area meets the requirements. It should be understood that in the above-mentioned repetitive transcranial magnetic stimulation intervention system, the step of determining whether the positioning accuracy of the target brain area meets the requirements is crucial. It is the key to ensure that rTMS stimulation acts on the expected target brain area and avoids misstimulation of other functional areas in the brain, reducing potential risks and side effects, which is crucial to protecting the safety of patients. Specifically, in the process of determining whether the positioning accuracy of the target brain area meets the requirements, the technical concept of the present application is to process the basic near-infrared spectrum data and the updated infrared spectrum data together through a data processing and analysis algorithm based on artificial intelligence and signal processing, so as to capture the basic near-infrared spectrum image semantics and the updated near-infrared spectrum image semantics, as well as the fine-grained contrast feature representation information between the two, so as to judge the positioning accuracy of the target brain area to determine whether the positioning accuracy of the target brain area meets the requirements. In this way, the positioning accuracy of the target brain area and whether the detection has deviated can be automatically judged based on the infrared spectral semantics and brain hemodynamic changes before and after rTMS stimulation, and a positioning adjustment prompt signal can be issued when there is a deviation. This ensures that the positioning accuracy of the target brain area can be guaranteed in each stimulation during repetitive transcranial magnetic stimulation in an automated manner, thereby improving the intelligence level of the repetitive transcranial magnetic stimulation intervention system.

[0028] In one embodiment, Figure 2 and Figure 3 As shown, the target brain area positioning detection module 40 includes: a near infrared spectrum feature extraction unit 41, which is used to perform near infrared spectrum feature extraction on the updated infrared spectrum data and the basic near infrared spectrum data respectively to obtain an updated near infrared spectrum image semantic coding feature map and a basic near infrared spectrum image semantic coding feature map; a near infrared spectrum explicit feature enhancement processing unit 42, which is used to perform explicit feature enhancement processing based on spatial domain conversion on the updated near infrared spectrum image semantic coding feature map and the basic near infrared spectrum image semantic coding feature map respectively to obtain an updated near infrared spectrum image semantic enhancement coding feature map and a basic near infrared spectrum image enhancement semantic coding feature map; a near infrared spectrum interactive coding unit 43, which is used to perform interactive coding based on fine-grained semantic sharing compensation on the updated near infrared spectrum image semantic enhancement coding feature map and the basic near infrared spectrum image enhancement semantic coding feature map to obtain an updated near infrared spectrum-basic near infrared spectrum semantic fine-grained comparison feature map; a positioning accuracy determination unit 44, which is used to determine whether the positioning accuracy of the target brain area meets the requirements based on the updated near infrared spectrum-basic near infrared spectrum semantic fine-grained comparison feature map.

[0029] Exemplarily, in the near-infrared spectral feature extraction unit 41, the updated infrared spectral data and the basic near-infrared spectral data are subjected to feature extraction based on the near-infrared spectral to obtain an updated near-infrared spectral image semantic coding feature map and a basic near-infrared spectral image semantic coding feature map. It should be understood that through feature extraction, complex NIRS data can be converted into feature representations with clear biological significance, which can better reflect the activity patterns and physiological changes of the brain in different states. Specifically, the basic near-infrared spectral data represents the hemodynamic characteristics of the target brain area at rest, while the updated infrared spectral data reflects the immediate response after rTMS stimulation. Comparing these two sets of features can help accurately evaluate whether rTMS has reached the intended target brain area and whether any deviations or abnormalities have occurred.

[0030] In one embodiment, the near-infrared spectral feature extraction unit is used to: input the updated infrared spectral data and the basic near-infrared spectral data into a near-infrared spectral feature extractor based on the Mobile-Former model to obtain the updated near-infrared spectral image semantic coding feature map and the basic near-infrared spectral image semantic coding feature map. It should be understood that Mobile-Former is a lightweight but efficient neural network architecture that combines the advantages of convolutional neural networks (CNNs) and Transformers, and can capture complex spatial and temporal features, which is critical for parsing subtle changes in NIRS data. Therefore, by encoding with the near-infrared spectral feature extractor based on the Mobile-Former model, the infrared spectral image semantic feature information in the updated infrared spectral data and the basic near-infrared spectral data can be extracted respectively.

[0031] Exemplarily, in the near infrared spectrum explicit feature enhancement processing unit 42, the updated near infrared spectrum image semantic coding feature map and the basic near infrared spectrum image semantic coding feature map are respectively subjected to explicit feature enhancement processing based on spatial domain conversion to obtain the updated near infrared spectrum image semantic enhancement coding feature map and the basic near infrared spectrum image enhancement semantic coding feature map. Considering that the updated near infrared spectrum image semantic coding feature map and the basic near infrared spectrum image semantic coding feature map respectively contain the semantic features of the infrared spectrum image before and after rTMS stimulation, the features of the two respectively reflect the specific location and pattern of the target brain area activity changes such as brain area hemodynamic changes and metabolic activity. However, the original updated near infrared spectrum image semantic coding feature map and the basic near infrared spectrum image semantic coding feature map will contain some pixel coupling relationships and feature information with low contribution to the semantic comparison of the infrared spectrum images of the two and the subsequent target brain area positioning detection tasks, and the influence of this information will result in the semantic comparison of the infrared spectrum image before and after stimulation being not targeted and distinguishable. Therefore, in order to enhance the accuracy and robustness of the fine-grained feature comparison between the semantics of the two infrared spectral images, in the technical solution of the present application, the updated near-infrared spectral image semantic coding feature map and the basic near-infrared spectral image semantic coding feature map are further subjected to explicit feature enhancement processing based on spatial domain conversion to obtain an updated near-infrared spectral image semantic enhanced coding feature map and a basic near-infrared spectral image enhanced semantic coding feature map.

