JP2009063508A - Cell image analyzer and cell image analysis software - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a cell image analyzer and a cell image analysis software capable of automating complicated cell classification to some degree, while maximumly saving trouble or labor of an experimenter, and reflecting a desire of a researcher who is a user so that automatic analysis can be progressed with flexibility in some degree. <P>SOLUTION: The analyzer has a function for performing a prescribed processing to an image acquired from an image acquisition means for acquiring a light image emitted from cells as the image, and automatically extracting a cell group having a similar characteristic to specific cells which are interesting to the user among images including a plurality of cells. The device also has a function enabling the user to designate an optional cell as an interesting cell. Preferably, the device has a function for automatically setting a parameter group comprising characteristic quantity parameters by which a cell group having a similar characteristic to specific cells designated by the user can be distinguished from other cell groups, and automatically setting a threshold of each characteristic quantity parameter included in the parameter group. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

The present invention relates to an apparatus and software for analyzing a cell image, and more specifically, in an experiment for analyzing a cell including cells of various cell types and various states, a specific one of these various cell types and cell states is specified. The present invention relates to a cell image analysis apparatus and a cell image analysis software having a function of automatically extracting a state cell (group) from an image and classifying the cell.
Cell image analyzers are roughly classified into two types. One is a so-called high-throughput fully automatic analyzer that presupposes a large amount of predetermined processing such as drug screening, and the other is a semi-automatic analyzer system for research support applications. It is.
However, the cell image analysis apparatus and the cell image analysis software of the present invention relate to a semi-automatic analysis system for the purpose of research support.

Conventionally, this type of cell image analyzer automatically acquires a certain number of cell images, and extracts the cells with features that the researcher is interested in (interests) from the acquired cell images. It is possible to provide data including a graph of data obtained by quantifying the number of these cells and characteristic quantities. As such a cell image analysis apparatus, for example, an apparatus described in the following Patent Document 1 has been proposed.
JP 2001-307066 A

  By the way, the cells to which the researchers pay attention generally differ depending on the experimental (research) system. For example, when studying the process of cell death, it is necessary to distinguish between healthy cells and dead cells. Further, for example, in an experimental system in which a state in which a plurality of types of cells are cultured is observed and the change is quantified, it is necessary to classify the cells for each type of cell.

  Since the classification of each type of cell is relatively complicated, it is usually performed based on visual observation. However, since classification based on visual observation takes time and effort, it is desirable to be able to automatically perform classification for each cell type via the cell image analyzer.

However, with a simple experimental system, it is possible to relatively easily automate the classification for each cell type via the cell image analyzer.
For example, in order to evaluate the effectiveness of a drug that has the effect of killing cells, in an experimental system in which the viability of cells is determined and the ratio of the number of cells is detected, only dead cells are stained specifically. A life-and-death determination is made using a fluorescent substance. In this case, since the criteria for classifying the cells, such as cells that emit fluorescence and cells that do not emit fluorescence, are relatively simple, classification by cell type can be performed automatically via a cell image analyzer. .

  However, in general research applications, an experimental system that can perform quantitative evaluation by fully automatically classifying each cell type via a cell image analyzer using such a relatively easy criterion. Is a small number, and in most experimental systems, it is difficult to extract features to be used as criteria for cell classification, or it is difficult to predetermine features to be used as criteria for cell classification Is most.

  The present invention has been made in view of the above-described conventional problems, and can perform complicated cell classification to some extent automatically while saving labor and labor of an experimenter as much as possible, and with a certain degree of freedom. An object of the present invention is to provide a cell image analysis device and a cell image analysis software capable of reflecting a researcher who is a user so that automatic analysis can proceed.

  In order to achieve the above object, a cell image analysis apparatus according to the present invention includes a plurality of cells by performing predetermined processing on an image obtained from an image obtaining unit that obtains an image of light emitted from a cell as an image. A device having a function of automatically extracting a cell group having characteristics similar to a specific cell of interest to a user from an image, and having a function of allowing a user to designate an arbitrary cell as the cell of interest It is characterized by that.

