JP2003107081A - Image analyzing method, analyzer, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image analyzing method, an analyzer, and a recording medium for avoiding the overlapping between objects to be analyzed which can happen in a specimen (for example, the coupling between organism cells) on images and correctly and highly accurately performing image analysis on the objects to be analyzed (for example, nuclear cells). SOLUTION: The image analyzer for performing image analysis on the objects to be analyzed on the basis of the images of the objects to be analyzed obtained by a microscope is provided with the microscope (A) for observing the plurality of objects to be analyzed arranged in a microregion within a magnified field of view, an image input means (10) for capturing the images from the microscope (A), a threshold value setting means (31) for setting two-stage reference threshold values to the images of the objects to be analyzed captured by the image input means (10), and an analyzing means (34) for analyzing the images of the nuclear cells by either of the two-stage reference threshold values according to a ratio of an area shift of the objects to be analyzed between the images obtained by the two-stage reference threshold values.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to image analysis of a plurality of analysis objects having substantially uniform shapes and dimensions, for example, nucleated cells of an organism, and particularly, in nucleated cells induced by environmental mutagens. The present invention relates to an image analysis method, an apparatus, and a recording medium for analyzing abnormal chromosomes.

[0002]

2. Description of the Related Art It is known that various chromosomal abnormalities can occur in all organisms having chromosomes due to genetic factors or environmental mutagens. Much of the detection of chromosomal abnormalities relies on analysis of karyotype. To analyze karyotype, cells in the metaphase of the cell cycle of nucleated cells are required.

When preparing a sample of metaphase cells,
All of the individual chromosome groups within a cell require careful preparation so that they are separated without overlap and that individual chromosomes are fully extended. However, for example, it is difficult to obtain many complete specimens as described above for human chromosomes consisting of 24 different chromosome groups. Further, the karyotype analysis requires the work of verifying all of the chromosomes that are individually separated in the cell, and classifying the types of abnormalities by defining a karyotype different from the normal karyotype of a healthy person as a chromosomal abnormality.

On the other hand, when morphological abnormalities such as deletion, cleavage, and non-separation occur in the chromosomes of metaphase cells, a part of the cell nucleus in the subsequent interphase is separated to form a minute nucleus. Are known to form
Such a small nucleus is called a micronucleus.
s). On the other hand, the parent nucleus that gave rise to a micronucleus is referred to as the main nucleus to distinguish them. Therefore, the appearance of micronuclei means the occurrence of chromosomal abnormalities in interphase cells.

In humans, lymphocytes in peripheral blood are mainly used for the study of micronuclei. In normal lymphocytes, the appearance frequency of micronuclei is extremely low, but it is known that the appearance frequency is dramatically increased by stimulation with harmful drugs and the like (Heddle, JA, et al., Mutat. Res. 4
4, 63-69, 1977). Therefore, the appearance and frequency of micronuclei can be an indicator that chromosomal abnormalities due to DNA damage have occurred due to the influence of some genetic element, environmental mutagen, virus or the like. In addition, today, micronucleus tests using rodents and cultured cells have contributed to research for experimentally predicting the presence and extent of mutagenicity in humans.

The above-described conventional analysis of micronuclei has been performed on the basis of microscopic observation. In this case, since the judgment of the appearance frequency of micronuclei by the eyes of the observer is a subjective judgment that each person has, the data may vary and reliable data may not always be obtained. Absent. Further, there is a problem that the observation with the naked eye requires considerable labor and man-hours. Therefore, in order to quantitatively evaluate the DNA damage in cells by the frequency of micronucleus appearance, a more objective and efficient micronucleus detection technique, namely,
Automatic analysis technology is required.

Up to now, as an automatic micronucleus analysis technique for non-nucleated rodent red blood cells, Japanese Patent Application Laid-Open No. 9-242242 has been disclosed.
-89886, an analysis method using microscope image processing,
Further, Japanese Patent No. 2893613 discloses an analysis method using flow cytometry. However, many somatic cells including humans are nucleated, and the above-described analysis technique for anucleated cells cannot be applied to micronucleus analysis in such nucleated cells.

On the other hand, conventionally, a Giemsa-stained specimen has been used to visually identify micronuclei in nucleated cells. Microscopic image analysis technology for such specimens has also been devised (Tates AD., Et al., Int.
J. Radiat. Biol. 58: 813-825,
1990; Castelain PP. , Et a
l. , Mutagenesis 8: 285-293,
1993, Verhaegen F .; , Et al. ，
Cytometry 17: 119-127, 199.
4; Bocker W. , Et al. , Cytome
try 19: 283-294, 1995; Friea.
uff W. et al. , MutationRe
s. , 413: 57-68, 1998).

However, today, the cytoplasm (RN) in cells is
A) / nucleus (DNA), acridine orange (AO) is preferred as a more specific fluorescent dye.
Observation of O-stained specimens is becoming mainstream (Matsuo
ka A. et al. , Mutation Re
s. , 272: 223-236, 1993), analysis of micronuclei in fluorescently stained nucleated cell specimens is emphasized.