[0032] In one embodiment, Figure 4 As shown, the near-infrared spectrum dominant feature enhancement processing unit 42 includes: a near-infrared spectrum pixel granularity feature decoupling subunit 421, which is used to perform pixel granularity feature decoupling on the updated near-infrared spectrum image semantic encoding feature map to obtain a set of updated near-infrared spectrum pixel granularity decoupling feature vectors; a near-infrared spectrum structural semantic saliency measurement subunit 422, which is used to calculate the fine-grained structural semantic saliency measurement coefficient of each updated near-infrared spectrum pixel granularity decoupling feature vector in the set of updated near-infrared spectrum pixel granularity decoupling feature vectors to obtain an updated near-infrared spectrum structural semantic saliency coefficient. distribution matrix; a near-infrared spectral feature explicit hierarchical precipitation processing subunit 423, which is used to perform feature explicit hierarchical precipitation processing on the updated near-infrared spectral structure semantic significance distribution matrix based on a multi-layer mask function to obtain an updated near-infrared spectral structure semantic significance distribution weight matrix; a position point multiplication processing subunit 424, which is used to calculate the position point multiplication between the updated near-infrared spectral structure semantic significance distribution weight matrix and each feature matrix along the channel dimension of the updated near-infrared spectral image semantic coding feature map to obtain the updated near-infrared spectral image semantic enhancement coding feature map.

[0033] Exemplarily, in the near infrared spectrum pixel granularity feature decoupling subunit 421, the updated near infrared spectrum image semantic coding feature map is subjected to pixel granularity feature decoupling to obtain a set of updated near infrared spectrum pixel granularity decoupling feature vectors. Specifically, the process can be expressed as follows:

[0034] in, for updating the semantic encoding feature map of the near infrared spectral image, For Decouple pixel-level features. , , and are the first, second, and third in the set of pixel granularity decoupling feature vectors for updating near-infrared spectrum. and Update the near-infrared spectral pixel granularity decoupling feature vector.

[0035] That is, by finely parsing the feature information at each pixel level, the redundancy and unnecessary coupling relationships in the original feature map are eliminated, so that the features of each pixel can be processed separately. This not only reduces the redundancy of the feature space, but also enhances the model's ability to understand local features. Specifically, during rTMS treatment, changes in brain hemodynamics and metabolic activity are reflected in the near-infrared spectral data. When these changes are converted into image form, each pixel actually represents a physiological response at a specific location. However, the original updated near-infrared spectral image semantic encoding feature map may contain some pixel coupling relationships and feature information that contribute less to the target brain area positioning detection task. This information may interfere with subsequent analysis, resulting in the semantic comparison of infrared spectral images before and after stimulation being less targeted and distinguishable. Through pixel-level feature decoupling, the system can effectively remove these interference factors, making the features of each pixel more independent and representative. This means that for rTMS treatment, it is possible to more accurately identify the specific location and pattern of changes in activity in the target brain area. For example, this decoupling operation can help the system focus on local patterns and features that are critical to the accuracy of positioning the target brain area, ensuring that the stimulation acts on the correct target area. In addition, the decoupled feature vector set provides a purer data foundation for subsequent steps, such as explicit feature enhancement processing and interactive encoding. These steps further improve the quality of feature representation, ensuring that the output feature map can highlight the most representative and discriminative feature information, providing a solid foundation for verifying the accuracy of target brain area positioning.

[0036] In one embodiment, the near-infrared spectrum structure semantic saliency measurement subunit 422 is used to: input each updated near-infrared spectrum pixel granularity decoupling feature vector in the set of the updated near-infrared spectrum pixel granularity decoupling feature vectors into a hyperbolic space mapper to obtain a set of updated near-infrared spectrum hyperbolic space modulation pixel granularity decoupling feature vectors; calculate the fine-grained structural semantic saliency measurement coefficient of each updated near-infrared spectrum hyperbolic space modulation pixel granularity decoupling feature vector in the set of the updated near-infrared spectrum hyperbolic space modulation pixel granularity decoupling feature vectors to obtain the updated near-infrared spectrum structure semantic saliency distribution matrix formed by arranging multiple updated near-infrared spectrum fine-grained structural semantic saliency measurement coefficients. Specifically, the process can be expressed as follows:

[0037]

[0038] in, and are the first weight matrix and the second weight matrix respectively, for The corresponding updated near-infrared spectral hyperbolic spatial modulation pixel granularity decoupling feature vector, is the square of the one-norm of the vector, represents the logarithmic function value with base 2, is to update the semantic saliency distribution matrix of near infrared spectral structure The corresponding updated near-infrared spectral fine-grained structure semantic saliency metric coefficient.

[0039] That is, first, by inputting the decoupled feature vector into the hyperbolic space mapper, the spatial transformation of feature representation can be achieved. Hyperbolic space is a non-Euclidean geometric space that is particularly suitable for representing hierarchical or tree-structured data, and is used here to better capture the complex relationships between different brain regions. This mapping helps to reveal subtle differences that are not easily perceived in the original feature space, allowing the system to focus on truly important local features. The set of feature vectors after hyperbolic space mapping, that is, the set of updated near-infrared spectroscopy hyperbolic space modulation pixel granularity decoupled feature vectors, provides a more refined and discriminative data basis for subsequent analysis. Next, the fine-grained structural semantic saliency metric coefficients of these feature vectors are calculated to quantify the importance of each feature and construct an updated near-infrared spectroscopy structural semantic saliency distribution matrix composed of multiple updated near-infrared spectroscopy fine-grained structural semantic saliency metric coefficients. This matrix reflects the response intensity and relative importance of each location point in the target brain area after stimulation, helping the system to identify local patterns and features that are critical to the positioning accuracy of the target brain area. For example, certain specific areas may show higher changes in hemoglobin concentration or other physiological indicators, which indicate that the area is more strongly activated or inhibited. In this way, the system can more accurately determine whether the brain area actually stimulated is consistent with the predetermined target area and detect any deviation in a timely manner.

[0040] Exemplarily, in the near infrared spectrum feature explicit hierarchical precipitation processing subunit 423, the updated near infrared spectrum structure semantic saliency distribution matrix is ​​subjected to feature explicit hierarchical precipitation processing based on a multi-layer mask function to obtain an updated near infrared spectrum structure semantic saliency distribution weight matrix. Specifically, the process can be expressed as follows:

[0041] in is the preset threshold, For masking operation, To update the near infrared spectral structure semantic saliency distribution weight matrix The corresponding updated near-infrared spectral structure semantic significance distribution weight value.

