Neural regulation intervention method and device based on brain network development deviation and medium

Abstract

The application discloses a neural regulation intervention method and device based on brain network development deviation and a medium, wherein the method comprises the following steps: acquiring multi-modal brain image data and age of a target object; determining a measured brain network integration degree of the target object according to the multi-modal brain image data; inputting the measured brain network integration degree and the age into a brain network development evaluation model to obtain a brain health development index of the target object; wherein the brain health development index is used to represent a standardized position parameter of the measured brain network integration degree of the target object in a same-age healthy population; determining a target stimulation intensity interval of the target object according to a score deviation between the brain health development index and a reference brain health development index corresponding to the age; determining an actual scalp stimulation target point of the target object according to the multi-modal brain image data; and outputting the target stimulation intensity interval and the actual scalp stimulation target point.

Description

Technical Field

[0001] This invention relates to the field of brain science, and more specifically, to a method, device, and medium for neuromodulation intervention based on deviations in brain network development. Background Technology

[0002] Transcranial stimulation techniques, including transcranial electrical stimulation (tES), transcranial magnetic stimulation (TMS), and transcranial light stimulation (tLS), have gradually become important non-pharmacological neuromodulation methods. These techniques act on specific brain regions through different forms of physical energy, regulating neural excitability, synaptic plasticity, and brain network function. They have demonstrated significant application value in regulating prefrontal cognitive function and assisting in the intervention of neuropsychiatric disorders such as attention deficit hyperactivity disorder (ADHD).

[0003] Current transcranial stimulation (TCS) intervention procedures typically involve: target localization and parameter setting based on multimodal signals (such as EEG and near-infrared spectroscopy) or individual imaging data (such as structural and functional magnetic resonance imaging); application of specific wavelengths and power densities of TCS at fixed scalp locations using a stimulation matrix; and assessment of the regulatory effect based on neural signals or metabolic responses before and after stimulation. However, existing methods have significant shortcomings: a cross-age comparable neural baseline reference system is lacking, and differences in neurodevelopment among individuals of different ages make it difficult to quantify the pre-stimulation state; power density often relies on experience or fixed parameter settings, failing to establish a quantitative mapping between neural index deviations and power density, resulting in dose inequality and discrete therapeutic effects; and target selection does not fully incorporate individual brain network connectivity characteristics, making it difficult to stably target the attentional network at fixed scalp points. Summary of the Invention

[0004] One objective of this disclosure is to provide a neuromodulation intervention scheme based on brain network developmental deviations, so as to achieve cross-age standardized characterization of individual neural states, and calibrate the stimulation intensity and network-guided target points according to the quantification, thereby improving the scientific nature, consistency, repeatability and predictability of long-term efficacy of the intervention.

[0005] According to a first aspect of this disclosure, a neuromodulation intervention method based on brain network developmental deviations is provided, the method comprising: Acquire multimodal brain imaging data and age of the target object; wherein, the multimodal brain imaging data includes brain structural imaging data, diffusion tensor imaging data, and functional magnetic resonance imaging data; Based on the multimodal brain imaging data, the measured brain network integration degree of the target object was determined; The measured brain network integration degree and the age are input into the brain network development assessment model to obtain the brain health development index of the target object; wherein, the brain health development index is used to characterize the standardized location parameter of the measured brain network integration degree of the target object in a healthy population of the same age; The target stimulus intensity range for the target object is determined based on the score deviation between the brain health development index and the reference brain health development index corresponding to the age. Based on the multimodal brain imaging data, the actual scalp stimulation target points of the target object are determined; Output the target stimulation intensity range and the actual scalp stimulation target point.

[0006] Optionally, determining the measured brain network integration degree of the target object based on the multimodal brain imaging data includes: Based on a pre-defined brain region template, the brain is divided into multiple brain regions; Based on the diffusion tensor image data, the fiber connection strength between white matter fibers between any two brain regions is determined to obtain a structural connectivity matrix, and the average value of all connection strengths in the structural connectivity matrix is ​​calculated as the structural connectivity strength of the target object. Based on the functional magnetic resonance imaging data, the correlation coefficient between the BOLD time series corresponding to any two brain regions is determined to obtain the functional connectivity matrix, and the average value of all correlation coefficients in the functional connectivity matrix is ​​calculated as the functional connectivity strength of the target object. By fusing the structural connectivity strength with the functional connectivity strength, the measured brain network integration degree of the target object is obtained.

[0007] Optionally, the brain network development assessment model is constructed through the following steps: Obtain a set of multimodal brain imaging data corresponding to a first sample group of any sample age; wherein, the set of multimodal brain imaging data includes multimodal brain imaging data corresponding to each first sample object in the first sample group; Based on the sample multimodal brain imaging data of each first sample object in the sample multimodal brain imaging data set, the sample brain network integration degree of the first sample object is determined, and the sample brain network integration degree set corresponding to the sample age is obtained. The brain network development assessment model is determined based on the set of brain network integration degrees corresponding to different sample ages.

[0008] Optionally, determining the target stimulus intensity range for the target object based on the score deviation between the brain health development index and the reference brain health development index corresponding to the age includes: Obtain the preset maximum adjustment range and adjustment ratio coefficient; The score deviation is determined based on the brain health development index and the reference brain health development index corresponding to the age; The power density adjustment value is determined based on the fractional deviation, the adjustment ratio coefficient, and the maximum adjustment range; Based on the power density adjustment value, the initial power density, and the preset safety tolerance, the target stimulus intensity range for the target object is constructed.

[0009] Optionally, determining the actual scalp stimulation target point of the target object based on the multimodal brain imaging data includes: Based on the functional magnetic resonance imaging data, cortical stimulation targets were identified from the candidate regions of the prefrontal cortex. Based on the structural image data of the target object, the cortical stimulation target points are projected onto the scalp surface to obtain the actual scalp stimulation target points.

