Microscope, observation method

Abstract

The microscope of the present application moves the position of a collimator lens to make light passing through an objective lens converge on the surface of a cover glass, detects reflected light from a sample as a focus error signal indicating the amount of deviation of the convergence point of the light passing through the objective lens from the focusing object plane (see FIG. 1).

Description

Technical Field

[0001] This invention relates to microscopes. Background Technology

[0002] Pathology is a field of medicine that diagnoses diseases by examining tissues under a microscope or by surgically removing organs and body fluids. In recent years, with the widespread use of digital imaging, especially the development and popularization of WSI (whole slide imaging) technology, traditional pathology has evolved into digital pathology. WSI is a system that uses a high-magnification microscope to pre-image the entire specimen on a glass slide as a digital image, imports it into a computer, and displays it on a monitor for observation. The device used to acquire these digital images is called a WSI scanner.

[0003] In the widespread application of digital pathology, specimen scanning is a major bottleneck in the workflow, requiring increased throughput (shortened scanning time).

[0004] However, in WSI scanner microscopes, autofocus control is used to automatically adjust the objective lens to the optimal position in order to obtain a clear image of the object. From the perspective of focus detection methods, autofocus methods for WSI scanners can be broadly divided into two types: reflection-based and image-based. Reflection-based autofocus uses a dedicated light source to receive reflected light from the sample and detects the height of a reference surface on the sample. Image-based autofocus detects the focus height based on image quality evaluation values ​​(contrast, etc.) of the captured image.

[0005] Astigmatism, a type of reflective focus detection method, involves focusing a laser beam onto the surface of the object being focused on, receiving the reflected light, and thus detecting focus shift (defocus). The detection system in this method consists of a converging lens, a cylindrical lens, and a detector. Defocus is detected by utilizing the change in the shape of the light spot on the detector according to the amount of defocus. This astigmatism method uses a dedicated light source and detection system for focus detection, therefore, it does not require image processing as in image-based focus detection methods. Furthermore, it can be implemented through simple signal processing, resulting in excellent high-speed focus tracking, which is beneficial for shortening the scanning time of WSI scanners.

[0006] Patent Document 1 described below describes an example of an autofocusing device using astigmatism. The document, with the subject matter of "providing an autofocusing device and microscope apparatus capable of shortening the time required to focus the objective lens on an object," describes the following technology: "An autofocusing device 20 includes: a drive unit 25 that moves the objective lens 11 in the optical axis direction; a detection unit 28 that detects whether an object is positioned at a detection position DP offset from the focal plane of the objective lens 11 by a predetermined reference distance; and a control unit 29 that, when the detection unit 28 detects that the object is positioned at the detection position DP, controls the drive unit 25 to move the objective lens 11 by a reference distance, thereby focusing the objective lens 11 on the object." (See abstract).

[0007] Existing technical documents

[0008] Patent documents

[0009] Patent Document 1: Japanese Patent Application Publication No. 2022-137838 Summary of the Invention

[0010] The problem that the invention aims to solve

[0011] One way to photograph a glass slide specimen for a WSI scanner is by sealing the tissue section onto the slide with a coverslip using a sealing agent. For astigmatism-based autofocus to function ideally, i.e., with high autofocus sensitivity, a laser beam for autofocus (hereinafter referred to as the autofocus beam) needs to be focused onto the surface of the specimen. When applying astigmatism to autofocus on glass slide specimens, the objective lens of a microscope is used to illuminate the specimen with a laser, and the surface of the coverslip becomes the primary reflector of the irradiated light.

[0012] However, when photographing tissue sections, there are limitations regarding the objective lens position (height), which must be adjusted to focus on the tissue section. Therefore, it may not be possible to converge the focusing beam onto the coverslip surface, which is the object of focus. In most cases, the focusing beam is a parallel beam, and the convergence point of the objective lens is the location of the tissue section. Therefore, on the coverslip surface, the beam becomes diffused, and the focusing sensitivity—that is, the change in the focusing error signal relative to the amount of defocus—is significantly reduced compared to ideal conditions. This reduction in focusing sensitivity leads to a decrease in focusing accuracy.

[0013] The purpose of this invention is to solve the above-mentioned problems and provide a microscope with an attached autofocus mechanism that balances high-speed focus tracking and focusing accuracy.