  In the cell image analysis apparatus of the present invention, a parameter group consisting of feature parameters that automatically distinguish a cell group having characteristics similar to the specific cell specified by the user from other cell groups is automatically set. In addition, it is preferable to provide a function for automatically setting the threshold value of each feature parameter included in the parameter group.

  In the cell image analysis apparatus of the present invention, a predetermined number of cells are automatically listed, and each listed cell is included in a cell group having characteristics similar to the specific cell designated by the user. It is preferable to provide an automatic response function that asks the user whether or not, and automatically changes the threshold value of each feature parameter included in the parameter group according to the answer.

  The cell image analysis apparatus of the present invention preferably has a function of providing the parameter group to the user or providing the user with a cell distribution for each feature parameter included in the parameter group.

  Further, in the cell image analysis apparatus of the present invention, the ratio of the cell group having a feature similar to the specific cell designated by the user, the average value of each feature parameter included in the parameter group, and each feature value It is preferable to provide a function of providing at least one of comparison results with other cell groups regarding parameters.

  In addition, the cell image analysis software according to the present invention performs a predetermined process on an image obtained from an image acquisition unit that acquires an image of light emitted from a cell as an image, and from among images including a plurality of cells, Cell image analysis software used in a cell image analysis device having a function of automatically extracting a cell group having characteristics similar to a specific cell of interest to the user, and the user designates any cell as the cell of interest It features a function that can be.

  Further, in the cell image analysis software according to the present invention, a parameter group composed of feature parameters that automatically distinguish a cell group having characteristics similar to the specific cell designated by the user from other cell groups is automatically set. In addition, it is preferable to provide a function for automatically setting the threshold value of each feature parameter included in the parameter group.

  In the cell image analysis software according to the present invention, a predetermined number of cells are automatically listed, and each listed cell is included in a cell group having characteristics similar to the specific cell designated by the user. It is preferable to provide an automatic response function that asks the user whether or not, and automatically changes the threshold value of each feature parameter included in the parameter group according to the answer.

  The cell image analysis software according to the present invention preferably has a function of providing the user with the parameter group, or providing the user with a distribution of cells for each feature parameter included in the parameter group.

  In the cell image analysis software according to the present invention, the ratio of the cell group having characteristics similar to the specific cell designated by the user, the average value of each feature parameter included in the parameter group, and each feature value It is preferable to provide a function of providing at least one of comparison results with other cell groups regarding parameters.

  According to the present invention, it is possible for a user to perform a complicated cell classification to some extent automatically while saving labor and labor of an experimenter as much as possible, and to proceed with an automatic analysis with a certain degree of freedom. A cell image analysis apparatus and cell image automatic analysis software that can reflect the demands of a person and can easily and accurately classify and analyze cells.

First Embodiment FIG. 1 is a block diagram showing an overall schematic configuration of a cell image analyzing apparatus according to a first embodiment of the present invention, and FIG. 2 is a user interface via a personal computer in the cell image analyzing apparatus of the first embodiment. FIG. 3 is a flowchart showing a processing procedure in the cell image analyzer of the first embodiment, and FIG. 4 is a cell acquired in the cell image analyzer of the first embodiment. It is explanatory drawing which shows an example of the function which classify | categorizes a cell based on the magnitude | size of a cell from an image, and displays it on a display screen, Comprising: (a) is the cell created using the information obtained from the several cell image FIG. 7B is an explanatory diagram showing an example of a cell image corresponding to the graph of FIG. FIG. 5 is an explanatory diagram showing an example of a function for designating a cell of interest from the cell image acquired in the cell image analyzer of the first embodiment and displaying it on the display screen. When the cell of interest is specified, the specified cell is cut out and displayed, (b) is a display example that displays the cell type that includes the specified cell, and (c) is the cell type that includes the specified cell. The other example of a display which displays is shown. FIG. 6 is an explanatory view showing a display example of correspondence between cells in a cell image acquired by the cell image analyzer of the first embodiment and positions on the graph, and (a) shows a frequency with respect to the size of the cell. The explanatory view showing the state where the position of the feature amount of the graph showing the distribution of the cell and the cell in the cell image corresponding thereto are displayed by arrows, (b) designates a predetermined position on the graph shown in (a) or cell An explanatory diagram showing a state in which a cell in a corresponding cell image or a position on a graph is displayed when a predetermined cell is designated on the image, and (c) is classified on the cell image shown in (a). It is explanatory drawing which shows the state which displayed the position on the graph corresponding to when the cell which is not suitable is designated.