[0010] Optionally, determining cortical stimulation target points from candidate regions of the prefrontal cortex based on the functional magnetic resonance imaging data includes: Based on the functional magnetic resonance imaging data, the BOLD time series of all voxels in the dorsal attention network are extracted, and the average value of the BOLD time series is calculated as the network time series of the dorsal attention network. The prefrontal cortex region is divided into voxel cubes of multiple spatial scales. The average value of the BOLD time series within each voxel cube is taken to obtain the candidate region time series for each voxel cube. Calculate the correlation coefficient between the candidate region time series of each voxel cube and the network time series of the back-side attention network; The cortical location corresponding to the voxel cube with the highest correlation coefficient is taken as the cortical stimulation target.

[0011] Optionally, the method further includes: Obtain the first symptom score data of the target subject before transcranial stimulation; Obtain the second symptom score data of the target subject after transcranial stimulation; The first symptom score data and the age are input into the symptom scoring model to obtain a first severity score, and the second symptom score data and the age are input into the symptom scoring model to obtain a second severity score; The difference between the second severity score and the first severity score is calculated as the symptom improvement rate of the target subject, and the symptom improvement rate is output.

[0012] Optionally, the symptom scoring model is determined through the following steps: Obtain the set of sample severity scores corresponding to the second sample group for any given sample age; The symptom scoring model is determined based on the set of severity scores corresponding to different sample ages.

[0013] According to a second aspect of this disclosure, a neuromodulation intervention device based on brain network developmental deviation is also provided, comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions being executed by the processor to implement the steps of the neuromodulation intervention method based on brain network developmental deviation as described in the first aspect.

[0014] According to a third aspect of this disclosure, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the neuromodulation method based on brain network developmental deviation patterns as described in the first aspect.

[0015] One beneficial effect of this disclosure is that the method can obtain the target stimulation intensity range and the actual scalp stimulation target point of the target object, which solves the problems of difficult replication of stimulation schemes and large differences in efficacy in the prior art. It provides an operable and repeatable technical basis for the implementation of individualized transcranial stimulation intervention for ADHD symptoms, efficacy evaluation and long-term strategy optimization.

[0016] Other features and advantages of the embodiments of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments of the present disclosure and, together with their description, serve to explain the principles of the embodiments of the present disclosure.

[0018] Figure 1 This is a hardware configuration block diagram of a neuromodulation intervention device based on brain network developmental deviation that can be used to implement the embodiments of this disclosure; Figure 2 This is a flowchart illustrating a neuromodulation intervention method based on brain network developmental deviation according to an embodiment of this application. Detailed Implementation

[0019] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention.

[0020] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0021] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0022] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0023] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0024] <Hardware Configuration> Figure 1 This is a hardware configuration block diagram of a neuromodulation intervention device based on brain network developmental deviation that can be used to implement embodiments of this disclosure.

[0025] like Figure 1 As shown, the neuromodulation intervention device 1000 based on brain network development deviation may include a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, a display device 1500, an input device 1600, a speaker 1700, a microphone 1800, etc.

[0026] The processor 1100 executes computer programs, which can be written using instruction sets of architectures such as x86, Arm, RISC, MIPS, and SSE. The memory 1200 includes, for example, ROM (Read-Only Memory), RAM (Random Access Memory), and non-volatile memory such as a hard disk. The interface device 1300 includes, for example, a USB interface and a headphone jack. The communication device 1400 is capable of wired or wireless communication. The communication device 1400 may include at least one short-range communication module, such as any module for short-range wireless communication based on protocols such as Hilink, WiFi (IEEE 802.11), Mesh, Bluetooth, ZigBee, Thread, Z-Wave, NFC, UWB, and LiFi. The communication device 1400 may also include a long-range communication module, such as any module for WLAN, GPRS, or 2G / 3G / 4G / 5G long-range communication. The display device 1500 is, for example, an LCD screen or a touch screen. The input device 1600 may include, for example, a touchscreen or a keyboard. Speaker 1700 and microphone 1800 are used for input / output of voice signals.

[0027] The neuromodulation intervention device 1000 based on brain network developmental deviations can be any type of electronic device with computing capabilities, without limitation. For example, the electronic device can be a tablet computer, laptop computer, handheld computer, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc., and can also be a server, network attached storage (NAS), personal computer (PC), etc.

[0028] Those skilled in the art should understand that, although in Figure 1 The present invention illustrates multiple devices of a neuromodulation intervention device 1000 based on brain network developmental deviations. However, the neuromodulation intervention device 1000 based on brain network developmental deviations in this embodiment may involve only some of the devices, for example, only the processor 1100 and the memory 1200. This is well known in the art and will not be described further here.

[0029] In this embodiment, the memory 1200 stores computer program instructions that control the processor 1100 to perform a neuromodulation intervention method based on brain network developmental deviations according to any embodiment of this disclosure. Those skilled in the art can design these instructions based on the disclosed scheme. How these instructions control the processor 1100 to operate is well known in the art and will not be described in detail here.

[0030] <Method Implementation> Figure 2 This is a flowchart illustrating a neuromodulation intervention method based on brain network developmental deviation according to an embodiment of this application.

[0031] like Figure 2 As shown, the neuromodulation intervention method based on brain network developmental deviation in this embodiment may include the following steps S210~S260: Step S210: Obtain multimodal brain imaging data and age of the target subject.

[0032] The multimodal brain imaging data includes brain structural imaging data, diffusion tensor imaging data, and functional magnetic resonance imaging data.

[0033] In this embodiment, the target subject may be an individual with attention deficit hyperactivity disorder (ADHD) who needs transcranial stimulation intervention or other individuals who need intervention, and there is no limitation here.

[0034] Among them, brain structural imaging data can be structural images (such as T1-weighted structural images), which are used to obtain brain structural morphological parameters (such as gray matter volume, cortical thickness, etc.).

[0035] Diffusion tensor imaging data can be diffusion tensor imaging (DTI) data, used to calculate white matter structure connectivity indices. Functional magnetic resonance imaging data can be functional magnetic resonance time series.

[0036] Functional magnetic resonance imaging (fMRI) data can include resting-state fMRI data. Resting-state fMRI data is obtained by keeping the subject awake with eyes closed and avoiding deliberate cognitive activity, with a scan duration of 8-12 minutes.