[0014] Methods for solving problems

[0015] The microscope of the present invention moves the position of the collimating lens so that the light passing through the objective lens converges on the surface of the coverslip, and detects the reflected light from the sample as a focusing error signal representing the amount of offset of the convergence point of the light passing through the objective lens relative to the focusing surface.

[0016] Invention Effects

[0017] According to the present invention, a microscope with an autofocus mechanism that combines high speed in focus tracking with focusing accuracy can be provided. Other issues, structures, advantages, etc., of the present invention will become clear from the following description of embodiments. Attached Figure Description

[0018] Figure 1 This section illustrates a structural example of the imaging optical system of the microscope described in Embodiment 1.

[0019] Figure 2 This diagram illustrates the principle of using astigmatism to detect focusing errors, specifically the offset of the convergence point of the illumination beam from the surface of the object being focused.

[0020] Figure 3 This is a diagram schematically representing one way in which the focusing detection beam 139 converges.

[0021] Figure 4 It means to Figure 3 The state is set to defocus = 0, and the shape of the light spot on the photodetector 153 is obtained when the position of the objective lens 121 is changed based on this.

[0022] Figure 5 It describes the Figure 3 The graph shows the relationship between defocus and focus error signal (FES) when the state is set to defocus = 0.

[0023] Figure 6 This is a diagram illustrating another way of showing the convergence state of the focus detection beam 139.

[0024] Figure 7 It means to Figure 6 The state is set to defocus = 0, and the shape of the light spot on the photodetector 153 is obtained when the position of the objective lens 121 is changed based on this.

[0025] Figure 8 It describes the Figure 6 The graph shows the relationship between defocus and focus error signal (FES) when the state is set to defocus = 0.

[0026] Figure 9 This is a diagram schematically showing the convergence state of the focus detection beam 139.

[0027] Figure 10It means to Figure 9 The state is set to defocus = 0, and the shape of the light spot on the photodetector 153 is obtained when the position of the objective lens 121 is changed based on this.

[0028] Figure 11 It describes the Figure 9 The graph shows the relationship between defocus and focus error signal (FES) when the state is set to defocus = 0.

[0029] Figure 12 This is a flowchart illustrating the process of implementing autofocus control by adjusting the position of the collimating lens 132 to match any glass slide specimen.

[0030] Figure 13 This is a conceptual diagram illustrating the principle of focusing error caused by uneven thickness of the cover glass.

[0031] Figure 14 This is a top view showing an example of the configuration of the focus learning point.

[0032] Figure 15 This is a flowchart illustrating the process of creating a focus error signal correction value mapping and using the focus error signal corrected according to this data to implement automatic focus control.

[0033] Figure 16 This is a flowchart illustrating the process of creating a collimating lens position mapping and dynamically controlling the position of the collimating lens 132 based on this data while implementing automatic focus control.

[0034] Figure 17 This is a diagram illustrating an example of the structure of the imaging optical system of the microscope in Embodiment 6. Detailed Implementation

[0035] <Implementation Method 1: Basic Structure (Adjustment of Beam Focusing Point Position)>

[0036] <Structure of the Overall Optical System>

[0037] Figure 1 This section illustrates a structural example of the imaging optical system of the microscope according to Embodiment 1 of the present invention. This imaging optical system includes: a microscope optical system 1 for imaging specimens, a focus detection optical system 2 for detecting focus error signals for implementing autofocus, and a processing unit 200. Regarding the coordinate axes, the optical axis direction of the microscope optical system 1 is defined as the Z-axis, the direction horizontal to the paper surface in the in-plane direction of the glass slide specimen (described later) is defined as the X-axis, and the direction perpendicular to the paper surface is defined as the Y-axis. The processing unit 200 controls each part of the microscope and implements the flowchart described later.

[0038] <Structure of a Microscope>

[0039] The structure of the microscope optical system 1 is classified as a transmission-type bright-field microscope. The illumination optical system 3, used to illuminate the object being photographed, includes: a light-emitting diode 101 as a light source, a light-collecting lens 102, and a condenser lens 103. The light-collecting lens 102 forms a real image of the light source on the condenser lens 103, and the condenser lens 103 forms a real image of the light-collecting lens 102 on the sample surface. That is, the illumination method of the illumination optical system 3 is Köhler illumination, which uniformly illuminates the sample surface. Furthermore, the illumination optical system 3 has a field stop 104 and an aperture stop 105. The illumination range on the sample surface is adjusted by the field stop 104, and the numerical aperture of the illumination light is adjusted by the aperture stop 105.

