JP5891719B2 - Cell cross-section analysis apparatus, cell cross-section analysis method, and cell cross-section analysis program - Google Patents

Description

  The present invention relates to a cell cross-section analysis technique, and particularly to a technique for detecting exfoliated cells by analyzing cross-sectional images of exfoliated cells.

  In the field of regenerative medicine, when self cells such as skin are proliferated and transplanted to the human body, or when stem cells such as ES cells and iPS cells are differentiated into specific functions such as the myocardium, etc. On the other hand, for planar tissues such as the cornea and skin, a method is used in which cells are grown in a sheet form and transplanted using a culture dish such as a petri dish. In this method, cells are seeded in a culture dish and cultured for a certain period, and a group of cells formed in a sheet shape is peeled off from a substrate such as a bottom plate of a petri dish and used for transplantation or the like.

As a method for accelerating the detachment of the cell group from the substrate, a method of detaching the cell group using a cell detachable drug, on the surface of the substrate coated with a temperature-responsive polymer whose cell adhesion changes depending on temperature conditions A method of detaching a cell group by culturing cells and controlling the temperature (Patent Document 1 below) is known. When a cell group is detached by applying exfoliation stimulus such as drug or temperature control, if there are more than a certain percentage of cells that remain attached without being detached from the substrate, the formed sheet shape is destroyed and can be used for transplantation. Disappear. Therefore, for quality control, it is required to non-invasively measure and manage the cell peeling performance when a peeling stimulus is applied to the surface of the cell culture substrate.
Therefore, analysis of a cross section viewed from the side of the cell, so-called cross section analysis is important.

  Possible methods for physically observing the cross-sectional state of cultured cells include inverting the microscope and observing the petri dish from the side, or submerging an ultra-small underwater camera in the petri dish and observing from the side. It is done. Since the former is difficult to realize with an ordinary circular petri dish, it is necessary to prepare and transfer a thin special square petri dish only at the time of observation, and there is a problem of invading cultured cells. On the other hand, with the latter ultra-small underwater camera, it is difficult to obtain a high-magnification image comparable to a microscope at present.

  In this way, it is difficult to realize a method of photographing the culture dish from the side direction. Therefore, the culture dish is sliced in the Z-axis direction, and multiple frames are photographed, and the culture dish is virtually observed from the side direction by image reconstruction. A method of indirectly creating a cross-sectional image can be considered.

As a basic technique for realizing such a method, a method for measuring a step in the depth direction by controlling the focus of an optical microscope is realized as follows.
(1) Z-direction measurement method using a general optical microscope Patent Document 2 “Step measurement device for fine pattern”
This method is a proposal of a method of performing three-dimensional measurement of a processing step of an electronic device by performing focus control for each photographing point using a general optical microscope.
(2) Patent Document 3 “Color Confocal Microscope”
This is a proposal for a three-dimensional measuring microscope suitable for level difference measurement.
When focus control is performed with an optical microscope, image blurring occurs. Therefore, a method has been proposed in which a step is measured with high accuracy in a three-dimensional direction while two-dimensionally scanning a light source in a beam shape.
(3) Patent Document 4 “Shooting Method Using Confocal Laser Scanning Microscope”
As a method for acquiring an object existing in a blind spot, an imaging method using a confocal laser scanning microscope using a light refraction effect has been proposed.
(4) Patent document 5 “Optical image reconstruction device”
Using a general optical microscope to control the focus and capture multiple images to obtain a cross-sectional image, the specimen is stored in a tubular container in order to obtain the shape of the blind spot part of the specimen. However, a technique for performing multi-slice imaging while rotating has been proposed.
(5) Patent Document 6 “Confocal laser scanning microscope and optical tomographic image display method thereof”
Using a laser scanning microscope, the objective lens is infiltrated into the culture solution, and the cross-sectional image of the acquired pixel unit is corrected in consideration of the optical refractive index of the culture solution and the sample. Proposal. A cross-sectional image can be acquired in real time for each one-dimensional scanning line.

Japanese Patent No. 3441530 JP-A-8-194737 Japanese Patent No. 4065129 JP-A-7-199073 JP-A-6-259533 JP-A-7-199073

  By the way, in the technique described in Patent Document 2, since each of the acquired slice images is out of focus, it has been difficult to apply to the use of imaging a cross-sectional image.

  Patent Document 3 is an improvement of Patent Document 2, and a confocal laser scanning microscope has been proposed. However, it is more expensive than an optical microscope, and using a transmissive light source like an optical microscope, There is a problem that a slice image on the back side that becomes a blind spot cannot be easily acquired.

  Patent Document 4 proposes a photographing method using a confocal laser scanning microscope, but has a problem that it is considerably different from a transmission image obtained by a normal optical microscope for cell observation.

  In patent document 5, it cannot apply to the floating object like a detached cell. In addition, although a method for acquiring a tomographic image using an optical microscope has been proposed, it is not applicable to cultured cells because the sample is imaged while the sample is fixed to a cylindrical container and the container is rotated.

  In Patent Document 6, since it is difficult to perform pixel-by-pixel correction with a normal optical microscope that is not a scanning type, it is difficult to apply. In addition, it is necessary to invade the culture medium, and there is a problem that noninvasive analysis is difficult.

In order to be able to virtually construct an image in a cross-sectional direction by photographing a normal culture dish with multi-slice in the depth direction using a normal optical microscope, there are the following problems.
(1) Most of the acquired slice images are out of focus. The in-focus state where the focus is achieved is a part, and most of the portions are not in focus, and the image becomes unclear.
(2) The resolution in the Z direction decreases due to the influence of the depth of field. Not only is the focus surface unclear, but also images above and below the Z axis are reflected.
(3) The image of the object in the direction opposite to the objective lens becomes a blind spot. It is impossible to observe the shape of the back side of the object or another object that overlaps the back side of the object.
An object of the present invention is to solve the above-mentioned problems and to observe a cross section of a cell with high accuracy.