[0010]

An automatic micronucleus detection technique for a fluorescent microscope image of nucleated cells is disclosed in Japanese Patent Laid-Open No. 10-185911, but the microscope image is captured as a color image. Rare, then DNA and RN
Since image processing is performed by dividing the image into each A image, remarkable image deterioration occurs, and it becomes difficult to detect very small micronuclei. In addition, quantitative analysis of chromaticity, brightness, and saturation of fluorescence
Since it is performed for images with a certain threshold set, it is not possible to accurately reflect the variation in fluorescence brightness peculiar to biological samples, and as a result, close cells, main nuclei, and micronuclei are accurately separated and separated. It will be unrecognizable. Therefore, there is a big problem that a significant difference occurs even when compared with the visual determination result.

The object of the present invention is to overlap the objects to be analyzed that may occur in a sample (for example, to combine cells of organisms).
The object of the present invention is to provide an image analysis method, device, and recording medium that avoids the above in an image and performs an accurate and highly accurate image analysis of an analysis target (for example, a nucleated cell).

[0012]

In order to solve the problems and achieve the object, the image analysis method, apparatus and recording medium of the present invention are configured as follows.

(1) In the image analysis method of the present invention, a step of setting a plurality of analysis objects having substantially uniform shapes and dimensions within a predetermined measurement visual field, and at least one measurement object within the measurement visual field. There is a step of obtaining an image and a step of image-analyzing the plurality of analysis objects in the measurement visual field, and in the step of image-analyzing, two or more analysis objects overlapping in the visual field are preset. The method further includes a step of selectively separating based on the selected shape, and a step of obtaining necessary measurement data for at least one of the separated analysis objects.

(2) The image analysis method of the present invention is an image analysis method for analyzing an abnormality occurring in a chromosome based on an image of a nucleated cell obtained by a microscope, and the cell cycle corresponds to interphase. The cytoplasm and nuclei of the nuclear cells are labeled with a fluorescent dye as an observation target, and in the setting of the reference threshold for the fluorescence image of the observation target obtained by a microscope, the setting of the two-step reference threshold for the image of the nucleus is performed. The image of the nucleated cell is analyzed using one of the two-step reference thresholds according to the ratio of the area transition of the nucleus between the images obtained by the two-step reference thresholds.

(3) The image analysis method of the present invention is based on the above (2).
And the ratio of the area transition is obtained from each binarized image of the nucleus obtained by the two-step reference threshold value, and when the ratio of the area transition is a predetermined value or more, When the ratio of the area transition is less than the predetermined value, the lower standard threshold is analyzed among the two standard thresholds.

(4) The image analysis method of the present invention is based on the above (2).
Alternatively, in the method according to (3), and when the cytoplasm is in contact with the cytoplasm image, the cytoplasm is separated by contour shape analysis, and the separated coordinates are used for analysis of the nucleus image.

(5) The image analysis apparatus of the present invention performs image analysis on a plurality of analysis objects having substantially uniform shapes and dimensions based on the image of the analysis object obtained by a microscope. In the apparatus, a microscope for observing the plurality of analysis objects arranged in a minute region in an enlarged field of view, an image input means for capturing an image from the microscope, and the analysis captured by the image input means Applying a threshold for measurement regarding the shape of the analysis object to the image of the object, and selectively separating even when there is an overlap in the plurality of analysis objects, these separated analysis And an analysis unit that performs analysis on the measurement data by obtaining necessary measurement data from at least one object.

(6) The image analysis apparatus of the present invention is an image analysis apparatus for analyzing an abnormality occurring in a chromosome based on an image of a nucleated cell obtained by a microscope, in which the cell cycle corresponds to interphase. A microscope for observing an observation object in which the cytoplasm and nuclei of a nuclear cell are labeled with a fluorescent dye, image input means for capturing an image from the microscope, and two steps for the image of the nucleus captured by the image input means The threshold value setting means for setting the reference threshold value, and the nucleated cell at any one of the two-step reference threshold values according to the ratio of the area transition of the nucleus between the images obtained by the two-step reference threshold values. And an analyzing means for analyzing the image of.

(7) The recording medium of the present invention is computer-readable, and a plurality of threshold value setting means for accessing the image signal of the object to be analyzed obtained by a microscope to set a threshold value for measurement by a computer. Functions as an analysis means for performing analysis on measurement data by selectively separating images with overlapping analysis objects and obtaining necessary measurement data from at least one of the separated analysis objects. It records the program to make it happen.

(8) The recording medium of the present invention can be read by a computer, and a computer sets a threshold setting means for setting a two-step reference threshold for an image of a nucleus of a nucleated cell obtained by a microscope. According to the ratio of the area transition of the nucleus between the images obtained by the two-step reference threshold,
A program for functioning as an analysis unit that analyzes the image of the nucleated cell using one of the two-stage reference threshold values is recorded.

As a result of taking the above-mentioned means, the following effects are achieved.

According to the image analysis method of the present invention, in the image analysis of nucleated cells, a two-step reference threshold value is set, and the ratio of the area transition of the nucleus is compared with a predetermined value to obtain an appropriate reference threshold value. It is possible to set automatically. As a result, cell binding that may occur in the specimen can be avoided on the image, micronuclei can be accurately determined even in non-isolated cells, and chromosomal abnormalities in nucleated cells are automatically analyzed with high accuracy. be able to.

According to the image analysis method of the present invention, the ratio of the area transition of the nucleus obtained from each binarized image obtained by setting the two-step reference threshold values is compared with a predetermined value to occupy the nucleus. When the ratio is equal to or more than the predetermined value, the reference threshold value that is set high is selected, and when the ratio is less than the predetermined value, the reference threshold value that is set low is selected and measurement is performed. As a result, a reference threshold value adapted to changes in cell shape can be set, and objective and efficient analysis can be performed.

