Image analyzer, image analysis method, and program for image analysis

Abstract

To provide an image analyzer for analyzing an emulsion image of a microchannel chip and efficiently acquiring an image of correct circular droplet region corresponding to the actual droplet region, an image analysis method, and a program for image analysis.SOLUTION: An image analyzer according to the present invention comprises: an initial image generation unit for generating an initial circular droplet region image; a first processing image generation unit for removing, from the initial circular droplet region image, a circular droplet region whose duplication rate is greater than or equal to a first prescribed duplication rate among circular droplet regions in a first particle diameter range, and generating a first circular droplet region image; and a second processing image generation unit for removing, from the first circular droplet region image, a circular droplet region whose duplication rate is greater than or equal to a second prescribed duplication rate smaller than the first duplication rate among the circular droplet regions in a second particle size range larger the first particle diameter range, and generating a second circular droplet region image.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an image analysis device, an image analysis method, and an image analysis program, and more particularly to an image analysis device, an image analysis method, and an image analysis program for analyzing an image of an emulsion present in an emulsion holding portion of a microchannel chip. [Background technology]

[0002] The microdroplet method, which divides a reaction solution into microcompartments and performs reactions independently, is known as a technique for dividing a reaction solution into microcompartments and performing reactions independently. This method is expected to be applied to, for example, the production of micro- and nano-particles, and in particular, it is used for digital measurement to measure the presence or absence of a target molecule by the presence or absence of a signal by dividing a target molecule into microcompartments using a microfluidic device and performing a reaction in the microdroplet, thereby performing absolute quantification of the number of target molecules.

[0003] Microdroplet techniques generally use emulsions that consist of a continuous phase, such as oil, and droplets of an aqueous solution dispersed in the continuous phase.

[0004] Non-Patent Document 1 discloses a centrifugal step droplet generation method. The document describes that the inlet of the device is filled with oil, and the oil is sent to a droplet collection chamber by centrifugation, and then a sample solution is introduced from the same inlet and droplets are generated by centrifugation.

[0005] In contrast to such droplet generation methods, a method is known in which a dispersed phase liquid such as a reaction liquid and a continuous phase liquid such as an oil are supplied to a microchannel chip via separate supply parts and then merged within the chip to generate an emulsion.

[0006] US Patent No. 5,999,633 discloses such a system and method for generating droplets suitable for droplet assays. The document describes transporting the generated droplets to an outlet area consisting of a pipette tip or a droplet well. The document also describes an air trap that substantially separates the sample and oil until application of a fluid driving force (such as negative or positive pressure).

[0007] In addition, in recent years, droplet array measurement, in which droplets are aligned in a monolayer in a detection area to easily measure signals, has been attracting attention as a method for simplifying and speeding up the device.

[0008] Patent Documents 2 and 3 disclose a microchannel chip having a channel for forming droplets and a droplet holder for holding droplets. Patent Document 2 describes that droplets are formed by joining two or more reaction solutions together and then contacting them with an immiscible liquid that is not miscible with the reaction solutions.

[0009] Non-Patent Document 2 describes a method for generating droplets on a chip and an apparatus for doing so. The method described in this document includes an operation of filling a droplet array part with oil (filling operation) before liquid transfer. [Prior art documents] [Patent documents]

[0010] [Patent Document 1] European Patent No. 2550528 [Patent Document 2] JP 2019-170363 A [Patent Document 3] JP 2020-169911 A [Non-patent literature]

[0011] [Non-Patent Document 1] Centrifugal step emulsification applied for absolute quantification of nucleic acids by digital droplet RPA, Lab Chip, 2015, 15, 2759-2766 [Non-Patent Document 2] 1-Million droplet array with wide-field fluorescence imaging for digital PCR, Lab on a Chip, 2011, 11, 3838-3845 Summary of the Invention [Problem to be solved by the invention]

[0012] In the microdroplet method as described above, in order to perform absolute quantification of the number of target molecules, it is preferable to obtain an image of the emulsion holding portion that holds the emulsion, identify individual droplets (emulsion particles) based on the image, and measure the presence or absence of target molecule signals for each of these droplets.