[0042] That is, during rTMS treatment, changes in brain hemodynamics and metabolic activity are reflected in near-infrared spectral data. Through the previous steps, an updated near-infrared spectral structure semantic saliency distribution matrix has been obtained, which records in detail the response intensity of each location point after stimulation and its relative importance. However, this information may still contain some features that are less relevant or redundant to the target brain area localization detection task. In order to further enhance the model's ability to understand key features and reduce the impact of irrelevant information, a feature-explicit hierarchical precipitation process based on a multi-layer mask function is introduced. This process applies a series of mask functions to the saliency distribution matrix to screen and emphasize those features that best represent the activity change pattern of the target brain area layer by layer. Each layer of mask function selectively retains certain specific features according to preset thresholds or rules, while suppressing other unimportant parts. As the number of layers increases, this selectivity and emphasis gradually accumulates, eventually forming an updated near-infrared spectral structure semantic saliency distribution weight matrix. This weight matrix not only highlights the most representative and discriminative feature information, but also provides a purer data basis for subsequent analysis. For example, in actual operation, some locations may show higher changes in hemoglobin concentration or other physiological indicators than other areas, indicating that they are more strongly activated or inhibited. Through multi-layer mask function processing, the features of these key locations will be given higher weights in the final weight matrix to ensure that they are fully valued in subsequent positioning detection and effect evaluation. On the contrary, those locations that have little effect on the positioning accuracy of the target brain area will be appropriately weakened to avoid interfering with the analysis results.

[0043] Exemplarily, in the position point multiplication processing subunit 424, the position point multiplication between the updated near infrared spectral structure semantic significance distribution weight matrix and each feature matrix along the channel dimension of the updated near infrared spectral image semantic encoding feature map is calculated to obtain the updated near infrared spectral image semantic enhancement encoding feature map. Specifically, the process can be expressed as follows:

[0044] in, is the updated near infrared spectral structure semantic saliency distribution weight matrix is matrix multiplication, It is the updated near infrared spectral image semantic enhancement encoding feature map.

[0045] That is, during the rTMS treatment, an updated near-infrared spectral structure semantic saliency distribution weight matrix and an updated near-infrared spectral image semantic encoding feature map have been obtained through a series of processing steps. The former reflects the response intensity and relative importance of each position point after stimulation, while the latter contains the spatial and temporal information that has been preliminarily parsed. In order to further improve the quality of these feature representations, the two need to be effectively combined. By calculating the positional point multiplication operation between the two matrices, that is, element-by-element multiplication, each feature in the original encoding feature map can be weighted and adjusted according to the weight value in the saliency distribution weight matrix while maintaining the spatial position correspondence. This weighted fusion makes the key position point features emphasized by the saliency distribution weight matrix more prominent in the final generated updated near-infrared spectral image semantic enhancement encoding feature map, while relatively unimportant features are appropriately weakened. As a result, the generated feature map not only retains the rich details in the original data, but also enhances the ability to understand the activity change pattern of the target brain area.

[0046] In summary, the explicit feature enhancement processing based on spatial domain conversion is to perform feature enhancement by using explicit modeling and hierarchical mask modulation of the fine-grained features of each infrared spectral image pixel in the feature map, which effectively improves the quality of the updated near-infrared spectral image semantic coding feature map and the basic near-infrared spectral image semantic coding feature map. Specifically, for the updated near-infrared spectral image semantic coding feature map, the explicit feature enhancement processing based on spatial domain conversion can eliminate unnecessary coupling relationships in the original updated near-infrared spectral image semantic coding feature map through pixel granularity feature decoupling operation, so that the features of each pixel can be processed separately. This not only reduces the redundancy of the feature space, but also enhances the model's ability to understand local features. For rTMS treatment, this means more accurately identifying the specific location and pattern of changes in the activity of the target brain area. Furthermore, hyperbolic space mapping and calculating the fine-grained structure semantic saliency metric coefficient can quantify the importance of each fine-grained feature in the feature map, providing a scientific basis for subsequent feature selection. In the context of rTMS, this step helps the system focus on identifying local patterns and features that are critical to the localization accuracy of the target brain area, ensuring that the stimulation acts on the correct target area. Then, a multi-layer mask function is used to achieve selective amplification of features at different levels. This process is similar to the attention mechanism, allowing the system to dynamically adjust the degree of attention to the semantic features of different pixel granularity in the updated near-infrared spectral image. For rTMS treatment, this helps to highlight the key features that have a significant impact on the treatment effect, while suppressing noise or irrelevant features and improving the quality of the final feature representation. Finally, element-by-element multiplication is used to generate the final updated near-infrared spectral image semantic enhancement encoding feature map, ensuring that the output feature map can highlight the most representative and discriminative feature information, providing a solid foundation for the subsequent verification of the accuracy of the target brain area positioning.

[0047] Exemplarily, in the near infrared spectrum interactive coding unit 43, the updated near infrared spectrum image semantic enhancement coding feature map and the basic near infrared spectrum image enhancement semantic coding feature map are interactively coded based on fine-grained semantic sharing compensation to obtain an updated near infrared spectrum-basic near infrared spectrum semantic fine-grained contrast feature map. It should be understood that since the updated near infrared spectrum image semantic enhancement coding feature map and the basic near infrared spectrum image enhancement semantic coding feature map contain the infrared spectrum image semantic feature information before and after stimulation after the dominant feature enhancement, some of the feature semantics between the two are common, and some are differential feature semantics, which are crucial for the semantic contrast and positioning detection tasks of the outer spectrum images before and after rTMS stimulation. However, traditional feature contrast and interaction methods often focus on the combination and contrast of overall features, ignoring the subtle differences in local areas in the feature map. This may cause the loss of key details between the outer spectrum image semantics before and after rTMS stimulation, affecting the accuracy of the final positioning detection result. In addition, the traditional method mainly relies on the information provided by the input features themselves, and is powerless for the potential semantic information that is not explicitly expressed, and may ignore the implicit features that are crucial to the target brain area positioning and identification task. Based on this, in the technical solution of the present application, the updated near-infrared spectrum image semantic enhancement coding feature map and the basic near-infrared spectrum image enhanced semantic coding feature map are further interactively encoded based on fine-grained semantic sharing compensation to obtain the updated near-infrared spectrum-basic near-infrared spectrum semantic fine-grained comparison feature map.