  For the problem (1), a method of sharpening by image processing with sharpness correction is proposed.

  For the above problem (2), a method of subtracting a predetermined amount of the components of slice images adjacent in the vertical direction from each other is proposed for each slice image.

  For the above problem (3), a method is proposed in which each slice plane is imaged with a transmission light source, and for each slice image, a component of the slice image adjacent to the back side is enhanced by a predetermined amount by image processing. In other words, multi-slice imaging is performed by changing the depth direction by controlling the focus of the optical microscope using a transmission light source, and sharpening image processing is applied to each acquired slice image to improve defocusing. Do.

  Furthermore, a difference enhancement process is performed between adjacent slice images to improve the problem of a decrease in resolution in the Z direction due to the influence of the depth of field, and a blurred image entering the blind spot is enhanced. Pixels for constructing a horizontal or vertical tomographic image at a specified position are extracted from the multi-slice image corrected in this way.

  According to an aspect of the present invention, a system including cells and a liquid is moved in a first direction while being focused and observed with a microscope, and a plurality of in-focus images having different vertical positions are obtained. A plurality of two-dimensional planar images demarcated by a second direction that intersects the first direction and a third direction that intersects the first direction and the second direction, and imaging by the imaging mechanism An image correcting unit that performs a correction process for sharpening an image on each of the plurality of focused images, and the two-dimensional image based on the plurality of focused images corrected by the image correcting unit. An image processing unit that reconstructs a cross section in either the second direction or the third direction so as to connect the plurality of focused images and generates a pseudo cross-sectional image through an arbitrary point on the plane And a cell cross-section solution characterized by comprising Apparatus is provided.

  After sharpening correction is performed on the multi-slice image acquired by the imaging mechanism, the multi-slice image is joined in the moving direction, so that a clearer and more accurate pseudo-sectional image can be obtained.

  The imaging mechanism may repeatedly capture the focused image at a predetermined time interval to obtain a plurality of moving images, and the image processing unit may generate a pseudo sectional moving image based on the plurality of moving images. preferable. As the multi-slice image, a moving image obtained by repeatedly capturing the focused image at a predetermined time interval can be used to obtain a more accurate pseudo-sectional moving image.

The image correction unit creates a defocused image that is more out of focus than the plurality of focused images by an algorithm based on an unsharp mask as a process for sharpening the plurality of focused images. It is preferable to perform correction by adding a difference image between the focused image and the corresponding defocused image by a predetermined ratio.
By considering the difference image, a clearer image can be obtained.

  In addition to the sharpening process, the image correcting unit is adjacent to the plurality of focused images by one frame in the first direction in the vertical direction or in the direction opposite to the first direction. Each image is created with a difference image from the frame of the image or movie, and the difference image corresponding to the frame of the still image or movie is added to the frame of each still image or movie by a predetermined ratio. May be.

  By adding / subtracting the difference between the previous image and the next image in the first direction to the current image, it is possible to suppress the influence of reflection or the like.

  The imaging mechanism sets a Z1 position that is in focus with respect to the bottom plate surface of the culture dish that contains the culture medium as the liquid, and a Z2 position that is in focus with respect to the culture liquid level in the culture dish, A plurality of images with different positions in the Z direction are picked up by moving the focal point to a plurality of locations at predetermined intervals within the range between the Z1 position and the Z2 position by an optical microscope.

  For example, a plane including an arbitrary point on the two-dimensional plane designated by the operation unit and formed by the second direction and the first direction or the third direction and the first direction It is preferable to include a display control unit that displays on the display screen at least one of the pseudo cross-sectional images cut by the plane formed by the above-described method so as to correspond to the two-dimensional plane image. By displaying the positions of the pseudo cross-sectional image and the planar image in association with each other, it becomes possible to analyze the cells more easily.

  According to another aspect of the present invention, a system including cells and liquid is relatively focused in a first direction while being moved relatively in the first direction, and is observed with a plurality of different positions in the first direction. Imaging that captures a focal image as a plurality of two-dimensional planar images defined by a second direction that intersects the first direction and a third direction that intersects the first direction and the second direction. An image correcting step for performing a correction process for sharpening an image on each of the plurality of focused images captured by the imaging mechanism, and the plurality of focused images after being corrected by the image correcting unit. Based on an image, it passes through an arbitrary point on the two-dimensional plane and reconstructs a cross section in either the second direction or the third direction so as to connect the plurality of focused images. An image processing step for generating a cross-sectional image; Cells sectional analysis method characterized by having a are provided. “Focus with a microscope while relatively moving a system including cells and liquid in the first direction” means that at least one of the system and the microscope is moved.

  The present invention may be a cell cross-section analysis program for causing a computer to execute the cell cross-section image analysis method described above, or a computer-readable recording medium for recording the program.

  According to the present invention, it is possible to artificially observe a cross section of a cell that is physically difficult to observe from the lateral direction by generating a pseudo cross section image.

  In addition, by improving the accuracy of creating the cell cross-sectional image, it is possible to determine whether the cells in the cell culture container are adherent cells or detached cells, and to easily evaluate the cell peeling performance on the cell adhesion surface.