[0013] In this case, there is a method for detecting multiple circular droplet regions in an emulsion based on the brightness of the image to obtain a circular droplet region image. However, the circular droplet region image does not correspond to the actual droplet region, and is prone to being an erroneous circular droplet region image in which multiple circular droplet regions overlap each other.

[0014] Therefore, in order to obtain a correct circular droplet area image that corresponds to the actual droplet area, it is necessary to remove the erroneous circular droplet areas from the image among the multiple overlapping circular droplet areas. If such an operation is performed based on experience, the results may vary depending on the operator, and the operation is time-consuming.

[0015] In response to this, the present invention provides an image analysis device, an image analysis method, and an image analysis program that efficiently obtain an image of a correct circular droplet area that corresponds to the actual droplet area. [Means for solving the problem]

[0016] <Aspect 1> An image analyzer for analyzing an image of an emulsion present in an emulsion holding portion of a microfluidic chip, comprising: an initial image generating unit that detects a plurality of circular droplet regions in the emulsion based on the brightness of an image of the emulsion present in the emulsion holding unit, and generates an initial image of the circular droplet regions; a first processed image generating unit that calculates an overlap rate with other circular droplet regions for each of the circular droplet regions in a first particle size range among the plurality of circular droplet regions in the initial circular droplet region image, and removes from the initial circular droplet region image the circular droplet regions in the first particle size range whose calculated overlap rate is equal to or greater than a first predetermined overlap rate, thereby generating a first circular droplet region image; a second processing image generation unit that calculates an overlap rate with other circular droplet regions for each of the circular droplet regions in a second particle size range among the multiple circular droplet regions in the first circular droplet region image, and removes from the first circular droplet region image the circular droplet regions in the second particle size range whose calculated overlap rate is equal to or greater than a second predetermined overlap rate, thereby generating a second circular droplet region image, wherein the second particle size range is a particle size range larger than the first particle size range, and the second overlap rate is smaller than the first overlap rate. <Aspect 2> The device of embodiment 1, further comprising: a third image generating unit that calculates an overlap rate with other circular droplet regions for each of the circular droplet regions in a third particle size range among the multiple circular droplet regions in the second circular droplet region image, and removes from the second circular droplet region image the circular droplet regions in the third particle size range whose calculated overlap rate is equal to or greater than a third predetermined overlap rate, thereby generating a third circular droplet region image, wherein the third particle size range is a particle size range larger than the second particle size range, and the third overlap rate is smaller than the second overlap rate. <Aspect 3> The apparatus of aspect 1 or 2, wherein the initial image generation unit binarizes the image to obtain a binarized image, and determines the center and particle size of the circular droplet region by performing a generalized Hough transform on a set of edge pixels of the binarized image, thereby detecting multiple circular droplet regions in the emulsion and generating the initial circular droplet region image. <Aspect 4> The device according to any one of aspects 1 to 3, wherein the overlap rate used in at least one of the first processed image generation unit, the second processed image generation unit, and the third image generation unit is the following (a) and / or (b): (a) the maximum value of the overlap rate with any one of the other circular droplet regions; (b) The integrated value of the overlap rate with all the other circular droplet regions. <Aspect 5> The device according to any one of aspects 1 to 4, further comprising a measuring unit that measures the presence or absence of a target molecule signal in the circular droplet region of the second circular droplet region image, the third circular droplet region image, or a circular droplet region image for counting obtained by further performing any image processing on them, and determines the proportion of the circular droplet region having the target molecule signal. Aspect 6 An image analysis method for analyzing an image of an emulsion present in an emulsion holding portion of a microchannel chip, the method comprising the steps of: detecting a plurality of circular droplet regions in the emulsion based on the brightness of an image of the emulsion present in the emulsion holding portion, and generating an initial circular droplet region image; calculating an overlap rate with other circular droplet regions for each of the circular droplet regions in a first size range among the plurality of circular droplet regions in the initial circular droplet region image, and removing the circular droplet regions in the first size range whose calculated overlap rate is equal to or greater than a first predetermined overlap rate from the initial circular droplet region image to generate a first circular droplet region image; calculating an overlap rate with other circular droplet regions for each of the circular droplet regions in a second particle size range among the plurality of circular droplet regions in the first circular droplet region image, and removing from the first circular droplet region image those circular droplet regions in the second particle size range whose calculated overlap rate is equal to or greater than a second predetermined overlap rate, thereby generating a second circular droplet region image, wherein the second particle size range is a particle size range larger than the first particle size range, and the second overlap rate is smaller than the first overlap rate. Aspect 7 A computer program for image analysis that analyzes an image of an emulsion present in an emulsion holding portion of a microfluidic chip, the computer program causing a processor to execute a process including the steps of: detecting a plurality of circular droplet regions in the emulsion based on the brightness of an image of the emulsion present in the emulsion holding portion, and generating an initial circular droplet region image; calculating an overlap rate with other circular droplet regions for each of the circular droplet regions in a first size range among the plurality of circular droplet regions in the initial circular droplet region image, and removing the circular droplet regions in the first size range whose calculated overlap rate is equal to or greater than a first predetermined overlap rate from the initial circular droplet region image to generate a first circular droplet region image; calculating an overlap rate with other circular droplet regions for each of the circular droplet regions in a second particle size range among the plurality of circular droplet regions in the first circular droplet region image, and removing from the first circular droplet region image those circular droplet regions in the second particle size range whose calculated overlap rate is equal to or greater than a second predetermined overlap rate, thereby generating a second circular droplet region image, wherein the second particle size range is a particle size range larger than the first particle size range, and the second overlap rate is smaller than the first overlap rate. Effect of the Invention

