JP5206562B2 - Network analysis device between objects - Google Patents

Description

  The present invention relates to a technique for analyzing a connection relationship between captured objects in a captured image.

  Things in nature often have some kind of connection relationship with each other, and attempts have been made to explore the connection relationship of objects in images obtained by photographing. In particular, since cells are components of tissues, they rarely exist alone and may adhere to nearby cells. Depending on the culture conditions, the presence or absence of adhesion, the direction of adhesion, adhesion to multiple cells, etc. There is a request to evaluate. Regarding the relationship between cells, there is a case where a link between a mother cell and a daughter cell is set using a Nomarski type 4D microscope in the creation of a cell lineage that summarizes the cell division and development process from a fertilized egg (Patent Literature). 1).

WO2002-45018

  However, the link in Patent Document 1 is a genetic link and is not related to a physical link. Therefore, the physical link between cells in the captured image must be set manually, but generally, when there are N objects (nodes), the number of links that can be set is N. Since the order is squared, the burden of manual work is enormous. In addition, there are various interpretations of the link setting at the branch point and the like.

  Therefore, an object of the present invention is to provide a network analysis apparatus between objects that can efficiently analyze the connection relationship between objects shown in a photographed image.

  In order to solve the above-mentioned problem, in the first aspect of the present invention, an object setting means for selecting a plurality of objects from image data and setting object markers for clearly indicating the positions of the objects on the image data; For the set object marker A, a link is made to another object marker B located in the vicinity on the same image data, and an identification mark clearly indicating that the object marker B is linked Object group extraction means for extracting one or a plurality of object groups connected by the object link setting means to be added to the object marker A and links set in the plurality of objects. And a branch point analyzing means for extracting, as a branch point, an object linked from a plurality of objects among the objects configured for each object group. To provide a network analyzer between the object to be.

  According to the first aspect of the present invention, each object marker is set to each of a plurality of objects in an image, each object marker is linked to another object marker, and each object group connected by the link Since the object to which the link is attached is extracted as a branch point, it is possible to efficiently analyze the connection relationship between the objects shown in the photographed image.

  In the second aspect of the present invention, in the network analysis device according to the first aspect of the present invention, the inter-object link setting means accepts links from a plurality of other object markers. The feature is that only one link to another object marker can be set.

  According to the second aspect of the present invention, in the first aspect of the present invention, each object marker accepts a plurality of links, but only one link can be established, so there are various interpretations. It is easy to specify the branch point to be performed, and it is difficult to prevent setting omission and duplicate setting.

  Further, in the third aspect of the present invention, in the network analysis device according to the second aspect of the present invention, the inter-object link setting means is a predetermined object marker A for which a link to another object marker is not set. When the edge point is close to a predetermined edge point of another object marker B within a predetermined distance, a link from the object marker A to the object marker B is set.

  According to the third aspect of the present invention, in the second aspect of the present invention, a predetermined edge point of the object marker A is close to a predetermined edge point of another object marker B within a predetermined distance. In this case, since the link from the object marker A to the object marker B is set, the link setting between the objects can be easily and quickly performed.

  Further, in the fourth aspect of the present invention, in the network analysis device according to the third aspect of the present invention, the inter-object link setting means has already set a link to another object marker B. When the predetermined edge point of the object marker C is close to the predetermined edge point of the object marker C for which no link to another object marker is set within a predetermined distance, the object marker A to the object marker A It is characterized by setting a link to.

  According to the fourth aspect of the present invention, in the third aspect of the present invention, a link to another is already established, and it is not possible to duplicate a link to another based on the second aspect of the present invention. Even so, it is possible to accept multiple links from other sources. Instead, the link is set in the opposite direction from the other, so it is always a link between objects that are close within a certain distance. Can be stretched.

  In the fifth aspect of the present invention, in the network analysis device according to the third aspect of the present invention, the inter-object link setting means sets the object marker A for which a link to another object marker B has already been set. The predetermined edge point of the target marker A is close to the predetermined edge point of the target marker C for which a link to another target marker D has already been set within a predetermined distance, and the predetermined edge of the target marker A When the point is separated from the predetermined edge point of the object marker D by a predetermined distance or more, the link from the object marker A to the object marker B is released, and the link from the object marker B to the object marker A is released. And a link from the object marker A to the object marker C is set.

