Quantitative analysis method and system for neuron reconstruction result

Abstract

The invention discloses a quantitative analysis method and system for a neuron reconstruction result. The method comprises the following steps: acquiring a predetermined neuron reconstruction node tree; de-weighting nodes in a predetermined neuron reconstruction node tree according to the coordinates to obtain a point set containing a plurality of node coordinates; and calculating according to the point set containing the plurality of node coordinates to obtain a minimum envelope ball containing all nodes, wherein the center coordinate and radius of the minimum envelope ball are the characteristics of the neuron reconstruction node tree. According to the method, for a node tree with given coordinates, the space geometric center and the minimum envelope sphere radius of the node coordinate tree are calculated through a welzl minimum envelope sphere algorithm, so that the characteristics of the neuron morphology are extracted for subsequent subtype classification and other researches; and the downward diffusion range of each node can be calculated as an index for evaluating the influence range of errors.

Description

Technical Field

[0001] The present invention relates to the technical field of neuron morphological feature extraction, and in particular to a quantitative analysis method and system for neuron reconstruction results. Background Art

[0002] Neuronal morphology reconstruction is a common analytical method in the field of neuroscience. Figure 1 After the whole brain imaging is completed, a low-resolution whole-brain image is obtained. Based on this image, a whole-brain coordinate system is constructed, and all or specific neurons in the target area are tracked and reconstructed. Using neuron reconstruction software such as Lychnis and Vaa3D, the neurons in the target area are tracked by combining points and lines, and finally a set of directed line segments (files) that completely record individual neurons is generated, which can be used for morphological classification and other research. Such a set of directed line segments is the digital reconstruction of neurons, that is, the neurons are "stripped" from the original image and abstracted into a description of some three-dimensional coordinate points with a branching order.

[0003] For neuron feature extraction, neuroXiv, a website jointly developed by the Institute of Brain Science and Intelligence Technology at Southeast University and Tencent's Ailab, provides some data metrics (select a neuron and click "Analyze" to view the distribution of related feature metrics). However, these statistical feature metrics mostly focus on length and cannot directly assess the neuron's diffusion range. In neuron analysis, it is important to understand the spatial extent of subsequent nodes that may be affected if a node fails. Furthermore, understanding the specific spatial distribution of neurons in the whole-brain coordinate system (where the clustering center is) provides more information for subsequent analysis.

[0004] For example, if the experimenter knows the wrong node coordinate A, then the existing technology can quickly calculate the length of the subsequent wrong nodes and the error ratio and other information. However, it is difficult to accurately and intuitively know the specific impact range of this error in the whole brain coordinate system.

[0005] Therefore, the existing technology still needs to be improved and developed. Summary of the Invention

[0006] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide a quantitative analysis method and system for neuron reconstruction results, aiming to solve the problem that the existing neuron feature extraction method cannot directly obtain an assessment of the diffusion range of neurons.

[0007] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows:

[0008] A quantitative analysis method for neuronal reconstruction results, comprising the steps of:

[0009] Obtaining a predetermined neuron reconstruction node tree;

[0010] De-duplication of nodes in the predetermined neuron reconstruction node tree is performed based on coordinates to obtain a point set containing several node coordinates;

[0011] The minimum envelope sphere containing all nodes is calculated based on a point set containing several node coordinates. The center coordinates and radius of the minimum envelope sphere are the characteristics of the neuron reconstructed node tree.

[0012] A quantitative analysis method for neuron reconstruction results, wherein a point set containing a plurality of node coordinates contains one of 0 points, 1 point, 2 points, 3 points, or more than 3 points.

[0013] A quantitative analysis method for neuron reconstruction results, wherein when a point set containing several node coordinates contains zero points, an empty envelope sphere is calculated based on the point set containing several node coordinates.

[0014] A quantitative analysis method for neuron reconstruction results, wherein when a point set containing several node coordinates contains one point, an envelope sphere with the point as the sphere center and a radius of 0 is calculated based on the point set containing several node coordinates.

[0015] A quantitative analysis method for neuronal reconstruction results, wherein, when a point set containing several node coordinates includes two points, an envelope sphere with the midpoint of the two points as the sphere center and the distance between the two points as the diameter is calculated based on the point set containing several node coordinates.