According to the image analysis method of the present invention, when the cytoplasms are in contact with each other in the fluorescence image, the contour shape analysis is used to obtain the coordinates for separating the cytoplasms, and the coordinates are used to determine the area of the corresponding nucleus. Find the transition. As a result, not only isolated cells but also bound cells are effectively measured, and the accuracy of automatic analysis is improved.

According to the image analysis apparatus of the present invention, the binding of cells that may occur in the specimen can be avoided on the image, the micronucleus can be accurately determined even in the non-isolated cell, and the It is possible to automatically analyze chromosomal abnormalities with high accuracy.

According to the recording medium of the present invention, a computer can avoid cell binding that may occur in a specimen on an image and accurately determine micronuclei even in unisolated cells. It is possible to analyze chromosomal abnormalities inside with high accuracy automatically.

[0027]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram showing the configuration of an image analysis apparatus according to an embodiment of the present invention. The image analyzer shown in FIG. 1 analyzes an abnormality that has occurred in a chromosome based on an image of a nucleated cell whose cell cycle corresponds to the interphase obtained by a microscope. This apparatus is mainly composed of a fluorescence microscope A which is an image input unit for microscopically imaging cells and a personal computer (PC) B which is an image analysis unit.

The main body 1 of the fluorescence microscope A is provided with a stage 2 which is automatically scanned in two dimensions, and this stage 2 is composed of cells labeled with a fluorescent substance (biological staining). A sample S is placed. Further, an autofocus device 3 that drives the stage 2 in the observation optical axis a direction so as to focus on the sample S is connected to the main body 1 and the stage 2.

On the observation optical axis a of the fluorescence microscope A, the sample S
, The objective lens 5 attached to the revolver 4, the dichroic mirror 6 provided in the main body 1, the observation light for specifically detecting the fluorescence emitted from the cytoplasm (RNA) / nucleus (DNA). A filter conversion unit 7 for converting a filter arranged on the axis a, a half mirror 9 provided in the lens barrel 8, and a C for inputting a microscope image.
A CD camera 10 is arranged. An eyepiece lens 11 is attached to the lens barrel 8. Further, the main body 1 is provided with a light source 12 for projecting excitation light onto the sample S.

The personal computer (PC) B is
The control unit 20 and the cytoplasm / nucleus measurement unit 30 are provided. The cytoplasm / nucleus measurement unit 30 includes a threshold setting unit 31 and a blob analysis unit 3.
2, the contour shape analysis unit 33, and the image determination unit 34. The cytoplasm / nucleus measurement unit 30 is stored as a program in a memory (not shown) and is executed by the control unit 20.

The control unit 20 is connected to the above-mentioned memory that stores the cytoplasm / nucleus measurement unit 30, and is also connected to the input unit 21 and the display unit 22. In addition, the control unit 20
A stage 2, an autofocus device 3, a filter conversion unit 7, and a CCD camera 10 are connected to the.
Further, the memory storing the cytoplasm / nucleus measurement unit 30 is connected to the CCD camera 10.

When the light emitted from the light source 12 is reflected by the dichroic mirror 6 and the sample S is excited through the objective lens 5, the fluorescence emitted by the sample S is converted into the objective lens 5,
It is taken into the CCD camera 10 via the dichroic mirror 6 and the half mirror 9.

The image information of the cells taken in by the CCD camera 10 is sent to the cytoplasm / nucleus measuring section 30 and the analysis processing described in detail later is executed. The threshold setting unit 31 sets a reference threshold value and a multi-step threshold value, the blob analysis unit 32 performs a conventionally known blob analysis on a cell image, and the contour shape analysis unit 33 analyzes a cell contour shape. Then, the image determination unit 34 counts the interphase cells forming micronucleus in the cell group analyzed by the contour shape analysis unit 33 in association with the main nucleus.

The control unit 20 is composed of a CPU and continuously executes the operations of all the constituent units of the image analysis apparatus. The input unit 21 is a keyboard, a touch panel,
A mouse or the like is used to add or change the type of information to be analyzed to the control unit 20. The display unit 22 is C
An RT, a printer, and the like, which display statistical data statistically processed by the control unit 20.

Further, if the control unit 20 is configured to be capable of transmitting / receiving various data to / from an inspection center, hospital, university, maker, etc. at a remote place through a communication line as appropriate, a more advanced The analysis software can be executed and the analysis method and information exchange by a skilled inspector or the like can be carried out quickly, and the configuration becomes more effective.

2 and 3 are flow charts showing the processing procedure of the image analysis method by the image analysis apparatus having the above-mentioned configuration. Further, FIG. 4 is a diagram schematically showing the processing steps according to the flowcharts of FIGS. 2 and 3.

Hereinafter, the cytoplasm (RN
The image analysis method of A) will be described with reference to FIG. First, in step S1, the control unit 20 causes the filter conversion unit 7 to arrange a filter (red filter) for specifically absorbing / exciting the fluorescence wavelength of cytoplasm (RNA) on the observation optical axis a. Thereby, the cytoplasm (RN
The image of A) is captured by the CCD camera 10 and input as image data to the cytoplasm / nucleus measurement unit 30.