  According to the fifth aspect of the present invention, in the third aspect of the present invention, each of the two objects has already established a link to the other, and based on the second aspect of the present invention, both directions are bidirectional. Even if it is not possible to create a duplicate link, the link will be changed so that one feature marker will release the link to the other and the other will set the link to the feature. Since it did in this way, it becomes possible to link between the objects which always adjoin within the predetermined distance.

  According to a sixth aspect of the present invention, in the network analysis device according to any one of the second to fifth aspects of the present invention, the object marker has a rhombus shape, and two vertices in a major axis direction are defined as the predetermined points. One of the two vertices in the long axis direction of the object marker A that is not set as a link to another object marker is one of the two vertices in the long axis direction of the object marker B. When close to each other within a predetermined distance, a link from the object marker A to the object marker B is set.

  According to the sixth aspect of the present invention, in any one of the second to fifth aspects of the present invention, two points in the major axis direction of the diamond-shaped object marker are used as edge points, and the link is performed using these edge points as a reference. Therefore, it is possible to set a link in consideration of cell directionality.

  Further, in the seventh aspect of the present invention, in the network analysis device according to any one of the first to sixth aspects of the present invention, the branch point analyzing means includes a length of each object marker configured for each object group. A total length value, a maximum width value, an average width value, and a total area value are calculated on the basis of geometric shape parameters including length, width, and area.

  According to the seventh aspect of the present invention, in any one of the first to sixth aspects of the present invention, the total length value and the maximum width of the groups connected on the basis of the geometric shape parameter of each object marker. Since the value, the average value of the width, and the total value of the areas are calculated, it is possible to quickly analyze the form of the whole connected group.

  According to the present invention, there is an effect that it is possible to efficiently analyze a connection relationship between objects captured in a captured image.

It is a block diagram of the network analysis apparatus which concerns on this invention. It is a figure which shows an example of the object marker which the network analysis apparatus of this invention generates. It is a figure which shows the mode before the link setting between target markers, and the state after a link setting in the condition where many target markers exist. It is a figure which shows the mode in the case of changing a link setting manually. It is a figure which shows the rough flow of the network topology analysis between objects. It is a figure which shows the detail of the tree analysis by the target object group extraction means 43. FIG. 5 is a flowchart showing details of branch point analysis and geometric shape analysis of each tree by a branch point analysis means 44.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(1. Cell culture and imaging)
The network analysis apparatus according to the present invention can analyze the connection relationship of various objects in a captured image, but a particularly promising object is a case where a cell is an object. In particular, it is useful when analyzing the network structure of vascular cells such as vascular endothelial cells. When analyzing the connection relationship of cells, a step of culturing cells in advance and photographing the cultured cells is required. In this embodiment, a cell culture substrate having a surface provided with a cell adhesion variable region / cell non-adhesion region is prepared, and cells are cultured on this substrate. For example, when cells are seeded on a substrate where the cell adhesion variable region / cell non-adhesion region has a line pattern with a width of 50 microns / 100 microns, the cells adhere only to the cell adhesion variable region by culturing under predetermined conditions And grow. After adhering for a predetermined time, the cells can be collected while maintaining the cell pattern by exposing to an environment in which the surface of the cell adhesion variable region changes and contacting with a biomaterial such as gelatin. The collected cell pattern can be sandwiched between other cell sheets to create a pseudo three-dimensional tissue in vitro and analyze its behavior. At that time, research on how the blood vessel pattern constructed in a line shape behaves in the pseudo three-dimensional tissue is also actively conducted. Even in the case of a pre-designed pattern, there are cases where the pattern is maintained and cases where it is not maintained in a three-dimensional organization environment. In other words, it is also conceivable that string-like cells that have constructed a line pattern at a certain interval mutually stimulate each other (paracline effect) and cause a bridge between cells. When bridging occurs, the number of branch points and branches increases, and many studies systematically analyze the relationship between pattern design (line width, spacing) and the resulting vascular network structure. Therefore, it is necessary to prepare a pseudo three-dimensional tissue, perform a process of specifically staining vascular cells, and acquire an image with a confocal laser microscope (Carl Zeiss, LSM5, etc.).