[0017] According to the image analysis device, image analysis method, and image analysis program of the present invention, an image of a circular droplet area that corresponds to an actual droplet area can be efficiently obtained. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 is a diagram illustrating an example of a hardware configuration of the disclosed image analyzing device. [Diagram 2] FIG. 2 is an example of a functional block diagram of a processor related to image processing of the present invention. [Diagram 3] FIG. 3 shows (A) a fluorescence image of the emulsion, (B) a binarized image of the emulsion, (C) an image of the initial circular droplet area of ​​the emulsion, and (D) a reference diagram for determining the center and particle size of the circular droplet area. [Figure 4] FIG. 4 shows (A) an initial circular droplet area image of an emulsion, (B) a first circular droplet area image of an emulsion, and (C) a second circular droplet area image of an emulsion. [Diagram 5] FIG. 5 shows (A) an initial circular droplet area image of an emulsion, (B) a first circular droplet area image of an emulsion, and (C) a second circular droplet area image of an emulsion. [Figure 6] FIG. 6 is a reference diagram for explaining a method for calculating the overlap rate of the circular droplet regions. [Figure 7]FIG. 7 is a flow chart of the image analysis method and computer program for image analysis of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0020] Image analysis device FIG. 1 is a diagram showing the hardware configuration of an image analysis device according to the present invention.

[0021] As shown in FIG. 1, an image analysis device 10 of the present invention may have an external communication interface 11, a memory 12, and a processor 13.

[0022] The communication interface 11, the memory 12 and the processor 13 may each be configured as separate circuits, or may be integrated into a single integrated circuit.

[0023] The communication interface 11 has an interface circuit for connecting the image analysis device 10 of the present invention to other devices. The communication interface 11 can transfer images of the emulsion present in the emulsion holding unit, particularly fluorescent images, to the processor 13 via a network. The communication interface 11 can also output images such as the initial image generating unit, the first circular droplet region image, and the second circular droplet region image received from the processor 23 to an external device such as a display.

