A sleep disorder transcranial magnetic stimulation target point positioning system and method based on thalamic dorsal medial nucleus functional connection

Abstract

This invention discloses a transcranial magnetic stimulation (TMS) target localization system and method for sleep disorders based on the functional connectivity of the dorsomedial nucleus of the thalamus. The system comprises a data acquisition module that acquires structural magnetic resonance imaging (SMRI) images and resting-state functional magnetic resonance imaging (fMRI) images of the subject; an image preprocessing module that preprocesses the image data; a seed point definition module that segments the dorsomedial nucleus of the thalamus as a seed point in the subject's brain space; a search space definition module that defines the dorsolateral prefrontal cortex as the target search region in the subject's cerebral cortex; a functional connectivity calculation module that extracts the average time series of the seed point and calculates the correlation coefficient between the average time series and the time series of each voxel in the target search region to construct a functional connectivity map; and a target determination module that identifies the voxel coordinates with the strongest functional connectivity to the seed point in the functional connectivity map as the stimulation target for repetitive transcranial magnetic stimulation. This invention uses the dorsomedial nucleus of the thalamus as an anchor point, which is more consistent with the neuropathological mechanism of insomnia and provides stronger mechanism targeting.

Description

Technical Field

[0001] This invention belongs to the fields of biomedical engineering and neuromodulation technology, specifically relating to a transcranial magnetic stimulation target localization system and method for sleep disorders based on the functional connectivity of the dorsomedial nucleus of the thalamus. Background Technology

[0002] Sleep disorders, especially chronic insomnia disorder (CID), have become a global public health problem. Long-term sleep deprivation can lead to cognitive decline, mood regulation disorders, and an increased risk of cardiovascular disease.

[0003] Repetitive transcranial magnetic stimulation (rTMS), a non-invasive physical therapy, uses magnetic fields to penetrate the skull and generate induced currents in the cortex, modulating neuronal excitability. It has shown potential in treating depression, anxiety, and sleep disorders. However, the efficacy of rTMS is highly dependent on the precise localization of the stimulation target.

[0004] In existing technologies, target localization methods for rTMS mainly include: Method 1, the 5cm method: This method involves stimulating the contralateral abductor pollicis brevis muscle to locate the hotspots in the primary motor cortex, then measuring 5cm forward along the scalp to determine the location of the dorsolateral prefrontal cortex (DLPFC). This method ignores individual differences in brain structure and has extremely low accuracy.

[0005] Method 2, the EEG International 10-20 System Method (Traditional Target Group): Using the EEG 10-20 electrode placement system or positioning cap as a reference, select a nearby electrode position, such as the F3 lead in the left DLPFC. Although head circumference is considered, it is still not based on intracranial brain structure, resulting in a relatively large positioning error.

[0006] Method 3, Structural MRI navigation: Based on T1-weighted images, the target point is located in a specific anatomical gyrus (such as the left middle frontal gyrus). Although this method takes into account structural anatomy, it ignores individual differences in functional anatomy.

[0007] In recent years, personalized localization techniques based on functional magnetic resonance imaging (fMRI) have emerged. For example, in the treatment of depression, determining target sites by calculating the functional connectivity between the dorsolateral prefrontal cortex (DLPFC) and the subgenual cingulate gyrus (sgACC) has become a research hotspot. However, the pathological mechanisms of sleep disorders differ significantly from those of depression. Insomnia primarily involves the "hyperarousal" system, and its core pathological basis lies in the dysfunction of the sleep-wake loop, especially the dorsolateral thalamus (MD), which is responsible for maintaining wakefulness and cognitive integration.

[0008] Most existing target localization technologies follow the sgACC strategy used for depression, failing to optimize for the specific pathological circuits of sleep disorders, resulting in poor treatment outcomes for some insomnia patients. Therefore, there is an urgent need for a technology that can accurately locate rTMS individualized targets suitable for the treatment of sleep disorders based on the functional connectivity characteristics of the thalamic MD nucleus. Summary of the Invention

[0009] To address the issues of lack of specificity and low individualized accuracy in target localization during rTMS treatment of sleep disorders in existing technologies, a transcranial magnetic stimulation target localization system and method for sleep disorders based on the functional connectivity of the dorsomedial nucleus of the thalamus is provided.