(2. Device configuration)
When the image file is acquired, the network analysis device according to the present invention analyzes the connection relationship of the objects in the image file. First, the configuration of the network analysis apparatus according to the present invention will be described. FIG. 1 is a configuration diagram of a network analysis apparatus according to the present invention. In FIG. 1, 10 is an image data storage unit, 20 is an image display unit, 30 is an input instruction unit, 40 is an arithmetic processing unit, 41 is an object setting unit, 42 is an inter-object link setting unit, and 43 is an object group. Extraction means 44 is a branch point analysis means.

  The image data storage means 10 stores an image file (cell-photographed image) obtained by preparation, and stores various data generated by network analysis processing, and is connected to or built in a computer such as a hard disk. This is realized by the storage device. Specifically, an image file processed by Metamorph (Molecular Device) or the like is prepared. Further, the image file in the image data storage means 10 may be a moving image file that exists as a set of a plurality of still images (frames). Specifically, a moving image file prepared by Olympus software DPManager, NIH software ImageJ, and its plug-in software MTtrackJ is prepared. In addition, the image data storage unit 10 stores the scale information of the captured image in order to convert the distance based on the coordinate value on the image into the distance of the SI unit system. Further, the image data storage means 10 is an image in which each still image is actually taken at what time interval as time-lapse data information when a moving image file is stored in the image data storage means 10. Is stored as a frame rate.

  The image display means 20 displays each frame constituting the image file or moving image file stored in the image data storage means 10 as a still image, and is realized by various display devices connected to a computer such as a liquid crystal display. Is done. The input instruction means 30 is used to give various input instructions to the arithmetic processing unit 40, and is realized by an input instruction device such as a keyboard and a mouse.

  The arithmetic processing unit 40 includes an object setting unit 41, an inter-object link setting unit 42, an object group extracting unit 43, and a branch point analyzing unit 44. The CPU, video memory, drawing accelerator and main memory of the computer It is realized by. The object setting unit 41, the inter-object link setting unit 42, the object group extracting unit 43, and the branch point analyzing unit 44 read a dedicated program into the main memory, and the CPU issues a predetermined command to the drawing accelerator as appropriate. It is realized by executing while writing predetermined data in the memory. The network analysis apparatus shown in FIG. 1 is actually realized by incorporating a dedicated program into a general-purpose computer.

(3. Processing operation)
Next, the processing operation of the network analysis apparatus shown in FIG. 1 will be described. When the system is activated, the object setting means 41 detects the object on the image data and sets an object marker on the detected object.

  For detection of the object in the image data, a known technique for detecting the object from the still image can be employed. There are various methods for specifying the detected object. Here, four methods of an object's x coordinate maximum value, x coordinate minimum value, y coordinate maximum value, and y coordinate minimum value on the image data are shown. The rectangular area specified by these four values is set as the object area. The identification of the object in the image data may not be detected by the computer. The object displayed on the image display means 20 by the object setting means 41 is confirmed with the naked eye, and the user inputs the input instruction means. 30 may be used to specify the object on the image data.

  Subsequently, the object setting unit 41 generates a contour line that clearly indicates the ranges in the X direction and the Y direction of the pixels constituting the object area detected based on the xy maximum and minimum values of the specified object. The contour line is displayed and output as an object marker superimposed on the microscope image.

  Next, details of the generation of the object marker will be described. First, the barycentric coordinates are calculated based on the average value of the XY coordinate values of the pixels constituting the object area. Then, a long axis is defined as a straight line direction where a straight line extending in two directions opposite to each other by 180 degrees from the barycentric coordinate and a pixel constituting the object region intersect at a position farthest from the barycentric coordinate. Further, the short axis is a straight line direction in which a straight line that is orthogonal to the long axis and extends in two directions opposite to each other by 180 degrees from the barycentric coordinate and a pixel that constitutes the object region intersects at a position farthest from the barycentric coordinate To do. Using the long axis and the short axis thus obtained, a rhombus composed of the long axis and the short axis is generated as an object marker.

  An example of the object marker generated by the object setting means 41 in this way is shown in FIG. As shown in FIG. 2A, in the initial state, the object marker is set in a state in which a contour line and a long axis are displayed. One end point of the long axis is the axis start point, and the other end point is the axis end point. The distance in the direction perpendicular to the major axis between the other two points not located on the major axis is defined as the width direction.