As shown in FIG. 1, the cell image analysis apparatus according to the first embodiment includes an illumination optical system 1, an imaging optical system 2, an electric stage 3 for placing a sample and carrying it to an observation position, and an imaging device such as a CCD camera. 4 and a personal computer (hereinafter, “PC”) 5 connected to the microscope.
The microscope is configured such that an image of light emitted from a cell can be acquired as an image via the CCD camera 4.
The personal computer 5 performs a predetermined process on the image obtained through the CCD camera 4 of the microscope, and from among the image including a plurality of cells, a group of cells having characteristics similar to a specific cell of interest to the user Has a function of automatically extracting.
In addition, the cell image analysis apparatus according to the first embodiment includes a user interface by the personal computer 5. Specifically, as shown in FIG. 2, the function of displaying the cell image captured on the display screen of the monitor 5a, the number of cells in the cell image, and the like, for example, the change in the distribution of the cell number for each cell image. Software having an image display function and a graph display function, such as a graph display function, and an input function for instructing operation control of the microscope apparatus via the input means 5b such as a keyboard and a mouse shown in FIG. Have.

In the cell image analysis apparatus according to the first embodiment, as shown in FIG. 3, imaging of a cell image (step S1), cell image analysis: quantification (step S2), cell image analysis: classification of cells and the like (steps) Image analysis processing is configured to be performed in the order of S3) and analysis result display (step S4).
Here, the outline of the process at each stage of the image analysis process will be briefly described.

In the cell image imaging process (step S1), imaging is automatically performed according to a predetermined procedure. Specifically, the operation control of the microscope is set in advance through software for instructing the operation control of the microscope apparatus. Then, in accordance with preset operation control, for example, the position of the stage 3 is moved to a predetermined position, the focus position of the imaging optical system 2 is automatically adjusted, imaged by the CCD camera 4, and stored as image data in a personal computer. A plurality of cell images are acquired by repeatedly performing the procedure.

Cell image analysis: quantification Cell image analysis: In quantification treatment (step S2), cell analysis is automatically performed to some extent. Specifically, the area of cells in the image is specified mainly through software for image processing and image analysis, or the cell size, brightness, roundness, and brightness of different fluorescent dyes. Multiple features such as ratios are automatically calculated and quantified.
These feature values are processed statistically. In other words, the feature values obtained for each cell in each image in a plurality of images are averaged and normalized.

Cell image analysis: Classification Cell image analysis: In classification processing (step S3), cells are analyzed for each predetermined feature using data averaged and normalized through cell image analysis: quantification processing (step S2). Classify.
For example, blood cells and cells released from cancer tissue are quite different in size and shape. Therefore, the cells are classified according to these characteristics (size, shape). In this example, the former cells are often small and round, and the latter cells are often large and distorted. For example, when the cell size distribution is shown in a graph, two characteristic peaks appear as shown in FIG. Therefore, the cells can be classified for each group of cells distributed in the vicinity of these two characteristic peaks.

  Moreover, in the cell image analyzer of the first embodiment, further analysis is performed depending on the experimental system. For example, when only cells stained with substances having different fluorescence characteristics are selected, the cells can be classified according to the amount of fluorescence brightness in each cell region.

Analysis result display In the analysis result display process (step S4), according to the classification in the cell image analysis: classification process (step S3), for example, the ratio of the cells, the fluorescence emitted to some extent or not, is not so. The ratio to the object is displayed on the display screen in a display form such as a graph or a numerical value.