Next, the cytoplasm / nucleus measurement unit 30 performs image brightness correction on the input cytoplasmic (RNA) image,
In step S2, the threshold value setting unit 31 sets a reference threshold value (reference brightness), and as shown in the image 41 of FIG.
The analysis target image is extracted by binarizing the image data of (NA) 411, 412.

Next, in step S3, the blob analysis unit 32 uses the conventional blob analysis method to determine the centroid, area (sum of the number of pixels), volume (sum of the fluorescence brightness of each pixel), and true of the images of the cytoplasm 411, 412. Analyze the roundness and the ratio of length to width. Here, from the roundness and the ratio of length to width,
For the case where multiple cells are suspected of overlapping (cytoplasm 412) instead of a single cell (cytoplasm 411),
In step S4, the contour shape analysis unit 33 performs the contour shape analysis described later.

As a result, in step S5, as shown in the image 42 of FIG. 4, a plurality of adjacent cells are separated into independent individual cells. For example, as shown by p in FIG.
The adjacent cytoplasm is separated into two cytoplasms by contour shape analysis. Here, the contour shape analysis unit 33 extracts the respective coordinates of the positions of the two peaks (constrictions) detected by the contour shape analysis as described later, as the separated coordinates of the cytoplasm, and records them in the nucleus (DNA) image. . As a result, in the analysis target image, each nucleus (DNA) in the adjacent cells is
Image is separated.

Hereinafter, the nucleus (DNA) obtained by the image analysis apparatus
The image analysis method will be described with reference to FIG. First step S
11, the control unit 20 causes the filter conversion unit 7 to execute the kernel (D
A filter (green filter) for specifically absorbing / exciting the fluorescence wavelength of (NA) is arranged on the observation optical axis a. As a result, the nucleus (main nucleus / micronucleus: DN) in the sample S
The image of A) is captured by the CCD camera 10 and input as image data to the cytoplasm / nucleus measurement unit 30.

Next, in step S12, the cytoplasm / nucleus measuring section 30 performs the extracted separation when the adjacent cytoplasm is separated in step S5 by the contour shape analysis of the cytoplasm (RNA) image in step S4. Input the same coordinates as the coordinates to the nuclear (DNA) image.

Next, the cytoplasm / nucleus measuring unit 30 performs image brightness correction on the input nuclear (DNA) image, and in step S13, the threshold setting unit 31 sets the first reference threshold value (brightness). Then, as shown in the image 43 of FIG.
A) The image to be analyzed is extracted by binarizing the image data of 413 to 417.

Next, in step S14, the blob analysis unit 32
Analyzes the rectangle, area, volume, circularity, ratio of length to width, etc. of the images of the nuclei 413 to 417 by the conventional blob analysis method. Here, the area ratio (cytoplasm / nucleus) is calculated, and cells having a certain value or more are classified as cells to be observed.

Further, in step S15, the threshold setting unit 31
Sets a second reference difference threshold having a predetermined negative difference with respect to the first reference threshold (first reference threshold> second
The image to be analyzed is extracted by binarizing the image data of the nucleus (DNA) and the reference difference threshold value of 1.). Next, in step S16, the blob analysis unit 32 performs the blob analysis in the same manner as in step S14 to classify the observation target cells.

Next, in step S17, the image determination unit 34
Is the transition of the area of the analysis target image obtained by the first reference threshold value and the analysis target image obtained by the second reference difference threshold value (area ratio (%) = area of the analysis target image obtained by the first reference threshold value). / Analyze the area of the analysis target image obtained by the second reference difference threshold × 100).

When the change of the area is equal to or larger than the predetermined judgment reference value (for example, 80%), the image judgment unit 34 determines in step S18 the image to be analyzed obtained by the first reference threshold among the two-step thresholds. The volume ratio (nucleus / cell) is calculated from the analysis by the blob analysis method, and those having a certain value or more are labeled as main nuclei, and those having less than the certain value are labeled as micronuclei. As a result, a plurality of adjacent main nuclei are separated as shown in the image 44 of FIG. For example, as shown by q in FIG. 4, the adjacent main nuclei are separated into two main nuclei by the contour shape analysis. Then, the cytoplasm / nucleus counting section 30 recognizes the number of the main nuclei, and the number of the main nuclei is 1 as shown by 45 in FIG.
Individual cells, two or more cells are classified, and step S1
At 9, they are determined as cells to be observed.

Then, in step S20, if a micronucleus is present as a result of the above-described image analysis processing, the image determination unit 34 associates the number of the micronucleus with the main nucleus as shown by 45 in FIG. The main nucleus and the micronucleus are associated with the cytoplasm subjected to image analysis, the micronucleus is counted for each cell of each observation target, and the appearance frequency of interphase cells having the micronucleus among all the observation target cells is totaled.

Thereafter, in step S21, a predetermined difference is sequentially added to the first reference threshold value until the image analysis processing for the preset number of steps (for example, 5 steps) is completed. As a new threshold, the above step S
The image analysis process of 18 to S20 is performed. By performing the analysis of the nucleus with the multi-step threshold value as described above, the result that the micronucleus is more present in the cell is acquired. For example, the micronucleus 416 not detected in the image 43 of FIG.
Will be detected as shown in.