[0010] The present invention adopts the following technical solution: On one hand, the present invention provides a transcranial magnetic stimulation target localization system for sleep disorders based on the functional connectivity of the dorsomedial nucleus of the thalamus, the system comprising: The data acquisition module acquires structural magnetic resonance imaging (MRI) images and resting-state functional magnetic resonance imaging (fMRI) images of the subjects. The image preprocessing module preprocesses the acquired image data; The seed point definition module segments the dorsomedial thalamic nucleus in the subject's brain space as a seed point. The search space definition module defines the dorsolateral prefrontal cortex as the target search area in the subject's cerebral cortex. The functional connectivity calculation module extracts the average time series of seed points and calculates the correlation coefficient between the average time series and the time series of each voxel in the target search area to construct a functional connectivity map. The target determination module identifies the voxel coordinates with the strongest functional connectivity to the seed point in the constructed functional connectivity map and uses them as the stimulation target for repetitive transcranial magnetic stimulation.

[0011] Preferably, the seed point definition module uses AAL3 mapping to segment the dorsolateral thalamic nucleus in the subject's brain space as a seed point.

[0012] Preferably, the image preprocessing module performs motion correction, spatial normalization, and noise reduction preprocessing on the resting-state functional magnetic resonance imaging.

[0013] Furthermore, the system also includes a data interface connected to a neural navigation device for transmitting the coordinates of the stimulation target obtained in the target determination module to the neural navigation device and guiding the placement of the magnetic stimulation coil.

[0014] On the other hand, the present invention also provides a method for locating transcranial magnetic stimulation targets for sleep disorders based on the functional connectivity of the dorsomedial nucleus of the thalamus. The method involves acquiring structural magnetic resonance images and resting-state functional magnetic resonance images of the subject, and preprocessing the obtained image data; determining the anatomical location of the dorsomedial nucleus of the thalamus as a seed point; calculating the voxel-level functional connectivity strength between the seed point and the entire dorsolateral prefrontal cortex; identifying the coordinates of the voxel with the highest functional connectivity strength, and using it as the stimulation target output for repetitive transcranial magnetic stimulation.

[0015] The preprocessing of the obtained image data specifically includes: The first 10 time points were removed to eliminate magnetic saturation effects; rigid body transformation was used to align all resting-state functional magnetic resonance images; the aCompCor denoising method was used to automatically segment white matter and cerebrospinal fluid masks, and the first 5 principal components were extracted as covariates for regression to obtain the regression average signal; a bandpass filter of 0.01-0.08Hz was used to bandpass filter the regression average signal to retain low-frequency fluctuations.

[0016] Preferably, the AAL3 map is used to calculate the nonlinear deformation field from the MNI standard space to the subject's individual space; the nonlinear deformation field is used to inversely map the dorsomedial nucleus (DNU) label in the AAL3 map to the subject's individual brain space; the DNU region is identified and segmented as a seed point.

[0017] Preferably, the specific method for calculating the stimulation target is: The dorsolateral prefrontal cortex was defined as the target search area in the subject's cerebral cortex. Calculate the Pearson correlation coefficient r between the time series Tv of each voxel v in the target search area and the average time series TMD of the dorsomedial thalamus. The correlation coefficient r was subjected to Fisher-Z transform to make it conform to a normal distribution. Traverse the entire target search area and find the coordinates with the largest Z value to obtain the location of the stimulation target.

[0018] The present invention also provides a computer device, including a processor and a memory for storing a computer program and magnetic resonance imaging data of a subject, wherein the computer program, when executed by the processor, implements any of the methods described above.

[0019] The technical solution of the present invention has the following advantages: A. This invention is the first to propose using the dorsomedial nucleus (MD) of the thalamus, which is closely related to sleep-wake regulation, as an anchor point. This is different from the sgACC anchor point used to treat depression, and is more in line with the neuropathological mechanism of insomnia, with a stronger mechanism targeting.