[0048] In one embodiment, Figure 5 As shown, the near-infrared spectrum interactive coding unit 43 includes: a near-infrared spectrum feature decomposition unit 431, which is used to perform channel-dimensional feature decomposition on the updated near-infrared spectrum image semantic enhancement coding feature map and the basic near-infrared spectrum image enhancement semantic coding feature map to obtain a set of updated near-infrared spectrum image semantic local feature matrices and a set of basic near-infrared spectrum image semantic local feature matrices; a near-infrared spectrum fine-grained semantic sharing compensation subunit 432, which is used to perform fine-grained semantic sharing compensation and interaction on the set of updated near-infrared spectrum image semantic local feature matrices and the set of basic near-infrared spectrum image semantic local feature matrices to obtain the updated near-infrared spectrum-basic near-infrared spectrum semantic fine-grained comparison feature map.

[0049] Exemplarily, in the near infrared spectral feature decomposition unit 431, the updated near infrared spectral image semantic enhancement coding feature map and the basic near infrared spectral image enhanced semantic coding feature map are respectively subjected to feature decomposition based on the channel dimension to obtain a set of updated near infrared spectral image semantic local feature matrices and a set of basic near infrared spectral image semantic local feature matrices. Specifically, the process can be expressed as follows:

[0050]

[0051] in, and are respectively the updated near infrared spectral image semantic enhancement coding feature map and the basic near infrared spectral image enhanced semantic coding feature map, To perform feature dissociation on the feature map, , , and They are the first, second, and third sets of semantic local feature matrices of the near infrared spectral image. and Update the semantic local feature matrix of the near infrared spectral image, , , and They are the first, second, and third in the set of semantic local feature matrices of basic near infrared spectral images. and A basic near infrared spectral image semantic local feature matrix.

[0052] That is, through the feature dissociation process, the original enhanced coding feature map is decomposed into multiple local feature matrices according to different channel dimensions. Each local feature matrix captures information in a specific spatial position or frequency range, reflecting the change pattern of physiological parameters such as hemoglobin concentration and oxygenation level in the region. This decomposition method allows for a closer focus on the specific details of the internal structure and activity pattern of the brain, providing more accurate data support for subsequent fine-grained comparison and positioning detection. Specifically, in the rTMS intervention system, the updated near-infrared spectral image semantic enhanced coding feature map records the state of the target brain area after stimulation, while the basic near-infrared spectral image enhanced semantic coding feature map represents the baseline state of the same area when not stimulated. By performing feature dissociation based on the channel dimension on these two sets of data, complex brain activity signals can be split into several relatively independent but interrelated components. This not only helps to identify which parts are due to specific changes caused by rTMS stimulation, but also distinguishes natural fluctuations and other non-specific interference factors.

[0053] In one embodiment, Figure 6 As shown, the near-infrared spectrum fine-grained semantic sharing compensation subunit 432 includes: a semantic sharing processing secondary subunit 4321, which is used to semantically share the updated near-infrared spectrum image semantic local feature matrix and the basic near-infrared spectrum image semantic local feature matrix of each group of corresponding channel dimensions in the set of the updated near-infrared spectrum image semantic local feature matrix and the set of the basic near-infrared spectrum image semantic local feature matrix to obtain an updated near-infrared spectrum-basic near-infrared spectrum fine-grained local shared semantic feature matrix set; a semantic feature compensation processing secondary subunit 4322, which is used to update the The set of near-infrared spectrum-basic near-infrared spectrum fine-grained local shared semantic feature matrices is subjected to semantic feature compensation to obtain a set of updated near-infrared spectrum-basic near-infrared spectrum semantic compensation text semantic coding feature matrices; the semantic enhancement fusion processing secondary sub-unit 4323 is used to perform semantic enhancement fusion processing on the set of updated near-infrared spectrum-basic near-infrared spectrum fine-grained local shared semantic feature matrices based on the set of updated near-infrared spectrum-basic near-infrared spectrum semantic compensation text semantic coding feature matrices to obtain the updated near-infrared spectrum-basic near-infrared spectrum semantic fine-grained comparison feature map.

[0054] Exemplarily, in the semantic sharing processing secondary subunit 4321, the updated near infrared spectrum image semantic local feature matrix and the basic near infrared spectrum image semantic local feature matrix of each corresponding channel dimension in the set of the updated near infrared spectrum image semantic local feature matrix and the set of the basic near infrared spectrum image semantic local feature matrix are semantically shared to obtain a set of updated near infrared spectrum-basic near infrared spectrum fine-grained local shared semantic feature matrices. Specifically, the process can be expressed as follows:

[0055] in, , and They are point-by-point addition, point-by-point multiplication, and point-by-point subtraction. For cascade processing, is point convolutional coding, yes function, yes and The updated NIRS-based NIRS fine-grained local shared semantic feature matrix.

[0056] That is, semantic sharing processing is not just a simple superposition or averaging, but a deep exploration of the information in each local feature matrix to find out those features that remain consistent before and after stimulation (i.e., commonalities) and those features that have changed significantly (i.e., differences). For the consistent parts, this indicates that these areas may not be affected by rTMS stimulation, or they are stable functional areas; while for the changed parts, they may indicate changes in neuronal activity levels or other physiological responses. In this way, the system can more accurately identify which areas are truly affected by the stimulation and which areas remain unchanged, thereby confirming whether the stimulation accurately acts on the expected target brain area. The generated collection of updated near-infrared spectroscopy-basic near-infrared spectroscopy fine-grained local shared semantic feature matrices provides more detailed data support for subsequent positioning detection. These matrices not only retain the rich details in the original data, but also highlight the most critical changes before and after stimulation, enabling the system to evaluate the effect of rTMS at a finer granularity.

[0057] In one embodiment, the semantic feature compensation processing secondary subunit 4322 is used to: input each updated near infrared spectrum-basic near infrared spectrum fine-grained local shared semantic feature matrix in the set of updated near infrared spectrum-basic near infrared spectrum fine-grained local shared semantic feature matrices into a semantic compensation decoding module based on a large language model to obtain a set of updated near infrared spectrum-basic near infrared spectrum semantic compensation text descriptions; input each updated near infrared spectrum-basic near infrared spectrum semantic compensation text description in the set of updated near infrared spectrum-basic near infrared spectrum semantic compensation text descriptions into a semantic encoder based on a text convolutional neural network model to obtain a set of updated near infrared spectrum-basic near infrared spectrum semantic compensation text semantic encoding feature matrices. Specifically, the process can be expressed as follows:

[0058]

[0059] in, For semantic compensation decoding operation, for The corresponding updated NIR-basic NIR semantic compensation text description, is the text convolutional encoding, for The corresponding updated NIRS-basic NIRS semantic compensation text semantic encoding feature matrix.