[0024] The memory 12 is an example of a storage unit and includes, for example, a volatile semiconductor memory and a non-volatile semiconductor memory. The memory 12 can store various data used in the image processing executed by the processor 13.

[0025] The processor 13 has one or more CPUs (Central Processing Units) and their peripheral circuits. The processor 13 may further have other arithmetic circuits such as a logic arithmetic unit, a numerical arithmetic unit, or a graphic processing unit. The processor 13 executes image processing.

[0026] 2 is an example of a functional block diagram of the processor 13 related to image processing. The processor 13 may have an initial image generating unit 21, a first processed image generating unit 22, and a second processed image generating unit 23. The processor 13 may also optionally have a third image generating unit 24, a further image generating unit (not shown), and a measurement unit 25. Each of these units included in the processor 13 may be, for example, a functional module realized by a computer program running on the processor 13. Alternatively, each of these units included in the processor 13 may be a dedicated arithmetic circuit provided in the processor 13.

[0027] <Initial Image Generation Unit> The initial image generating section detects a plurality of circular droplet regions in the emulsion based on an image of the emulsion present in the emulsion holding section and the brightness of a fluorescent image of the emulsion present in the emulsion holding section, and generates an initial image of the circular droplet regions.

[0028] The generation of the image can be done with any camera capable of acquiring an image of the emulsion, in particular a fluorescence image of the emulsion. Figure 3(A) shows a fluorescence image of the emulsion thus generated.

[0029] The initial image generating unit detects a plurality of circular droplet regions in the emulsion based on the brightness and darkness of such an image, and generates an initial circular droplet region image.

[0030] Here, as shown in FIG. 3, a fluorescent image (FIG. 3(A)) is binarized to obtain a binarized image (FIG. 3(B)), and a generalized Hough transform is performed on a set of edge pixels of the binarized image to determine the center and particle size of the circular droplet region, thereby detecting multiple circular droplet regions in the emulsion and generating an initial circular droplet region image (FIG. 3(C)). In determining the center and particle size of the circular droplet region, as shown in FIG. 3(D), a number is assigned to each pixel according to the distance from the edge region E, and the pixel with the highest number among the surrounding pixels is determined as the center of the circular droplet region, and the number of this pixel is determined as the particle size (radius) of the circular droplet region.

[0031] First processed image generating unit As shown in Figures 4 and 5, the first processed image generation unit calculates the overlap rate with other circular droplet regions for each of the multiple circular droplet regions of a first particle size range in the initial circular droplet region image (Figures 4(A) and 5(A)), and removes from the initial circular droplet region image those circular droplet regions of the first particle size range whose calculated overlap rate is equal to or greater than a first predetermined overlap rate, to generate a first circular droplet region image (Figures 4(B) and 5(B)).

[0032] This overlap rate can be calculated, for example, as shown in Figure 6. Here, the overlapping area between the circular areas is the area shown by diagonal lines in Figure 6(A). In the embodiment of Figure 6(A), if the center-to-center distance (Rc) and radii (r1, r2) of the two circles are known, the magnitudes of x and y in Figure 6(A) can be found from the following formula (1), and therefore the overlapping area can be calculated: y 2 =r1 2 -(Rc-x) 2 =r2 2 -x 2 … Formula (1)

[0033] Specifically, in the embodiment shown in FIG. 6(A), the area of ​​the left side portion of the overlapping region (the region shown by diagonal lines in FIG. 6(B)) can be calculated by the following formula (2): The left part of the overlapping area = Area of ​​the sector of the central angle (2θ) – Area of ​​a triangle with base x and height y × 2 =r2 2 ×π×2θ / 2π-x×y / 2×2 =r2 2 ×tan -1 (y / x)-x×y … Equation (2)

[0034] The area of ​​the right side of the overlapping region can be determined in the same manner as the area of ​​the left side of the overlapping region.