[0037] In one example, functional magnetic resonance imaging (fMRI) data may also include task-state fMRI data with added Go / No-Go tasks or reward-delayed tasks to enhance the functional signals of the dorsal attention network (DA network). The task-state scan duration can be adjusted to 5-8 minutes depending on the task design.

[0038] Before performing step S220, the multimodal brain imaging data undergoes preprocessing such as time correction, head motion correction (e.g., head translation less than 2 mm, rotation less than 2°), spatial normalization (registered to a normalized brain spatial reference template), smoothing (e.g., Gaussian kernel 4-6 mm), and bandpass filtering (e.g., frequency range of 0.01-0.08 Hz), and quality control is implemented (e.g., samples with excessive head motion or excessive signal-to-noise ratio are removed) to ensure the stability and consistency of subsequent calculations.

[0039] It should be noted that the acquisition of multimodal brain imaging data must follow clinical imaging acquisition standards.

[0040] Step S220: Determine the measured brain network integration degree of the target object based on the multimodal brain imaging data.

[0041] In this embodiment, the measured brain network integration degree is a core indicator reflecting the level of structural and functional connectivity in the target subject's brain. The role of the measured brain network integration degree is to quantify the overall synergistic level of the target subject's brain network, and its value directly reflects the health of the target subject's brain development.

[0042] In one embodiment of this application, step S220, which determines the measured brain network integration degree of the target object based on the multimodal brain imaging data, includes steps S220.1 to S220.4.

[0043] Step S220.1: Based on a preset brain region template, the brain is divided into multiple brain regions.

[0044] In this embodiment, the selection of brain region templates should be based on research needs, prioritizing widely applicable and precisely zoning templates. Once a brain region template is selected, all objects (target objects and sample objects) must use that template to ensure the comparability of indicators. Furthermore, brain regions within the template can be merged or subdivided according to actual needs to further optimize the computational efficiency and accuracy of the connectivity matrix.

[0045] The preset brain region template can be any existing brain region template, such as an automatic anatomical marker template (dividing the brain into 90 brain regions, 45 in each hemisphere), a Harvard-Oxford brain region template, or a brain network atlas template, etc. There are no restrictions here.

[0046] The number of brain regions after division is denoted as N, which lays the foundation for the subsequent construction of N×N structural connectivity matrix and functional connectivity matrix, and avoids index deviation caused by inconsistent brain region division.

[0047] Step S220.2: Based on the diffusion tensor image data, determine the fiber connection strength between white matter fibers between any two brain regions among the plurality of brain regions, obtain the structural connectivity matrix, and calculate the average value of all connection strengths in the structural connectivity matrix as the structural connectivity strength of the target object.

[0048] In this embodiment, white matter fibers between different brain regions are tracked in diffusion tensor imaging data. Furthermore, fiber tractography techniques (such as deterministic tracing and probabilistic tracing) are used to statistically analyze the number or strength of fiber connections between any two brain regions (connection strength can be quantified using indicators such as the number of fiber tracts or the average anisotropy fraction), thereby constructing an N×N structural connectivity matrix. Each element C in this matrix (i and j are brain region numbers, i≠j) corresponds to the structural connectivity strength between the i-th and j-th brain regions. The diagonal elements (i=j) are set to 0 (no connection within the same brain region). The structural connectivity strength of the target object is obtained by calculating the average of all connectivity strengths in this matrix.

[0049] Structural connectivity strength reflects the overall integrity and strength level of white matter connectivity in a target object and is an important component of the measured brain network integration.

[0050] It should be noted that the calculation of structural connection strength must exclude abnormal fiber bundles (such as false connections caused by noise) to ensure the authenticity of the indicators.

[0051] Step S220.3: Based on the functional magnetic resonance imaging data, determine the correlation coefficient between the BOLD time series corresponding to any two brain regions to obtain the functional connectivity matrix, and calculate the average value of all correlation coefficients in the functional connectivity matrix as the functional connectivity strength of the target object.

[0052] In this embodiment, firstly, brain region signals are extracted from the functional magnetic resonance imaging data (preprocessed functional magnetic resonance imaging data), and the average value of the blood oxygen level dependent signal time series (referred to as BOLD time series) of all voxels in each brain region is taken to obtain the BOLD time series TS (i=1,2,...,N) for each brain region.

[0053] Then, the correlation coefficient (e.g., Pearson correlation coefficient) between two BOLD time series corresponding to any two brain regions is calculated to quantify the synchronicity of functional activities in different brain regions, thereby constructing an N×N functional connectivity matrix. Each element R (i and j are brain region numbers, i≠j) in this functional connectivity matrix corresponds to the functional connectivity strength between the i-th brain region and the j-th brain region.

[0054] Then, the average value of all correlation coefficients in the functional connectivity matrix is ​​calculated to obtain the functional connectivity strength of the target object.

[0055] Functional connectivity strength can reflect the level of coordination of functional activities in the target brain region.

[0056] It should be noted that after obtaining the functional connectivity strength, Fisher's Z-transform can be performed to reduce data skewness and improve the accuracy of subsequent statistical analysis.

[0057] Step S220.4: The structural connectivity strength and the functional connectivity strength are fused to obtain the measured brain network integration degree of the target object.

[0058] In this embodiment, the fusion method can be a feature fusion function: ,in, The measured brain network integration of the target subjects. For the structural connection strength of the target object, For the functional connection strength of the target object, This is the feature fusion function.

[0059] In one specific embodiment, a weighted linear combination form can be adopted, as detailed in formula (1): (1) Among them, α and β are preset weight coefficients, which can be determined based on experience (the range of values ​​for α and β is 0 < α, β < 1, and α + β = 1, for example α = 0.4, β = 0.6, which can be adjusted according to the model validation results).

[0060] The fusion method can also use nonlinear fusion functions (such as the sigmoid function or neural network fusion), but the weighted linear combination method has the advantages of simple calculation and strong interpretability, and is more suitable for clinical translational applications.

[0061] By combining the anatomical connectivity features of the brain reflected by structural connectivity strength with the functional synergistic features of the brain reflected by functional connectivity strength, high-dimensional structural and functional network features are mapped to a single scalar measured brain network integration degree, reducing the complexity of subsequent modeling and analysis. At the same time, it comprehensively reflects the overall state of the target object's brain network, ensuring that the indicator can accurately characterize the target object's brain development level.