On the other hand, in step S17, if the transition of the area is less than the predetermined determination reference value (for example, 80%), in step S22, the image determination unit 34 determines the second of the two-step threshold values.
The volume ratio (nucleus / cell) is calculated from the analysis by the blob analysis method for the image to be analyzed obtained with the reference difference threshold value of 1. Label as. And cytoplasm / nucleus measurement unit 30
Recognizes the number of main nuclei, classifies cells having 1, 2, or more main nuclei, and determines them as cells to be observed in step S23.

Then, as a result of the above-described image analysis processing, if micronuclei are present, in step S24, the image determination unit 34
Associates the number of the micronucleus with the main nucleus, associates the main nucleus and the micronucleus with the cytoplasm subjected to image analysis, counts the micronucleus for each observation target cell, and determines the micronucleus in all the observation target cells. The frequency of appearance of interphase cells carrying That is, when the transition of the area is less than the predetermined value, the image analysis with the multi-step threshold value is not performed.

Among the above processing procedures, the contour shape analysis will be described below. The contour shape analysis is a method of analyzing the contour shape of the target image. In this embodiment, this method is applied to the image of a biological sample, and particularly applied to shape recognition of cytoplasm and nucleus.

FIG. 5 is a diagram showing the basic principle of contour shape analysis. In the contour shape analysis, the analysis target 5 in the image 51
Pixels on the two contours (hatched portions in the figure) are traced, and a coordinate list showing coordinates on the contours is generated. Next, a feature amount derived using an arbitrary one point Pi in the coordinate list and two points Pi + s and Pi-s, which are respectively separated by s before and after the point Pi, from a total of three points, the feature amount of the coordinate list is calculated. The transition pattern is calculated by calculating for all coordinates.

Here, the feature amount is an arbitrary one point P described above.
i is the area of a triangle formed by two points Pi + s and Pi-s, and the direction of the apex is defined as positive or negative when the straight line connecting the two points Pi + s and Pi-s is the base. Further, the above-mentioned transition pattern is based on the number Tt of consecutive portions that exceed the threshold value Tv with respect to the size of the above-mentioned feature amount, and the portions that exceed and fall below the threshold value Tv in the transition of the pattern. It serves as a reference for determining the unevenness of the detection site (the contact site of a plurality of cytoplasms or a plurality of nuclei) based on the appearance order.

FIGS. 6A and 6B are diagrams showing examples of the results of the contour shape analysis of the main nucleus / micronucleus. As shown in (a) of FIG. 6, in the transition pattern of the contour shape of the nucleus in the cell 61 having one nucleus, the feature amount is always below the threshold Tv, but as shown in (b) of FIG. In the transition pattern of the contour shape of the nucleus in the cell 62 having two adjacent nuclei, two peaks (constrictions) exceeding the threshold Tv occur in the feature amount. In addition, as shown in FIG. 6 (c), cells 6 having three adjacent nuclei
In No. 3, three peaks (constrictions) in which the feature amount exceeds the threshold Tv occur.

By using this method, as shown in FIG. 6D, even a micronucleus in the cell 64 close to the main nucleus can be detected by the characteristic of the transition pattern of the contour shape. In this case, the interval between the two peaks generated in the main nucleus and the micronucleus is 2 between the main nuclei shown in (b) of FIG.
It becomes narrower than the interval of peaks at some points.

The number of constrictions (lacks) detected in the images analyzed as described above and the existence modes of main nuclei / micronuclei can be classified as follows, for example.

[0059]

[Table 1]

If the "number of constrictions" is "other", the fact is reserved and stored in the memory.
After that, the operator visually determines the cell classification.

Further, in the present embodiment, the main core
Image analysis is performed by setting multi-step thresholds for micronucleus images, but this is a means for solving the problem of variations in fluorescence brightness between cells. For example, when image processing is performed with a high threshold value, small nuclei having low luminance with respect to the main nucleus will not be recognized as an image. Thus, in order to eliminate erroneous determinations due to variations in brightness between the main nucleus and micronuclei or between cells, by analyzing images with multiple thresholds, micronuclei can be overlooked extremely. It is possible to obtain less results.

In this embodiment, the contour shape analysis is performed and the multi-step threshold value is set, and the image analysis is performed. However, only the contour shape analysis or only the multi-step threshold value is set. Even in this case, it is possible to accurately separate and recognize the adjacent cytoplasm and the main nucleus / micronucleus as compared with the conventional method.

According to the present embodiment, firstly, when the cytoplasmic (RNA) image and the nuclear (DNA) image are separately input at the time of inputting the image of the AO-stained nucleated cell specimen, as in the conventional case. Since image compression is avoided, image deterioration can be reduced and subsequent image processing can be performed smoothly. Second,
In the step of identifying the main nucleus and the micronucleus by the image analysis process, the threshold of the image is made variable, whereby the variation in fluorescence caused by the biological sample can be eliminated. Third, by adopting a unique algorithm that analyzes contour information of the cytoplasm and nucleus in the image analysis process, it is possible to efficiently separate and recognize cells and nuclei in close proximity. As a result, the micronucleus can be determined more objectively and the measurement can be performed more efficiently than in the conventional case.

In the present invention, in addition to human lymphocytes used for ordinary micronucleus analysis, any nucleated cell obtained from living tissue or various cell lines can be applied. In addition, as the analysis target, even if cells derived from a test organism in which chromosomal abnormality may have already occurred for some reason,
It may be a cell exposed to a particular environmental mutagen.

