A localization system for personalized neural regulation targets

Abstract

The present invention provides a positioning system for individualized neural regulation targets, comprising: a data acquisition module for acquiring functional magnetic resonance (FMR) data of a target brain region of a subject; a target brain region functional network construction module for constructing a functional connection network of the target brain region according to the FMR data; a target brain region segmentation module for segmenting the target brain region into a plurality of sub-brain regions based on the functional connection network of the target brain region; a sub-brain region functional network construction module for calculating a second functional connection value between each sub-brain region and the remaining sub-brain regions, and constructing a sub-brain region functional connection network of the sub-brain region relative to the whole brain; and an abnormality positioning module for calculating an abnormality degree index of the functional connection network of each sub-brain region of the subject by using an abnormal value detection method, comparing the abnormality degree index of the sub-brain regions of each target brain region, and positioning the sub-brain region with the largest abnormality degree index as the abnormal target brain region.

Description

Technical Field

[0001] The present invention mainly relates to the field of medical instruments, and in particular to a positioning system for individualized neural regulation targets based on functional magnetic resonance imaging data. Background Art

[0002] With the continuous development of science and technology, modern neuromodulation technologies such as deep brain stimulation and transcranial magnetic stimulation, which are minimally invasive, reversible and adjustable, have gained more and more recognition in the fields of clinical psychiatric diseases, neurological diseases and rehabilitation, and have shown good prospects. However, current neuromodulation technology still faces huge challenges. On the one hand, the location of neuromodulation targets and the intensity of stimulation are mainly determined by doctors based on their clinical experience. There are even contradictions between the judgment results of different doctors. If the wrong choice is made, serious consequences will occur. On the other hand, the target selection and the setting of control parameters in the general treatment plan cannot achieve good therapeutic effects in all individual patients, and may even lead to unpredictable side effects. Therefore, it is hoped that the neuromodulation targets can be objectively selected for individual patients, and the control parameters can be set to obtain the optimal intervention effect. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a positioning system for positioning neural regulation targets according to individual conditions.

[0004] In order to solve the above technical problems, the present invention provides a positioning system for individualized neural regulation targets, characterized in that it includes: a data acquisition module, which is used to acquire functional magnetic resonance data of the target brain area of ​​the subject; a target brain area functional network construction module, which is used to construct a target brain area functional connection network based on the functional magnetic resonance data, and the target brain area functional connection network includes a first functional connection value between the target brain areas; a target brain area segmentation module, which is used to segment the target brain area into multiple sub-brain areas based on the target brain area functional connection network; a sub-brain area functional network construction module, which is used to calculate the second functional connection value between each of the sub-brain areas and the remaining sub-brain areas, and construct a sub-brain area functional connection network of the sub-brain area relative to the whole brain; and an abnormality positioning module, which is used to calculate the abnormality degree index of each sub-brain area functional connection network of the subject by using an abnormal value detection method, compare the abnormality degree index of each sub-brain area of ​​the target brain area, and locate the sub-brain area with the largest abnormality degree index as the abnormal target brain area.

[0005] In one embodiment of the present invention, the target brain area functional network construction module extracts the time series of each voxel in each of the target brain areas, calculates the Pearson correlation coefficient between the time series of any voxels in every two target brain areas, and uses Fisher-Z transformation to transform the Pearson correlation coefficient into a z value, and uses the z value as the first functional connectivity value between related voxels.

[0006] In one embodiment of the present invention, the target brain region segmentation module utilizes an unsupervised machine learning algorithm to segment the target brain region into a plurality of sub-brain regions.

[0007] In one embodiment of the present invention, the unsupervised machine learning algorithm includes a k-means clustering method.

[0008] In one embodiment of the present invention, the sub-brain region functional network construction module is also used to extract the time series of each voxel in each of the sub-brain regions, calculate the Pearson correlation coefficient between the average time series of each sub-brain region in each target brain region and the average time series of the remaining brain regions in the whole brain, and use Fisher-Z transformation to transform the Pearson correlation coefficient into a z value, and use the z value as the second functional connectivity value between the sub-brain region and the remaining brain regions.

[0009] In one embodiment of the present invention, the whole brain region includes 116 brain regions divided according to the AAL template, and the brain regions include the target brain regions.

[0010] In one embodiment of the present invention, the outlier detection method includes the Grubbs method, wherein the sub-brain region functional connection network of a normal person is taken as a normal distribution, and the outlier detection method calculates an abnormality index of the sub-brain region functional connection network of the subject relative to the normal distribution.

[0011] In one embodiment of the present invention, the abnormality localization module is further used to calculate the average value of the abnormality degree index of each of the sub-brain regions of the subject relative to the remaining brain regions in the whole brain, and use the average value as the abnormality degree index of the sub-brain region.

[0012] In one embodiment of the present invention, the abnormality localization module is also used to divide each of the target brain areas into a left side and a right side, calculate the left-right functional connection value between each voxel on the left side and each voxel on the right side, and compare the left-right functional connection values ​​of the subject with those of the normal person.