[0062] Step S230: Input the measured brain network integration degree and the age into the brain network development assessment model to obtain the brain health development index of the target object.

[0063] The brain health development index is used to characterize the standardized location parameters of the measured brain network integration degree of the target object in a healthy population of the same age.

[0064] In this embodiment, the brain network development assessment model reflects the distribution characteristics of brain network integration in healthy individuals at different ages, thereby providing a reference standard for brain network integration in healthy individuals of different ages and achieving standardized quantification of the measured brain network integration of the target subjects.

[0065] By inputting the measured brain network integration degree and age of the target subject into the model, the brain network state of the target subject can be mapped to the distribution of healthy people of the same age, thereby obtaining the relative position of the brain network state in the same age group, judging the degree of deviation between the brain development level of the target subject and that of healthy people of the same age, and providing a core reference for the individualized calibration of subsequent stimulus intensity.

[0066] In one embodiment of this application, the brain network development assessment model is constructed through the following steps S110 to S130: Step S110: Obtain the sample multimodal brain imaging data set corresponding to the first sample group of any sample age; wherein, the sample multimodal brain imaging data set includes the sample multimodal brain imaging data corresponding to each first sample object in the first sample group.

[0067] In this embodiment, the sample age can be any age between 6 and 25 years old. That is, healthy individuals aged between 6 and 25 years old (i.e., individuals excluding those with neurological diseases, mental illnesses, and contraindications to magnetic resonance imaging; healthy individuals are also referred to as the first sample subjects) are selected to obtain the first sample group.

[0068] Each first sample object in the first sample group corresponds to a sample multimodal brain imaging data.

[0069] To ensure the statistical reliability of the brain network development assessment model, ages 6 to 25 can be stratified, with each year representing one age group. Each age group corresponds to a primary sample population. The sample size of the primary sample population should be no less than 30 cases (a sample size of less than 30 cases in the primary sample population will lead to excessive model fitting bias, affecting the accuracy of the assessment).

[0070] The multimodal brain imaging data of the samples is basically the same as the multimodal brain imaging data of the target object in step S210, and will not be described again here. Similarly, the multimodal brain imaging data of the samples are preprocessed and quality controlled (such as head movement correction, standardization, filtering, etc.).

[0071] It should be noted that the first sample group for any sample age must include healthy individuals of different genders and regions to avoid sample homogeneity leading to insufficient model generalization ability; at the same time, sample collection must obtain ethical approval to ensure compliance with medical research ethics.

[0072] Step S120: Based on the sample multimodal brain image data of each first sample object in the sample multimodal brain image data set, determine the sample brain network integration degree of the first sample object, and obtain the sample brain network integration degree set corresponding to the sample age.

[0073] In this embodiment, the process of determining the integration degree of the sample brain network is basically the same as the process of determining the integration degree of the target object's measured brain network in step S220, and will not be described here.

[0074] Step S130: Determine the brain network development assessment model based on the brain network integration degree set corresponding to different sample ages.

[0075] In this embodiment, a Generalized Additive Models for Location, Scale and Shape (GAMLSS) or other suitable nonlinear fitting methods are used. The sample brain network integration degree is used as the response variable, and sample age as the independent variable to fit the distribution characteristics of sample brain network integration degree under different age conditions. During the model fitting process, a smoothing function (such as a cubic spline function) is used to fit the relationship between distribution parameters (mean, scale, skewness, and kurtosis) and age.

[0076] Subsequently, the maximum likelihood estimation method was used to solve for the model parameters, and stable model parameters were obtained through iterative optimization algorithms (such as the Newton-Raphson algorithm). Finally, a brain network development assessment model was constructed that reflects the distribution characteristics of brain network integration in healthy individuals at different ages. This model can simultaneously characterize the nonlinear trend of brain network integration with age and its distribution morphology, providing a reliable population reference standard for calculating the brain health development index of the target population.

[0077] It should be noted that after the model is built, cross-validation (such as 10-fold cross-validation) should be used to evaluate the model performance to ensure the model's fit and prediction accuracy. If the model performance is not up to standard, the sample size should be increased or the model parameters should be adjusted for refitting.

[0078] Step S240: Determine the target stimulus intensity range for the target object based on the score deviation between the brain health development index and the reference brain health development index corresponding to the age.

[0079] In this embodiment, a target level (such as the 50th or 60th percentile) in the distribution of brain network integration among healthy individuals of the same age is preset as a reference benchmark for the brain health development index of the target subject, i.e., the reference brain health development index of the target subject.

[0080] Score deviation can be the difference between the target subject's current brain health development index and the reference brain health development index, used to quantify the degree of deviation between the target subject and the healthy population.

[0081] This deviation is then converted into a specific basis for adjusting the stimulation power density, ensuring that the stimulation intensity matches the target subject's current brain health development index, thus achieving safe and controllable individualized intervention.

[0082] In one embodiment of this application, step S240 determines the target stimulus intensity range for the target object based on the score deviation between the brain health development index and the reference brain health development index corresponding to the age, including steps S240.1 to S240.4.

[0083] Step S240.1: Obtain the preset maximum adjustment range and adjustment ratio coefficient.

[0084] In this embodiment, the preset maximum adjustment range is used to limit the adjustment range of the power density of transcranial stimulation, avoiding safety risks caused by excessive adjustment (such as damage to the scalp or brain tissue due to excessive power). The specific value can be, for example, ±5 mW / cm². The value of the maximum adjustment range can be adjusted according to clinical safety standards. The safe power density range for transcranial stimulation is 0.5-10 mW / cm², therefore the maximum adjustment range needs to be controlled within this range.

[0085] The adjustment ratio coefficient is used to control the degree of influence of fractional deviation on the power density adjustment value. The specific value can be, for example, 0.5. This value can be optimized based on sample data fitting or clinical experience. The adjustment ratio coefficient ranges from 0.3 to 0.7, thereby balancing adjustment sensitivity and safety.

[0086] The maximum adjustment range and adjustment ratio can also be fine-tuned according to the age and gender of the target group (e.g., the maximum adjustment range for children can be appropriately reduced to ±3 mW / cm²).