  The object setting means 41 assigns a marker ID for identifying each object marker to the generated object marker, and associates it with the coordinate values of the four vertices (edge points) of the rhombus shape to store image data. Register with means 10.

  Next, the inter-object link setting means 42 sets a link between the set object markers. From each object marker, it is possible to set a link to only one other object marker. There is no limit on the number of links set from other object markers. As a specific link setting rule, first, either the axis start point or the axis end point of the object marker A for which no link to another object marker is set is the axis of the other object marker B. When it is close to either the start point or the axis end point within a predetermined distance, a link from the object marker A to the object marker B is set.

  In addition, an object in which either the long axis start point or the axial end point of the object marker A for which a link to another object marker B has already been set does not set a link to another object marker Since the link from the object marker A to the object marker B remains unchanged when the axis C is close to either the axis start point or the axis end point of the marker C, the object marker A to the object marker C The link to cannot be set by the above-mentioned link setting rule. Therefore, conversely, a link from the object marker C to the object marker A is set.

  Here, FIG. 2B and FIG. 2C show examples of object markers that are not linked to other object markers and object markers that are not linked to other object markers, respectively. . As shown in FIG. 2 (b), the object marker that is not linked is one in which a line segment having a half length of the major axis is generated within the rhombus. Yes. In addition, as shown in FIG. 2 (c), the object marker that is linked is a line segment that connects the other two rhombus vertices that are not located on the long axis to the object marker that is not linked. Is added. After the link setting process by the inter-object link setting means 42, since two types of object markers as shown in FIGS. 2B and 2C are displayed on the photographed image, the link setting status becomes obvious at a glance. .

  The specific link setting is performed by the inter-object link setting unit 42 registering the marker ID of the link target object marker as the link destination marker ID in association with the marker ID for specifying the object marker. Is called.

  Usually, since there are a large number of objects in the captured image, a large number of object markers are set. FIG. 3 shows a state before and after the link setting between the object markers in a situation where there are many object markers. FIG. 3A shows a state before link setting, and FIG. 3B shows a state after link setting. In FIG. 3A, two points indicated by dotted circles indicate that either the axis start point or the axis end point of the object marker is within a predetermined distance from either the axis start point or the axis end point of another object marker. The adjacent parts are shown. As a result of the link setting process performed by the inter-object link setting means 42 on the object marker group shown in FIG. 3A, the object marker group is in a state as shown in FIG. Be changed. In FIG. 3A, the center object marker in which both the axis start point and the axis end point are close to each other is the other three objects as the axis end points in the vicinity of the right three object markers. A link is set from the object marker, and a link is set to another single object marker as an axis start point at a location close to the left object marker. In the left proximity point, the left end object marker is set as a link from the right object marker, so the left end is set as the axis start point and the right end is set as the axis end point.

  As described above, the object marker is set in the object region that is determined to be the object by executing the object detection process by the object setting unit 41, but is set from other object markers. Although there is no restriction on the number of links to be linked, it is possible to set a link from each object marker to only one other object marker. For this reason, even if the object markers are close to each other within a predetermined distance, there is a case where a link cannot be established as it is. In order to cope with such a situation, this system has a function of automatically correcting link settings such as link direction change and re-linkage.

  FIG. 4 is a diagram illustrating a case where it is necessary to automatically change the link setting. As shown in FIG. 4A, when the object marker D sets a link to the object marker F and the object marker E sets a link to the object marker G, the object marker D and the object Even if either the axis start point or the axis end point of the object marker E is close, no link is set between the object marker D and the object marker E. Even in such a case, when the object to which the object marker D to the object marker G is attached is confirmed on the photographed image, a link should be set between the object marker D and the object marker E. Exists.

  In this case, the inter-object link setting means 42 first checks whether any of the axis start point and the axis end point of the object marker F and the object marker G is close to each other. Instead of setting a link between the object marker D and the object marker E, a link from the object marker F to the object marker G or a link from the object marker G to the object marker F is set instead. If neither the axis start point nor the axis end point of the object marker F and the object marker G are close to each other, whether or not there is a link set by the object marker F with respect to another object marker When it is confirmed that the object does not exist, the inter-object link setting means 42 removes the link from the object marker D to the object marker F, and conversely, from the object marker F to the object marker D. Reset the link. As a result, the link relationship between the object markers is changed to a state as shown in FIG.