  Further, in the cell image analysis apparatus of the first embodiment, the same image analysis processing as described above is performed, for example, by changing the culture conditions or using different samples and comparing the analysis results. Can also be done.

Next, a detailed configuration of the user interface in the analysis system of the cell image analysis apparatus of the first embodiment will be described.
In the cell image analysis apparatus of the first embodiment, as a function that allows the user to designate an arbitrary cell as a cell of interest, for example, an interface for designating a cell in an image, cutting out these cell regions, and displaying them is displayed. It is provided as software.

For example, as shown in FIG. 5, a cell in which the user is interested is selected from the cell images displayed on the display screen via the cell image analysis software, and designated by an operation such as a screen click.
FIG. 5 (a) shows an example in which a larger one of two types of cells (large and small) included in the cell image is designated.

  By this operation, for example, similar cells can be selected from the cell image and displayed on the image or displayed as a graph with respect to the cell designated by the user.

  FIG. 5B shows an example in which a predetermined type of cells including selected cells are displayed by a display method such as changing the display color of only large cells designated as cells of interest on the cell image. Yes.

  As another display method, for example, as shown in FIG. 5C, the extracted cells can be cut out on another image panel and displayed as a list.

  However, simply having the function to specify the desired cell by clicking on the cell image, the cell specified by clicking on the screen is specified to classify the cell when it does not have characteristics suitable for classification. There is a risk that it will become unsuitable for the user's purpose of designating cells.

Therefore, in the cell image analysis apparatus of the present invention, for example, the following functions are added in order to make it easy to specify cells suitable for cell classification.
As one of the functions, in the cell image analysis apparatus of the first embodiment, the designation of the cell on the cell image is linked to the designation of the cell on the graph.
For example, as shown in FIGS. 6 (a) and 6 (b), when an arbitrary cell in the cell image displayed on the display screen is clicked, the size of a predetermined feature amount related to the cell is displayed on the graph distribution. Displayed as a position. Accordingly, it is possible to determine what kind of characteristic the arbitrary clicked cell has compared to other cells and how well the characteristic is displayed while looking at the overall distribution. As a result, for example, as shown in FIGS. 6 (a) and 6 (b), in the case of a cell image having a distribution having two peaks, a cell located at an intermediate position between the two peaks is shown in FIG. As can be seen from the graph of), the frequency is low and it is not suitable as a standard for cell classification.

  Also, for example, by specifying one point on the cell distribution graph, it is possible to specify a characteristic cell. Thereby, for example, a cell close to the center position of the peak in the graph shown in FIG. 6 can be selected, and more accurate cell classification becomes possible.

  In addition, when selecting a cell on the graph, the cell corresponding to (approximating) the selected designated cell is the one closest to the value of the predetermined feature value of the selected cell on the currently displayed cell image. It is also possible to choose.

  Cell distribution usually spans multiple images. For this reason, for example, the cell image analysis apparatus of the first embodiment extracts only the position of the designated cell included in the graph from the points on the distribution displayed in the graph, and displays the specified cell with blinking or different colors. It also has a function of selecting a cell having a feature value closest to the designated position in the image (in the currently displayed image).

  According to the cell image analysis apparatus of the first embodiment configured as described above, the feature of the cell is designated by a plurality of methods and referring to a plurality of results such as the cell image and the distribution of the feature amount. be able to.

Second Embodiment FIG. 7 is an explanatory view of the feature parameter set by the cell image analyzer of the second embodiment of the present invention, where (a) is the cell size, (b) is the roundness of the cell, (c ) Shows the luminance of the cell, and (d) shows an example of a cell image made up of an image of each cell having a predetermined feature parameter among the feature parameters of (a) to (c). FIG. 8 is an explanatory diagram showing a combination of feature parameter and an example of cell classification corresponding to the combination. (A) is an example of cell classification by appropriate feature parameter combination, and (b) is an inappropriate feature. Examples of cell classification based on combinations of quantity parameters are shown.
In addition to the configuration of the cell image analysis device of the first embodiment, the cell image analysis device of the second embodiment can distinguish a cell group having characteristics similar to a specific cell designated by the user from other cell groups. And a function for automatically setting a parameter group of feature quantity parameters, and automatically setting a threshold value of each feature quantity parameter included in these parameter groups.