[0040] The glass slide specimens used for photography are prepared by sealing tissue sections placed on the slide with a coverslip 112 using a sealing agent.

[0041] Objective lens 121 and imaging lens 123 constitute an infinity-corrected optical system, which images the tissue section within the glass slide specimen onto the sensor surface of imaging element 124. Objective lens position adjustment mechanism 122 adjusts the position of the optical axis of objective lens 121.

[0042] <Structure of the Focus Detection Optical System>

[0043] The focus detection optical system 2 detects focus errors using the astigmatism method. A laser diode 131, serving as the light source, emits linearly polarized infrared laser light with a wavelength of 785 nm. The emitted laser light is collimated by a collimating lens 132 to become a roughly parallel beam, and then the beam diameter is limited to a predetermined size by an aperture 134, becoming the focus detection beam 135. The focus detection beam 135 passes through a λ / 2 plate 136 and is incident on a polarizing beam splitter 137.

[0044] The polarization beam splitter 137 has the function of allowing approximately 100% transmission of p-polarized light (polarized light with an electric field component parallel to the incident plane) incident on the splitting surface, and approximately 100% reflection of s-polarized light (polarized light with an electric field component perpendicular to the incident plane). By adjusting the rotation angle of the λ / 2 plate 136 about the optical axis relative to the focus detection beam 135, the polarization direction of the light transmitted through the λ / 2 plate 136 can be arbitrarily changed, and the intensity ratio of transmitted light to reflected light in the polarization beam splitter 137 can be arbitrarily adjusted. In this embodiment, the rotation angle of the λ / 2 plate 136 is adjusted so that the light transmitted through the λ / 2 plate 136 becomes p-polarized light, and the focus detection beam 135 transmits approximately 100% through the polarization beam splitter 137.

[0045] The focusing detection beam 135, which passes approximately 100% through the polarizing beam splitter 137, is transformed into circularly polarized light by passing through a λ / 4 plate 138, whose fast axis is tilted at 45 degrees relative to the polarization direction of the incident beam, and is then guided to the microscope optical system 1. The focusing detection beam 135 is coupled to the optical path of the microscope optical system 1 through a dichroic filter 141 disposed in the optical path of the microscope optical system 1. Then, the focusing detection beam 135 is converged by the objective lens 121 and illuminates the glass slide specimen, which is the object to be photographed.

[0046] The light beam reflected from the surface of the coverslip 112 of the glass slide specimen (hereinafter, reflected beam 139 or focusing detection beam 139) is restored to a generally parallel beam by the objective lens 121 and to linearly polarized light by the λ / 4 plate 138. At this time, the rotation direction of the circularly polarized light is reversed due to reflection on the surface of the coverslip 112, so the direction of the linearly polarized light becomes s-polarized light, which has been rotated 90 degrees from the path. Therefore, the reflected beam 139 of the s-polarized light transmitted through the λ / 4 plate 138 is approximately 100% reflected by the polarizing beam splitter 137 and directed toward the converging lens 151.

[0047] The reflected beam 139 is converged by the converging lens 151 and simultaneously incident on the cylindrical lens 152. The cylindrical lens 152 is configured such that its cylindrical axis is tilted at 45 degrees relative to the Y-axis in the XY plane. The reflected beam 139 is astigmatically atomized by the cylindrical lens 152 and incident on the photodetector 153. The photodetector 153 is a 4-segment photodiode with its segmentation lines arranged along the X and Y axes. The focus error signal (FES), described later, is generated based on the output signals of each element (A, B, C, and D) of the photodetector 153.

[0048] <Principle of Focus Detection Based on Astigmatism>

[0049] Figure 2 This diagram illustrates the principle of using astigmatism to detect focusing errors, specifically the offset of the convergence point of the illumination beam from the surface of the object being focused.

[0050] The reflected beam 139 is converged by the converging lens 151 while being subjected to astigmatism by the cylindrical lens 152. The cylindrical lens 152 functions as a lens only in a single direction; its direction orthogonal to the parallel plate is the same, and it does not function as a lens. Therefore, the focal length differs in each direction, resulting in astigmatism. In the focus detection optical system 2 of this embodiment, the focal length of the beam observed from the non-lensing section of the cylindrical lens 152 and the focal length (f) of the converging lens 151 are... det The two are roughly the same. On the other hand, the beam of light observed from the cross section that acts as a lens is not focused here, resulting in a long, thin line image (focal line).