[0060] That is, a collection of fine-grained local shared semantic feature matrices has been obtained through a series of processing of the update and base near-infrared spectral images. These matrices capture the subtle changes in the target brain area before and after rTMS stimulation, but they are still numerical feature representations and lack direct interpretability. In order to better understand the biological significance behind these changes and supplement potential important information, a semantic compensation decoding module based on a large language model is introduced. By inputting each fine-grained local shared semantic feature matrix into the semantic compensation decoding module, the corresponding semantic compensation text description can be generated. This step utilizes the powerful expression power of the large language model and is able to generate biologically interpretable text descriptions based on the contextual information provided by the feature matrix. For example, for a specific location point, if its hemoglobin concentration increases significantly, the decoding module may generate a text description describing the enhanced neuronal activity or vascular dilation in the area. This conversion not only makes the features more intuitive and easy to understand, but also provides a rich semantic background for subsequent analysis. Next, these semantic compensation text descriptions are input into the semantic encoder based on the text convolutional neural network (TextCNN) model. TextCNN is good at extracting meaningful features from text data and converting them into compact and efficient vector representations. After being processed by the encoder, a set of updated NIRS-basic NIRS semantically compensated text semantic encoding feature matrices was obtained. These encoding feature matrices fuse the spatial and temporal information of the original NIRS data with the semantic description generated by the large language model to form a richer and more comprehensive data representation. The generated semantically compensated text semantic encoding feature matrix not only retains the key information in the original feature matrix, but also enhances the understanding of physiological changes through text descriptions. This helps to more accurately identify which areas are truly affected by rTMS stimulation and which areas remain unchanged, thereby confirming whether the stimulation accurately acts on the expected target brain area.

[0061] In one embodiment, the semantic enhancement fusion processing secondary subunit 4323 is used to: perform fine-grained semantic interaction compensation on each corresponding group of updated near infrared spectrum-basic near infrared spectrum fine-grained local shared semantic feature matrices and updated near infrared spectrum-basic near infrared spectrum semantic compensation text semantic coding feature matrices in the set of the updated near infrared spectrum-basic near infrared spectrum fine-grained local shared semantic feature matrices and the set of the updated near infrared spectrum-basic near infrared spectrum semantic compensation text semantic coding feature matrices to obtain a set of updated near infrared spectrum-basic near infrared spectrum fine-grained local interactive semantic enhancement feature matrices; aggregate the set of updated near infrared spectrum-basic near infrared spectrum fine-grained local interactive semantic enhancement feature matrices along the channel dimension to obtain the updated near infrared spectrum-basic near infrared spectrum semantic fine-grained contrast feature map. Specifically, the process can be expressed as follows:

[0062]

[0063] in, for The transposed matrix of is matrix multiplication, yes The scale is the matrix width multiplied by the matrix height, yes function, and are weighting coefficients, , , and They are the first, second, and third in the set of near infrared spectrum-basic near infrared spectrum fine-grained local interactive semantic enhancement feature matrices. and Updated near infrared spectrum-basic near infrared spectrum fine-grained local interaction semantic enhancement feature matrix, is the number of feature matrices in the set of updated near infrared spectrum-basic near infrared spectrum fine-grained local interactive semantics enhanced feature matrices, To aggregate a collection of feature matrices along the channel dimension, It is the updated near infrared spectrum-basic near infrared spectrum semantic fine-grained contrast feature map.

[0064] That is, semantic interaction compensation is not just a simple superposition or averaging, but a deep exploration of the relationship between each fine-grained local shared semantic feature matrix and its corresponding semantic compensation text semantic encoding feature matrix. In this way, the feature performance of the same location point in different states can be analyzed more carefully, and those features that remain consistent before and after stimulation (i.e., commonalities) and those features that change significantly (i.e., differences) can be found. For the consistent parts, this indicates that these areas may not be affected by rTMS stimulation or are stable functional areas; while for the changed parts, it may indicate changes in neuronal activity levels or other physiological responses. The generated updated near infrared spectrum-basic near infrared spectrum fine-grained local interactive semantic enhancement feature matrix not only retains the key information in the original feature matrix, but also enhances the understanding of physiological changes through semantic description. For example, some specific location points may show higher changes in hemoglobin concentration or other physiological indicators, and the semantic description can provide additional explanations, such as the specific mechanism of enhanced neuronal activity or vasodilation in this area. Through the analysis of these interactive semantic enhancement feature matrices, the rTMS intervention system can more accurately determine whether the brain area actually stimulated is consistent with the predetermined target area, and detect any deviation in time. Finally, in order to form a holistic representation, these interactive semantically enhanced feature matrices are aggregated along the channel dimension, and finally an updated near-infrared spectrum-basic near-infrared spectrum semantic fine-grained comparison feature map is obtained.

[0065] In summary, the interactive encoding method based on fine-grained semantic shared compensation can use prompt learning to perform fine-grained compensation in the process of feature interaction between the updated near-infrared spectral image semantics and the basic near-infrared spectral image semantics, thereby using the prior information of the large model to perform fine-grained semantic compensation for feature interaction comparison, so as to use prior knowledge to supplement potential important information to generate a richer and more accurate infrared spectral semantic comparison feature representation. In addition, the introduction of prompt learning also improves the interpretability of feature interaction. This method ensures that even the slightest semantic changes in infrared spectral images will not be ignored, allowing for a more detailed analysis of the interactive comparison of the semantics of the external spectral image before and after rTMS stimulation, and improving the accuracy of subsequent positioning detection tasks.