[0013] In one embodiment of the present invention, the abnormality localization module also calculates the abnormality degree indicators of the left side and the right side respectively to obtain the left-right difference of the target brain area, and compares the left-right difference of the target brain area of ​​the subject with the left-right difference of the target brain area of ​​the normal person.

[0014] In one embodiment of the present invention, the target brain regions include the left and right posterior combined brain regions and the left and right temporal-parietal combined brain regions.

[0015] In one embodiment of the present invention, it also includes an output module for aligning the sub-brain region with the largest abnormality index and the voxel with the strongest signal in the sub-brain region to the brain space of the subject, and displaying the sub-brain region with the largest abnormality index and the voxel with the strongest signal in the sub-brain region in the image of the brain space.

[0016] The positioning system of the present invention is based on the physiological significance of the functional connections between brain regions, and refines the target brain region into sub-brain regions. It can accurately locate the neural regulation target for an individual, and can assist in optimizing the parameters of neural regulation, thereby obtaining a better regulation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are included to provide a further understanding of the present application, and they are included and constitute a part of the present application. The accompanying drawings illustrate embodiments of the present application and together with the present specification serve to explain the principles of the present invention. In the accompanying drawings:

[0018] Figure 1 is a block diagram of a localization system for individualized neural regulation targets according to an embodiment of the present invention;

[0019] Figure 2 is a schematic diagram of the position of the target brain area of ​​the subject obtained by the positioning system according to an embodiment of the present invention;

[0020] Figure 3 is a relationship diagram between Silhouette parameters and the number of cluster centers obtained by using a k-means clustering method in a target brain region segmentation module in a positioning system according to an embodiment of the present invention;

[0021] Figure 4 is a schematic diagram of the result of clustering the second brain region into two sub-brain regions by the positioning system according to an embodiment of the present invention;

[0022] Figure 5 is a schematic diagram of t distribution of standardized statistical values ​​of sub-brain region functional connectivity values ​​obtained by an abnormality localization module of a localization system according to an embodiment of the present invention;

[0023] Figure 6 It is a schematic diagram of an image output and displayed by an output module of a positioning system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0025] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0026] Unless otherwise specifically stated, the relative arrangement, numerical expressions and numerical values ​​of the parts and steps set forth in these embodiments do not limit the scope of the present application. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to those of ordinary skill in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be considered as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so that once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0027] In addition, it should be noted that the use of words such as "first" and "second" to define components is only for the convenience of distinguishing the corresponding components. If not otherwise stated, the above words have no special meaning and cannot be understood as limiting the scope of protection of this application. In addition, although the terms used in this application are selected from well-known and commonly used terms, some terms mentioned in the specification of this application may be selected by the applicant at his or her discretion, and their detailed meanings are explained in the relevant parts of the description of this article. In addition, it is required to understand this application not only by the actual terms used, but also by the meaning implied by each term.

[0028] Minimally conscious state is a state of severe altered consciousness with tiny but very clear behavioral evidence to prove the ability to perceive self and environment. It is a special state of consciousness between coma and conscious awareness. Minimally conscious state is a necessary stage for all patients with chronic disorders of consciousness to awaken. At present, there is a lack of effective treatment for patients with minimally conscious state. Neuromodulation can be used as a wake-up method by targeting the relevant brain regions. Studies have shown that neurological and psychiatric diseases may be caused by dysfunctional connections between brain regions. Neuromodulation can reverse abnormal connections of brain networks by adjusting brain network connections, thereby achieving the effect of assisting in the treatment of diseases.

[0029] In this specification, the term "functional magnetic resonance data" refers to the image data obtained by scanning using magnetic resonance imaging technology. The term "neuromodulation" is a high-end application of neurointerventional technology in the field of neuroscience. It is a biomedical engineering technology that uses implantable or non-implantable technology, physical means (such as electrical stimulation, magnetic stimulation) or drug means (implantation of micropumps) to change the activity of the central nervous system, peripheral nervous system or autonomic nervous system to improve the symptoms of the sick population and improve the quality of life. Compared with traditional brain damage and resection surgery, it emphasizes regulation, that is, the process is reversible and the treatment parameters can be adjusted in vitro. The term "neuropsychiatric disease" refers to cognitive, sensory, and motor disorders caused by abnormal neural circuits, such as autism in children, middle-aged affective disorders (depression, obsessive-compulsive disorder, addiction, anorexia) and elderly neurodegenerative diseases (Parkinson's disease, Alzheimer's disease), etc. The neural circuit of a normal body is an inherently balanced system composed of electrical stimulation and chemical signals (i.e., a normal brain network), but diseases (including congenital and acquired factors) break this balance, resulting in sensory, motor or cognitive impairment (i.e., brain network abnormalities). Although the causes of these diseases are complex, they have one thing in common, that is, they are usually accompanied by functional disorders of the brain's neural circuits. The treatment of such diseases requires overall regulation of the neural circuits to restore the normal functioning of the brain network, thereby achieving the effect of curing the disease. Therefore, based on the scientific hypothesis that "neuromodulatory methods can achieve the effect of assisting the treatment of diseases by adjusting the neural circuits in the brain network, radiating and reversing the network properties of the entire brain network." combined with a large amount of clinical evidence at home and abroad, the above-mentioned brain diseases can be effectively treated by intervening with physical (electrical, magnetic, etc.) means at appropriate targets.