[0051] The light beam observed from the section of cylindrical lens 152 that acts as a lens converges to the focal length of the combined lens of converging lens 151 and cylindrical lens 152. However, the light beam observed from the section without a lens function does not focus and becomes a focal line. The focal length f of the combined lens... comb It is represented by Equation 1. f det It is the focal length of the converging lens 151, f. cyl It is the focal length of the cylindrical lens 152, D lens It is the interval between the converging lens 151 and the cylindrical lens 152. The beam is circular in the middle of the two focal lines.

[0052] [Mathematical Expression 1]

[0053]

[0054] When using astigmatism to measure defocus, such as Figure 2 As shown in (2), the center of the photodetector 153 is aligned with the reflected light beam from the focal plane of the objective lens 121 to form a circular position. The photodetector 153 is composed of four segmented photodiodes, and the light signals detected by each element are converted into electrical signals. The signal processing circuit calculates the focus error signal FES using Equation 2 based on the four outputs A, B, C, and D of the photodetector 153.

[0055] [Mathematical Expression 2]

[0056]

[0057] When the height of the cover glass 112, which serves as the focusing surface, coincides with the focal plane (focal length) of the objective lens 121, FES = 0. For example... Figure 2 As shown in (1), when the plane of the object being focused is lower than the focal plane, the beam tilts to the left and becomes horizontally elongated, and FES > 0. On the other hand, as Figure 2 As shown in (3), when the plane of the object being focused is higher than the focal plane, the beam is tilted to the right and becomes horizontally elongated, and FES < 0.

[0058] The sign of FES depends on the setting angle centered on the optical axis of the cylindrical lens 152. In this embodiment, it is set at a 45-degree angle relative to the vertical direction, centered on the optical axis. The optical axis is defined as the direction of light propagation. Thus, the amount of offset from the cover glass surface is detected based on the focusing error signal.

[0059] <Focus Detection Sensitivity>

[0060] Figure 3This diagram schematically illustrates one method of converging the focusing detection beam 139. A sealing agent 302 is used to seal the tissue section 301. In this state, the focusing detection beam 139 is a parallel beam, and the objective lens 121 is positioned so that the focusing detection beam 139 converges onto the surface of the coverslip 112. That is, this is a method in which astigmatism ideally functions.

[0061] Figure 4 It means to Figure 3 The state is set to defocus = 0, and the shape of the light spot on the photodetector 153 is obtained when the position of the objective lens 121 is changed based on this. Figure 4 (1), (2), and (3) represent the cases where the defocus is -20, 0, and +20 μm, respectively. The size of the photodetector is 0.5 mm × 0.5 mm. The depicted area is the same as the size of the photodetector, 0.5 mm × 0.5 mm. Thus, relative to the change in defocus, the shape of the light spot on the photodetector 153 changes elongatedly in the tilt direction.

[0062] Figure 5 It describes the Figure 3 The graph shows the relationship between defocus and focus error signal (FES) when the state is set to defocus = 0. Thus, the focus error signal changes linearly relative to defocus between positive and negative peaks, indicating that the astigmatism method is functioning ideally.

[0063] However, in Figure 3 In the state shown, the focus of objective lens 121 is aligned with the surface of coverslip 112, and therefore the original purpose of the WSI scanner, such as obtaining a clear image of tissue section 301, is not achieved.

[0064] Figure 6 This is a diagram schematically illustrating another way in which the focus detection beam 139 converges. The focus detection beam 139 and... Figure 3 The situation is the same as before, with a parallel beam. On the other hand, the position of the objective lens 121 is adjusted so that the focus is on the tissue section 301. That is, this represents the actual operating state of the WSI scanner. However, in this state, the focusing detection beam 139 converges on the tissue section 301 and extends on the surface of the coverslip 112, which serves as the focusing target surface.

[0065] Figure 7 It means to Figure 6 The state is set to defocus = 0, and the shape of the light spot on the photodetector 153 is obtained when the position of the objective lens 121 is changed based on this. Figure 7 (1), (2), and (3) represent the defocusing values ​​of -20, 0, and +20 μm, respectively. The depicted range is... Figure 3The difference is that it is 2.0mm × 2.0mm. In the diagram, the size of the photodetector (0.5mm × 0.5mm) is represented by a white dashed line. Thus, in Figure 6 In this state, the light spot on the photodetector 153 expands significantly and protrudes from the photodetector 153. Therefore, the focus error signal hardly changes relative to the defocusing change, making defocusing detection difficult. Even when using a photodetector capable of receiving the entire size of the light spot, the change in the light spot shape relative to the defocusing change is significant. Figure 3 It is also significantly smaller than the ideal state.