[0066] Exemplarily, in the positioning accuracy determination unit 44, based on the updated near infrared spectrum-basic near infrared spectrum semantic fine-grained contrast feature map, it is determined whether the positioning accuracy of the target brain area meets the requirements. In one embodiment, the positioning accuracy determination unit is used to: input the updated near infrared spectrum-basic near infrared spectrum semantic fine-grained contrast feature map into the positioning identifier based on the classifier to obtain a positioning result, and the positioning result is used to indicate whether the positioning accuracy of the target brain area meets the requirements. That is to say, the fine-grained contrast feature information between the updated near infrared spectrum image semantics and the basic near infrared spectrum semantics is used for classification processing, so as to perform positioning identification, thereby judging whether the positioning accuracy of the target brain area meets the requirements. In this way, the positioning accuracy of the target brain area can be automatically judged based on the infrared spectrum semantics and brain area hemodynamic changes before and after rTMS stimulation, and whether the deviation occurs, and a positioning adjustment prompt signal is issued when the deviation occurs, so that each stimulation can ensure the positioning accuracy of the target brain area in an automated manner during the repetitive transcranial magnetic stimulation process, thereby improving the intelligence level of the repetitive transcranial magnetic stimulation intervention system. In one embodiment, the classifier-based positioning identifier uses an SVM model.

[0067] Preferably, inputting the updated near infrared spectrum-basic near infrared spectrum semantic fine-grained comparison feature map into a classifier-based positioning identifier to obtain a positioning result comprises: Determine the median eigenvalue, maximum eigenvalue and minimum eigenvalue in the updated near infrared spectrum-basic near infrared spectrum semantic fine-grained comparison feature map, and divide the median eigenvalue by the difference between the maximum eigenvalue and the minimum eigenvalue to obtain the infrared spectrum semantic fine-grained comparison distribution probability value, that is, ,in, , , respectively represent the median eigenvalue, the maximum eigenvalue and the minimum eigenvalue in the updated near infrared spectrum-basic near infrared spectrum semantic fine-grained comparison feature map, Indicates the probability value of the infrared spectrum semantic fine-grained contrast distribution; The updated near infrared spectrum-basic near infrared spectrum semantic fine-grained contrast feature map is normalized to obtain an infrared spectrum semantic fine-grained contrast probability feature map. ; The power function of each eigenvalue of the infrared spectrum semantic fine-grained contrast probability feature map is calculated by subtracting the infrared spectrum semantic fine-grained contrast distribution probability value as an exponent to obtain a first infrared spectrum semantic fine-grained contrast convergence feature map. ,in, represents the infrared spectrum semantic fine-grained contrast probability feature map, It represents the power function calculated by subtracting the infrared spectrum semantic fine-grained contrast distribution probability value from one as an exponent, Represents the first infrared spectrum semantic fine-grained contrast convergence feature map; The infrared spectrum semantic fine-grained contrast distribution probability value is used as an exponent to calculate the power function of each eigenvalue of the point difference feature map between the unit feature map and the infrared spectrum semantic fine-grained contrast probability feature map to obtain a second infrared spectrum semantic fine-grained contrast convergence feature map ,in, It means subtracting by position. represents the unit feature map, represents the infrared spectrum semantic fine-grained contrast probability feature map, The power function is calculated using the infrared spectrum semantic fine-grained contrast distribution probability value as an exponent, Represents the semantic fine-grained contrast convergence feature map of the second infrared spectrum; The infrared spectrum semantic fine-grained comparison probability feature map is multiplied by a difference value minus the infrared spectrum semantic fine-grained comparison distribution probability value to obtain a first infrared spectrum semantic fine-grained comparison limited feature map. ,in, represents the infrared spectrum semantic fine-grained contrast probability feature map, represents the probability value of the infrared spectrum semantic fine-grained contrast distribution, represents dot product, represents a first infrared spectrum semantic fine-grained contrast-qualified feature map; The point difference feature map is multiplied by the infrared spectrum semantic fine-grained comparison distribution probability value to obtain a second infrared spectrum semantic fine-grained comparison definition feature map. ,in, represents the infrared spectrum semantic fine-grained contrast probability feature map, It means subtracting by position. represents the unit feature map, represents the probability value of the infrared spectrum semantic fine-grained contrast distribution, represents dot product, Representing a second infrared spectrum semantic fine-grained contrast qualification feature map; After dot-multiplying the first infrared spectrum semantic fine-grained comparison convergence feature map with the second infrared spectrum semantic fine-grained comparison convergence feature map, dot-multiplying the first infrared spectrum semantic fine-grained comparison limit feature map and the second infrared spectrum semantic fine-grained comparison limit feature map to obtain an optimized updated near infrared spectrum-basic near infrared spectrum semantic fine-grained comparison feature map ,in, Represents the first infrared spectrum semantic fine-grained contrast convergence feature map, Represents the second infrared spectrum semantic fine-grained contrast convergence feature map, represents the first infrared spectrum semantic fine-grained contrast-qualified feature map, Represents the second infrared spectrum semantic fine-grained contrast qualification feature map, represents dot product, It means adding by position. Represents the optimized updated near infrared spectrum-basic near infrared spectrum semantic fine-grained contrast feature map; The optimized updated near infrared spectrum-basic near infrared spectrum semantic fine-grained contrast feature map is input into a classifier-based positioning identifier to obtain a positioning result.

[0068] Here, considering that the updated near-infrared spectral image semantic enhancement coding feature map and the basic near-infrared spectral image enhanced semantic coding feature map respectively represent the infrared spectral features of the updated infrared spectral data of the target brain area after stimulation and the infrared spectral features of the basic near-infrared spectral data of the target brain area without any stimulation, when performing feature enhancement interactive coding based on cross-domain fine-grained semantic compensation, due to the cross-domain fine-grained semantic compensation differences caused by the differences in the distribution of infrared spectral semantic features, the updated near-infrared spectrum-basic near-infrared spectrum semantic fine-grained comparison feature map will have a probability convergence and divergence based on different feature enhancement interactions, thereby affecting the accuracy of the positioning result obtained by the input classifier-based positioning identifier.

[0069] Based on this, by using the cross entropy form power series of the distribution analysis probability of the updated near infrared spectrum - basic near infrared spectrum semantic fine-grained contrast feature map as the probability distribution convergence limit, the class probability convergence approximation of the updated near infrared spectrum - basic near infrared spectrum semantic fine-grained contrast feature map is performed on the basis of the combination of the feature set distribution and the probability density distribution of the updated near infrared spectrum - basic near infrared spectrum semantic fine-grained contrast feature map, so as to avoid the conservative convergence control of the fatal potential divergence of the updated near infrared spectrum - basic near infrared spectrum semantic fine-grained contrast feature map, and further use the probability distribution cross entropy of the updated near infrared spectrum - basic near infrared spectrum semantic fine-grained contrast feature map as the objective function to guide the limited recovery strategy, so as to achieve the common agility of the convergence of the probability convergence and divergence of the updated near infrared spectrum - basic near infrared spectrum semantic fine-grained contrast feature map to the probability density distribution space, and improve the accuracy of the positioning result obtained by inputting the updated near infrared spectrum - basic near infrared spectrum semantic fine-grained contrast feature map into the classifier-based positioning identifier.