[0035] For the present invention, the "overlap ratio" of each circular droplet region of a first size range with other circular droplet regions can be determined as the ratio of said overlapping area to the area of ​​each of the circular droplet regions.

[0036] Furthermore, as shown in FIG. 6C, when one of the circular droplet regions in the first particle size range overlaps with a plurality of other circular droplet regions, the “overlap rate” can be calculated as (a) and / or (b) below: (a) The maximum overlap rate with any one of the other circular droplet regions; (b) The integrated overlap rate with all other circular droplet areas.

[0037] When the "overlap rate" is determined as the maximum value of the overlap rates with any one of the other circular droplet regions (a), the overlap rates between one of the circular droplet regions in the first particle size range and each of the multiple other circular droplet regions are determined, and the maximum value among them can be adopted as the "overlap rate" to be used in at least one of the first processing image generation unit, the second processing image generation unit, and the third image generation unit.

[0038] In addition, when the "overlap rate" is calculated as the integrated value of the overlap rates with all other circular droplet regions (b), the ratio of the total overlap area (weighted overlap area) to the area of ​​each circular droplet region is calculated, and the ratio can be used as the "overlap rate" to be used in at least one of the first processed image generating unit, the second processed image generating unit, and the third image generating unit. In this case, the "overlap rate" may exceed 100%.

[0039] <Second processed image generating unit> As shown in Figures 4 and 5, the second processed image generation unit calculates the overlap rate with other circular droplet regions for each of the multiple circular droplet regions in the second particle size range in the first circular droplet region image (Figures 4(B) and 5(B)), and removes from the first circular droplet region image those circular droplet regions in the second particle size range whose calculated overlap rate is equal to or greater than a second predetermined overlap rate, thereby generating a second circular droplet region image (Figures 4(C) and 5(C)).

[0040] Here, the overlap rate can be calculated, for example, as described with respect to the first processed image generating unit.

[0041] The second particle size range is a particle size range larger than the first particle size range, and the second overlap rate is smaller than the first overlap rate. That is, the second processed image generating unit is for performing processing to remove, from the circular droplet region image, the circular droplet region having a particle size range larger than the circular droplet region removed by the first processed image generating unit.

[0042] In the image processing device of the present invention, by performing a process of removing the circular droplet area images in order, starting with the relatively small circular droplet areas, an image of the circular droplet area that corresponds to the actual droplet area can be efficiently obtained.

[0043] Specifically, for example, the first to fifth particle size ranges and the first to fifth overlap rates (maximum overlap rate and weighted overlap rate) used in the above-mentioned first processed image generating unit, the second processed image generating unit, the below-mentioned third processed image generating unit, the below-mentioned fourth processed image generating unit, and the below-mentioned fifth processed image generating unit can be set as follows:

[0044] [Table 1]

[0045] <Third processed image generating unit> The third processed image generation unit calculates an overlap rate with other circular droplet regions for each of the multiple circular droplet regions in a third particle size range among the second circular droplet region image, and removes from the second circular droplet region image those circular droplet regions in the third particle size range whose calculated overlap rate is equal to or greater than a third predetermined overlap rate, thereby generating a third circular droplet region image.

[0046] Here, the overlap rate can be calculated, for example, as described with respect to the first processed image generating unit.

[0047] The third particle size range is a particle size range larger than the second particle size range, and the third overlap rate is smaller than the second overlap rate. That is, the third processed image generating unit is for performing processing to remove, from the circular droplet region image, the circular droplet regions having a particle size range larger than the circular droplet regions removed by the second processed image generating unit.

[0048] Further Processing Image Generation Unit The image processing device of the present invention has further processed image generation units, such as a fourth processed image generation unit, a fifth processed image generation unit, a sixth processed image generation unit, a seventh processed image generation unit, etc., and can process, in a similar manner to the second and third processed image generation units, to sequentially remove circular droplet regions in a relatively large particle size range from the circular droplet region image.

