Functional Connectivity and Gradient Analysis System and Method of Preclinical Functional Magnetic Resonance Imaging

Abstract

The present invention relates to a system and method for analyzing functional connectivity and gradients of preclinical functional magnetic resonance imaging. According to the present invention, the system for analyzing functional connectivity and gradients of preclinical functional magnetic resonance imaging comprises: a neuroimaging preprocessing unit that performs preprocessing on an input neuroimaging image; a neuroimaging signal extraction unit that extracts neuroimaging signals from the preprocessed neuroimaging image; a normalization unit that performs normalization for noise removal of the extracted neuroimaging signals; a functional network generation unit that receives the normalized signal and generates a functional network; and a functional connectivity and gradient analysis unit that analyzes the network of functional connectivity and the gradients of functional connectivity. As such, according to the present invention, the generation of local networks by function is facilitated by configuring the entire brain network. Furthermore, by providing advanced neuroimaging analysis techniques such as brain functional connectivity analysis, network attribute analysis, and gradient analysis, it is possible to utilize them effectively in brain imaging-based preclinical brain science research. Moreover, research can be conducted efficiently by providing an efficient and convenient analysis method through a general-purpose system capable of automatically performing the entire process from the input of brain imaging data to the analysis results.

Description

Technology Field

[0001] The present invention relates to a system and method for analyzing functional connectivity and gradients in preclinical functional magnetic resonance imaging. More specifically, it relates to a system and method for analyzing functional connectivity and gradients in preclinical functional magnetic resonance imaging that enables efficient and convenient research by analyzing functional connectivity and gradients through the construction of global brain networks and local networks utilizing fMRI image data. Background Technology

[0002] Functional Magnetic Resonance Imaging (fMRI) refers to a method of capturing brain activity signals by primarily using an MRI device to detect changes in blood oxygen levels (Blood-Oxygen-Level Dependent, BOLD) during brain activity, and is utilized for translational research ranging from rodents to humans.

[0003] In particular, conventional techniques for analyzing brain functional imaging in small animals involve preprocessing brain neuroimaging data to observe brain neural activation in localized regions or to investigate functional connectivity between regions in brain networks composed of brain regions of interest. However, there are limitations in that localized regions to observe brain neural activation must be re-selected or brain networks must be reconstructed depending on the disease animal model.

[0004] For example, in the case of animal models of Parkinson's disease, when the region of interest is selected as the motor region, there is a difficulty in having to reset the region of interest and the brain network if the activated region changes during the study. Prior art literature

[0005] Wang, Q., Ding, SL, Li, Y., Royall, J., Feng, D., Lesnar, P., Graddis, N., Naeemi, M., Facer, B., Ho, A., Dolbeare, T., Blanchard, B., Dee, N., Wakeman, W., Hirokawa, K.E., Szafer, A., Sunkin, SM, Oh, SW, Bernard, A., Phillips, J. W., Hawrylycz, M., Koch, C., Zeng, H., Harris, J. A. and Ng, L. (2020). The Allen Mouse Brain CommonCoordinate Framework: A 3D Reference Atlas. Cell 181, 936-953 e920.Rubinov, M. and Sporns, O. (2010).Complex network measures of brain connectivity: uses and interpretations.Neuroimage 52, 1059-1069. The problem to be solved

[0006] As such, according to the present invention, the purpose is to provide a system and method for analyzing functional connectivity and gradients of preclinical functional magnetic resonance imaging, which enables efficient and convenient research by utilizing fMRI image data to construct global and local networks and analyzing functional connectivity and gradients of functional connectivity. means of solving the problem

[0007] According to an embodiment of the present invention for achieving such technical challenges, a system for analyzing functional connectivity and gradients of preclinical functional magnetic resonance imaging comprises: a brain image preprocessing unit that performs preprocessing on an input brain image; a brain signal extraction unit that extracts brain signals from the preprocessed brain image; a normalization unit that performs normalization to remove noise from the extracted brain signals; a functional network generation unit that receives the normalized signal and generates a functional network; and a functional connectivity and gradient analysis unit that analyzes the network of functional connectivity and the gradient of functional connectivity, and may further include an analysis result visualization unit that visualizes the analysis results by the functional connectivity and gradient analysis unit.