[0087] Step S240.2: Determine the score deviation based on the brain health development index and the reference brain health development index corresponding to the age.

[0088] In this embodiment, the sign of the fractional deviation directly determines the direction of power density adjustment, providing a clear logical basis for subsequent power density adjustments. If the fractional deviation is negative, it indicates that the target subject's brain network integration is lower than the reference level for healthy individuals of the same age (i.e., the brain health development index for the same age), and the stimulation power density needs to be appropriately increased to enhance the intervention effect. If the fractional deviation is positive, it indicates that the target subject's brain network integration is higher than the reference level for healthy individuals of the same age, and the stimulation power density needs to be appropriately reduced to avoid overstimulation. If the fractional deviation is 0, it indicates that the target subject's brain network integration is consistent with the reference level for healthy individuals of the same age, and no adjustment to the initial power density is required, providing a clear basis for subsequent power density adjustments.

[0089] The larger the absolute value of the score deviation, the more significant the deviation between the target object and the reference level of healthy people of the same age. The adjustment range of power density needs to be increased accordingly, but it should not exceed the maximum adjustment range.

[0090] Step S240.3: Determine the power density adjustment value based on the fractional deviation, the adjustment ratio coefficient, and the maximum adjustment range.

[0091] In this embodiment, the power density adjustment value is calculated using formula (2): (2) in, Here, α is the power density adjustment value, and α is the adjustment ratio coefficient. For fractional deviation, This is the maximum adjustment range. Simultaneously, the absolute value of the power density adjustment must not exceed the maximum adjustment range. This means that when the power density adjustment value is greater than the maximum adjustment range, the maximum adjustment range is used. When the power density adjustment value is less than the maximum adjustment range, the power density adjustment value is used to avoid safety risks caused by excessive adjustment range.

[0092] This formula transforms the standardized index of fractional deviation into a specific power density adjustment value, ensuring that the adjustment value accurately matches the neural state deviation of the target subject. This avoids the uncertainty brought about by empirical adjustments, provides a unified quantitative standard for the adjustment of stimulation power density for different target subjects, and improves the repeatability of intervention programs.

[0093] Step S240.4: Based on the power density adjustment value, the initial power density, and the preset safety tolerance, construct the target stimulus intensity range for the target object.

[0094] In this embodiment, the target stimulus intensity range of the target object can be calculated using the following formula (3). : (3) in, For the target power density, For the initial power density, A preset safety tolerance (e.g., 0.5 mW / cm²) is used to ensure the range that can be finely adjusted later.

[0095] This process enables individualized optimization of the initial power density, allowing the target power density to accurately match the neural state of the target subject. It ensures that the brain network integration of the target subject after stimulation can approach the reference range of healthy people of the same age, solving the dose inequivalence problem caused by fixed power density in the existing technology, and providing a guarantee for the stability and repeatability of subsequent intervention effects.

[0096] Based on the above, individualized calibration of transcranial stimulation power density can be achieved, solving the problems of stimulation power density dependence on experience setting and dose inequivalence in existing technologies.

[0097] Step S250: Determine the actual scalp stimulation target points of the target object based on the multimodal brain imaging data.

[0098] In this embodiment, based on the multimodal brain imaging data of the target object and combined with the functional characteristics of the dorsal attention network (DA network), cortical target points that can effectively regulate the DA network are screened and accurately mapped onto the scalp surface to ensure that the stimulation can accurately act on the DA network that is closely related to the symptoms of attention deficit hyperactivity disorder (ADHD).

[0099] The DA network mainly includes regions such as the prefrontal cortex and parietal cortex.

[0100] In one embodiment of this application, step S250 determines the actual scalp stimulation target of the target object based on the multimodal brain imaging data, including steps S250.1 to S250.2.

[0101] Step S250.1: Based on the functional magnetic resonance imaging data, determine the cortical stimulation target from the candidate region of the prefrontal cortex.

[0102] In this embodiment, the prefrontal cortex is involved in higher cognitive functions such as attention regulation and impulse control, and its functional abnormalities are closely related to ADHD symptoms. The prefrontal cortex is also a core intervention area for transcranial stimulation.

[0103] The extent of the prefrontal candidate region can be determined based on brain region templates (such as the prefrontal region in an automated anatomical marker template) or based on the spatial distribution of the DA network, ensuring that the prefrontal candidate region covers all prefrontal regions that may be involved in DA network regulation.

[0104] By analyzing the functional synergy between candidate regions in the prefrontal cortex and the DA network in functional magnetic resonance imaging data, the optimal cortical target points are selected, providing accurate cortical location references for subsequent scalp mapping and avoiding blind target selection.

[0105] By determining the cortical stimulation target based on the functional connectivity strength of the DA network, it is ensured that the target can effectively modulate the DA network (the key functional network for goal-oriented attention and attention maintenance).

[0106] In one embodiment of this application, step S250.1 determines cortical stimulation target points from the candidate region of the prefrontal cortex based on the functional magnetic resonance imaging data, including steps S250.11 to S250.14.

[0107] Step S250.11: Based on the functional magnetic resonance imaging data, extract the BOLD time series of all voxels in the dorsal attention network, calculate the average value of the BOLD time series, and use it as the network time series of the dorsal attention network.

[0108] In this embodiment, firstly, the voxel range of the DA network is determined according to the DA network spatial template (such as the DA network partition in the brain network atlas template).

[0109] Then, based on the functional magnetic resonance imaging data, the BOLD time series corresponding to all voxels within the voxel range are extracted. The average of all BOLD time series corresponding to all voxels is calculated to obtain the network time series of the DA network.

[0110] Network time series can reflect the overall functional activity rhythm of the DA network, providing a benchmark for subsequent calculation of the functional connectivity strength between candidate prefrontal regions and the DA network.

[0111] It should be noted that when extracting the voxel range of the DA network, spatial registration of the functional magnetic resonance imaging data is required to ensure that the DA network template is aligned with the brain imaging data of the target object and to avoid voxel extraction bias.

[0112] Step S250.12: Divide the prefrontal cortex region into multiple spatial scale voxel cubes, and average the BOLD time series within each voxel cube to obtain the candidate region time series for each voxel cube.