It is a figure which shows an example of the object of the analysis technique of the cell cross-sectional image by the 1st Embodiment of this invention. It is a figure which shows the principle of the cell image analysis technique by this Embodiment, and is a figure which shows the outline | summary of the multidimensional cell cross-section analysis method. FIG. 3 is a principle diagram of a multi-dimensional cell cross-sectional analysis method and corresponds to FIG. 2. It is a figure which shows the example of 1 structure of the cross-sectional image analysis system by this Embodiment, and is a figure which shows a required structure in addition to the structure shown in FIG. It is a functional block diagram which shows one structural example of the cross-sectional image analyzer by this Embodiment, and is a figure which shows the structure centering on the analysis mechanism among the structures shown in FIG. It is a flowchart figure which shows a flow in the process in the cross-sectional image analysis system by this Embodiment. It is a flowchart figure which shows the flow of the process which performs the process which increases a blur further by sharpness correction | amendment (unsharpness mask). It is a flowchart figure which shows the flow of the process which performs the process which increases a blur further by sharpness correction | amendment (unsharpness mask). It is a flowchart figure which shows the flow of the process which performs the process which increases a blur further by sharpness correction | amendment (unsharpness mask). It is a flowchart figure which shows the detail of the flow of a difference emphasis process with the Z direction adjacent image of step S6. It is a flowchart figure which shows the detailed flow of the difference emphasis process with respect to each still image. It is a figure which shows an example of the microscope image before correction | amendment, and is a figure which shows the image from Z = 1 to 62. It is a figure which shows the image after the correction | amendment corresponding to FIG. The corrected planar image (Z = 30) shown in FIG. 13 and the cross-sectional structure cut by the horizontal straight line L1 and the vertical straight line L2 are obtained by analyzing the central portion and the above-described image processing. Is shown on the lower side and the right side. It is a principle figure of the multidimensional cell cross-section analysis method by the 2nd Embodiment of this invention, and is a figure corresponding to FIG. It is a flowchart figure which shows the flow of a process of the multidimensional cell cross-section analysis method by this Embodiment.

Hereinafter, terms in the present embodiment will be defined.
In the present invention, the “system containing cells and liquid” refers to a system containing cells and a fluid such as a liquid that is a continuous phase around the cells. The cells of the system may include cells that are free in the liquid, ie, free cells, and cells that are fixed in position within the system, ie, fixed cells. The “system containing cells and liquid” is typically a cell-culture solution system containing cells and culture solution contained in a cell culture instrument. In the cell-culture system, cells existing in a state adhered to the surface of the cell culture instrument correspond to fixed cells, and cells existing in a state released in the culture medium correspond to free cells. In one embodiment of the present invention, the cell-culture system is a state in which cells are cultured together with the culture medium in a cell culture device and then a peeling stimulus is applied to detach the cells from the cell adhesion surface of the device. Can be used.

  The cell culture device can be any shape having a portion capable of holding cells and culture fluid. Typically, a dish-shaped cell culture device, a multi-well plate in which a plurality of recesses capable of containing cells and a culture solution are connected, a plate-shaped or film-shaped cell culture device, a plate shape with a recess formed on the surface Or a film-shaped cell culture instrument etc. are mentioned.

  At least a portion of the cell culture device capable of holding the cells and culture medium, typically the bottom surface, forms the cell adhesion surface.

  Examples of the material constituting the cell culture instrument include materials such as glass, plastics, and ceramics that are usually used for cell culture, but are not limited to these as long as they are materials that allow cell culture and cell observation. For example, it can be composed of synthetic resin such as polyester, polyethylene, polypropylene, polystyrene, polycarbonate, polyvinyl chloride, polytetrafluoroethylene, hydroxyapatite ceramics, alumina ceramics, glass or the like.

  In addition, it is also possible to perform evaluation without using a container in a general sense using a culture solution itself such as a droplet containing cells. In addition, as long as it is an instrument which can hold | maintain the system containing the system containing a cell and a liquid, what kind of form may be sufficient and what is necessary is just a structure which can hold | maintain a droplet instead of a container.

  In the present specification, the cell adhesion surface to be evaluated for cell detachment performance may be processed according to the purpose of evaluation. Typically, the cell adhesion surface is coated with a substance capable of changing the cell adhesion by a predetermined stimulus, such as a temperature responsive polymer, a pH responsive polymer, an ion responsive polymer, etc. ing. The surface on which cell detachment performance is to be evaluated is not limited to the viewpoint of “detachment”, and may be, for example, a surface that is coated with a hydrophilic polymer to reduce cell adhesion and is controlled so that cells do not adhere. .

  The temperature-responsive polymer can be appropriately selected from those usable for cell culture according to the purpose of the test. Typical temperature-responsive polymers include poly-N-isopropylacrylamide (temperature changing from hydrophobic to hydrophilic (critical solution temperature (T) in water) = 32 ° C.), poly-Nn-propylacrylamide ( T = 21 ° C.), poly-Nn-propylmethacrylamide (T = 32 ° C.), poly-N-ethoxyethylacrylamide (T = about 35 ° C.), poly-N-tetrahydrofurfurylacrylamide (T = about 28) ° C), poly-N-tetrahydrofurfuryl methacrylamide (T = about 35 ° C.), poly-N, N-diethylacrylamide (T = 32 ° C.) and the like.

  The pH-responsive polymer and the ion-responsive polymer can be appropriately selected from those usable for cell culture according to the purpose of the test.

  As the peeling stimulus, when the above-described substance is used for coating the cell adhesion surface, a predetermined peeling stimulus such as a temperature change, a pH change, an ion concentration change and the like can be employed depending on the substance to be used.

  As the peeling stimulus, other stimuli may be employed depending on the purpose of the evaluation. For example, it can be appropriately selected from chemical stimulation by a cell detachment drug, mechanical stimulation by the flow of a culture solution, and the like.