[0030] Figure 1 FIG. 1 is a block diagram of a positioning system for individualized neural regulation targets according to an embodiment of the present invention. Figure 1 As shown, the positioning system 100 of this embodiment includes a data acquisition module 110, a target brain region functional network construction module 120, a target brain region segmentation module 130, a sub-brain region functional network construction module 140 and an abnormality positioning module 150.

[0031] The data acquisition module 110 is used to acquire functional magnetic resonance data of the target brain region of the subject. The present invention does not impose any specific restrictions on the subject, and the subject may include abnormal persons with abnormal brain function and normal persons with normal brain function.

[0032] In some embodiments, the abnormal subject is a patient with a psychiatric neurological disease.

[0033] In some embodiments, the psychiatric neurological disorder comprises a minimally conscious state.

[0034] In some embodiments, a normal person is a healthy person who is clinically diagnosed as not suffering from a neuropsychiatric disease.

[0035] In some embodiments, the fMRI data includes resting-state fMRI data.

[0036] The present invention does not limit the manner in which the data acquisition module 110 acquires functional magnetic resonance data. The data acquisition module 110 may be part or all of a magnetic resonance device, and is directly used to measure the functional magnetic resonance data of the subject. The data acquisition module 110 may also be an independent device, which acquires the functional magnetic resonance data of the subject from the magnetic resonance device. The positioning system 100 may establish a communication connection with the magnetic resonance device in a wired or wireless manner, and the magnetic resonance device may send the collected functional magnetic resonance data to the data acquisition module 110 in the positioning system 100.

[0037] The present invention does not limit the format of functional magnetic resonance data. In some embodiments, the format includes DICOM format and NIFTI format.

[0038] In some embodiments, the target brain region of the subject includes any brain region in the AAL template. According to the AAL (Anatomical Automatic Labeling) template, the human cerebral cortex is divided into 116 brain regions, as shown in Table 1 below. The present invention does not limit the specific location and number of brain regions corresponding to the functional magnetic resonance data to be acquired.

[0039] Table 1: Standard brain partition information based on the AAL template

[0040]

[0041]

[0042]

[0043]

[0044]

[0045] Figure 2 It is a schematic diagram of the position of the target brain area of ​​the subject obtained by the positioning system according to an embodiment of the present invention. Figure 2 A schematic diagram of a brain magnetic resonance image of a subject is shown, wherein four brain regions are encircled by ellipses, namely, the first brain region 210, the second brain region 220, the third brain region 230, and the fourth brain region 240. Table 2 lists the names of the brain regions corresponding to the four brain regions and the number of voxels contained in the magnetic resonance image. In an embodiment of the present invention, a voxel is the smallest unit of medical image segmentation in three-dimensional space that can be provided by a magnetic resonance scanning device.

[0046] Table 2: Figure 2 The target brain region information shown in

[0047]

[0048] refer to Figure 1 As shown, the target brain region functional network construction module 120 is used to construct a target brain region functional connection network based on the functional magnetic resonance data obtained by the data acquisition module 110, and the target brain region functional connection network includes the first functional connection values ​​between the target brain regions. It can be understood that a target brain region functional connection network is constructed for each subject.

[0049] In some embodiments, the target brain region functional network construction module 120 extracts the time series of each voxel in each target brain region, calculates the Pearson correlation coefficient between the time series of any voxels in every two target brain regions, and transforms the Pearson correlation coefficient into a z value using Fisher-Z transformation, and uses the z value as the first functional connection value between the related voxels. In these embodiments, the constructed target brain region functional connection network is composed of the first functional connection values ​​between the voxels of different target brain regions.

[0050] Taking the four brain regions shown in Table 2 as an example, the functional connection network of the target brain region includes the first functional connection value between each voxel in the first brain region 210 and 408 voxels in the remaining three brain regions, the first functional connection value between each voxel in the second brain region 220 and 425 voxels in the remaining three brain regions, and so on.

[0051] According to these embodiments, the target brain region functional network construction module 120 can use the following steps to construct the target brain region functional connection network.

[0052] Step S11: transforming the functional magnetic resonance data into a standard space for standard calculation.

[0053] Specifically, the standard space in step S11 can be the MNI space. The MNI space is a coordinate system established by the Montreal Neurological Institute based on a series of magnetic resonance images of normal human brains. The magnetic resonance images of different subjects are not comparable in the original space. By aligning the images of all subjects to a standardized template and unifying the dimensions, origin, voxels and other parameters, the functional magnetic resonance data of different subjects can be made comparable for multi-sample analysis. Using the MNI standard template, the functional magnetic resonance data of different subjects can be converted to the MNI space for analysis and comparison. Figure 2 Shown is the magnetic resonance image of the subject in the MNI space.