[0008] The above preprocessing unit can perform at least one of MRI image orientation correction, voxel scaling, slice timing correction, motion correction, distortion correction, and co-registration / normalization.

[0009] The visualization unit of the above analysis results can visualize functional connectivity or the gradient of functional connectivity as at least one of a graph, mapping, or rendering.

[0010] According to an embodiment of the present invention, a method for analyzing functional connectivity and gradients of preclinical functional magnetic resonance imaging comprises: a brain image preprocessing step for performing preprocessing on an input brain image; a brain signal extraction step for extracting a brain signal from the preprocessed brain image; a normalization step for performing normalization to remove noise from the extracted brain signal; a functional network generation step for receiving the normalized signal and generating a functional network; and a functional connectivity and gradient analysis step for analyzing the network of functional connectivity and the gradient of functional connectivity, and may further include an analysis result visualization step for visualizing the analysis results by the functional connectivity and gradient analysis unit. Effects of the invention

[0011] As such, according to the present invention, the generation of local networks by function is facilitated by configuring the entire brain network. Furthermore, by providing advanced neuroimaging analysis techniques such as brain functional connectivity analysis, network attribute analysis, and gradient analysis, it is possible to utilize them effectively in brain imaging-based preclinical brain science research. Moreover, research can be conducted efficiently by providing an efficient and convenient analysis method through a general-purpose system capable of automatically performing the entire process from the input of brain imaging data to the analysis results. Brief explanation of the drawing

[0012] FIG. 1 is a configuration diagram of a functional connectivity and gradient analysis system for preclinical functional magnetic resonance imaging according to one embodiment of the present invention. Figure 2 is a diagram illustrating an example of analyzing preclinical functional connectivity and gradients by the analysis system shown in Figure 1. FIG. 3 is a drawing illustrating an example of visualizing functional connectivity according to an embodiment of the present invention. FIG. 4 is a drawing illustrating an example of visualizing the slope of functional connectivity in two-dimensional and three-dimensional spaces according to an embodiment of the present invention. FIG. 5 is a flowchart of a method for analyzing functional connectivity and gradient of preclinical functional magnetic resonance imaging according to another embodiment of the present invention. Specific details for implementing the invention

[0013] Preferred embodiments according to the present invention will be described in detail below with reference to the attached drawings. In this process, the thickness of lines or the size of components shown in the drawings may be exaggerated for clarity and convenience of explanation.

[0014] Furthermore, the terms described below are defined in consideration of their functions within the present invention, and these may vary depending on the intent or practice of the user or operator. Therefore, the definitions of these terms should be based on the content throughout this specification.

[0015] FIG. 1 is a configuration diagram of a functional connectivity and gradient analysis system for preclinical functional magnetic resonance imaging according to one embodiment of the present invention.

[0016] Referring to FIG. 1, a preclinical functional connectivity and gradient analysis system (100) according to one embodiment of the present invention may be configured to include a brain neural image preprocessing unit (110), a brain neural signal extraction unit (120), a normalization unit (130), a functional network generation unit (140), a functional connectivity and gradient analysis unit (150), and an analysis result visualization unit (160).

[0017] The brain neural image preprocessing unit (110) can perform preprocessing on the input brain neural image (i.e., fMRI image), and can perform at least one of, for example, MRI image orientation correction, voxel scaling, slice timing correction, head movement correction, distortion correction, and image co-registration / normalization, but is not limited thereto.

[0018] Specifically, in MRI image orientation correction, the image orientation can be corrected according to the anterior-posterior axis, the superior-inferior axis, and the left-right axis.

[0019] Generally, brain function imaging data reconstructed in the axial plane, sagittal plane, and coronal plane is composed of a three-dimensional volume. Since the image may be flipped along the anterior-posterior, superior-inferior, or left-right axes depending on the fMRI image reconstruction direction of the MRI equipment, MRI image orientation correction can be performed to correct these problems.