  The type of cell analyzed by the present invention is not particularly limited and may be appropriately selected according to the purpose of the test. For example, the technique of the present invention in a test using various cells collected from warm-blooded animals such as humans, mice, rats, guinea pigs, hamsters, chickens, rabbits, pigs, sheep, cows, horses, dogs, cats, monkeys, etc. Can be applied. Examples of the warm-blooded animal cells include keratinocytes, spleen cells, neurons, glial cells, pancreatic β cells, mesangial cells, Langerhans cells, epidermal cells, epithelial cells, endothelial cells, fibroblasts, fiber cells, Muscle cells, adipocytes, synoviocytes, chondrocytes, bone cells, osteoblasts, osteoclasts, breast cells, hepatocytes or stromal cells, or precursor cells of these cells, stem cells or adhesion-dependent cancer cells Is mentioned. Embryonic stem cells can also be used.

  In this embodiment, a case where a video photographed by a video camera is used as an example of a moving image will be described. Of course, still images may be taken continuously or intermittently as moving images, and the types of images are not limited. The moving image may be taken with a video camera or the like, or may be obtained with a still camera or the like every predetermined time.

  In the following embodiments, a technique for detecting exfoliated cells (substrate exfoliated cells) in cells by image processing using a culture dish containing cells and a culture solution as a cell culture instrument will be described as an example.

  Hereinafter, a cell image analysis technique according to an embodiment of the present invention, particularly a cell image analysis apparatus for detecting detached cells will be described with reference to the drawings.

  In this specification, the pseudo cross-sectional image refers to an image obtained by the cell cross-sectional analysis processing according to the present embodiment. Although it is not an image obtained by directly observing a cross section, an image that is substantially close to a cross-sectional image is obtained as described later.

  FIG. 1 is a diagram showing an example of an object of a cell cross-sectional image analysis technique according to the first embodiment of the present invention. As shown in FIG. 1, after the temperature is lowered, in the culture solution 3 in the culture dish 1 in which the predetermined cell group is cultured with the culture solution, among the predetermined cell group, the adherent cell 7 is The exfoliated cells 5 are floating in the culture solution 3 and are stationary in the culture solution 3 due to the frictional force of the culture solution 3.

  By observing the cross section of the cell from the side surface of the culture dish 1, the detached cell 5 and the adherent cell 7 can be distinguished. As a technique for observing in this state, there is a method of taking a close-up image from the side by, for example, enclosing an ultra-small underwater camera in the culture dish. However, this method is not practical because it is difficult to obtain the magnification necessary for cell observation and it is necessary to observe from a side using a special apparatus with a special technique.

  In addition, there is a method in which the microscope barrel is inverted and the culture dish is photographed from the side, but it is actually difficult to observe through a circular petri dish.

  FIG. 2 is a diagram showing the principle of the cell image analysis technique according to this embodiment, and is a diagram showing an outline of the multidimensional cell cross-sectional analysis method. This method performs cross-sectional analysis by performing multi-slice imaging by controlling the focus of a microscope. As shown in FIG. 2, a cell culture dish 1 containing cells 5 and 7 in a culture solution 3 is placed on an XYZ stage 11, irradiated with light from a transmission light source 15 from above, and an objective lens from below. The cells are observed with a light microscope having 17.

  Here, from the position of the top surface of the bottom plate 1b of the culture dish to the position of the water surface 3a of the culture solution, the focal surfaces 1 to N (N is an integer of 2 or more) which are different positions in the Z direction, which is the normal direction of the water surface 3a. Until the observation surface of the optical microscope is taken with a digital video camera. Here, a plurality of still images from at least 1 to N are acquired. Then, based on the acquired plurality of still images, a cross-sectional image in the depth direction in the horizontal direction and the vertical direction at a specified position in the still image is created.

  For this purpose, the culture dish 1 is magnified to a predetermined magnification and observed with an optical microscope. At this time, in the optical microscope, for example, a plurality of focused still images with different positions in the Z direction are obtained by focusing on the focal plane 1 to the focal plane N while moving to a plurality of locations at intervals equal to N−1. Take an image. Here, the range of movement in the Z direction is between the upper surface of the bottom plate 1b and the water surface 3a of the culture solution 3, and is at least within the range in which cells are included. Assuming that the Z direction is the first direction, the X direction is the second direction, and the Y direction is the third direction, the first direction and the second direction are orthogonal to each other, and both the first direction and the third direction are It is orthogonal and the second direction and the third direction are also orthogonal.

  FIG. 3 is a principle diagram of the multidimensional cell cross-sectional analysis method, and corresponds to FIG. In the focused still image, N = 5, and imaging slice images from 1 to 5 spread on the XY plane are obtained. This imaging slice image is represented by the following equation.

x=0,…,Xs-1; y=0,…,Ys-1; z=0,…,Zs-1; c=0,1,2(RGB)

x = 0, ..., Xs-1; y = 0, ..., Ys-1; z = 0, ..., Zs-1; c = 0,1,2 (RGB)

  Here, the reconstructed cross-sectional image in the X direction obtained by cutting the imaging slice image along the XZ plane indicated by reference numeral S1 is represented by the following expression.

  On the other hand, a reconstructed cross-sectional image in the Y direction obtained by cutting the imaging slice image along the YZ plane indicated by reference numeral S2 is represented by the following expression.

  Based on the above principle, a desired cross-sectional image in the X direction, the Y direction, or the like can be reconstructed.

  By increasing N, it is possible to reconstruct a higher-definition cross-sectional image and obtain a pseudo cross-sectional image.

  FIG. 4 is a diagram showing a configuration example of the cross-sectional image analysis system according to the present embodiment, and is a diagram showing a necessary configuration in addition to the configuration shown in FIG.

  As shown in FIG. 4, the cross-sectional image analysis system A is provided at an XYZ stage 11 that enables the culture dish 1 to move in the XYZ directions, and at a position where the cells in the culture dish 1 are irradiated with light. A transmission light source 15 and a moving image capturing video camera 23 provided with an objective lens 17 at a position where the cells irradiated with the transmission light source can be observed are provided. The video camera 23 can be replaced with a digital still camera that captures still images.