[0049] <Measurement section> The image processing device of the present invention may further have a measurement unit, in which the presence or absence of a target molecule signal is measured for the circular droplet region of the second circular droplet region image, the third circular droplet region image, or the counting circular droplet region image obtained by further performing any image processing on them, and the percentage of the circular droplet region having the target molecule signal can be determined.

[0050] Thus, by determining the percentage of the circular droplet area that has a target molecule signal, absolute quantification of the number of target molecules can be performed.

[0051] "Image analysis method and computer program for image analysis" The image analysis method and computer program for image analysis of the present invention are an image analysis method and computer program for image analysis that analyze an image of an emulsion present in an emulsion holding portion of a microchannel chip, by causing a processor, particularly a computer having a processor, to execute processing including specified steps.

[0052] FIG. 7 is a flow chart of the image analysis method and computer program for image analysis of the present invention.

[0053] As shown in FIG. 7, in the method and program of the present invention, first, after the start of processing (S70), a number of circular droplet regions in the emulsion are detected based on the brightness of the image of the emulsion present in the emulsion holding section, and an initial circular droplet region image is generated (S71).

[0054] Then, for each of the multiple circular droplet regions in the initial circular droplet region image, an overlap rate with other circular droplet regions is calculated, and the circular droplet regions in the first particle size range whose calculated overlap rate is equal to or greater than a first predetermined overlap rate are removed from the initial circular droplet region image to generate a first circular droplet region image (S72).

[0055] Then, for each of the circular droplet regions in the second particle size range among the multiple circular droplet regions in the first circular droplet region image, an overlap rate with other circular droplet regions is calculated, and the circular droplet regions in the second particle size range whose calculated overlap rate is equal to or greater than a second predetermined overlap rate are removed from the first circular droplet region image to generate a second circular droplet region image (S73). Here, the second particle size range is a particle size range larger than the first particle size range, and the second overlap rate is smaller than the first overlap rate.

[0056] The method and program of the present invention can then optionally repeat the same process as above for a larger particle size range and a smaller overlap rate to generate a third circular droplet area image, a fourth circular droplet area image, a fifth circular droplet area image, etc.

[0057] In addition, the method and program of the present invention can then optionally measure the presence or absence of a target molecule signal in the circular droplet area of ​​the second circular droplet area image, the third circular droplet area image, or the counting circular droplet area image obtained by further performing any image processing on them, and determine the percentage of the circular droplet area having a target molecule signal.

[0058] For details of the method and program of the present invention, please refer to the description of the apparatus of the present invention.

[0059] The computer program of the present invention can be provided in a form recorded on a computer-readable portable recording medium such as a semiconductor memory, a magnetic recording medium, or an optical recording medium.

[0060] Those skilled in the art may make various modifications to the embodiments within the scope of the present invention.

Claims

1. An image analyzer for analyzing an image of an emulsion present in an emulsion holding portion of a microchannel chip, comprising: an initial image generating unit that detects a plurality of circular droplet regions in the emulsion based on the brightness of an image of the emulsion present in the emulsion holding unit, and generates an initial image of the circular droplet regions; a first processed image generating unit that calculates an overlap rate with other circular droplet regions for each of the circular droplet regions in a first particle size range among the plurality of circular droplet regions in the initial circular droplet region image, and removes from the initial circular droplet region image the circular droplet regions in the first particle size range whose calculated overlap rate is equal to or greater than a first predetermined overlap rate, thereby generating a first circular droplet region image; a second processing image generation unit that calculates an overlap rate with other circular droplet regions for each of the circular droplet regions in a second particle size range among the multiple circular droplet regions of the first circular droplet region image, and removes from the first circular droplet region image the circular droplet regions in the second particle size range whose calculated overlap rate is equal to or greater than a second predetermined overlap rate, thereby generating a second circular droplet region image, wherein the second particle size range is a particle size range larger than the first particle size range, and the second overlap rate is smaller than the first overlap rate.