  At this stage, since there is no link that the object marker D has set with respect to other object markers, the inter-object link setting means 42 subsequently establishes a link from the object marker D to the object marker E. Perform the setting process. When neither the axis start point nor the axis end point is in contact like between the object marker D and the object marker E, the link setting means 42 sets the link between the object markers at the same time as setting the link. The object marker D and the object marker E are moved so that the axis end points touch each other. As a result, the link relationship between the object markers is changed to a state as shown in FIG.

  Also, when neither the axis start point nor the axis end point of the object marker F and the object marker G are close to each other, and there is a link set by the object marker F with respect to another object marker The link from the target marker E to the target marker G is confirmed by checking whether or not there is a link set by the target marker G with respect to another target marker. On the contrary, the link from the object marker G to the object marker E is reset. At this stage, since there is no link set by the object marker E with respect to other object markers, the inter-object link setting unit 42 sets a link from the object marker E to the object marker D. I do.

  Furthermore, when neither the axis start point nor the axis end point of the object marker F and the object marker G are close to each other, there is a link set by the object marker F with respect to another object marker, And when the link which the object marker G has set with respect to the other object marker exists, the inter-object link setting means 42 sets the link to the object marker F or the object marker G. For each target marker, check whether there is a link set for another target marker in the same way. If it is confirmed that there is no link, reverse the link direction. The correction process is performed in the same manner.

  When the link setting between the object markers by the inter-object link setting unit 42 is completed, the object group extracting unit 43 and the branch point analyzing unit 44 perform network topology analysis between the objects. FIG. 5 shows a rough flow of network topology analysis between objects. First, the object group extraction unit 43 extracts a group of objects connected by a link. Tree analysis is used to extract this group. Specifically, a series of object marker groups connected by links is set as a tree, and N object markers set in the captured image are classified into M trees (S100). This tree corresponds to a group of objects.

  Details of the tree analysis by the object group extraction means 43 are shown in FIG. In the tree analysis, when N object markers S (i) (i = 0, 1, 2,..., N−1) are linked to other markers S (j), L ( If i) = j and no link is set, L (i) =-1. Further, the number of trees to be obtained is M, and the tree ID given to the N object markers is T (i). In the state defined as described above, first, i = 0 and M = 0, and at j = 0, 1, 2,..., N−1, T (j) = 0 is initialized (S101). ).

  Next, it is determined whether or not the tree ID is 0 (S102). When the tree ID is 0, it means that no tree ID is set for the object marker, so the variable M indicating the number of trees is incremented by 1, and the variable M is set as the tree ID of the object marker. (S103). When the tree ID is not 0 in S102 and when M is set as the tree ID in S103, the tree ID of the target marker linked to the target marker is 0, that is, T (L (i)) = 0. Is determined (S104). Since the tree to which a certain object marker belongs and the tree to which the object marker linked to that object marker belongs are the same, it is necessary to set the tree IDs of both to be the same. Therefore, when the tree ID of the target marker at the link destination is 0, T (L (i)) that is the tree ID is set to the same value as T (i) that is the tree ID of the link source (S105). ).

  In S104, when it is determined that the tree ID of the target marker linked to the target marker is not 0, the tree ID of the target marker linked to is already set. Even in this case, the tree to which a certain object marker belongs and the tree to which the object marker to which the object marker is linked belong are the same, and therefore it is necessary to set both tree IDs to be the same. Therefore, in S106 to S109, processing for unifying the young tree ID is performed.

  First, it is determined which is larger, T (L (i)), which is the tree ID of the link target object marker, or T (i), which is the tree ID of the link target object marker (S106). In S107 and S108, as a result of the determination in S106, the smaller one is set to T1, and the larger one is set to T2. Specifically, if T (L (i)) is larger, T (i) is set to T1, T (L (i)) is set to T2 (S107), and T (i) is more If larger, T (L (i)) is set to T1, and T (i) is set to T2 (S108).