Furthermore, an XYZ stage controller / autofocus unit 21 that performs movement control and position control of the XYZ stage 11, a video frame memory 25 that stores moving image data shot by the moving image shooting video camera 23, and an XYZ stage controller / autofocus. A control personal computer 27 that controls the unit 21, the video frame memory 25, and the like, and a video monitor 31 that can observe an image are provided.
Although the system for photographing cells from below has been described here, the cells may be photographed from above.

  FIG. 5 is a functional block diagram showing a configuration example of the cross-sectional image analysis apparatus according to the present embodiment, and is a diagram showing a configuration around the analysis mechanism in the configuration shown in FIG. As shown in FIG. 5, the cross-sectional image analysis apparatus according to the present embodiment includes a microscope 51 that observes cells, an imaging unit 53 that captures an observation image of the microscope, and a moving mechanism that moves the objective lens of the culture dish or microscope. 55, a sharpness correction unit 57 that is a first image correction unit that performs correction processing of an image that has been moved by a moving mechanism, a difference enhancement processing unit 61 that is a second image correction unit, and a cross-sectional image. A cross-sectional image reconstruction unit 62 that reconstructs and obtains a pseudo cross-sectional image, a cross-sectional image display processing unit 63, an output unit 65, an operation unit 67, and a memory 68 are provided. The microscope 51, the imaging unit 53, and the moving mechanism 55 constitute an imaging mechanism 50.

  The microscope 51 is an optical microscope and has an enlargement function and a focusing function. The imaging unit 53 is, for example, the moving image imaging video camera of FIG. 4, and the moving mechanism 55 is the XYZ stage controller 21 and the autofocus unit 21. A sharpness correction unit 57 that is a first image correction unit, a difference enhancement processing unit 61 that is a second image correction unit, a cross-sectional image reconstruction unit 62, and a cross-sectional image display processing unit 63 include a control personal computer. 27 corresponds to a control processing unit according to the control program in 27. The output unit 65 corresponds to the video monitor 31 or a printer, and the operation unit 67 corresponds to a mouse, a keyboard, or the like. The memory 68 is a storage medium such as a hard disk, and stores obtained data and the like.

  FIG. 6 is a flowchart showing a flow of processing in the cross-sectional image analysis system according to the present embodiment. The processing content includes processing in the microscope body, image processing, and interactive display of cross-sectional images. First, in step S1, the microscope 51 is set up. More specifically, the focus is set to an appropriate field of view and position so that cells in the culture dish can be observed. In step S2, a still image is captured by the imaging unit 53, and the captured image is stored in the memory 68 or a storage medium.

In step S3, the moving mechanism 55 moves the XYZ stage stepwise. In step S1, since the focal plane is initially set on the surface of the bottom plate 1b, the focal plane is moved in the direction of the water surface 3a from the surface of the bottom plate 1b. In step S4, it is determined whether or not the focal plane is within the water surface. This determination can be made based on position information in the Z direction. If Yes in step S4, the process returns to step S2 to perform the next shooting. Finally, in step S4, when the focal plane exceeds the water surface (No), the process proceeds to step S5. Thus, in the culture dish 1, the Z1 position where the focal point is on the surface of the bottom plate 1b and the Z2 position where the focal point is on the water surface 3a in the culture dish 1 are set. A plurality of still images with different positions in the Z direction are captured while being moved to a plurality of locations at predetermined intervals within a range between the Z1 position and the Z2 position.
In step S5, the sharpness correction unit 57 corrects the sharpness using an unsharp mask.

FIG. 7 to FIG. 9 are flowcharts showing the flow of processing for further blurring processing by sharpness correction (unsharpness mask). In FIG. 7, a correction target voxel image group is prepared in step S5-1. That is,

  In step S5-2, z = 0 is set. In step S5-3, a defocus image Image2 (x, y, c) is created based on Image (z, x, y, c).

  In step S5-4, sharpness correction is performed by adding a difference image between the original image Image (z, x, y, c) and the out-of-focus image Image2 (x, y, c) to the original image. In step S5-5, z is incremented by 1. In step S5-6, z and Zs-1 are compared. If z is smaller, the process returns to step S5-3 to continue the processing. If it becomes above, a process will be complete | finished (End: step S5-7). Here, Zs is the distance between the bottom surface 1b and the water surface 3a.

  FIG. 8 is a flowchart showing the flow of the process of creating a defocus image (step S5-3). First, y = 0 is set in step S5-3-1, x = 0 is set in step S5-3-2, and c = 0 is set in step S5-3-3. c = 0 is, for example, R of RGB. In step S5-3-4, J = −r, vs = 0, and ws = 0.

  In step S5-3-5, i = −r. Here, r is a blur radius and is expressed in units of one pixel, for example.

Next, in step S5-3-6,

  Next, i is incremented by 1 in step S5-3-7, and i and r are compared in step S5-3-8. If i is equal to or smaller than r, the process returns to step S5-3-6. If i is larger than r, the process proceeds to step S5-3-9, and j is incremented by one. Next, in step S5-3-10, j and r are compared. If j is equal to or less than r, the process returns to step S5-3-5. If j is greater than r, in step S5-3-11. The following calculation is performed.

  Next, in step S5-3-12, c is incremented by 1. In step S5-3-13, c and 2 are compared. If c is 2 or more, the process returns to step S5-3-4. If x is 2 or less, x is incremented by 1 in step S5-3-14. Next, in step S5-3-15, x and Xs are compared. If x is smaller than Xs, the process returns to step S5-3-3, and if x is greater than or equal to Xs, the process proceeds to step S5-3-16. Increment y by 1. Next, in step S5-3-17, y and Ys are compared. If y is smaller than Ys, the process returns to step S5-3-2, and if y is equal to or greater than Ys, the process ends (End). . With the above processing, an image with a defocus can be created.