[0016] A quantitative analysis method for neuron reconstruction results, in which, when a point set containing several node coordinates includes three points, a set of equations is listed based on the equal distances from the three points to the center of the envelope sphere to calculate the center and radius of the envelope sphere.

[0017] A quantitative analysis method for the neuron reconstruction results, wherein when the point set containing the coordinates of several nodes contains more than 3 points, 3 points are randomly selected from the point set as the boundary point set, and the remaining points are regarded as the external point set;

[0018] Set a point in the outer point set as the focus, and recursively call the algorithm to calculate the envelope sphere of the point set excluding the focus and the boundary point set.

[0019] A quantitative analysis method for neuronal reconstruction results, wherein if the focus is within the envelope sphere, the envelope sphere is the minimum envelope sphere;

[0020] Alternatively, if the focus is not within the envelope sphere, the focus is transferred from the outer point set to the boundary point set, and a recursive call algorithm is used to calculate a new envelope sphere containing the focus until all points are contained in the new envelope sphere. The new envelope sphere is then the minimum envelope sphere.

[0021] A quantitative analysis method for neuron reconstruction results, wherein the predetermined neuron reconstruction node tree is neuron node tree data stored in SWC format after whole-brain imaging and morphological reconstruction of neurons in a whole-brain coordinate system.

[0022] A quantitative analysis system for neuronal reconstruction results, comprising:

[0023] A target acquisition module is used to obtain a predetermined neuron reconstruction node tree;

[0024] The data cleaning module is used to remove duplicate nodes in the predetermined neuron reconstruction node tree according to their coordinates to obtain a point set containing several node coordinates;

[0025] The feature extraction module is used to calculate the minimum envelope sphere containing all nodes based on a point set containing several node coordinates. The center coordinates and radius of the minimum envelope sphere are the features of the neuron reconstruction node tree.

[0026] Beneficial effects: The present invention provides a quantitative analysis method and system for neuron reconstruction results. For a node tree with given coordinates, the spatial geometric center and the minimum envelope radius of the node coordinate tree are calculated by the Welzl minimum sphere envelope algorithm, thereby extracting the characteristics of neuron morphology for subsequent subtype classification and other studies. In addition, by calculating the downward diffusion range of each node, it can be used as an indicator for evaluating the impact range of the error, and then it can be quickly evaluated which part of the neurons the neuron will affect. Combined with the standard atlas, the affected brain area can be further inferred. Compared with the previous statistics based on the length between nodes, the method of the present invention can better describe the spatial distribution characteristics of the neuron reconstruction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0028] Figure 1 It is a low-resolution whole-brain image obtained after the existing whole-brain imaging is completed.

[0029] Figure 2 This is a flow chart of a quantitative analysis method for neuron reconstruction results provided by an embodiment of the present invention.

[0030] Figure 3 A schematic diagram of the neuron reconstruction process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] For neuron feature extraction, neuroXiv, a website jointly developed by the Institute of Brain Science and Intelligence Technology at Southeast University and Tencent's Ailab, provides some data metrics (select a neuron and click "Analyze" to view the distribution of related feature metrics). However, these statistical feature metrics mostly focus on length and cannot directly assess the neuron's diffusion range. In neuron analysis, it is important to understand the spatial extent of subsequent nodes that may be affected if a node fails. Furthermore, understanding the specific spatial distribution of neurons in the whole-brain coordinate system (where the clustering center is) provides more information for subsequent analysis.

[0034] Based on the above technical problems, the present invention provides a quantitative analysis method and system for neuron reconstruction results. Figure 2 , which comprises the steps of:

[0035] S100, obtaining a predetermined neuron reconstruction node tree;

[0036] S200, deduplicating nodes in a predetermined neuron reconstruction node tree according to coordinates to obtain a point set containing coordinates of several nodes;

[0037] S300 , calculating a minimum envelope sphere containing all nodes based on a point set containing several node coordinates, wherein the center coordinates and radius of the minimum envelope sphere are the features of the node tree reconstructed by the neuron.