  If T (j), which is the tree ID of all object markers, is T2, the value is changed to T1 (S109). This is a process for unifying all the tree IDs of the object markers having the same tree ID as the larger one among the tree IDs of the link source and link target object markers. Further, in S109, if the value of T (j), which is the tree ID of all the object markers, is larger than T2, the value of the tree ID is decreased by 1. This is a process in which T2 is missing because the tree ID having the value of T2 is changed to T1, and the subsequent tree ID values are reduced by one. When the process in S109 is completed, the variable M indicating the number of trees is reduced by 1 (S110).

  After unifying the link source and link destination tree IDs by the process in S105 or the processes in S106 to S110, the variable i is incremented by 1 to process the next object marker (S111). Then, it is determined whether i has reached N (S112). If i is less than the total number of markers N, the process returns to S102 and the process is repeated. If i reaches the total number of markers N, the processing has been performed for all the target markers, so the process ends. To do. As a result of the processing according to the flowchart of FIG. 6 performed by the object group extraction unit 43, each object marker is assigned a tree ID, and object markers having the same tree ID belong to the same tree. Classified into groups.

  When the tree analysis by the object group extraction unit 43 is completed, the branch point analysis unit 44 performs a branch point analysis and a geometric shape analysis for each tree. Specifically, as shown in FIG. 5, for each tree, the number of object markers linked by two or more object markers is analyzed (S200), and the objects to be configured for each tree. The total value of the length and area of the object marker is calculated and output, and the maximum value of the width of the target object marker is output (S300).

  The branch point analysis means 44 performs initial setting for the analysis. Specifically, for N object markers, the number of links set by other object markers (this is referred to as “the number of branches”) is B (i) (i = 0, 1, 2, ..., N-1). In addition, the initial state is set to B (j) = 0 at j = 0, 1, 2,..., N−1, and then L ( i) In the case of ≧ 0, the operation of B (L (i)) = B (L (i)) + 1 is sequentially executed. Thereby, the branch number B (L (i)) of the object marker L (i) at each link destination is calculated.

  Details of the branch point analysis and geometrical shape analysis of each tree by the branch point analysis means 44 are shown in FIG. For each of the N object markers, the length of the object marker is D (i), the area of the object marker is A (i), and the width of the object marker is W (i) (however, i = 0, 1, 2,..., N−1). For each tree, the branch point analyzing means 44 first calculates the total tree branch number Bt (j), the tree length Dt (j), the tree area At (j), and the maximum tree width Wt (j). After the setting, i = 0, and j = 0,..., M−1, Bt (j) = Dt (j) = At (j) = Wt (j) = 0 is performed (S201). ).

  Subsequently, for each tree, the branch point analyzing means 44, for each tree, the total number of branches Bt (T (i)), the length Dt (T (i)) of the tree, the area At (T (i)) of the tree, Processing for obtaining the maximum tree width Wt (T (i)) is performed (S301). Specifically, if B (i)> 1, it means that the target marker S (i) is a link from a plurality of target markers and is a branch point. The total branch number Bt (T (i)) is increased by 1. Further, the length D (i) of the object marker S (i) is added to the length Dt (T (i)) of the tree. Further, the area A (i) of the object marker S (i) is added to the area At (T (i)) of the tree. Further, if B (i)> 1 and the width W (i) of the object marker S (i) satisfies W (i)> Wt (T (i)), the maximum width of the tree A process of replacing Wt (j) with the width W (i) of the object marker S (i) is performed.

  In S301, if W (i) ≦ Wt (T (i)), there is no change in the maximum width of the tree, so no processing is performed on Wt (T (i)). Further, if B (i) ≦ 1, it means that the object marker S (i) is not branched, so that Bt (j), Dt (j), area At (j) in S301, The update process for Wt (j) is not performed.

  When the process of S301 for the object marker S (i) is completed, the value of i is incremented by 1, and the process of S301 is executed for the next object marker S (i). In this way, the process of S301 is executed for all N object markers S (i), and when the process of S301 is completed for N object markers S (i), branch point analysis for each tree, The geometric shape analysis process ends.

  In the processing in S301, the total number of branches Bt (T (i)), the tree length Dt (T (i)), the tree area At (T (i)), and the maximum tree width Wt (T Although the processing for obtaining (i)) has been performed, an average value of the width W (i) of each object marker S (i) may be further obtained. In this case, it is not necessary to impose the condition of W (i)> Wt (T (i)).