  As this sharpening process, first, each of the defocused images that are out of focus from each still image is created by an algorithm based on an unsharp mask.

  Next, each corrected image or each corrected moving image is created by adding a difference image between the still image and the corresponding defocused image to each still image by a predetermined ratio.

Returning to FIG. 6, in step S <b> 6, difference enhancement processing is performed on adjacent pixels in the Z direction.
Difference images from still images adjacent to each still image by +1 in the Z direction are respectively created, and a difference image corresponding to the still image is added to each still image by a predetermined ratio.

FIG. 9 is a flowchart showing the flow of the difference image addition process (step S5-4) between the defocused image and the original image.
First, y = 0 is set in step S5-4-1, x = 0 is set in step S5-4-2, and c = 0 is set in step S5-4-3. Next, in step S5-4-4, the difference d is obtained.

  In step S5-4-5, the absolute value of d is compared with Sl. Here, S1 is a threshold value of the addition process, and is “0”, for example. If the absolute value of d is greater than or equal to S1, the process proceeds to step S5-4-6. If the absolute value of d is smaller than S1, the process proceeds to step S5-4-7. In step S5-4-6, the following processing is performed.

  Here, Sv is an addition processing level (real value), for example, 1.0.

  Next, in step S5-4-7, c is incremented by 1, and c and 2 are compared in step S5-4-8. If c is greater than 2, the process returns to step S5-4-4, and if c is 2 or less, the process proceeds to step S5-4-9. Next, in step S5-4-10, x and Xs are compared. If x is smaller than Xs, the process returns to step S5-4-3. If x is greater than or equal to Xs, the process proceeds to step S5-4-11, and y is incremented by one. Next, in step S5-4-12, y and Ys are compared. If y is smaller than Ys, the process returns to step S5-4-2. If y is equal to or greater than Ys, the process is terminated (step S5- 4-13: End). With the above processing, a sharp image can be obtained.

Next, details of step S6 will be described.
Here, in addition to the sharpening process, the image processing means creates a difference image of each still image or each moving image frame that is adjacent to the still image or moving image frame by +1 in the Z direction. A corrected image is created by adding the difference image corresponding to the still image by a predetermined ratio. Instead of the above processing, a difference image is created for each still image from a still image adjacent to the still image by −1 in the Z direction, and a difference image corresponding to the still image is predetermined for each still image. It is also possible to create a corrected image by adding only the ratio.

  FIG. 10 is a flowchart showing details of the flow of the difference enhancement process with the Z-direction adjacent image in step S6. In step S6-1, Image (z, x, y, c) that is a correction target voxel image and Image2 (z, x, y, c) that is a copy voxel image are defined, and z = 0 , ..., Zs-1; x = 0, ..., Xs-1; y = 0, ..., Ys-1; c = 0,1,2 (RGB).

  In step S6-2, z = 1. In step S6-3, the difference image between the target image Image2 (z, x, y, c) and the adjacent image Image2 (z-1, x, y, c), and the target image Image2 (z, x, y, c) ) And an adjacent image Image2 (z + 1, x, y, c) in the z direction are added to the original image Image (z, x, y, c). In step S6-4, z is incremented by 1. In step S6-5, z and Zs-1 are compared. If the former is small, the process returns to step S6-3, and z is greater than or equal to Zs-1. If there is, the process ends (step S6-6). As described above, the difference between the previous image and the next image in the z direction is obtained and added to the target image, thereby reducing the influence of reflection in the image.

  FIG. 11 is a flowchart showing a detailed flow of difference enhancement processing for each still image. First, in step S6-3-1, y = 0 is set, and in step S6-3-2, x = 0 is set. In step S6-3-3, c = 0 is set. In step S6-3-4, the following calculation is performed.

  Here, Sb and Sf are levels from the addition ns, and are real values, for example, 1.0.

  Next, in step S6-3-5, c is incremented by 1. In step S6-3, c and 2 are compared. If c is larger than 2, the process returns to step S6-3-4. If c is 2 or less, the process proceeds to step S6-3-7, and x is set to 1. Increment. In step S6-3-8, x and Xs are compared. If x is smaller than Xs, the process returns to step S6-3-3. If x is equal to or greater than Xs, in step S6-3-9. Increment y by 1.

Next, in step S6-3-10, y and Ys are compared. If y is smaller than Ys, the process returns to step S6-3-2, and if y is equal to or greater than Ys, the process is terminated (step S6). -3-11).
With the above processing, it is possible to perform processing for emphasizing a difference with respect to each still image.

  Next, returning to FIG. 6 again, in step S7 and subsequent steps, interactive display processing of the cross-sectional image is performed. First, in step S7, the display position of the cross-sectional image on the screen is designated. Next, in step S8, a pseudo cross-sectional image is created by stitching the corrected images obtained in step S6 as reconstructed images in the X and Y directions in the Z direction. Next, in step S9, the pseudo cross-sectional image is output to the output unit 65 as the reconstructed image in each of the X and Y directions created in step S8. In step S10, if another position is desired to be displayed (Yes), the process returns to step S7. If not (No), the process ends in step S11 (End).

  In step S9, a plurality of corrected images having different positions in the Z direction to output the pseudo cross-sectional image to the output unit 65 as the reconstructed images in the X and Y directions created in step S8, one selected Z As an XY display position portion that displays a corrected image corresponding to the position on the screen and allows an arbitrary point on the screen to be designated as the XY display position, the operation unit 67 of FIG. 5, specifically, a mouse or the like Can be used.