  The classification of the cells is obtained by comparing the entire distribution for each of the plurality of feature value values. For example, in an experimental system for classifying cell types, the feature parameter 1 is the average cell size, the feature parameter 2 is the average cell roundness, and the feature parameter 3 is the average cell brightness. . For example, as shown in FIG. 7, among these three feature parameters, the cell size and the roundness of the cell are highly dependent on the cell type, and the distribution varies greatly depending on the cell type. On the other hand, regarding the brightness of the cells, it is assumed that there is no significant feature in the distribution depending on the cell type. In this case, the feature parameter 3 is not a feature parameter suitable for cell classification.

The following method can be considered as a method for determining the effect on the classification of each feature parameter.
The feature parameter is p1, p2, p3,..., And the specified feature value of the cell is ap1, ap2, ap3,. Here, an upper limit value and a lower limit value each having an appropriate width (for example, a value including 5% of cells centered on the value) are set as threshold values for each feature value, and a set of upper limit values of the feature values is set. Assume that a group of cells satisfying alp1>axp1> amp1 is obtained when a set of lower limit values of amp1, amp2, amp3,..., alp1, alp2, alp3,.
The cell group satisfying alp1>axp1> amp1 means cells in the upper and lower 5% each having a value close to the feature value of the designated cell with respect to the feature parameter 1 among the cells included in the entire cell image. ing. Here, it is considered that the upper and lower 5% cell groups having values close to the value of the feature quantity of the designated cell are often different for each of the feature quantity parameters 1, 2, and 3. If the cells can be effectively classified by a combination of some predetermined feature parameters, the group of cell groups classified based on the some predetermined feature parameters should match. That is, among the cell feature parameter p1, p2, p3,..., A combination of feature parameters that are effective for cell classification is, for example, a cell group of upper and lower 5% each having a value close to a designated cell. Can be defined as a parameter group with many overlaps (see FIG. 8).

  For example, when the combination of the feature parameter 1 and the feature parameter 2 is appropriate, as shown in FIG. 8A, the group of cell groups classified by the feature parameter 1 and the feature parameter 2 The group of cell groups classified by (1) substantially coincides with the distribution of cell groups. On the other hand, when the combination of the feature parameter 1 and the feature parameter 2 is inappropriate, as shown in FIG. 8 (b), a group of cell groups classified by the feature parameter 1 and the feature parameter The group of cell groups classified by 2 does not match, and the distribution of the cell groups is not clearly separated.

The threshold value is 5% in the upper and lower directions in the above example. However, in the cell image analysis apparatus of the second embodiment, for example, the parameter group determined as described above and the cell determined for the parameter group are used. With respect to the group, the cell group can be automatically determined by a method such as taking a value where the degree of overlap of the cell group is the highest.
According to the cell image analysis apparatus of the second embodiment configured as described above, it becomes easy to designate a cell suitable for classification.

Third Embodiment FIG. 9 is an explanatory diagram showing a procedure by a feedback function for cell designation provided in the cell image analysis apparatus according to the third embodiment of the present invention. (A) is an approximation based on the cell designated by the user. (B) is a step in which a user designates a cell that does not approximate a specific cell from a plurality of cells displayed on the screen, and (c) is a specific cell by the user. The steps of changing the threshold value of the feature parameter based on the designation of the cells that are not approximate to are shown.

  In addition to the configuration of the cell image analyzer of the second embodiment, the cell image analyzer of the third embodiment automatically lists several cells in order to narrow down the characteristics of the cells specified by the user. Is included in the cell group of interest, and an automatic response function is provided to automatically change the threshold value of each feature parameter included in the parameter group according to the answer.