[0070] Exemplarily, in the positioning adjustment module 50, in response to the positioning accuracy of the target brain area not meeting the requirements, a prompt for adjusting the positioning of the target brain area is generated. It should be understood that accurate positioning of the target brain area is crucial during rTMS treatment. If the stimulation position is inaccurate, the expected therapeutic effect may not be achieved, and may even lead to side effects or ineffective treatment. Therefore, real-time monitoring and timely adjustment of positioning are essential for maintaining the effectiveness of treatment and reducing potential risks. Through dynamic monitoring and timely adjustment, the needs of personalized treatment can be better met, the pertinence and effect of treatment can be improved, and reliable data support can also be provided for scientific research. Specifically, if it is found that the positioning deviation exceeds the acceptable range, the system will immediately generate a positioning adjustment prompt. This prompt can be conveyed to the operator in a visual, auditory or tactile form. According to the prompt information, the technician can manually or with the help of an automated tool to recalibrate the position of the rTMS coil until it is restored to the correct stimulation position. Some advanced systems also have automatic adjustment capabilities and can complete this process without manual intervention. After the adjustment is completed, a short period of monitoring is performed again to confirm whether the new positioning is accurate. All relevant operations and data should be fully recorded for future review and further analysis.

[0071] In summary, according to the repetitive transcranial magnetic stimulation intervention system of the embodiment of the present application, it is explained that the basic near-infrared spectral data and the updated infrared spectral data are processed together through a data processing and analysis algorithm based on artificial intelligence and signal processing, so as to capture the basic near-infrared spectral image semantics and the updated near-infrared spectral image semantics, as well as the fine-grained contrast feature representation information between the two, so as to make a positioning accuracy judgment on the target brain area to determine whether the positioning accuracy of the target brain area meets the requirements. In this way, the positioning accuracy of the target brain area can be automatically judged based on the infrared spectral semantics and brain area hemodynamic changes before and after rTMS stimulation, and whether a deviation occurs, and a positioning adjustment prompt signal is issued when a deviation occurs, so that each stimulation can ensure the positioning accuracy of the target brain area in an automated manner during the repetitive transcranial magnetic stimulation process, thereby improving the intelligence level of the repetitive transcranial magnetic stimulation intervention system.

[0072] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0073] It should be understood that the specific examples in this article are only intended to help those skilled in the art better understand the embodiments of the present application, rather than to limit the scope of the embodiments of the present application.

[0074] It should also be understood that in the various embodiments of the present application, the size of the serial number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0075] It should also be understood that the various implementation methods described in this specification can be implemented individually or in combination, and the embodiments of the present application are not limited to this.

[0076] Unless otherwise stated, all technical and scientific terms used in the embodiments of the present application are the same as the meanings generally understood by those skilled in the art of the technical field of the present application. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the scope of the present application. The term "and / or" used in the present application includes any and all combinations of one or more related listed items. The singular forms of "a kind of", "above" and "the" used in the embodiments of the present application and the appended claims are also intended to include majority forms, unless the context clearly indicates other meanings. In addition, the terms "first", "second" etc. are only used for description purposes and cannot be understood as indicating or suggesting relative importance.

[0077] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0078] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0079] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0080] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0081] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A repetitive transcranial magnetic stimulation intervention system, characterized in that: include: A brain region initial positioning data acquisition module is used to record the initial positioning data of the target brain region; A basic near-infrared spectral data acquisition module, used to acquire basic near-infrared spectral data of the target brain area without any stimulation using a NIRS monitoring device; An updated infrared spectrum data acquisition module is used to control the rTMS device to stimulate the target brain area, and at the same time, continuously monitor the near-infrared spectrum changes of the target brain area using the NIRS device to obtain updated infrared spectrum data; A target brain region positioning detection module, used to determine whether the positioning accuracy of the target brain region meets the requirements based on the comparison between the updated infrared spectrum data and the basic near-infrared spectrum data; The positioning adjustment module is used to generate a target brain area positioning adjustment prompt in response to the positioning accuracy of the target brain area not meeting the requirements.

2. The repetitive transcranial magnetic stimulation intervention system according to claim 1, characterized in that: The target brain region positioning detection module includes: A near infrared spectrum feature extraction unit, used for performing feature extraction based on the near infrared spectrum on the updated infrared spectrum data and the basic near infrared spectrum data respectively to obtain an updated near infrared spectrum image semantic coding feature map and a basic near infrared spectrum image semantic coding feature map; A near infrared spectrum explicit feature enhancement processing unit is used to perform explicit feature enhancement processing based on spatial domain conversion on the updated near infrared spectrum image semantic coding feature map and the basic near infrared spectrum image semantic coding feature map to obtain an updated near infrared spectrum image semantic enhancement coding feature map and a basic near infrared spectrum image enhanced semantic coding feature map; A near infrared spectrum interactive coding unit, used for interactively coding the updated near infrared spectrum image semantic enhancement coding feature map and the basic near infrared spectrum image enhanced semantic coding feature map based on fine-grained semantic shared compensation to obtain an updated near infrared spectrum - basic near infrared spectrum semantic fine-grained contrast feature map, wherein the near infrared spectrum interactive coding unit comprises: a near infrared spectrum feature decomposition unit, used for performing feature decomposition based on the channel dimension on the updated near infrared spectrum image semantic enhancement coding feature map and the basic near infrared spectrum image enhanced semantic coding feature map respectively to obtain a set of updated near infrared spectrum image semantic local feature matrices and a set of basic near infrared spectrum image semantic local feature matrices; a near infrared spectrum fine-grained semantic shared compensation subunit, used for performing fine-grained semantic shared compensation and interaction on the set of updated near infrared spectrum image semantic local feature matrices and the set of basic near infrared spectrum image semantic local feature matrices to obtain the updated near infrared spectrum - basic near infrared spectrum semantic fine-grained contrast feature map; The positioning accuracy determination unit is used to determine whether the positioning accuracy of the target brain area meets the requirements based on the updated near-infrared spectrum-basic near-infrared spectrum semantic fine-grained comparison feature map.