[0020] In addition, voxel size adjustment allows the size of the voxels to be enlarged by several to tens of times. Generally, since the brain size of small animals is very small, data captured at a size of less than a millimeter (mm) can be enlarged and adjusted by a preset magnification (e.g., 10 times) for smooth image preprocessing.

[0021] In addition, slice timing correction can correct the time difference between multiple slices acquired at different times.

[0022] In addition, head movement correction can correct positional changes caused by head movements that occur during brain function imaging. In other words, the accuracy of image analysis can be improved by correcting positional changes of head movements over time, and six parameters related to head movement—namely, translation parameters x, y, and z, and rotation parameters pitch, roll, and yaw—can be corrected based on the first acquired image.

[0023] In addition, image distortion correction can correct geometric distortion of the image or signal loss caused by the non-uniformity of the magnetic field.

[0024] In addition, in image registration, functional images can be registered to improve the resolution of functional images and for quantitative analysis.

[0025] For example, if the disease animal model is a small animal, functional images can be co-registered with structural images to improve their resolution. Additionally, since structural positional information varies by individual, non-linear transformation techniques can be used to normalize the images to a standard space for quantitative analysis.

[0026] The brain neural image preprocessing unit (110) can apply various preprocessing techniques known to a person skilled in the art to perform preprocessing of fMRI images as described above, so a detailed description thereof is omitted.

[0027] The brain neural signal extraction unit (120) can extract brain neural signals from a preprocessed brain neural image. Here, the brain neural signal extraction unit (120) can perform the extraction of brain neural signals from a brain neural image by applying techniques known to a person skilled in the art, and a detailed description thereof is omitted.

[0028] The normalization unit (130) can perform normalization to remove noise from the extracted brain neural signal.

[0029] According to one embodiment, the normalization unit (130) can perform filtering operations in the brain neural signal time series using the following nuisance parameters to unify the brain neural signals extracted from each brain region into a single band. Normalization can be performed using six parameters of a rigid body for correcting head movement in brain functional imaging (three translation parameters and three rotation parameters for the X, Y, and Z axes) and their derivative values, three principal components for each tissue in the white matter and cerebrospinal fluid mask obtained from brain structure segmentation, linear and quadratic regressors, and band-pass filtering at 0.01-3 Hz.

[0030] The functional network generation unit (140) can generate a functional network by receiving a normalized signal.

[0031] According to one embodiment, the functional network generation unit (140) can construct an M×M functional connection matrix by using the Pearson correlation coefficient method to generate a brain functional network, calculating correlation coefficients for all pairs of time series signals from which noise has been removed in multiple regions, and converting them into Fisher's Z scores. Here, M represents the number of regions.

[0032] The functional connectivity and slope analysis unit (150) can analyze the network of functional connectivity and analyze the slope of functional connectivity.

[0033] According to one embodiment, the functional connectivity and gradient analysis unit (150) can calculate global and local network attributes by performing graph theory analysis of functional connectivity using a publicly available technology (e.g., Brain Connectivity Tool box) to analyze the functional connectivity network.

[0034] For example, the functional connectivity and gradient analysis unit (150) can calculate the global node degree and global node strength to determine global network properties. The node degree represents the total number of edges connected to a single node (n), and the global node degree ( ) is defined as the average value of the node degree of all nodes as shown in Equation 1 below.

[0035]

[0036] Here, N is the total number of nodes, and G represents the adjacency matrix.

[0037] Node strength is defined as the sum of the weights of all edges connected to a single node, and global node strength ( ) represents the average value of the node strength of all nodes and is defined as in Equation 2 below.

[0038]

[0039] Here, ω represents the weight of the edge between nodes i and j, N is the total number of nodes, and G represents the adjacency matrix.

[0040] In addition, the functional connectivity and gradient analysis unit (150) can calculate local efficiency and clustering coefficient to identify local network attributes. Local efficiency ( ) is the global efficiency between adjacent nodes as a ratio of path lengths. It is defined as the average of ) as shown in mathematical formula 3 below.

[0041]

[0042]

[0043] Here, d represents the shortest distance connecting nodes i and j.