According to the present invention, even if the main nuclei in the nucleated cells to be observed have the number of mononuclear, two or more nuclei, they are identified and the micronucleus-bearing cells are counted. it can.
For micronucleus analysis of human peripheral blood lymphocytes, a method in which a drug such as cytochalasin B is treated in a culture medium and induced binuclear cells are observed is often used (Fenech). M. and Morley AA.,
Mutation Res. 147: 29-36, 19
85) In particular, it is extremely effective in performing monitoring targeting human cells (evaluating the degree of health and the degree of exposure to mutagenic substances over time).

In the image analysis method by the image analysis device of the present embodiment described above, the image obtained by the microscope is targeted, and the blob analysis, the contour shape analysis, and the multi-step threshold analysis are effectively combined to perform the staining. Detected micronuclei.

However, when observing the image during the measurement, there may be a case in which the measurement becomes impossible when the cells are connected to each other, and the number of cells to be observed may decrease from the actual number of cells. That is, in the measurement with only a low threshold value, the nuclei are bound to each other on the image due to the binding of the cells to each other, and as a result, the nuclei protrude from the cytoplasm, resulting in a problem that the measurement is not performed accurately.

In order to solve such a problem, in the present embodiment, a two-step threshold value is set, and the combined cells are separated on the image according to the transition of the area of each obtained analysis target image. Is done automatically. By performing the image analysis according to the present embodiment, it is possible to perform accurate analysis even when nuclei are bound together, and it is possible to prevent a defect in micronucleus measurement caused by the binding between cells.

[0069]

EXAMPLE An example of the image analysis apparatus according to the present invention will be described below. In the image analysis method according to the above embodiment,
Based on the image of nucleated cells obtained by a microscope, the nucleated cells in the interphase of the cell cycle are labeled with a fluorescent dye for observation. Then, with respect to the fluorescence images of the cytoplasm and the nucleus, by using the distribution of the intermediate brightness or high brightness of the nucleus or the area of the nucleus as an index, it is determined whether or not the nucleated cell is subjected to cytotoxicity. Image analysis is performed by setting a reference threshold according to the result of identification.

However, when the cells are closely connected to each other, it may be determined that the cells cannot be measured. If you set a low threshold and measure,
This is because the nuclei are connected to each other on the image and are observed as one nucleus, and as a result, the nuclei are observed as if they are protruding from one cytoplasm. In this case, micronucleus counting is not performed on bound cells. Such a problem is prevented by the image analysis method according to the above-described embodiment, and an example will be described below. In this example, the test was carried out using human cultured cells treated with a chromosomal aberration-inducing substance.

As described in the above embodiment, in the present image analysis apparatus, the cytoplasm (RNA) of the fluorescently labeled cells is
Take an image and a nuclear (DNA) image. And the nucleus (DN
A) The first reference threshold value of the image is set to 200, the difference to this is set to -50, and the second reference difference threshold value is set to 150, and measurement is performed. In addition, the judgment reference value of the area change is 8
Set it to 0%. That is, when the area transition is 80% or more, the reference threshold value 200 is used, and when the area transition is less than 80%, the reference difference threshold value 150 is used.

FIG. 7 is a view showing a photograph of an image showing a determination result in the case where the area transition analysis is performed without performing the nuclear separation as described later. In FIG. 7, non-isolated cells were not detected as mononuclear cells (ce
llNG).

FIG. 8 shows the kernel (DN
It is a figure which carries out the photograph print of the image which shows the binarized image of A), and Table 2 has shown the area value of the nucleus (DNA) in the binarized image. FIG. 9 is a view showing an image showing a binarized image of nuclei (DNA) obtained with the reference difference threshold value 150 by printing, and Table 2 shows the area of nuclei (DNA) in the binarized image. Indicates the value.

[0074]

[Table 2]

[0075]

[Table 3]

In Table 2, the labels attached to the images a to h of the nucleus (DNA) in FIG. 8 and the area values of the images a to h are shown. Similarly, in Table 3, the nucleus (DNA) in FIG.
The labels attached to the images A to E and the area values of the images A to E are shown. That is, in FIGS. 8 and 9, the images a to
h, AE are images of nuclei (DNA), and images a to h, A
Unique labels are attached to ~ E.

FIG. 10 is a diagram of FIG.
FIG. 10 is a diagram showing an image showing an area transition image obtained from the two images of FIG. 9 by printing a photograph. The area transition in FIG.
The area ratio of the images (for example, a and A, d and B, and e and C) located at substantially the same positions in the two images of FIGS. 8 and 9 (image area obtained by the reference threshold 200 / reference difference threshold 150 The area of the image obtained by x 100 (%) is shown. That is, when the area transition of each image is obtained from the image in which the nuclei are close to each other as shown in FIG. 9, the area ratio has a low value. This is a result obtained by measuring without separating the image of the bound nucleus.

In order to separate the image of the bound nuclei, as described in the above embodiment, when the adjacent cytoplasms are separated in the processing of the cytoplasmic (RNA) image, the separation coordinates are extracted, and FIG. It is also recorded in the nuclear (DNA) image as shown. The maximum abscissa of the image shown in FIG. 11 is 659, and the maximum ordinate is 517.

FIG. 12 is a diagram showing a photograph of a binarized nuclear separation image of nuclei (DNA) obtained at the reference difference threshold 150, and Table 4 shows nuclei (DNA) in the binarized images. The area value of is shown.