[0038] Specifically, the present invention uses the Welzl minimum sphere envelope algorithm to calculate the spatial geometric center and minimum envelope radius of the node coordinate tree for a node tree with given coordinates, thereby extracting the characteristics of neuronal morphology for subsequent subtype classification and other studies. In addition, by calculating the downward diffusion range of each node, it can be used as an indicator for evaluating the impact range of the error, and then it can be quickly evaluated which part of the neurons the neuron will affect. Combined with the standard atlas, the affected brain area can be further inferred. Compared with the previous statistics based on the length between nodes, the method of the present invention can better describe the spatial distribution characteristics of the neuron reconstruction results.

[0039] In some embodiments, the point set containing the coordinates of the plurality of nodes includes one of 0 points, 1 point, 2 points, 3 points, or more than 3 points.

[0040] In some embodiments, when the point set containing the node coordinates contains 0 points, an empty envelope sphere is calculated based on the point set containing the node coordinates.

[0041] In some embodiments, when a point set containing several node coordinates includes one point, an envelope sphere with the point as the sphere center and a radius of 0 is calculated based on the point set containing the several node coordinates.

[0042] In some embodiments, when a point set containing several node coordinates includes two points, an enveloping sphere with the midpoint of the two points as the sphere center and the distance between the two points as the diameter is calculated based on the point set containing the several node coordinates.

[0043] In some embodiments, when the point set containing the coordinates of several nodes includes three points, a set of equations is listed based on the assumption that the distances from the three points to the center of the envelope sphere are equal to calculate the center and radius of the envelope sphere.

[0044] In some embodiments, when the point set containing the coordinates of several nodes contains more than 3 points, 3 points are randomly selected from the point set as the boundary point set, and the remaining points are used as the external point set;

[0045] Set a point in the outer point set as the focus, and recursively call the algorithm to calculate the envelope sphere of the point set excluding the focus and the boundary point set.

[0046] In some embodiments, if the focus is within the envelope sphere, the envelope sphere is a minimum envelope sphere;

[0047] Alternatively, if the focus is not within the envelope sphere, the focus is transferred from the outer point set to the boundary point set, and a recursive call algorithm is used to calculate a new envelope sphere containing the focus until all points are contained in the new envelope sphere. The new envelope sphere is then the minimum envelope sphere.

[0048] In some embodiments, the predetermined neuron reconstruction node tree is neuron node tree data stored in SWC format after whole-brain imaging and morphological reconstruction of neurons in a whole-brain coordinate system.

[0049] See also Figure 3 , a schematic diagram of the neuron reconstruction process. A is a schematic representation of a neuron image in the whole-brain coordinate system after whole-brain imaging. B is a digital reconstruction of neuron morphology based on neuron image A, "retracing" the neuron image using connected directed line segments. Green dots represent cell body nodes, red dots represent axon nodes, and blue dots represent dendrite nodes. C is the result of neuron reconstruction in the whole-brain coordinate system after image stripping. Node connectivity and node type reveal the region in which the neuron is located and the extent of its axons and dendrites. D is a file storing the node tree in C in the SWC (Stockley–Wheal–Cole) format, a format first proposed by EW Stockley, HV Wheal, and HMCole, and named after them.

[0050] The present invention also provides a quantitative analysis system for neuron reconstruction results, comprising:

[0051] A target acquisition module is used to obtain a predetermined neuron reconstruction node tree;

[0052] The data cleaning module is used to remove duplicate nodes in the predetermined neuron reconstruction node tree according to their coordinates to obtain a point set containing several node coordinates;

[0053] The feature extraction module is used to calculate the minimum envelope sphere containing all nodes based on a point set containing several node coordinates. The center coordinates and radius of the minimum envelope sphere are the features of the neuron reconstruction node tree.