[0054] Step S12: extracting the time signal of each voxel in each target brain region and the average time signal of all voxels in each target brain region.

[0055] In step S12, taking Table 2 as an example, the time signal of each voxel in the four target brain regions in Table 2 and the average time signal of all voxels in the four target brain regions are extracted. Taking the first brain region 210 as an example, in step S12, the time signal of each voxel in the first brain region 210 and the average time signal of all 57 voxels are extracted. In these embodiments, the time signal of the voxel is a time series.

[0056] Step S13: For each target brain region, the Pearson correlation coefficient r between the time signal of each voxel in the target brain region and the time signal of each voxel in the remaining target brain regions is calculated, and the r value is transformed into a z value using Fisher-Z transformation, as shown in the following formula (1):

[0057]

[0058] Step S14: The z value is used as the first functional connectivity value between the related voxels involved, so as to construct the functional connectivity matrix between each target brain region and the voxels of the remaining target brain regions. For the four brain regions shown in Table 2, the size of the functional connectivity matrix between the first brain region 210 and the voxels of the remaining target brain regions is 57×408, the size of the functional connectivity matrix between the second brain region 220 and the voxels of the remaining target brain regions is 40×425, the size of the functional connectivity matrix between the third brain region 230 and the voxels of the remaining target brain regions is 131×334, and the size of the functional connectivity matrix between the fourth brain region 240 and the voxels of the remaining target brain regions is 237×228. The size of the functional connectivity matrix between each brain region and the voxels of the remaining target brain regions is m×n, where m is equal to the number of voxels in the target brain region, and n is equal to the total number of voxels in the remaining target brain regions except the target brain region.

[0059] It is understood that what is shown in Table 2 is only an example. In other embodiments, any other target brain regions may be selected.

[0060] refer to Figure 1 As shown, the target brain region segmentation module 130 is used to segment the target brain region into multiple sub-brain regions based on the target brain region functional connection network.

[0061] In some embodiments, the target brain region segmentation module 130 uses an unsupervised machine learning algorithm to segment the target brain region into multiple sub-brain regions.

[0062] In some embodiments, the unsupervised machine learning algorithm includes a k-means clustering method. In these embodiments, the following steps are used to segment the target brain region.

[0063] Step S21: Use the k-means clustering method to cluster the functional connectivity values ​​in the target brain region functional connectivity network.

[0064] Continuing with the example shown in Table 2, in step S21, all 57×408, 40×425, 131×334, and 237×228 first functional connectivity values ​​of the voxels belonging to the four target brain regions are clustered using k-means, and the number of initial cluster centers is traversed from 2 to 8.

[0065] Step S22: traverse the number of initial cluster centers k from 2 to 8, and calculate the Silhouette parameter according to the following formulas (2) and (3):

[0066]

[0067] b(j)=min{e(j,Cs)},s=1,2,…k (3)

[0068] Among them, Cs (s = 1, 2, ... k) is the k clusters obtained after clustering, a (j) represents the average Euclidean distance between sample j in cluster Cs and other members in cluster Cs, e (j, Cs) represents the average Euclidean distance between sample j in cluster Cs and members in other clusters, and b (j) represents the minimum distance between sample j and other cluster members.

[0069] The larger the Silhouette parameter calculated in step S22 is, the greater the intra-class compactness and inter-class separability is, and the better the clustering quality is.

[0070] Figure 3 : This is a relationship diagram between the Silhouette parameters and the number of cluster centers obtained by the target brain region segmentation module in the positioning system of an embodiment of the present invention using the k-means clustering method. The horizontal axis is the number of cluster centers, ranging from 2 to 8; the vertical axis is the Silhouette parameters. Figure 3 As shown, when the number of cluster centers is 2, the corresponding Silhouette parameter is the largest, indicating that in this embodiment, the clustering quality is best when the number of cluster centers is equal to 2.

[0071] Figure 4 FIG. 4 is a schematic diagram showing the result of clustering the second brain region into two sub-brain regions according to a positioning system according to an embodiment of the present invention. Figure 4 As shown, the positioning system 100 of the present invention is used to Figure 2 The second brain region 220 shown in FIG. 1 , namely the PPC.C target brain region, is clustered into two sub-brain regions 410 and 420 . Figure 4The following are three views of the brain image of the same subject, of which the upper left view is the coronal view, the upper right view is the sagittal view, and the lower view is the transverse view. Figure 4 The pattern filled with blanks in the figure represents the sub-brain region 410, and the pattern filled with oblique lines represents the sub-brain region 420. The sub-brain regions 410 and 420 shown in different views represent the same sub-brain region 410 and 420, respectively, and therefore the same reference numerals are used.

[0072] Figure 4 The illustration is only an example and is not intended to limit the number of cluster centers in the clustering method of the present invention.

[0073] After k-means clustering of the four target brain regions in the example shown in Table 2, the first brain region 210 is divided into 2 sub-brain regions, the second brain region 220 is divided into 3 sub-brain regions, the third brain region 230 is divided into 2 sub-brain regions, and the fourth brain region 240 is divided into 2 sub-brain regions.