[0080]

[Table 4]

In Table 4, the labels attached to the images (a'-h ') of the nucleus (DNA) in FIG. 12 and the area values of the images (a'-h') are shown. By using the above-mentioned separation coordinates, an image in which the nuclei are separated as shown in FIG. 12 with respect to FIG. 9 can be obtained.

FIG. 13 shows the case of FIG.
It is a figure which prints and shows the area transition image obtained from two images of FIG. The area transition in FIG. 13 is the area ratio of the images (for example, a and a ′, b and b ′, c and c ′) located at substantially the same positions in the two images of FIGS. The area of the obtained image / the area of the nuclear separation image obtained by the reference difference threshold value 150 × 100 (%) is shown. As a result of measuring the area ratio as shown in FIG.
The determination result of the area transition analysis as shown in is obtained. Figure 1
In contrast to FIG. 7, the cells of No. 4 were detected as mononuclear cells (1 NUC) even in the non-isolated cells.

By performing the above-described measurement, even for the cells that are connected, by setting a lower reference difference threshold value out of the two-step reference threshold values, the connected nuclei are displayed on the image. Since it can be separated with, it will be analyzed correctly.

The present invention is not limited to the above-described embodiments and examples, and can be carried out by appropriately modifying it without departing from the scope of the invention.

For example, in the description of the above embodiment,
It was applied to sample observation when cells of a living organism were stained with two-color fluorescent dyes as the analysis target, but it seems that the analysis target is a substance other than the organism and its shape and dimensions are determined almost uniformly. It may be applied to other materials such as artificial particles, blood cells and the like. In addition, the reaction results from biological reactions (enzyme reaction, antigen-antibody reaction, nucleic acid hybridization reaction, nucleic acid synthesis reaction, etc.) can be obtained in a specified micro area using the analysis target such as cells, artificial particles, blood cells, etc. to be applied. The method and apparatus of the present invention and the storage medium may be applied for measuring in.

Further, according to the present invention, although it can be applied to an example using one color dye, in a preferred embodiment, a method for outputting an optimum analysis result by classifying different color types of dyes is provided. Therefore, it can be effectively applied even when three or more color dyes are used. Further, the image used for the analysis does not have to be a single still image, and a plurality of still images or a moving image of a desired time may be analyzed as appropriate to improve the reliability of measurement data or to analyze various analysis information. You may get sex. Further, according to the present invention, a plurality of types (two types of main nuclei and micronuclei in the embodiment) of analysis objects having substantially uniform shapes and dimensions are mixed as in the above-described embodiment. Since an image analysis method for various specimens is also provided, it is possible to simultaneously analyze images of arbitrary different types of analysis objects.

[0087]

According to the image analysis method, the apparatus and the recording medium of the present invention, it is possible to accurately separate the analysis objects having two or more overlaps based on the shapes and selectively acquire the measurement data. Therefore, it is possible to efficiently perform image analysis of a large number of closely arranged analysis objects without wasting it. Further, according to the present invention, it is possible to automatically set an appropriate reference threshold value from the transition of the area of the nucleus by the image analysis of the nucleated cells using the two-step reference threshold value, and possible cell-to-cell binding in the sample is displayed on the image. It is possible to avoid it and realize accurate and accurate image analysis. In other words, it is possible to eliminate the variation in fluorescence brightness between cells, and even when interphase cells are bound, efficient separation on the image, confirmation and automatic analysis significantly reduce the number of unmeasurable cells. To be done.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of an image analysis apparatus according to an embodiment of the present invention.

FIG. 2 is a flowchart showing a cytoplasmic (RNA) image processing procedure of the image analysis method by the image analysis apparatus according to the embodiment of the present invention.

FIG. 3 is a flowchart showing a nuclear (DNA) image processing procedure of the image analysis method by the image analysis apparatus according to the embodiment of the present invention.

FIG. 4 is a diagram schematically showing processing steps of an image analysis method by the image analysis device according to the embodiment of the present invention.

FIG. 5 is a diagram showing a basic principle of contour shape analysis according to the embodiment of the present invention.

FIG. 6 is a diagram showing an example of a result of contour shape analysis of main nuclei / micronuclei according to the embodiment of the present invention.

FIG. 7 is a diagram showing an image showing a determination result of an area transition analysis according to an embodiment of the present invention by printing a photograph. (No nuclear separation)

FIG. 8 is a diagram showing an image showing a binarized image at a core part reference threshold value 200 according to an embodiment of the present invention by printing a photograph.

FIG. 9 is a core reference difference threshold value 150 according to the embodiment of the present invention.
The figure which shows the image which shows the binary image in FIG.

FIG. 10 is a diagram showing a printed area transition image according to an embodiment of the present invention as a photograph.

FIG. 11 is a diagram showing an image showing cell detachment coordinates according to an embodiment of the present invention by printing a photograph.

FIG. 12 is a core reference difference threshold value 15 according to the embodiment of the present invention.
The figure which shows the image which shows the binarized nucleus separation image in 0 by photo-printing.