[0020] B. This invention overcomes the shortcomings of traditional target group and structural MRI navigation in ignoring individual functional network variations. It directly utilizes the patient's own functional connectivity data for localization, achieving "personalized" precision treatment with higher individual accuracy.

[0021] C. This invention introduces denoising and high-precision map segmentation techniques targeting the signal characteristics of deep thalamic nuclei, solving the technical problem that thalamic signals are easily interfered with by physiological noise, and has good technical robustness. Attached Figure Description

[0022] To more clearly illustrate the specific embodiments of the present invention, the accompanying drawings used in the specific embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is a flowchart of the overall system architecture provided by the present invention; Figure 2 This is a schematic diagram of the anatomical location of the MD nucleus seed point and the DLPFC search area provided by the present invention (based on the AAL3 atlas).

[0024] Figure 2 a is the MD kernel map in the AAL3 atlas (Automated Anatomical Labeling atlas 3); Figure 2 b is the DLPFC partitioning diagram in the Brodmann atlas; Figure 3 This is a distribution map of individual differences in the optimal MD-DLPFC connection points among different subjects. The red dots represent the traditional target group, which uses the national standard 10-20 method to locate the right DLPFC (F4) based on the national standard 10-20 scalp marker. The blue dots represent the precision target group, which uses the method provided in this invention to calculate the strongest connection point of the right MD-DLPFC.

[0025] Figure 4 The comparison of Pittsburgh Sleep Quality Index (PSQI) scores (mean ± standard deviation) between the traditional target group and the precise target group of this invention at different time points is shown. * indicates P < 0.05, and ns indicates no statistically significant difference. Figure 5 This is a logic block diagram of the functional connection calculation and target selection algorithm provided by the present invention. Detailed Implementation

[0026] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] like Figure 1 As shown, this invention provides a transcranial magnetic stimulation target localization system for sleep disorders based on the functional connectivity of the dorsomedial thalamus (MDN). The system includes a data acquisition module, a computer and an image preprocessing module stored in the computer, a seed point definition module, a search space definition module, a functional connectivity calculation module, a target point determination module, and an output module. The data acquisition module preferably uses a 3.0T MRI scanner to acquire structural MRI images and resting-state functional MRI images of the subject. The image preprocessing module preprocesses the acquired image data, performing motion correction, spatial normalization, and noise reduction on the rs-fMRI data. The seed point definition module uses AAL3 mapping to segment the dorsomedial thalamus (MDN) in the subject's brain space as seed ROIs. The search space definition module defines the dorsolateral prefrontal cortex (DLPFC) as the target search region (ROI) in the subject's cerebral cortex. The functional connectivity calculation module extracts the average time series of the seed points and calculates the correlation coefficient between the average time series and the time series of each voxel in the target search region to construct a functional connectivity map. The target determination module identifies the voxel coordinates (cortical location) with the strongest functional connectivity with the seed points (i.e., the largest absolute value of the correlation coefficient) in the constructed functional connectivity map and uses it as the stimulation target for repetitive transcranial magnetic stimulation. The output module outputs the three-dimensional spatial coordinates of the stimulation target to guide the physical therapy of sleep disorders. To this end, the system of this invention also includes a data interface that connects to a neuronavigation device to transmit the stimulation target coordinates obtained in the target determination module to the neuronavigation device and guide the placement of the magnetic stimulation coil.

[0028] Combination Figure 1 and Figure 5 As shown, this invention also provides a method for locating transcranial magnetic stimulation (TMS) targets for sleep disorders based on the functional connectivity of the dorsomedial nucleus of the thalamus. This method utilizes the aforementioned system to calculate the stimulation target points for an individual subject and outputs the target point coordinates. Specifically, it includes the following steps:

S01

[0029] This invention preferably uses a 3.0T magnetic resonance scanner to collect data from each subject, as detailed below: Structural magnetic resonance imaging: MPRAGE T1WI sequence, sagittal scan, TR=2000ms, TE=2.04ms, FOV=256mm×256mm, slice thickness=1mm, resolution=256×256, phase encoding direction is AP, flip angle=7°, voxel size=1mm×1mm×1mm.