[0074] refer to Figure 1 As shown, the sub-brain region functional network construction module 140 is used to calculate the second functional connectivity value between each sub-brain region and the remaining sub-brain regions, and to construct a sub-brain region functional connectivity network of the sub-brain region relative to the whole brain.

[0075] In some embodiments, the sub-brain region functional network construction module 140 is also used to extract the time series of each voxel in each sub-brain region, calculate the Pearson correlation coefficient between the sub-brain region average time series of each sub-brain region in each target brain region and the average time series of the remaining brain regions in the whole brain, and use Fisher-Z transformation to transform the Pearson correlation coefficient into a z value, and use the z value as the second functional connectivity value between the sub-brain region and the remaining brain regions.

[0076] In some embodiments, the whole brain region includes 116 brain regions divided according to the AAL template, and the 116 brain regions include the target brain region. In other words, the target brain region mentioned above is one or more of the 116 brain regions.

[0077] In this embodiment, assuming that the target brain region is brain region A among the 116 brain regions, which includes sub-brain regions A1 and A2, the sub-brain region functional network construction module 140 needs to extract the time series of each voxel in each sub-brain region A1, A2 in brain region A, and calculate the sub-brain region average time series At1, At2 of each sub-brain region A1, A2; it also needs to extract the time series of each voxel in the remaining 115 brain regions in the whole brain, which is used to calculate the average time series Ar of the 115 brain regions; then for sub-brain region A1, calculate the Pearson correlation coefficient between the sub-brain region average time series At1 and the average time series Ar, and for sub-brain region A2, calculate the Pearson correlation coefficient between the sub-brain region average time series At2 and the average time series Ar; and then transform all Pearson correlation coefficients into z values ​​through Fisher-Z transformation.

[0078] Continuing with the example shown in Table 2, for the four target brain regions, the sub-brain region functional network construction module 140 constructs a sub-brain region functional connection network respectively. According to the number of sub-brain regions of the four target brain regions, the sizes of the sub-brain region functional connection networks constructed by the sub-brain region functional network construction module 140 are 2×115, 3×115, 2×115 and 2×115 respectively, where 115 is the number of remaining brain regions after removing the target brain region corresponding to the sub-brain region from the 116 brain regions. It can be understood that the sub-brain region functional connection network also has a matrix form. The size of each sub-brain region functional connection network is the number of z values ​​contained therein.

[0079] In some embodiments, the positioning system 100 of the present invention also calculates brain features of sub-brain regions, which include low-frequency amplitude (Alff), local consistency (Reho), graph theory indicators, etc. Among them, the low-frequency amplitude is the amplitude of the low-frequency oscillation signal of the magnetic resonance data. By observing the low-frequency amplitude, the situation of the local low-frequency amplitude in the resting state can be understood, so as to reflect the changes in the degree of spontaneous activity of neurons. Local consistency is an indicator for measuring the functional synchronization strength of local brain regions. This method is based on data-driven, assuming that in an active or activated brain region, the signal values ​​of adjacent voxels have certain similarities or consistency in the time domain, which is quantitatively measured by calculating the Kendall harmony coefficient. Graph theory is a theory for studying point sets connected by lines. Characterizing the topological relationship of complex networks through graph theory methods is an important means to study different nodes, different edges and the overall characteristics of the network. Graph theory indicators are network characteristics calculated after the research object is modeled according to graph theory, including but not limited to node degree, node strength, node centrality, characteristic path length, and small world attributes.

[0080] In these embodiments, the positioning system 100 of the present invention can locate abnormal sub-brain regions by comparing the brain features of the sub-brain regions of the subject with the brain features of the corresponding sub-brain regions of normal persons.

[0081] refer to Figure 1 As shown, the abnormality localization module 150 is used to calculate the abnormality index of each sub-brain region functional connection network of the subject by adopting the abnormal value detection method, compare the abnormality index of the sub-brain region of each target brain region, and locate the sub-brain region with the largest abnormality index as the abnormal target brain region.

[0082] The outlier detection method refers to a method of analyzing outlier data from a large amount of data by mathematical statistics according to data characteristics. The present invention does not limit the outlier detection method used by the outlier location module 150, and any outlier detection method in the art can be used.

[0083] In some embodiments, the outlier detection method includes the Grubbs method, wherein the sub-brain region functional connection network of a normal person is taken as a normal distribution, and the outlier detection method calculates an abnormality index of the sub-brain region functional connection network of the subject relative to the normal distribution.

[0084] It can be understood that in order to locate the abnormal brain area of ​​an abnormal person, the positioning system 100 of the present invention has obtained the functional connection network of the sub-brain areas of a normal person.

[0085] In a related study of the present invention, functional magnetic resonance data of 47 healthy subjects were collected, and the data acquisition module 110, the target brain region functional network construction module 120, the target brain region segmentation module 130 and the sub-brain region functional network construction module 140 in the positioning system 100 of the present invention were used to process these functional magnetic resonance data, and the sub-brain region functional connection networks of these healthy subjects were obtained as the sub-brain region functional connection networks of normal people. The sub-brain region functional networks of multiple normal people have normal distribution.