[0044] The cohesion factor (C) is defined as the ratio of the number of edges connected to one node and adjacent nodes, as shown in Equation 4 below.

[0045]

[0046] Here, represents the node level of node i.

[0047] Meanwhile, the functional connectivity and slope analysis unit (150) can perform the following process to analyze the slope of the functional connectivity.

[0048] 1) fMRI Data Preparation

[0049] 2) Calculation of functional connectivity

[0050] 3) Diffusion embedding

[0051] 4) Search for related results in previous literature

[0052] 5) Investigation of computational mechanisms explaining fundamental functional organization

[0053] In the gradient analysis method described above, the first and second components of dimensionality reduction are used as gradient values, and changes in gradient values ​​represent variability within the data.

[0054] The analysis result visualization unit (160) can visualize the functional connectivity and the slope of the functional connectivity as at least one of a graph, mapping, or rendering.

[0055] FIG. 3 is a drawing illustrating an example of visualizing functional connectivity according to an embodiment of the present invention.

[0056] According to one embodiment, the analysis result visualization unit (160) can visualize functional connections by brain region as a circular map and a three-dimensional connectivity map.

[0057] Specifically, (a) of FIG. 3 shows the degree of interconnectedness between nodes by representing brain structural regions as nodes in a circular map, and (b) shows the degree of interconnectedness between the nodes shown in (a) in the midbrain, hypothalamus, and thalamus regions in a three-dimensional connectivity map.

[0058] FIG. 4 is a drawing illustrating an example of visualizing the gradient of functional connectivity in two-dimensional and three-dimensional spaces according to an embodiment of the present invention.

[0059] According to one embodiment, the analysis result visualization unit (160) can visualize the distribution graph of gradient values ​​and the degree of distribution by region of the brain as a three-dimensional surface rendering.

[0060] Specifically, Figure 4(a) shows the distribution of the first gradient value (gradient 1) and the second gradient value (gradient 2) of the sensorimotor area, visual regions, and brainstem areas of the brain as a two-dimensional graph, and (b) shows the area where the nodes of (a) are distributed in the actual brain as a three-dimensional surface rendering.

[0061] FIG. 5 is a flowchart of a method for analyzing functional connectivity and gradient of preclinical functional magnetic resonance imaging according to another embodiment of the present invention.

[0062] Referring to FIG. 5, a method for analyzing functional connectivity and gradient of preclinical functional magnetic resonance imaging according to another embodiment of the present invention may include a brain neural image preprocessing step (S10); a brain neural signal extraction step (S20); a brain neural signal normalization step (S30); a functional network generation step (S40); a functional connectivity and gradient analysis step (S50); and an analysis result visualization step (S60).

[0063] The brain neural image preprocessing step (S10) can perform preprocessing on the brain neural image.

[0064] The brain neural signal extraction step (S20) can extract brain neural signals from the preprocessed brain neural image.

[0065] The brain neural signal normalization step (S30) can perform normalization to remove noise from the extracted brain neural signal.

[0066] The functional network generation step (S40) can generate a functional network including a global network and a local network using a normalized signal.

[0067] The functional connectivity and gradient analysis step (S50) can analyze the functional connectivity and functional connectivity gradient of the generated global network and local network.

[0068] The analysis result visualization step (S60) can visualize the functional connectivity analysis results and the functional connectivity gradient analysis results.

[0069] The specific details of each step are as described with reference to FIGS. 1 to 4, so redundant explanations are omitted.

[0070] The present invention has been described with reference to embodiments illustrated in the drawings, but this is merely illustrative, and those skilled in the art will understand that various modifications and equivalent alternative embodiments are possible therefrom. Accordingly, the true technical scope of protection of the present invention should be determined by the technical spirit of the following claims. Explanation of the symbols

[0071] 110: Brain neural imaging preprocessing unit 120: Brain nerve signal extraction unit 130: Normalization section 140: Functional Network Generation Unit 150: Functional Connectivity and Gradient Analysis Section 160: Analysis Result Visualization Section