2. 10. The apparatus of claim 1 further comprising: a third image generating unit that calculates an overlap rate with other circular droplet regions for each of the circular droplet regions in a third particle size range among the multiple circular droplet regions of the second circular droplet region image, and removes from the second circular droplet region image the circular droplet regions in the third particle size range whose calculated overlap rate is equal to or greater than a third predetermined overlap rate, thereby generating a third circular droplet region image, wherein the third particle size range is a particle size range larger than the second particle size range, and the third overlap rate is smaller than the second overlap rate.

3. 3. The apparatus of claim 1, wherein the initial image generating unit binarizes the image to obtain a binarized image, and determines a center and a particle size of the circular droplet region by performing a generalized Hough transform on a set of edge pixels of the binarized image, thereby detecting a plurality of circular droplet regions in the emulsion and generating the initial circular droplet region image.

4. 3. The apparatus according to claim 1, wherein the overlap rate used in at least one of the first processed image generating unit, the second processed image generating unit, and the third image generating unit is the following (a) and / or (b): (a) the maximum value of the overlapping rate with any one of the other circular droplet regions; (b) The integrated value of the overlap rate with all the other circular droplet regions.

5. The device according to claim 1 or 2, further comprising a measurement unit that measures the presence or absence of a target molecule signal in the circular droplet region of the second circular droplet region image, the third circular droplet region image, or a circular droplet region image for counting obtained by further performing any image processing on them, and determines the proportion of the circular droplet region having the target molecule signal.

6. An image analysis method for analyzing an image of an emulsion present in an emulsion holding portion of a microchannel chip, the method comprising the steps of: detecting a plurality of circular droplet regions in the emulsion based on the brightness of the image of the emulsion present in the emulsion holding portion, and generating an initial circular droplet region image; calculating an overlap rate with other circular droplet regions for each of the circular droplet regions in a first size range among the plurality of circular droplet regions in the initial circular droplet region image, and removing the circular droplet regions in the first size range whose calculated overlap rate is equal to or greater than a first predetermined overlap rate from the initial circular droplet region image to generate a first circular droplet region image; Calculating an overlap rate with other circular droplet regions for each of the circular droplet regions in a second particle size range among the plurality of circular droplet regions in the first circular droplet region image, and removing from the first circular droplet region image those circular droplet regions in the second particle size range whose calculated overlap rate is equal to or greater than a second predetermined overlap rate, thereby generating a second circular droplet region image, wherein the second particle size range is a particle size range larger than the first particle size range, and the second overlap rate is smaller than the first overlap rate.

7. A computer program for image analysis that analyzes an image of an emulsion present in an emulsion holding portion of a microchannel chip, the computer program causing a processor to execute a process including the steps of: detecting a plurality of circular droplet regions in the emulsion based on the brightness of the image of the emulsion present in the emulsion holding portion, and generating an initial circular droplet region image; calculating an overlap rate with other circular droplet regions for each of the circular droplet regions in a first size range among the plurality of circular droplet regions in the initial circular droplet region image, and removing the circular droplet regions in the first size range whose calculated overlap rate is equal to or greater than a first predetermined overlap rate from the initial circular droplet region image to generate a first circular droplet region image; Calculating an overlap rate with other circular droplet regions for each of the circular droplet regions in a second particle size range among the plurality of circular droplet regions in the first circular droplet region image, and removing from the first circular droplet region image those circular droplet regions in the second particle size range whose calculated overlap rate is equal to or greater than a second predetermined overlap rate, thereby generating a second circular droplet region image, wherein the second particle size range is a particle size range larger than the first particle size range, and the second overlap rate is smaller than the first overlap rate.

Images