FIG. 13 is a diagram showing an area transition image according to an embodiment of the present invention by printing a photograph. (Nuclear separation)

FIG. 14 is a diagram showing an image showing a determination result of an area transition analysis according to an embodiment of the present invention by printing a photograph. (Nuclear separation)

[Explanation of symbols]

A ... Fluorescent microscope B: Personal computer (PC) 1 ... Main body 2 ... Stage 3 ... Auto focus device 4 ... Revolver 5 ... Objective lens 6 ... Dichroic mirror 7 ... Filter converter 8 ... lens barrel 9 ... Half mirror 10 ... CCD camera 11 ... Eyepiece 12 ... Light source 20 ... Control unit 21 ... Input section 22 ... Display 30 ... Cytoplasm / Nuclear Measurement Department 31 ... Threshold setting section 32 ... Blob analysis section 33 ... Contour shape analysis unit 34 ... Image determination unit S ... specimen

Front page continuation (51) Int.Cl. ⁷ Identification code FI theme code (reference) G01N 21/64 G01N 21/64 F 5L096 33/48 33/48 M Z 33/58 33/58 AZ G06T 1/00 295 G06T 1/00 295 7/00 300 7/00 300F F term (reference) 2G043 AA03 BA16 CA03 DA02 EA01 FA02 GA08 GB21 HA01 JA02 LA03 NA01 NA06 2G045 BB24 CB01 DA13 FA16 FB12 GC15 4B029 AA07 BB07 FA12B02 FA02 BB08 FA02 BB08 FA02 BB08 FA02 BB08 FA02 QA01 QA18 QA19 QQ02 QQ04 QQ05 QQ06 QQ07 QQ08 QQ09 QQ42 QQ52 QR66 QS03 QS36 QS39 QX02 5B057 AA10 BA02 CA12 CA16 CE12 DA12 DB02 DC04 DC36 5L096 BA06 FA04 FA06 FA14 FA59 FA64 GA51 JA11

Claims (8)

[Claims] 1. A step of setting a plurality of analysis objects having substantially uniform shapes and dimensions in a predetermined measurement visual field, a step of obtaining at least one measurement image in the measurement visual field, and the measurement visual field. A step of image-analyzing a plurality of the analysis objects in the image analysis step, wherein the step of image-analyzing selectively separates two or more of the analysis objects overlapping in the visual field based on a preset shape. And a step of obtaining necessary measurement data for at least one of the separated objects to be analyzed. 2. Based on an image of nucleated cells obtained by a microscope,
In an image analysis method for analyzing abnormalities occurring in chromosomes, the cytoplasm and nuclei of nucleated cells whose cell cycle corresponds to the interphase are labeled with a fluorescent dye as an observation target, and a fluorescence image of the observation target obtained by a microscope. In the setting of the reference threshold value for the nuclei, the two-step reference threshold value is set for the nucleus image, and the nuclei area transition ratio between the images obtained by the two-stage reference threshold value An image analysis method, characterized in that the image of the nucleated cell is analyzed using any one of the reference thresholds of steps. 3. The area transition rate is obtained from each binarized image of the nucleus obtained by the two-step reference threshold value, and if the area transition rate is a predetermined value or more, the two steps are performed. The higher standard threshold among the standard thresholds is analyzed, and when the ratio of the area transition is less than the predetermined value, the lower standard threshold is analyzed among the two-stage standard thresholds. Image analysis method described. 4. When the cytoplasm is in contact with the cytoplasm image, the cytoplasm is separated by contour shape analysis, and the separated coordinates are used for analysis of the nucleus image. The image analysis method described in. 5. An image analysis apparatus for performing image analysis on a plurality of analysis objects having substantially uniform shapes and dimensions based on an image of the analysis object obtained by a microscope, the image analysis apparatus being arranged in a minute area. A microscope for observing the plurality of analysis objects in an enlarged field of view, image input means for capturing an image from the microscope, and the analysis for the image of the analysis object captured by the image input means. A threshold for measurement relating to the shape of the object is applied, and even when there is an overlap in the plurality of analysis objects, the separation is selectively performed, and it is necessary from at least one of the separated analysis objects. An image analysis apparatus comprising: an analysis unit that executes analysis related to measurement data by obtaining measurement data. 6. Based on an image of nucleated cells obtained by a microscope,
In an image analyzer that analyzes abnormalities occurring in chromosomes, a microscope for observing the observation target in which the cytoplasm and nucleus of nucleated cells whose cell cycle corresponds to the interphase is labeled with a fluorescent dye, and an image is captured from this microscope Image input means, threshold setting means for setting a two-step reference threshold for the image of the nucleus captured by the image input means, and the nuclei of the nuclei between the images obtained by the two-step reference thresholds. An image analyzing apparatus comprising: an analyzing unit that analyzes the image of the nucleated cell using one of the two-stage reference threshold values according to a ratio of the area transition. 7. A threshold setting means for accessing the image signal of an analysis object obtained by a microscope to set a threshold for measurement by a computer, and selectively selecting an image in which a plurality of analysis objects overlap. A computer-readable recording medium in which a program for causing separation to obtain necessary measurement data from at least one of these separated analysis objects and performing analysis on the measurement data is recorded. 8. A threshold value setting means for setting a two-step reference threshold value with respect to an image of a nucleus of a nucleated cell obtained by a microscope, the method between the respective images obtained by the two-step reference threshold value. A computer-readable recording medium recording a program for functioning as an analyzing unit for analyzing an image of the nucleated cell with one of the two-step reference threshold values according to the ratio of the area transition of the nucleus.