  These parameters (total number of branches in tree, length, area, width) are very useful for quantifying vascular networks. Particularly in the field of regenerative medicine, an important issue of how to efficiently construct a blood vessel network in a pseudo three-dimensional tissue in a series of flows of constructing a pseudo three-dimensional tissue outside the living body and returning it to the living body. There is a lot of research. At that time, it is necessary to analyze how the blood vessel pattern arbitrarily designed in advance shows in the pseudo three-dimensional tissue. In addition, when the line patterns of the vascular tissue are close to each other to some extent, the patterns tend to be bridged by stimulating each other. When designing a pattern arbitrarily, in order to maintain the pattern, the line pattern is designed at a certain interval, so that the bridge hardly occurs. In order to maintain an arbitrarily designed pattern in the three-dimensional tissue, it is necessary to optimize the design. In such a case, research is also conducted in which the number of branches of the blood vessel network is evaluated as an index. From the above, these parameters are considered to be very useful information when a dynamic tissue such as a vascular system is constructed in vitro and its behavior is analyzed.

  As described above, as a result of processing performed by the network analysis device, detailed information of the tree (group) to which each object marker belongs can be obtained. Since this tree shows the connection relationship of each object marker, it can be estimated as the connection relationship between the objects on which each object marker is set. For this reason, as a result, it becomes possible to analyze the connection relation between objects.

  The preferred embodiments of the present invention have been described above. However, the present invention is not limited to the above embodiments, and various modifications can be made. For example, in the above-described embodiment, image data based on a microscopic image is used as a processing target, and the connection relationship is analyzed using cells as a target. -You may make it analyze networks, such as a road network and a track network, using the image data which imaged the track.

DESCRIPTION OF SYMBOLS 10 ... Image data storage means 20 ... Image display means 30 ... Input instruction means 40 ... Operation processing part 41 ... Object setting means 42 ... Inter-object link setting means 43 ... -Object group extraction means 44 ... Branch point analysis means

Claims (8)

An object setting means for selecting a plurality of objects from the image data and setting an object marker on the image data to clearly indicate the position of the object;
For the set object marker, a link is made to another object marker located in the vicinity on the same image data, and an identification mark that clearly indicates that the other object marker is linked An inter-object link setting means to be added to the object marker;
Object group extraction means for extracting one or a plurality of object groups connected by links set in the plurality of objects;
A branch point analyzing means for extracting, as a branch point, a target that is linked from a plurality of targets in the target configured for each target group;
A network analysis device between objects characterized by comprising: In claim 1,
The inter-object link setting means is configured such that a predetermined edge point of the object marker A for which a link to another object marker is not set is within a predetermined distance from a predetermined edge point of the other object marker B. A network analysis apparatus between objects, wherein a link from the object marker A to the object marker B is set when the objects are close to each other. In claim 2,
The inter-object link setting means is characterized in that each object marker receives a link from a plurality of other object markers, but only one link to the other object marker can be set. Network analysis device between objects. In claim 3,
The inter-object link setting means is configured such that a predetermined edge point of the object marker A for which a link to another object marker B has already been set does not set a link to another object marker. A network analysis apparatus between objects, wherein a link from the object marker C to the object marker A is set when the marker C is close to a predetermined edge point within a predetermined distance. In claim 3,
In the inter-object link setting means, a predetermined edge point of the object marker D that has already set a link to another object marker F has already set a link to another object marker G. When the predetermined edge point of the object marker E is close to the predetermined edge point within a predetermined distance, and the predetermined edge point of the object marker F is separated from the predetermined edge point of the object marker G by a predetermined distance or more, The link from the object marker D to the object marker F is canceled, the link from the object marker F to the object marker D is set, and the link from the object marker D to the object marker E is set. Network analysis device between objects. In any one of Claims 2-5,
The network analysis apparatus between objects, wherein the object marker has a rhombus shape, and two vertices in a major axis direction are the predetermined edge points. In any one of Claims 1-6,
The branching point analyzing means is based on a geometric shape parameter consisting of the length, width, and area of each object marker configured for each object group, and the total length value, the maximum width value, the width A network analysis device between objects characterized by calculating an average value and a total value of areas.   The program for functioning a computer as a network analysis apparatus in any one of Claims 1-7.

Images