  An example of an image after performing each of the above processes will be described. FIG. 12 is a diagram illustrating an example of a microscope image before correction, and is a diagram illustrating images from Z = 1 to 62. FIG. 13 is a diagram showing a corrected image corresponding to FIG. 12, where Sb = Sf = 0.5 in FIG. It can be seen that focusing is achieved at Z = 30. Comparing the flat image before correction in FIG. 12 and the corrected flat image in FIG. 13, it can be seen that the corrected image can clearly obtain the outline of the cell in a wider range with respect to Z.

  FIG. 14 shows the corrected planar image (Z = 30) shown in FIG. 13 and the cross-sectional structure cut along the horizontal straight line L1 and the vertical straight line L2, based on the central portion and the above image processing. It is the figure which showed the analyzed cross-sectional image on the lower side and the right side. Although the analysis cross-sectional image is different from the actual cross-sectional image, it is understood that a pseudo cross-sectional structure reflecting the cross-sectional structure can be obtained by performing image processing according to the present embodiment.

In the lower side of FIG. 14, a pseudo cross-sectional structure of seven cells clearly designated as Ch1 to Ch7 is clearly shown. The white line L3 drawn in the horizontal direction corresponds to the surface of the bottom plate 1b, and Ch1, Ch2, Ch3, Ch5, and Ch7 can be determined to be adhered to the surface of the bottom plate 1b because the cell base extends to the white line L3.
On the other hand, it can be determined that Ch4 and Ch6 are floating (detached) because the cell base does not reach the white line L3.

  Similarly, on the right side of FIG. 14, a pseudo cross-sectional structure of five cells clearly shown as Cv1 to Cv5 is clearly shown. The white line L4 drawn in the vertical direction corresponds to the surface of the bottom plate 1b, and Cv1, Cv2, and Cv3 can be determined to be adhered to the surface of the bottom plate 1b because the cell base extends to the white line L4. And Cv5 can be determined to be floating (detached) because the cell base does not reach the white line L4.

  In addition, in the pseudo cross-sectional structure of the cell, the base of the upper end or the lower end spreads radially due to light diffraction (reflection of the cell body), and the upper or lower end is an adjacent frame to be used for correction. Is not sufficiently corrected (deleted) in the above-described difference enhancement processing.

  In the image shown in FIG. 14, a pseudo sectional structure at an arbitrary position can be obtained by moving the horizontal straight line L1 and the vertical straight line L2 with, for example, a mouse.

  Next, a multidimensional cell cross-sectional analysis method according to the second embodiment of the present invention will be described. In the present embodiment, a plurality of still images are repeatedly acquired until a predetermined time elapses, and based on a plurality of moving images acquired in time series, horizontal and vertical directions at designated positions in the moving image are obtained. Create a cross-sectional image in the depth direction.

  FIG. 15 is a principle diagram of the multidimensional cell cross-sectional analysis method according to the present embodiment, and corresponds to FIG. In the present embodiment, a plurality of still images with different positions in the Z direction are repeatedly imaged at predetermined time intervals until a predetermined time elapses while moving the focal point to a plurality of locations at predetermined intervals. Get the video. In FIG. 15, unlike FIG. 3, time-plus photographing is performed. This time-plus imaging slice image is expressed by the following equation.

  Here, the reconstructed cross-sectional moving image in the X direction obtained by cutting the imaged slice image along the XZ plane indicated by reference numeral S11 is represented by the following expression.

  On the other hand, a reconstructed cross-sectional moving image in the Y direction obtained by cutting the imaged slice image along the YZ plane indicated by reference numeral S12 is represented by the following expression.

  Based on the above principle, a desired cross-sectional moving image in the X direction, the Y direction, or the like can be reconstructed to obtain a pseudo cross-sectional image.

  FIG. 16 is a flowchart showing a process flow of the multi-dimensional cell section analysis method according to the present embodiment.

First, in step S21, the microscope 51 is set up, the field of view is set, and timing is started. Next, in step S22, the focal point is set at the substrate position (bottom plate 1b). In step S <b> 23, a still image is captured by the imaging unit 53 and recorded in the memory 68. Next, in step S24, the XYZ stage 11 is moved stepwise, and in step S25, it is determined whether or not the focal plane is within the water surface 3a. If it is within the water surface (Yes), the process returns to step S23, and if it exceeds the water surface (No), the process proceeds to step S26, and after waiting for a specified time, the number of photographing is incremented. In step S27, it is determined whether or not the shooting time exceeds the upper limit. If the upper limit has not been reached (shooting time <upper limit), the process returns to step S22. If the upper limit is exceeded, that is, if the shooting time has reached the upper limit, the process proceeds to step S28. Here, the upper limit of the shooting time refers to the shooting time immediately before the focal plane comes out of the water surface when shooting while moving the Z stage.
Next, in steps S28 and 29, processing similar to that in steps S5 and S6 in FIG. 6 is performed.

  Thereafter, interactive display processing of the cross-sectional image is performed. First, in step S30, the display position of the cross-sectional image on the screen is designated. Next, in step S31, a pseudo sectional moving image is created as a reconstructed moving image in each of the X and Y directions. Next, in step S32, the reconstructed moving images in the X and Y directions created in step S31 are output to the output unit 65. In step S33, if another position is desired to be displayed (Yes), the process returns to step S30. If not displayed (No), the process ends (End).

  As a result, a two-dimensional YZ moving image is created by extracting all pixels corresponding to the X coordinate of the XY display position from a plurality of corrected moving images having different positions in the Z direction, and corresponding to the Y coordinate of the XY display position. All the pixels to be extracted are extracted to create a two-dimensional XZ moving image, and two two-dimensional YZ moving images or XZ moving images can be displayed on the screen.