[0030] Resting-state functional magnetic resonance imaging: EPI sequence, combined with simultaneous multi-slice (SMS) technology, transverse scanning, TR=2000ms, TE=30ms, FOV=220mm×220mm, slice thickness=2.5mm, number of slices=66, acquisition time point=240, phase encoding direction is AP, flip angle=90°, voxel size=2.5mm×2.5mm×2.5mm, subject is required to lie still with eyes open.

[0031]

S02

[0032] (2) Head movement correction (Realignment): All resting-state functional magnetic resonance images are aligned using rigid body transformation. If the head movement exceeds 3 mm or 3 degrees, the data will be marked as unacceptable.

[0033] (3) aCompCor denoising: The system automatically segments the white matter (WM) and cerebrospinal fluid (CSF) masks, extracts the first 5 principal components as covariates for regression, and obtains the regression average signal. This preserves the effective signal of the thalamus better than whole brain signal regression (GSR).

[0034] (4) Bandpass filtering: The regression average signal is bandpass filtered using a 0.01-0.08 Hz bandpass filter to retain low-frequency fluctuations.

[0035]

S03

[0036] This invention defines seed points based on the AAL3 map, enabling refined segmentation and reconstruction of thalamic nuclei, and calculating the nonlinear warp field from the MNI standard space to the subject's individual space. Figure 2 In Figure 2As shown in a, the MD nucleus seed point is a complete nucleus synthesized based on Thal_MDm and Thal_MDl in AAL3 atlas.

[0037] The dorsomedial thalamic nucleus (DNU) label from the AAL3 map was inversely mapped into the individual brain space of the subject using a nonlinear deformation field; the DNU region was identified and segmented as seed points.

[0038] In addition, the present invention etches one voxel into the mapped MD mask to further eliminate contamination from the peripheral cerebrospinal fluid signal and improve positioning accuracy.

[0039]

S04

[0040] The dorsolateral prefrontal cortex was defined as the target search area in the subject's cerebral cortex (including Brodmann's 9 / 46 area), such as Figure 2 As shown in b, the DLPFC ROI search area is obtained as the Brodmann 9 / 46 partition in the Brodmann atlas; Calculate the correlation coefficient r between the time series Tv of each voxel v within the target search region and the mean time series TMD of the dorsomedial thalamus: r=(cov(Tv,TMD)) / (σ(Tv)*σ(T MD )) The correlation coefficient r was subjected to Fisher-Z transform to make it conform to a normal distribution. Traverse the entire target search area and find the coordinate Popt(x,y,z) with the largest Z value to obtain the location of the stimulus target.

[0041]

S05

[0042] The system converts the target coordinates Popp(x,y,z) into a common neuronavigation data format (such as DICOM RTStruct or CSV) and generates a guidance file. The guidance file contains the target coordinates and the recommended coil placement angle (usually 45 degrees to the sagittal plane or perpendicular to the target brain gyrus).

[0043] Comparison Examples A retrospective analysis was conducted on data from 16 patients with insomnia.

[0044] Figure 3 This is a distribution map of individual differences in the optimal connectivity points of MD-DLPFC among different subjects, demonstrating the necessity of individualized localization. Figure 3As can be seen, the traditional target set and the precise target set of this invention have almost no overlap.

[0045]

[0046] Combining the above table 1 and Figure 4 As can be seen, there was no significant difference in the total PSQI score between the two groups at baseline. Both the post-treatment assessment and the assessment three months after treatment showed that the total PSQI score of the precision target group was significantly lower than that of the traditional target group.

[0047] Results: The target points obtained by this invention were discretely distributed among individuals, deviating from traditional target points by an average of approximately 31 mm. Patients in the precise target group showed a significantly lower PSQI after treatment compared to the traditional target group, demonstrating that the MD-guided target points are more physiologically significant.

[0048] In addition, the present invention also provides a computer device, including a processor and a memory for storing computer programs and magnetic resonance imaging data of a subject, wherein the computer programs, when executed by the processor, implement the above-described method.

[0049] Any aspects of this invention not described herein are applicable to existing technologies.