The cell classification can be determined without an automatic response function. However, more accurate classification can be performed if judgment and feedback are performed by the user.
Therefore, the cell image analysis apparatus according to the third embodiment is characterized in that a cell group having characteristics similar to a specific cell specified by the user in the cell image analysis apparatus according to the second embodiment can be distinguished from other cell groups. A parameter group consisting of parameters is automatically set, and the cell classification automatically extracted through the function of automatically setting the threshold value of each feature parameter included in the parameter group, for example, some cells If the user responds that the classification is wrong, the threshold value of each feature parameter is changed to make the cell classification more accurate. Is provided (see FIG. 9).
Specifically, as shown in FIG. 9 (a), first, cells similar to a specific cell designated by the user are automatically listed and displayed on the screen.
Next, as shown in FIG. 9B, the user designates a cell that does not approximate the specified cell from the cells displayed on the screen as a cell inappropriate for classification.
When the user designates an inappropriate cell, as shown in FIG. 9C, the feature value parameter threshold value is changed based on the designation.
Thereafter, the same process is repeated until the user no longer designates an inappropriate cell, whereby a more accurate feature parameter value is set.

Fourth Embodiment A cell image analysis apparatus according to the fourth embodiment includes feature amount parameters obtained via the cell image analysis apparatus according to the second embodiment in addition to the configuration of the cell image analysis apparatus according to the second embodiment. A parameter group is provided to the user as the best feature group, or a cell distribution for each feature parameter included in the parameter group is provided to the user. In this way, the user can easily make a determination.

As with the cell image analysis apparatus of the fourth embodiment, if the function of displaying the result of these classifications as well as the image, and displaying the distribution of cells with respect to the parameter group and each feature amount is provided to the user, it is accurate. Classification can be performed.
For example, the distribution as shown in FIG. 8 is displayed on the display screen. For this reason, the user can easily determine whether or not the parameter group set by designating the cell is appropriate.

Fifth Embodiment A cell image analysis device according to the fifth embodiment is a cell having characteristics similar to a specific cell designated by the user in addition to the configuration of the cell image analysis device according to any one of the first to fourth embodiments. It has a function of providing at least one of a group ratio, an average value of each feature parameter included in the parameter group, and a result of comparison with other cell groups regarding each feature parameter.

  These results can be used as experimental results and can also be used by the user to determine the accuracy of the cell classification.

  As described above, according to the cell image analysis devices of the first to fifth embodiments, a user who uses the cell image analysis device can easily and accurately classify and analyze cells. .

  The present invention relates to a cell (group) of a specific state among these various cell types and cell states in an experiment for analyzing cells containing cells of various cell types and various states for the purpose of supporting research. It is useful in the medical, medical, and biological fields where cells are automatically extracted from images and classified.

It is a block diagram which shows the schematic structure of the whole cell image analyzer concerning 1st embodiment of this invention. It is explanatory drawing which shows an example of the information displayed on a display screen as a user interface via the personal computer in the cell image analyzer of 1st embodiment. It is a flowchart which shows the process sequence in the cell image analyzer of 1st embodiment. FIG. 6 is an explanatory diagram illustrating an example of a function for classifying cells based on cell size from a cell image acquired in the cell image analysis apparatus of the first embodiment and displaying the cells on a display screen. FIG. 5B is an explanatory diagram showing an example of a cell image corresponding to the graph of (a), which is an example of a frequency distribution with respect to the cell size created using information obtained from the image. It is explanatory drawing which shows an example of the function which designates a cell of interest from the cell image acquired in the cell image analyzer of 1st embodiment, and displays it on a display screen, (a) is a cell of interest on a cell image (B) is a display example that displays the cell type that includes the specified cell, and (c) displays the cell type that includes the specified cell. Other display examples are shown. It is an explanatory view showing a display example of the correspondence between the cell in the cell image acquired in the cell image analysis apparatus of the first embodiment and the position on the graph, (a) is a distribution of the frequency with respect to the size of the cell The explanatory view showing the state where the position of the characteristic amount of the graph shown and the cell in the cell image corresponding thereto is displayed with an arrow, (b) is a designated position on the graph shown in (a) or on the cell image Explanatory diagram showing the state in which the cell or the position on the graph is displayed in the cell image corresponding to the specified cell, (c) is not suitable for classification on the cell image shown in (a) It is explanatory drawing which shows the state which displayed the position on the graph corresponding to it when a cell is designated. In the explanatory diagram of the feature parameter set in the cell image analysis apparatus of the second embodiment of the present invention, (a) is the size of the cell, (b) is the roundness of the cell, (c) is the brightness of the cell, ( d) shows an example of a cell image made up of an image of a cell having a predetermined feature parameter among the feature parameters of (a) to (c). It is explanatory drawing which shows the combination example of the feature parameter and the cell classification example corresponding to the combination, (a) is the cell classification example by the combination of the appropriate feature parameter, (b) is the inappropriate feature parameter An example of cell classification by combination is shown. It is explanatory drawing which shows the procedure by the feedback function about the designation | designated of the cell with which the cell image analyzer concerning 3rd embodiment of this invention is equipped, (a) is a list which automatically approximates the cell based on the cell designated by the user (B) is a step in which the user designates a cell that does not approximate a specific cell from among a plurality of cells displayed on the screen, and (c) is a step in which a user designates a cell that does not approximate a specific cell. The steps of changing the threshold value of the feature parameter based on each are shown.