[0086] In these embodiments, the outlier detection method includes the following steps:

[0087] Step S41: Use the following formulas (4) and (5) to calculate the mean value μ and standard deviation σ of the z values ​​in all sub-brain regions respectively:

[0088]

[0089]

[0090] Among them, i is the number of the second functional connectivity value of the sub-brain region, n is the number of the second functional connectivity values ​​of all sub-brain regions, and x i represents the i-th second functional connection value.

[0091] Step S42: Calculate the second functional connectivity value x of the sub-brain region using the following formula (6): i The standardized statistical value G i :

[0092]

[0093] Step S43: If G satisfies the following formula (7), the second functional connectivity value x of the sub-brain region is detected as an abnormal value:

[0094]

[0095] Here, t represents the critical value of the t distribution, which is a normal distribution.

[0096] Figure 5 1 is a schematic diagram of the t distribution of the standardized statistical values ​​of the sub-brain region functional connectivity values ​​obtained by the abnormal positioning module of the positioning system according to an embodiment of the present invention. The horizontal axis represents the t value, and the vertical axis represents the probability density function value. In this embodiment, the standardized statistical value G i The standard two-tailed t distribution is assumed, and the confidence level α is 0.01. Figure 5 As shown, three exemplary confidence intervals of 68.3%, 95.5%, and 99.7% are shown, corresponding to the cases where the functional connectivity values ​​of the sub-brain regions are 1, 2, and 3 standard deviations away from the mean, respectively.

[0097] In this embodiment, the t distribution can also give a p-value of the significance of the G value, and the p-value is used as a statistical value of the abnormal degree of sub-brain region connection. The smaller the p-value, the greater the abnormal degree. Figure 5 As shown, for normal people, the p value is in the middle of the t distribution curve, and for abnormal people, the p value is on both sides of the t distribution curve.

[0098] Therefore, the following formula (8) is used to define the abnormality index D of the sub-brain region connection:

[0099] D=1-p (8)

[0100] It can be understood that the larger the D is, the greater the abnormality of the functional connectivity value of the corresponding sub-brain region is.

[0101] Continuing with the example shown in Table 2, for each subject, a sub-brain region functional connection network of the four target brain regions can be obtained, where each sub-brain region has an abnormality index D relative to the rest of the brain regions. Then, the sizes of the abnormality matrices of the sub-brain region functional connection networks of the four target brain regions are: 2×115, 3×115, 2×115, and 2×115, respectively.

[0102] In some embodiments, the abnormality localization module 150 is also used to calculate the average value of the abnormality index of each sub-brain region of the subject relative to the remaining brain regions in the whole brain region, and use the average value as the abnormality index of the sub-brain region. Continuing with the example shown in Table 2, for each subject, the average value of the abnormality index of the sub-brain region in each target brain region can be calculated based on the abnormality matrix of the four target brain regions, and the sizes of the average value matrices are: 2×1, 3×1, 2×1, and 2×1. In other words, each sub-brain region has an average value of an abnormality index, that is, it has an abnormality index.

[0103] According to the above embodiment, the abnormality localization module 150 can obtain the sub-brain region with the largest abnormality index, and locate the sub-brain region as a new abnormal target brain region. It can be understood that the target brain region including the sub-brain region also belongs to the abnormal target brain region. According to the present invention, a more detailed result is obtained, and the sub-brain region is directly related to the individual condition of the subject as a new abnormal target brain region, and can be used as the object of subsequent neuromodulation intervention.

[0104] According to the positioning system of the present invention, the target brain area is subdivided based on the physiological significance of the functional connection network, which can determine more specific locations that need stimulation, and help to enhance the accuracy and effectiveness of neural regulation.

[0105] In some embodiments, the abnormality localization module 150 is further used to divide each target brain region into a left side and a right side, calculate the left-right functional connectivity value between each voxel on the left side and each voxel on the right side, and compare the left-right functional connectivity value of the subject with the left-right functional connectivity value of a normal person. In these embodiments, the time series of each voxel on the left side and the time series of each voxel on the right side can be extracted respectively, the Pearson correlation coefficient between the time series of each voxel on the left side and the time series of each voxel on the right side can be calculated, and the Pearson correlation coefficient can be transformed into a z value using Fisher-Z transformation, and the z value is used as the left-right functional connectivity value between the left side and the right side.

[0106] At the same time, the left and right functional connection values ​​of normal people are obtained, and the left and right functional connection values ​​of the subject are compared with those of normal people, and the comparison results are used as a reference for auxiliary neural regulation. For example, the difference between the left and right functional connection values ​​of the subject and the left and right functional connection values ​​of normal people is calculated, and the magnitude of the stimulation intensity, the length of the stimulation time, etc. are adjusted according to the difference between the two.