[0113] In this embodiment, the prefrontal cortex region is divided into multiple voxel cubes of this scale to ensure coverage of the entire prefrontal cortex candidate region, thereby achieving a fine division of the prefrontal cortex region and ensuring the comprehensiveness and representativeness of the candidate region.

[0114] The BOLD time series of all voxels within each voxel cube are averaged to obtain the candidate region time series for each voxel cube.

[0115] The time series of these candidate regions can reflect the functional activity rhythm of each prefrontal candidate region, providing functional characteristic data of individual candidate regions for subsequent calculation of functional connectivity strength with the DA network.

[0116] It should be noted that the division of voxel cubes should avoid overlap, ensuring that each voxel belongs to only one cube. At the same time, cubes containing non-prefrontal voxels should be removed to improve the accuracy of candidate regions.

[0117] Step S250.13: Calculate the correlation coefficient between the candidate region time series of each voxel cube and the network time series of the back-side attention network.

[0118] In this embodiment, the correlation coefficient can be selected as the Pearson correlation coefficient, which is used to quantify the functional connectivity strength between each prefrontal candidate region and the DA network.

[0119] The closer the correlation coefficient is to 1, the stronger the functional synergy between the candidate region and the DA network, and the more significant the regulatory effect on the DA network may be. The closer the correlation coefficient is to -1 or 0, the weaker the functional synergy between the candidate region and the DA network, and the worse the regulatory effect may be.

[0120] This step enables the screening and ranking of candidate regions in the prefrontal cortex, providing a quantitative basis for the selection of the optimal cortical target.

[0121] Step S250.14: The cortical location corresponding to the voxel cube with the largest correlation coefficient is taken as the cortical stimulation target.

[0122] In this embodiment, the voxel cube with the highest correlation coefficient is first selected, and then the center position of the voxel cube is used as the cortical stimulation target.

[0123] This cortical stimulation target point has the strongest functional connectivity with the DA network, ensuring that when transcranial stimulation is applied to this target point, it maximizes the regulation of the DA network's functional activity and improves ADHD symptoms.

[0124] In one example, the structural connectivity strength established through the above steps can be used to further constrain the functional connectivity strength, thereby further improving the accuracy of cortical target selection.

[0125] In one example, if there are multiple voxel cubes with large correlation coefficients, the average value of the center positions of these voxel cubes can be used as the target point for cortical stimulation, thus avoiding target deviation caused by a single cube.

[0126] By selecting cortical stimulation targets based on the individual brain network connectivity characteristics of the target object, the limitations of fixed scalp localization are avoided, and individualized selection of cortical targets is achieved.

[0127] Step S250.2: Based on the structural image data of the target object, project the cortical stimulation target point onto the scalp surface to obtain the actual scalp stimulation target point.

[0128] In this embodiment, the projection method can employ the Balloon Inflation Algorithm.

[0129] The core principle of the balloon inflation algorithm is as follows: starting from the cortical stimulation target point, the brain expands outward along the normal direction of the brain surface until it reaches the scalp surface. During the expansion process, it avoids non-scalp tissues such as the skull, and finally obtains the coordinates corresponding to the scalp surface, which is the actual scalp stimulation target point. This algorithm can achieve precise mapping of cortical target points to the scalp surface, solving the problem that cortical target points cannot be directly used for clinical stimulation.

[0130] After projection, it is necessary to verify the coordinates using the 10–20 system (a scalp coordinate system based on head circumference percentage) to ensure that the actual location of the scalp stimulation target can be determined by routine clinical positioning methods, which will facilitate clinical application.

[0131] This projection process transforms individualized cortical stimulation targets into actionable scalp stimulation targets, ensuring that subsequent transcranial stimulation interventions can be performed with precise scalp positioning devices (such as positioning caps or navigation systems) to accurately locate the stimulation position. This enables stable and repeatable interventions, providing reliable target information for clinical procedures and avoiding target deviations caused by individual anatomical differences (such as skull thickness and scalp thickness).

[0132] Based on the above, the core of determining the actual scalp stimulation target is to achieve individualized target selection and scalp mapping guided by the dorsal attention network (DA network), which solves the problem that fixed scalp positioning (such as the 10–20 system) in the existing technology cannot guarantee effective control of the target network.

[0133] Step S260: Output the target stimulation intensity range and the actual scalp stimulation target point.

[0134] In this embodiment, the target stimulation intensity range provides an individualized power density range for transcranial stimulation intervention, ensuring that the stimulation intensity matches the neural state of the target subject and achieving a standardized neural perturbation effect. The actual scalp stimulation target point provides a precise stimulation location for transcranial stimulation intervention, ensuring that the stimulation can effectively act on the DA network and improve the intervention effect.

[0135] The combination of these two technologies constitutes a complete system of individualized transcranial stimulation intervention parameters, which solves the problems of difficult replication of stimulation protocols and large differences in efficacy in existing technologies. It provides an operable and repeatable technical basis for the implementation, efficacy evaluation and long-term strategy optimization of individualized transcranial stimulation intervention for ADHD symptoms.

[0136] In one embodiment of this application, the method further includes steps S270 to S300.

[0137] Step S270: Obtain the first symptom score data of the target object before transcranial stimulation.

[0138] In this embodiment, the first symptom score data can be the symptom score before intervention, which is derived from a standardized attention deficit hyperactivity disorder (ADHD) symptom assessment scale. Such scales include the Connors Attention Deficit Hyperactivity Disorder Rating Scale, the Attention Deficit Hyperactivity Disorder Diagnostic Interview Scale, or the Swanson-Nolan-Pelham 4th Edition Rating Scale, etc. Scales with wide clinical application and high reliability and validity are preferred.

[0139] The scoring data can be obtained through assessments by medical staff, parents, or teachers (for children and adolescents) to ensure the objectivity and accuracy of the scores. The data collection time should be within 24-48 hours before the start of transcranial stimulation intervention to avoid excessive time intervals that may cause symptom fluctuations and affect the reliability of baseline data.