[0054] In summary, the present invention discloses a quantitative analysis method and system for neuron reconstruction results, the method comprising the steps of: obtaining a predetermined neuron reconstruction node tree; deduplicating nodes in the predetermined neuron reconstruction node tree according to coordinates to obtain a point set containing a number of node coordinates; calculating a minimum envelope sphere containing all nodes based on the point set containing the number of node coordinates, the coordinates and radius of the center of the minimum envelope sphere being the characteristics of the neuron reconstruction node tree. The present invention calculates the spatial geometric center and the minimum envelope sphere radius of the node coordinate tree by the Welzl minimum sphere envelope sphere algorithm for a node tree with given coordinates, thereby extracting the characteristics of the neuron morphology for subsequent subtype classification and other studies, and calculating the downward diffusion range of each node can be used as an indicator for evaluating the impact range of the error, thereby being able to quickly evaluate which part of the neurons the neuron will specifically affect, and combining with the standard atlas to further infer the affected brain area. Compared with the previous statistics based on the length between nodes, the method of the present invention can better describe the spatial distribution characteristics of the neuron reconstruction results.

[0055] The above are only preferred embodiments of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A quantitative analysis method for neuronal reconstruction results, characterized in that: Including steps: Obtaining a predetermined neuron reconstruction node tree; Deduplication of nodes in the predetermined neuron reconstruction node tree is performed based on coordinates to obtain a point set containing coordinates of several nodes; The minimum envelope sphere including all nodes is calculated based on the point set containing the coordinates of several nodes. The center coordinates and radius of the minimum envelope sphere are the characteristics of the neuron reconstruction node tree.

2. The quantitative analysis method of neuron reconstruction results according to claim 1, characterized in that: The point set containing the coordinates of the plurality of nodes includes one of 0 points, 1 point, 2 points, 3 points, or more than 3 points.

3. The method for quantitative analysis of neuronal reconstruction results according to claim 2, characterized in that: When the point set containing the coordinates of the plurality of nodes contains 0 points, an empty envelope sphere is calculated based on the point set containing the coordinates of the plurality of nodes.

4. The method for quantitative analysis of neuronal reconstruction results according to claim 2, characterized in that: When the point set containing the coordinates of the plurality of nodes includes one point, an envelope sphere with the point as the center and a radius of 0 is calculated based on the point set containing the coordinates of the plurality of nodes.

5. The method for quantitative analysis of neuronal reconstruction results according to claim 2, characterized in that: When the point set containing the coordinates of the plurality of nodes includes two points, an envelope sphere with the midpoint of the two points as the sphere center and the distance between the two points as the diameter is calculated based on the point set containing the coordinates of the plurality of nodes.

6. The method for quantitative analysis of neuronal reconstruction results according to claim 2, characterized in that: When the point set containing the coordinates of the plurality of nodes includes three points, a set of equations is listed based on the equal distances between the three points and the center of the envelope sphere to calculate the center and radius of the envelope sphere.

7. The method for quantitative analysis of neuronal reconstruction results according to claim 2, characterized in that: When the point set containing the coordinates of several nodes contains more than 3 points, 3 points are randomly selected from the point set as the boundary point set, and the remaining points are used as the external point set; Set a point in the outer point set as the focus, and recursively call the algorithm to calculate the envelope sphere of the point set excluding the focus and the boundary point set.

8. The method for quantitative analysis of neuron reconstruction results according to claim 7, characterized in that: If the focus is within the envelope sphere, then the envelope sphere is the minimum envelope sphere; Alternatively, if the focus is not within the envelope sphere, the focus is transferred from the outer point set to the boundary point set, and a new envelope sphere containing the focus is calculated using a recursive call algorithm until all points are contained in the new envelope sphere, and the new envelope sphere is the minimum envelope sphere.

9. The method for quantitative analysis of neuron reconstruction results according to claim 1, characterized in that: The predetermined neuron reconstruction node tree is neuron node tree data stored in SWC format after whole-brain imaging and morphological reconstruction of neurons in a whole-brain coordinate system.

10. A quantitative analysis system for neuron reconstruction results, characterized in that: include: A target acquisition module is used to obtain a predetermined neuron reconstruction node tree; A data cleaning module, configured to remove duplicate nodes in the predetermined neuron reconstruction node tree according to coordinates, and obtain a point set containing coordinates of several nodes; The feature extraction module is used to calculate the minimum envelope sphere containing all nodes based on the point set containing the coordinates of several nodes, and the center coordinates and radius of the minimum envelope sphere are the features of the neuron reconstruction node tree.

Images