[0107] In some embodiments, the abnormality localization module 150 is also used to calculate the abnormality index of the left and right sides to obtain the difference between the left and right sides of the target brain area, and compare the difference between the left and right sides of the target brain area of ​​the subject and the difference between the left and right sides of the target brain area of ​​a normal person. In these embodiments, the difference results obtained by comparison can be used as a reference for auxiliary neural regulation. For example, the magnitude of the stimulation intensity, the length of the stimulation time, etc. are regulated according to the magnitude of the difference results.

[0108] According to the above-mentioned abnormal positioning module 150, not only can the abnormal brain area to be stimulated be positioned in a personalized manner, but also the parameters of neural regulation can be assisted in adjusting to optimize the neural regulation parameters and obtain better regulation effects.

[0109] In some embodiments, the target brain regions include the left and right posterior combined brain regions and the left and right temporal-parietal combined brain regions. The left and right posterior combined brain regions include the first brain region 210 and the second brain region 220 described above, and the left and right temporal-parietal combined brain regions include the third brain region 230 and the fourth brain region 240 described above.

[0110] refer to Figure 1 As shown, in some embodiments, the positioning system 100 of the present invention also includes an output module 160, which is used to align the sub-brain region with the largest abnormality index and the voxel with the strongest signal in the sub-brain region to the brain space of the subject, and display the sub-brain region with the largest abnormality index and the voxel with the strongest signal in the sub-brain region in the image of the brain space.

[0111] The present invention does not limit the specific display method, and the output module 160 can be various hardware devices with display functions, such as but not limited to a display screen.

[0112] In some embodiments, the abnormality localization module 150 is further used to sort the connections between the target brain regions according to the degree of abnormality, and the output module 160 outputs the sorting result. According to these embodiments, the doctor can clearly see the degree of abnormality of the connection between the target brain regions, which helps to assist in formulating a personalized neuromodulation plan.

[0113] In some embodiments, the output module 160 is further used to output the overall signal average of the target brain region. The overall signal average includes one or more of the overall magnetic resonance signal average, the average of all connection values ​​of the target brain region, the average of abnormality index, etc.

[0114] Figure 6 FIG. 1 is a schematic diagram of an image output and displayed by an output module of a positioning system according to an embodiment of the present invention. Figure 6As shown, the image is a three-view image with different viewing angles in a standard space, where the upper left is the transverse plane, the upper right is the coronal plane, and the lower part is the sagittal plane. In each view, the white block 610 represents the sub-brain region with the largest abnormality index, and the dot 620 represents the voxel with the strongest signal in the sub-brain region. The magnetic resonance BOLD signal uses the dependence of measuring blood oxygen levels to represent the strength of brain activity. The voxel with the strongest signal in the sub-brain region represents a voxel point with the strongest brain activity in a small area with the greatest difference in brain function from a normal person. According to Figure 6 The method shown directly displays and marks the positioning results of the positioning system on the magnetic resonance image, which helps to reduce the errors caused by manual annotation by doctors. In some embodiments, the intensity of the nerve regulation stimulation can also be displayed or prompted at the same time, without the need for doctors to carry redundant data.

[0115] The positioning system of the present invention refines the target brain region into sub-brain regions, and the functional connection patterns within the sub-brain regions are consistent. In medical theory, it is assumed that stimulating these sub-brain regions can obtain similar therapeutic effects. Based on the physiological significance of functional connections, the positioning system of the present invention can accurately locate the neural regulation target for an individual, and can assist in optimizing the parameters of neural regulation, thereby obtaining a better regulation effect.

[0116] The positioning system of the present invention also has the following advantages:

[0117] (1) By segmenting the target brain regions of neuromodulation, it is used to analyze the differences in brain function between individual patients and normal people, and deepen researchers' understanding of the pathological mechanisms and neuromodulatory mechanisms of neuropsychiatric diseases.

[0118] (2) It realizes individualized neural regulation target positioning, which conforms to the trend of individualized precision medicine compared with the previous group average analysis method, thus better benefiting human society.

[0119] (3) Using data-driven algorithms, the location of the target brain region with the largest difference between individual patients and normal subjects and the index of the degree of abnormality of the difference are located. This process avoids the subjectivity of selecting the target location and stimulation intensity based solely on the doctor's judgment.

[0120] (4) The localization system of the present invention can analyze all brain regions in the whole brain, and can also be combined with pre-selected target brain regions, which has flexibility in scheme.

[0121] (5) When locating individual patients, the present invention chooses to introduce a large amount of more easily available normal person data to guide the model to locate individual patients with complex individual differences, making the system more reliable.

[0122] (6) The present invention can not only provide neural regulation target location, but also provide abnormality degree indicators to assist in adjusting the neural regulation intensity parameter selection.

[0123] (7) The present invention uses a machine learning algorithm to segment sub-brain regions based on the functional connection network of the target brain region. Compared with the rough selection of brain regions based on previous doctors' experience, it can provide more precise and accurate target positioning.

[0124] The basic concepts have been described above. Obviously, for those skilled in the art, the above invention disclosure is only used as an example and does not constitute a limitation of the present application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and amendments to the present application. Such modifications, improvements and amendments are suggested in the present application, so such modifications, improvements and amendments still belong to the spirit and scope of the exemplary embodiments of the present application.