[0140] The first symptom scoring data should include complete scale scores (including total score and scores for each dimension, such as inattention and impulsivity), as well as information such as the rater and the time of the rating, to facilitate subsequent tracking and comparative analysis. If the target subject has comorbidities (such as anxiety or depression), scores on comorbid-related scales should also be collected to exclude the interference of comorbidities on ADHD symptom scores. Step S280: Obtain the second symptom score data of the target object after transcranial stimulation.

[0141] In this embodiment, the second symptom score data is the post-intervention symptom score, used to reflect the changes in symptoms after transcranial stimulation intervention. It is compared with the first symptom score data and is the core input for calculating the extent of symptom improvement. Data collection requirements include: using the same standardized scale as the first symptom score data to ensure consistent scoring dimensions and standards, avoiding comparison bias due to different scales; prioritizing the same personnel collecting data as those collecting the first symptom score data to reduce rater bias; and data collection within 24-48 hours after the end of the transcranial stimulation intervention (a typical transcranial stimulation intervention lasts 2-4 weeks, 3-5 times per week, with each intervention lasting 20-30 minutes). The transcranial stimulation intervention uses uniform geometric parameters, maintaining consistent stimulation frequency and cycle, and keeping the stimulation power density constant throughout without dynamic adjustment. For multiple intervention courses, data can be collected once after each course for dynamic evaluation of the intervention effect.

[0142] When collecting second symptom score data, it is necessary to record the intervention completion status of the target subjects (such as the actual number of interventions, the power density of each intervention and whether the stimulation target points meet the preset parameters, the stimulation power density being the target stimulation intensity range, and the stimulation target points being the actual scalp stimulation target points). If there are any intervention interruptions or parameter adjustments, these should be noted in the score data to facilitate subsequent analysis of the correlation between the intervention effect and the intervention parameters. At the same time, it is necessary to exclude other factors that may affect the symptoms during the intervention (such as taking other medications or receiving other intervention treatments) to ensure that the symptom changes are mainly caused by transcranial stimulation intervention.

[0143] Behavioral data such as activity level, attention stability, and sleep rhythm before intervention can be collected simultaneously through wearable devices as auxiliary assessment indicators.

[0144] Step S290: Input the first symptom score data and the age into the symptom scoring model to obtain a first severity score, and input the second symptom score data and the age into the symptom scoring model to obtain a second severity score. In this embodiment, since the manifestations and scoring standards of ADHD symptoms differ at different age stages, such as the different criteria for judging the severity of symptoms in children and adolescents.

[0145] The role of the symptom scoring model is to map the original symptom score data (total scale score or dimension score) to standardized positional parameters in the same age ADHD group, such as symptom percentile, thereby eliminating the influence of age differences on symptom scores, ensuring that the severity scores before and after intervention are comparable, and thus accurately quantifying the extent of symptom improvement.

[0146] The first severity score corresponds to the severity of symptoms before intervention within the same age group with ADHD. The second severity score corresponds to the severity of symptoms after intervention within the same age group with ADHD. Both are scalar indicators, facilitating subsequent calculation of the difference.

[0147] The output of the symptom scoring model can simultaneously include severity levels (e.g., mild, moderate, severe), which, when used in conjunction with the severity score, more intuitively reflect the symptom changes of the target individual. If the symptom score data of the target individual contains missing values, it is necessary to use missing value imputation methods (e.g., mean imputation, regression imputation) to process the data before inputting it into the model to calculate the severity score, thus avoiding calculation bias caused by missing values. In one embodiment of this application, the symptom scoring model is determined through the following steps S1 and S2: Step S1: Obtain the set of sample severity scores corresponding to the second sample group for any sample age. In this embodiment, the second sample group includes multiple second sample subjects. The second sample subjects are individuals with ADHD. The sample age can be any age between 6 and 25 years old. That is, individuals with ADHD aged between 6 and 25 years old are selected to form the second sample group.

[0148] Each second sample object in the second sample group corresponds to a sample severity score.

[0149] To ensure the statistical reliability of the symptom scoring model, ages 6 to 25 can be stratified, with each year representing an age group. Each age group corresponds to a secondary sample group. The sample size of the secondary sample group should be no less than 50 cases (a sample size of less than 50 cases in the secondary sample group will lead to excessive model fitting bias, affecting the accuracy of the assessment).

[0150] The second sample group should include individuals with ADHD of different symptom severity (mild, moderate, and severe) to avoid insufficient generalization ability of the model due to sample concentration; at the same time, samples with severe neurological diseases, comorbid mental illnesses, or other intervention history should be excluded to ensure the relevance and reliability of the sample data. Step S2: Determine the symptom scoring model based on the set of severity scores corresponding to different sample ages.

[0151] In this embodiment, the method for constructing the symptom scoring model is basically the same as that for the brain network development assessment model, and will not be described here.

[0152] Step S300: Calculate the difference between the second severity score and the first severity score as the symptom improvement rate of the target subject, and output the symptom improvement rate. In this embodiment, the degree of symptom improvement should be positively correlated with the change in the Brain Health Development Index; that is, the greater the increase in the Brain Health Development Index, the greater the symptom improvement. If a negative correlation is found, it is necessary to investigate whether the intervention parameters are reasonable and whether the symptom score data is accurate. Simultaneously, the output results should include the change in severity level before and after the intervention, allowing medical staff and target subjects to intuitively understand the intervention effect and providing a basis for optimizing subsequent intervention strategies. For example, if symptom improvement is not significant, the stimulation parameters can be recalibrated based on the Brain Health Development Index, or the intervention course can be extended.

[0153] In one example, steps S210 to S300 can be performed on target subjects of different ages to obtain the symptom improvement magnitude for each age group. Analysis can then compare the age-related modulatory effect of transcranial stimulation on ADHD symptoms, assuming the neural stimulation effect has been individually calibrated. If the overall level of symptom improvement is significantly higher in a certain age range than in other age ranges, it indicates that transcranial stimulation can achieve better long-term symptom improvement within that age range. Conversely, if the symptom improvement remains relatively limited in other age groups, it indicates a weaker clinical response to transcranial stimulation in those age groups.