As described above, according to the present embodiment, a pseudo cross-sectional image of detached cells and adherent cells in a culture solution can be obtained without using a special device.
Therefore, there is an advantage that it is possible to easily distinguish between detached cells and adherent cells.

  In the above-described embodiment, the configuration and the like illustrated in the accompanying drawings are not limited to these, and can be appropriately changed within a range in which the effect of the present invention is exhibited. In addition, various modifications can be made without departing from the scope of the object of the present invention. It may be a program for causing a computer to execute the cell image processing method, or a computer-readable recording medium for recording the program.

  The present invention can be used as an image processing apparatus for detecting detached cells.

A ... sectional image analysis system, 1 ... culture dish, 1b ... bottom plate, 3 ... culture solution, 5 ... detached cell, 7 ... adherent cell, 11 ... XYZ stage, 15 ... transmission light source, 17 ... objective lens, 17 ... optical microscope 21 ... XYZ stage controller, autofocus unit, 23 ... video camera, 25 ... video frame memory, 27 ... control PC, 31 ... monitor, 50 ... imaging mechanism, 51 ... microscope, 53 ... imaging unit, 55 ... Movement mechanism, 57... Image correction unit 1 (sharpness correction unit), 61... Image correction unit 2 (difference enhancement processing unit), 62... Image processing unit (cross-sectional image reconstruction unit), 63. ... output unit, 67 ... operation unit, 68 ... memory.

Claims (8)

A system including cells and liquid in a cell culture instrument having a bottom plate is observed in focus with a microscope while moving in a first horizontal direction with respect to the bottom plate, and a plurality of in-focus positions with different vertical positions are observed. The image is defined by a second direction horizontal to the bottom plate intersecting the first direction and a third direction perpendicular to the bottom plate intersecting the first direction and the second direction. An imaging mechanism that captures images as a plurality of two-dimensional planar images,
An image correction unit that performs correction processing for sharpening an image for each of the plurality of focused images captured by the imaging mechanism;
Based on the plurality of in-focus images after being corrected by the image correction unit, it passes through an arbitrary point on the two-dimensional plane, and a cross section in either the second direction or the third direction, An image processing unit that reconstructs the plurality of focused images together to generate a pseudo cross-sectional image , and
Based on the pseudo cross-sectional image, the image in which the position in the third direction is different and the image subjected to the sharpening correction process are compared with the planar image before the sharpening correction process is performed. A cell cross-section analysis apparatus for determining whether the cell is an adherent cell or a detached cell depending on whether or not the cell extends to a line corresponding to the surface of the bottom plate . The imaging mechanism is
A plurality of moving images are obtained by repeatedly capturing the focused image at predetermined time intervals,
The cell section analysis device according to claim 1, wherein the image processing unit generates a pseudo section moving image based on the plurality of moving images.   The image correction unit creates a defocused image that is more out of focus than the plurality of focused images by an algorithm based on an unsharp mask as a process for sharpening the plurality of focused images. The cell cross-section analysis apparatus according to claim 1 or 2, wherein the correction is performed by adding a difference image between the focused image and the corresponding defocused image by a predetermined ratio. The image correction unit
In addition to the sharpening process, a frame of a still image or a moving image adjacent to the plurality of focused images by one frame in the first direction in the vertical direction or in the direction opposite to the first direction; Each of the difference images is created, and the difference image corresponding to the still image or moving image frame is added to the still image or moving image frame by a predetermined ratio to perform correction. Item 3. The cell cross-section analyzer according to Item 1 or 2. The imaging mechanism is
Set the said bottom Z1 position at which the focusing relative to the surface of the plate of the cell culture instrument add broth as the liquid, and a Z2 position at which the focusing respect culture surface in the cell culture instrument, Taking a plurality of images with different positions in the Z direction while moving the focal point to a plurality of locations at predetermined intervals in the range between the Z1 position and the Z2 position by an optical microscope ,
A difference image between a target image to be corrected by the image correction unit and a first adjacent image in the Z direction, and a difference image between the target image and a second adjacent image opposite to the first adjacent image in the Z direction. Is added to the original image . The cell cross-section analysis apparatus according to any one of claims 1 to 4, wherein   A plane including an arbitrary point on the specified two-dimensional plane, formed by the second direction and the first direction, or formed by the third direction and the first direction. 6. The display control unit according to claim 1, further comprising: a display control unit configured to display on the display screen at least one of the pseudo cross-sectional images cut by a plane in correspondence with the image of the two-dimensional plane. The cell cross-section analyzer described in 1. A system including cells and liquid in a cell culture instrument having a bottom plate is observed while being focused with a microscope while relatively moving in a first horizontal direction with respect to the bottom plate, and the position in the first direction A plurality of in-focus images differing in the vertical direction with respect to the bottom plate intersecting the first direction and the second direction in the horizontal direction with respect to the bottom plate intersecting with the first direction and the bottom direction intersecting with the first direction and the second direction. An imaging step of imaging as a plurality of two-dimensional planar images defined by the third direction;
An image correction step of performing correction processing for sharpening images for each of the plurality of focus image captured by an imaging mechanism,
Based on the plurality of focus image after being corrected by the images correcting unit, through an arbitrary point on the two-dimensional plane, one of the cross-section of the second direction or the third direction, It possesses an image processing step of generating a reconstructed pseudo cross-sectional image to piece together the plurality of focused image,
Based on the pseudo cross-sectional image, the image in which the position in the third direction is different and the image subjected to the sharpening correction process are compared with the planar image before the sharpening correction process is performed. A cell cross-sectional analysis method characterized by determining whether the image is an adhesive image or a peeled image depending on whether or not the image extends to a line corresponding to the surface of the bottom plate .   A cell cross-section analysis program for causing a computer to execute the cell cross-section analysis method according to claim 7.

Images