3. The repetitive transcranial magnetic stimulation intervention system according to claim 2, characterized in that: The near infrared spectrum feature extraction unit is used to: input the updated infrared spectrum data and the basic near infrared spectrum data into a near infrared spectrum feature extractor based on a Mobile-Former model to obtain the updated near infrared spectrum image semantic coding feature map and the basic near infrared spectrum image semantic coding feature map.

4. The repetitive transcranial magnetic stimulation intervention system according to claim 3, characterized in that: The near-infrared spectrum dominant feature enhancement processing unit comprises: A near infrared spectrum pixel granularity feature decoupling subunit is used to perform pixel granularity feature decoupling on the updated near infrared spectrum image semantic coding feature map to obtain a set of updated near infrared spectrum pixel granularity decoupling feature vectors; A near infrared spectrum structural semantic saliency measurement subunit, used for calculating the fine-grained structural semantic saliency measurement coefficient of each updated near infrared spectrum pixel granularity decoupling feature vector in the set of updated near infrared spectrum pixel granularity decoupling feature vectors to obtain an updated near infrared spectrum structural semantic saliency distribution matrix; A near-infrared spectrum feature explicit hierarchical precipitation processing subunit is used to perform feature explicit hierarchical precipitation processing based on a multi-layer mask function on the updated near-infrared spectrum structure semantic saliency distribution matrix to obtain an updated near-infrared spectrum structure semantic saliency distribution weight matrix; The position point multiplication processing subunit is used to calculate the position point multiplication between the updated near-infrared spectral structure semantic significance distribution weight matrix and the various feature matrices along the channel dimension of the updated near-infrared spectral image semantic encoding feature map to obtain the updated near-infrared spectral image semantic enhancement encoding feature map.

5. The repetitive transcranial magnetic stimulation intervention system according to claim 4, characterized in that: The near infrared spectral structure semantic saliency measurement subunit is used to: Inputting each updated near infrared spectrum pixel granularity decoupling feature vector in the set of updated near infrared spectrum pixel granularity decoupling feature vectors into a hyperbolic space mapper to obtain a set of updated near infrared spectrum hyperbolic space modulation pixel granularity decoupling feature vectors; The fine-grained structural semantic significance measurement coefficient of each updated near-infrared spectrum hyperbolic spatial modulation pixel granularity decoupling feature vector in the set of the updated near-infrared spectrum hyperbolic spatial modulation pixel granularity decoupling feature vectors is calculated to obtain the updated near-infrared spectrum structural semantic significance distribution matrix formed by arranging multiple updated near-infrared spectrum fine-grained structural semantic significance measurement coefficients.

6. The repetitive transcranial magnetic stimulation intervention system according to claim 5, characterized in that: The near infrared spectroscopy fine-grained semantic sharing compensation subunit includes: A semantic sharing processing secondary subunit is used to semantically share the updated near infrared spectrum image semantic local feature matrix and the basic near infrared spectrum image semantic local feature matrix of each corresponding channel dimension in the set of the updated near infrared spectrum image semantic local feature matrix and the set of the basic near infrared spectrum image semantic local feature matrix to obtain a set of updated near infrared spectrum-basic near infrared spectrum fine-grained local shared semantic feature matrices; A semantic feature compensation processing secondary subunit is used to perform semantic feature compensation on the set of updated near infrared spectrum-basic near infrared spectrum fine-grained local shared semantic feature matrices to obtain a set of updated near infrared spectrum-basic near infrared spectrum semantic compensation text semantic coding feature matrices; The semantic enhancement fusion processing secondary sub-unit is used to perform semantic enhancement fusion processing on the set of updated near infrared spectrum - basic near infrared spectrum fine-grained local shared semantic feature matrices based on the set of updated near infrared spectrum - basic near infrared spectrum semantic compensation text semantic encoding feature matrices to obtain the updated near infrared spectrum - basic near infrared spectrum semantic fine-grained comparison feature map.

7. The repetitive transcranial magnetic stimulation intervention system according to claim 6, characterized in that: The semantic feature compensation processing secondary subunit is used to: Inputting each updated near infrared spectrum-basic near infrared spectrum fine-grained local shared semantic feature matrix in the set of updated near infrared spectrum-basic near infrared spectrum fine-grained local shared semantic feature matrix into a semantic compensation decoding module based on a large language model to obtain a set of updated near infrared spectrum-basic near infrared spectrum semantic compensation text description; Each updated near infrared spectrum - basic near infrared spectrum semantic compensation text description in the set of updated near infrared spectrum - basic near infrared spectrum semantic compensation text description is input into the semantic encoder based on the text convolutional neural network model to obtain the set of updated near infrared spectrum - basic near infrared spectrum semantic compensation text semantic encoding feature matrix.

8. The repetitive transcranial magnetic stimulation intervention system according to claim 7, characterized in that: The semantic enhancement fusion processing secondary sub-unit is used to: Perform fine-grained semantic interaction compensation on each corresponding group of updated near infrared spectrum - basic near infrared spectrum fine-grained local shared semantic feature matrices and updated near infrared spectrum - basic near infrared spectrum semantic compensation text semantic encoding feature matrices in the set of the updated near infrared spectrum - basic near infrared spectrum fine-grained local shared semantic feature matrices and the set of the updated near infrared spectrum - basic near infrared spectrum semantic compensation text semantic encoding feature matrices to obtain a set of updated near infrared spectrum - basic near infrared spectrum fine-grained local interactive semantic reinforcement feature matrices; The set of the updated near infrared spectrum-basic near infrared spectrum fine-grained local interactive semantic reinforcement feature matrices is aggregated along the channel dimension to obtain the updated near infrared spectrum-basic near infrared spectrum semantic fine-grained comparison feature map.

9. The repetitive transcranial magnetic stimulation intervention system according to claim 8, characterized in that: The positioning accuracy determination unit is used to: input the updated near-infrared spectrum-basic near-infrared spectrum semantic fine-grained comparison feature map into a classifier-based positioning identifier to obtain a positioning result, and the positioning result is used to indicate whether the positioning accuracy of the target brain area meets the requirements.

Images