[0154] Therefore, based on the ADHD symptom scoring model, a mapping relationship between the symptom regulation effect of personalized stimulation parameters (including actual scalp stimulation target points and target stimulation intensity range) of the target subjects can be established, and the optimal intervention age range for transcranial stimulation to improve ADHD symptoms can be further defined.

[0155] <Media Example> In this embodiment of the disclosure, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the neuromodulation intervention method based on brain network developmental deviation patterns as described in any of the above embodiments.

[0156] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0157] This disclosure may be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement any of the methods in the foregoing embodiments of this disclosure.

[0158] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media may include, for example, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), compact disc-read-only memory (CD-ROM), digital versatile disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any combination thereof. The computer-readable storage medium used herein is not to be interpreted as a transient signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0159] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include one or more of copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to computer-readable storage media in the respective computing / processing device.

[0160] The computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source or object programs written in any combination of one or more programming languages, including object-oriented programming languages ​​(such as Smalltalk, C++, etc.) and conventional procedural programming languages ​​(such as the "C" language or similar programming languages). The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network (e.g., a local area network or a wide area network), or it may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays, or programmable logic arrays, can execute computer-readable program instructions to implement various aspects of the embodiments of this disclosure by utilizing state information from the computer-readable program instructions.

[0161] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0162] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0163] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0164] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It should be noted that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are all equivalent.

[0165] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of this disclosure is defined by the appended claims.

Claims

1. A neuromodulation intervention method based on deviations in brain network development, characterized in that, include: Acquire multimodal brain imaging data and age of the target object; wherein, the multimodal brain imaging data includes brain structural imaging data, diffusion tensor imaging data, and functional magnetic resonance imaging data; Based on the multimodal brain imaging data, the measured brain network integration degree of the target object was determined; The measured brain network integration degree and the age are input into the brain network development assessment model to obtain the brain health development index of the target object; wherein, the brain health development index is used to characterize the standardized location parameter of the measured brain network integration degree of the target object in a healthy population of the same age; The target stimulus intensity range for the target object is determined based on the score deviation between the brain health development index and the reference brain health development index corresponding to the age. Based on the multimodal brain imaging data, the actual scalp stimulation target points of the target object are determined; Output the target stimulation intensity range and the actual scalp stimulation target point; The step of determining the target stimulus intensity range for the target object based on the score deviation between the brain health development index and the reference brain health development index corresponding to the age includes: Obtain the preset maximum adjustment range and adjustment ratio coefficient; The score deviation is determined based on the brain health development index and the reference brain health development index corresponding to the age; The power density adjustment value is determined based on the fractional deviation, the adjustment ratio coefficient, and the maximum adjustment range; Based on the power density adjustment value, the initial power density, and the preset safety tolerance, the target stimulus intensity range for the target object is constructed.

2. The method according to claim 1, characterized in that, The step of determining the measured brain network integration degree of the target object based on the multimodal brain imaging data includes: Based on a pre-defined brain region template, the brain is divided into multiple brain regions; Based on the diffusion tensor image data, the fiber connection strength between white matter fibers between any two brain regions is determined to obtain a structural connectivity matrix, and the average value of all connection strengths in the structural connectivity matrix is ​​calculated as the structural connectivity strength of the target object. Based on the functional magnetic resonance imaging data, the correlation coefficient between the BOLD time series corresponding to any two brain regions is determined to obtain the functional connectivity matrix, and the average value of all correlation coefficients in the functional connectivity matrix is ​​calculated as the functional connectivity strength of the target object. By fusing the structural connectivity strength with the functional connectivity strength, the measured brain network integration degree of the target object is obtained.

3. The method according to claim 1, characterized in that, The brain network development assessment model is constructed through the following steps: Obtain a set of multimodal brain imaging data corresponding to a first sample group of any sample age; wherein, the set of multimodal brain imaging data includes multimodal brain imaging data corresponding to each first sample object in the first sample group; Based on the sample multimodal brain imaging data of each first sample object in the sample multimodal brain imaging data set, the sample brain network integration degree of the first sample object is determined, and the sample brain network integration degree set corresponding to the sample age is obtained. The brain network development assessment model is determined based on the set of brain network integration degrees corresponding to different sample ages.

4. The method according to claim 1, characterized in that, The step of determining the actual scalp stimulation target points of the target object based on the multimodal brain imaging data includes: Based on the functional magnetic resonance imaging data, cortical stimulation targets were identified from the candidate regions of the prefrontal cortex. Based on the structural image data of the target object, the cortical stimulation target points are projected onto the scalp surface to obtain the actual scalp stimulation target points.

5. The method according to claim 4, characterized in that, The step of determining cortical stimulation target points from candidate regions of the prefrontal cortex based on the functional magnetic resonance imaging data includes: Based on the functional magnetic resonance imaging data, the BOLD time series of all voxels in the dorsal attention network are extracted, and the average value of the BOLD time series is calculated as the network time series of the dorsal attention network. The prefrontal cortex region is divided into voxel cubes of multiple spatial scales. The average value of the BOLD time series within each voxel cube is taken to obtain the candidate region time series for each voxel cube. Calculate the correlation coefficient between the candidate region time series of each voxel cube and the network time series of the back-side attention network; The cortical location corresponding to the voxel cube with the highest correlation coefficient is taken as the cortical stimulation target.

6. The method according to claim 1, characterized in that, The method further includes: Obtain the first symptom score data of the target subject before transcranial stimulation; Obtain the second symptom score data of the target subject after transcranial stimulation; The first symptom score data and the age are input into the symptom scoring model to obtain a first severity score, and the second symptom score data and the age are input into the symptom scoring model to obtain a second severity score; The difference between the second severity score and the first severity score is calculated as the symptom improvement rate of the target subject, and the symptom improvement rate is output.

7. The method according to claim 6, characterized in that, The symptom scoring model is determined through the following steps: Obtain the set of sample severity scores corresponding to the second sample group for any given sample age; The symptom scoring model is determined based on the set of severity scores corresponding to different sample ages.

8. A neuromodulation intervention device based on brain network developmental deviation, characterized in that, It includes a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the steps of the neuromodulation intervention method based on brain network developmental deviation as described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method of any one of claims 1 to 7.

Images