Claims

Claim 1 A neuroimaging preprocessing unit that performs preprocessing on an input neuroimaging image; a neuroimaging signal extraction unit that extracts neuroimaging signals from the preprocessed neuroimaging image; a normalization unit that performs normalization to remove noise from the extracted neuroimaging signals; a functional network generation unit that receives the normalized signal and generates a functional network including a global network and a local network; and a functional connectivity and gradient analysis unit that analyzes the functional connectivity of the functional network and the gradient of said functional connectivity.and includes an analysis result visualization unit that visualizes the functional connectivity or the gradient of functional connectivity as an analysis result by the functional connectivity and gradient analysis unit as at least one of a graph, mapping, or rendering; the functional network generation unit receives the normalized signal as input, calculates the correlation coefficient for all pairs of time-series signals in multiple regions using the Pearson correlation coefficient, and generates an M×M functional connectivity matrix by applying a Fisher transform to the calculated correlation coefficient, where M is the number of regions; the functional connectivity and gradient analysis unit analyzes global network properties by calculating the global node degree and global node strength in the global network, and analyzes local network properties by calculating the local efficiency and clustering coefficient in the local network; calculates the gradient value of functional connectivity using the first and second components derived after applying dimensionality reduction to the functional connectivity matrix, and analyzes variability based on the calculated gradient value of functional connectivity; and the analysis result visualization unit [represents] brain structural regions A system for analyzing functional connectivity and gradients of preclinical functional magnetic resonance imaging, which visualizes functional connectivity by brain region—indicated by nodes to show the degree of interconnectedness between nodes—as a circular map or a 3D connectivity map, and visualizes the distribution of gradient values ​​of said functional connectivity mapped on a multidimensional space with said first component and said second component as axes, as a 2D graph or a 3D surface rendering. Claim 2 In claim 1, the brain neural image preprocessing unit performs at least one of MRI image orientation correction, voxel scaling, slice timing correction, motion correction, distortion correction, and co-registration / normalization, in a preclinical functional connectivity and gradient analysis system for functional connectivity and gradient of functional magnetic resonance imaging. Claim 3 delete Claim 4 delete Claim 5 A neuroimaging preprocessing step in which a neuroimaging preprocessing unit performs preprocessing on an input neuroimaging image; a neuroimaging signal extraction step in which a neuroimaging signal extraction unit extracts neuroimaging signals from the preprocessed neuroimaging image; a neuroimaging signal normalization step in which a normalization unit performs normalization to remove noise from the extracted neuroimaging signals; a functional network generation step in which a functional network generation unit receives the normalized signals and generates a functional network including a global network and a local network; and a functional connectivity and gradient analysis step in which a functional connectivity and gradient analysis unit analyzes the functional connectivity and the gradient of the functional connectivity of the functional network.The analysis result visualization unit includes a step of visualizing the functional connectivity or the gradient of functional connectivity as an analysis result by the functional connectivity and gradient analysis unit using at least one of a graph, mapping, or rendering; the step of generating the functional network includes receiving the normalized signal as input, calculating the correlation coefficient for all pairs of time-series signals in multiple regions using the Pearson correlation coefficient, and generating an M×M functional connectivity matrix by applying a Fisher transform to the calculated correlation coefficient, where M is the number of regions; the functional connectivity and gradient analysis step includes analyzing global network properties by calculating the global node degree and global node strength in the global network, analyzing local network properties by calculating the local efficiency and clustering coefficient in the local network, calculating the gradient value of functional connectivity using the first and second components derived after applying dimensionality reduction to the functional connectivity matrix, and analyzing variability based on the calculated gradient value of functional connectivity; and the visualization step includes brain structural A method for analyzing functional connectivity and gradients of preclinical functional magnetic resonance imaging, performed by a system for analyzing functional connectivity and gradients of preclinical functional magnetic resonance imaging, wherein regions are represented as nodes to visualize functional connectivity by brain region, indicating the degree of interconnectedness between nodes, as a circular map or a 3D connectivity map, and the degree of distribution of gradient values ​​of said functional connectivity mapped on a multidimensional space with said first component and said second component as axes is visualized as a 2D graph or a 3D surface rendering.; Claim 6 delete

Images