[0050] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A sleep disorder transcranial magnetic stimulation target positioning system based on thalamic dorsal medial nucleus functional connectivity, characterized in that, The system includes: The data acquisition module acquires structural magnetic resonance imaging (MRI) images and resting-state functional magnetic resonance imaging (fMRI) images of the subjects. The image preprocessing module preprocesses the acquired image data; The seed point definition module segments the dorsomedial thalamic nucleus in the subject's brain space as a seed point. The search space definition module defines the dorsolateral prefrontal cortex as the target search area in the subject's cerebral cortex. The functional connectivity calculation module extracts the average time series of seed points and calculates the correlation coefficient between the average time series and the time series of each voxel in the target search area to construct a functional connectivity map. The target determination module identifies the voxel coordinates with the strongest functional connectivity to the seed point in the constructed functional connectivity map and uses them as the stimulation target for repetitive transcranial magnetic stimulation.

2. The thalamic dorsomedial nucleus functional connectivity based sleep disorder transcranial magnetic stimulation target positioning system according to claim 1, characterized in that, The seed point definition module uses AAL3 mapping to segment the dorsolateral thalamic nucleus in the subject's brain space as a seed point.

3. The thalamic dorsomedial nucleus functional connectivity based sleep disorder transcranial magnetic stimulation target positioning system according to claim 1, characterized in that, The image preprocessing module performs motion correction, spatial normalization, and noise reduction preprocessing on resting-state functional magnetic resonance images.

4. The thalamic dorsomedial nucleus functional connectivity based sleep disorder transcranial magnetic stimulation target positioning system according to claim 1, characterized in that, The system also includes a data interface that connects to a neuronavigation device to transmit the coordinates of the stimulation target obtained in the target determination module to the neuronavigation device and guide the placement of the magnetic stimulation coil.

5. A sleep disorder transcranial magnetic stimulation target positioning method based on thalamic dorsal medial nucleus functional connectivity, characterized in that, Structural magnetic resonance imaging (SMRI) and resting-state functional magnetic resonance imaging (fMRI) of the subjects were acquired, and the obtained image data were preprocessed. The anatomical location of the dorsomedial nucleus of the thalamus was determined as a seed point. The voxel-level functional connectivity strength between the seed point and the entire dorsolateral prefrontal cortex was calculated. The coordinates of the voxel with the highest functional connectivity strength were identified and used as the stimulation target output for repetitive transcranial magnetic stimulation.

6. The thalamic dorsomedial nucleus functional connectivity-based sleep disorder transcranial magnetic stimulation target positioning method according to claim 5, characterized in that, The obtained image data were preprocessed, specifically including: removing the first 10 time points to eliminate magnetic saturation effects; aligning all resting-state functional magnetic resonance images using rigid body transformation; automatically segmenting white matter and cerebrospinal fluid masks using the aCompCor denoising method, extracting the first 5 principal components as covariates for regression, and obtaining the regression average signal; and applying a bandpass filter of 0.01-0.08Hz to the regression average signal to retain low-frequency fluctuations.

7. The method for locating transcranial magnetic stimulation targets for sleep disorders based on the functional connectivity of the dorsomedial nucleus of the thalamus, as described in claim 5, is characterized in that... Using the AAL3 map, a nonlinear deformation field was calculated from the MNI standard space to the subject's individual space. The nonlinear deformation field was used to inversely map the dorsomedial nucleus (DNU) label from the AAL3 map to the subject's individual brain space. The DNU region was identified and segmented as seed points.

8. The method for locating transcranial magnetic stimulation target points for sleep disorders based on the functional connectivity of the dorsomedial nucleus of the thalamus according to claim 5, characterized in that, The specific method for calculating the stimulus target is: The dorsolateral prefrontal cortex was defined as the target search area in the subject's cerebral cortex. Calculate the Pearson correlation coefficient r between the time series Tv of each voxel v in the target search area and the average time series TMD of the dorsomedial thalamus. The correlation coefficient r was subjected to Fisher-Z transform to make it conform to a normal distribution. Traverse the entire target search area and find the coordinates with the largest Z value to obtain the location of the stimulation target.

9. A computer device comprising a processor and a memory for storing computer programs and magnetic resonance imaging data of a subject, characterized in that, When the computer program is executed by the processor, it implements the method described in any one of claims 5-8.

Images