Explanation of symbols

1 Illumination optical system 2 Imaging optical system 3 Stage 3
4 CCD camera (image sensor)
5 Personal computer 5a Monitor 5b Keyboard

Claims (10)

A feature similar to a specific cell that is of interest to the user from among images containing a plurality of cells, by performing predetermined processing on the image obtained from the image acquisition means that acquires an image of light emitted from the cell as an image A device having a function of automatically extracting a cell group having
A cell image analysis apparatus provided with a function that allows a user to designate an arbitrary cell as the cell of interest.   The cell group having features similar to the specific cell designated by the user automatically sets a parameter group consisting of feature amount parameters that can be distinguished from other cell groups, and each feature included in the parameter group The cell image analysis apparatus according to claim 1, further comprising a function of automatically setting a threshold value of the quantity parameter.   A predetermined number of cells are automatically listed, and the user is asked whether the listed cells are included in a cell group having characteristics similar to the specific cells designated by the user, and the user responds to the answer. The cell image analysis apparatus according to claim 2, further comprising an automatic response function that automatically changes a dark value of each feature parameter included in the parameter group.   The cell image analysis apparatus according to claim 2, further comprising a function of providing the parameter group to a user or providing a user with a cell distribution for each feature parameter included in the parameter group.   At least a ratio of a cell group having characteristics similar to the specific cell designated by the user, an average value of each feature parameter included in the parameter group, and a comparison result with other cell groups regarding each feature parameter The cell image analysis apparatus according to claim 2, which has a function of providing any one of them. A feature similar to a specific cell that is of interest to the user from among images containing a plurality of cells, by performing predetermined processing on the image obtained from the image acquisition means that acquires an image of light emitted from the cell as an image Cell image analysis software used in a cell image analysis device having a function of automatically extracting a cell group having
A cell image analysis software comprising a function that allows a user to designate an arbitrary cell as the cell of interest.   The cell group having features similar to the specific cell designated by the user automatically sets a parameter group consisting of feature amount parameters that can be distinguished from other cell groups, and each feature included in the parameter group The cell image analysis software according to claim 6, further comprising a function of automatically setting a threshold value of the quantity parameter.   A predetermined number of cells are automatically listed, and the user is asked whether the listed cells are included in a cell group having characteristics similar to the specific cells designated by the user, and the user responds to the answer. The cell image analysis software according to claim 7, further comprising an automatic response function that automatically changes a dark value of each feature parameter included in the parameter group.   The cell image analysis software according to claim 7, further comprising a function of providing the parameter group to a user or providing a user with a cell distribution for each feature parameter included in the parameter group.   At least a ratio of a cell group having characteristics similar to the specific cell designated by the user, an average value of each feature parameter included in the parameter group, and a comparison result with other cell groups regarding each feature parameter The cell image analysis software according to any one of claims 7 to 9, further comprising a function of providing any one of them.

Images