[0125] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0126] Some aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". The processor may be one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DAPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, or combinations thereof. In addition, various aspects of the present application may be expressed as computer products located in one or more computer-readable media, which include computer-readable program codes. For example, computer-readable media may include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, tapes ...), optical disks (e.g., compact disks CDs, digital versatile disks DVDs ...), smart cards, and flash memory devices (e.g., cards, sticks, key drives ...).

[0127] A computer-readable medium may include a propagated data signal containing computer program code, such as in baseband or as part of a carrier wave. The propagated signal may have a variety of manifestations, including electromagnetic, optical, etc., or a suitable combination. A computer-readable medium may be any computer-readable medium other than a computer-readable storage medium, which may be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit a program for use. The program code on the computer-readable medium may be propagated via any suitable medium, including radio, cable, fiber optic cable, radio frequency signal, or similar medium, or any combination of the above mediums.

[0128] Similarly, it should be noted that in order to simplify the description of the disclosure of this application and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this application, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the object of this application are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0129] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0130] Although the present application has been described with reference to the current specific embodiments, ordinary technicians in this technical field should recognize that the above embodiments are only used to illustrate the present application, and various equivalent changes or substitutions may be made without departing from the spirit of the present application. Therefore, as long as the changes and modifications to the above embodiments are within the essential spirit of the present application, they will fall within the scope of the claims of the present application.

Claims

1. A localization system for individualized neural regulation targets, characterized in that: include: A data acquisition module, used to acquire functional magnetic resonance imaging data of a target brain region of a subject; A target brain region functional network construction module, used to construct a target brain region functional connection network according to the functional magnetic resonance imaging data, wherein the target brain region functional connection network includes first functional connection values ​​between the target brain regions; A target brain region segmentation module, used for segmenting the target brain region into multiple sub-brain regions based on the functional connection network of the target brain region; A sub-brain region functional network construction module, used for calculating the second functional connectivity value between each of the sub-brain regions and the remaining sub-brain regions, and constructing a sub-brain region functional connectivity network of the sub-brain region relative to the whole brain; as well as an abnormality localization module, used to calculate the abnormality index of each sub-brain region functional connection network of the subject by using an abnormal value detection method, compare the abnormality index of each sub-brain region of the target brain region, and locate the sub-brain region with the largest abnormality index as the abnormal target brain region; The target brain region functional network construction module extracts the time series of each voxel in each target brain region, calculates the Pearson correlation coefficient between the time series of any voxels in every two target brain regions, and transforms the Pearson correlation coefficient into a z value using Fisher-Z transformation, and uses the z value as the first functional connectivity value between related voxels; The sub-brain region functional network construction module is also used to extract the time series of each voxel in each sub-brain region, calculate the Pearson correlation coefficient between the average time series of each sub-brain region in each target brain region and the average time series of the remaining brain regions in the whole brain region, and transform the Pearson correlation coefficient into a z value using Fisher-Z transformation, and use the z value as the second functional connectivity value between the sub-brain region and the remaining brain regions; The outlier detection method includes the Grubbs method, wherein the sub-brain region functional connection network of a normal person is taken as a normal distribution, and the outlier detection method calculates an abnormality index of the sub-brain region functional connection network of the subject relative to the normal distribution.

2. The positioning system according to claim 1, characterized in that The target brain region segmentation module uses an unsupervised machine learning algorithm to segment the target brain region into multiple sub-brain regions.

3. The positioning system according to claim 2, characterized in that The unsupervised machine learning algorithm includes a k-means clustering method.

4. The positioning system according to claim 1, characterized in that: The whole brain regions include 116 brain regions divided according to the AAL template, and the brain regions include the target brain regions.

5. The positioning system according to claim 1, characterized in that: The abnormality localization module is also used to calculate the average value of the abnormality degree index of each of the sub-brain regions of the subject relative to the remaining brain regions in the whole brain, and use the average value as the abnormality degree index of the sub-brain region.

6. The positioning system according to claim 1, characterized in that The abnormality localization module is also used to divide each target brain area into a left side and a right side, calculate the left-right functional connection value between each voxel on the left side and each voxel on the right side, and compare the left-right functional connection values ​​of the subject with those of the normal person.

7. The positioning system according to claim 6, characterized in that The abnormality localization module also calculates the abnormality degree indexes of the left side and the right side to obtain the left-right difference of the target brain area, and compares the left-right difference of the target brain area of ​​the subject with the left-right difference of the target brain area of ​​the normal person.

8. The positioning system according to claim 1, characterized in that: The target brain regions include the left and right posterior combined brain regions and the left and right temporal-parietal combined brain regions.

9. The positioning system according to claim 1, characterized in that: It also includes an output module for registering the sub-brain region with the largest abnormality index and the voxel with the strongest signal in the sub-brain region to the brain space of the subject, and displaying the sub-brain region with the largest abnormality index and the voxel with the strongest signal in the sub-brain region in the image of the brain space.

Images