[0066] Figure 8 It describes the Figure 6 The state is set as defocus = 0, and the curve is the relationship between defocus and focus error signal (FES). In this way, the focus error signal hardly changes near defocus = 0, which is a state with significantly low focus detection sensitivity.

[0067] Figure 9 This diagram schematically illustrates the convergence state of the focusing detection beam 139 in this embodiment. The objective lens 121 is positioned so that the focus is aligned with the tissue section 301. That is, it represents the actual operating state of the WSI scanner. However, the focusing detection beam 139 and... Figure 3 Unlike other conditions, the light weakly converges and enters the objective lens 121, converging on the surface of the coverslip 112, which serves as the focusing target. According to this structure, the objective lens 121 can be positioned appropriately for shooting while allowing the astigmatic focus detection to function ideally.

[0068] Figure 10 It means to Figure 9 The state is set to defocus = 0, and the shape of the light spot on the photodetector 153 is obtained when the position of the objective lens 121 is changed based on this. Figure 10 (1), (2), and (3) represent the defocusing values ​​of -20, 0, and +20 μm, respectively. The depicted area is 0.5 mm × 0.5 mm, the same size as the photodetector. Thus, in Figure 9 In this state, the shape of the defocused light spot changes significantly, compared to... Figure 3 Changes in the shape of the light spot in the structure ( Figure 4 (Equal)

[0069] Figure 11 It describes the Figure 9 The graph shows the relationship between defocus and focus error signal (FES) when the state is set to defocus = 0. Thus, the slope in the straight line region between the positive and negative peaks of the focus error signal indicates the relationship between focus detection sensitivity and... Figure 3 Focusing error signal under ideal conditions ( Figure 5This means that, through the structure of this embodiment, focus detection based on astigmatism functions ideally.

[0070] <Adjustment of beam convergence point>

[0071] exist Figure 9 The focusing detection beam 139 shown is focused on the surface of the coverslip 112 even when the focus of the objective lens 121 is aligned with the tissue section 301. This focusing detection beam 139 is focused on the surface of the coverslip 112 by the following means.

[0072] about Figure 1 The position of the collimating lens 132 shown is based on the position where the laser beam passing through the collimating lens 132 becomes a parallel beam. The collimating lens position adjustment mechanism 133 moves the position of the collimating lens 132 along the optical axis in the direction of beam travel. As a result, the laser beam passing through the collimating lens 132 becomes a weakly converged beam traveling simultaneously. The weakly converged focus detection beam 135 is converged by the objective lens 121, and the position of its convergence point is closer to the focal plane (focal length) of the objective lens 121.

[0073] The collimating lens position adjustment mechanism 133 adjusts the position of the collimating lens 132 so that the focusing detection beam 135 is exactly focused on the surface of the cover glass 112.

[0074] <Implementation Method 2: Method for Adjusting the Position of the Beam Convergence Point>

[0075] <Adjust the focus error signal to near zero>

[0076] Embodiment 2 of the present invention relates to the process of adjusting the collimating lens 132 to a position suitable for focus detection according to the astigmatism method. The structure of the microscope is the same as that of Embodiment 1. First, the setting process of the focus detection optical system 2 will be described. Figure 1 The photodetector 153 in the focus detection optical system 2 is positioned such that the focus error signal is zero when the focus detection beam 135 converges exactly on the surface of the cover glass 112. The following illustrates an example of the process of positioning the photodetector 153 in such a position.

[0077] First, the position of objective lens 121 is adjusted so that its focal plane aligns with the surface of coverslip 112. This is achieved, for example, by adjusting the position of objective lens 121 while observing the image captured by microscope optical system 1 so that minute scratches or dust particles on the surface of coverslip 112 are clearly visible. Next, with the focus detection beam 135 parallel, the photodetector 153 is positioned where the focus error signal is zero. That is, the photodetector 153 is positioned where the reflected beam 139 is circular. Through the above process, the focus detection optical system 2 is set to a state where the focus error signal becomes zero when the focus detection beam 135 converges precisely on the surface of coverslip 112.

[0078] Figure 12 This is a flowchart illustrating the process of implementing automatic focusing control by adjusting the position of the collimating lens 132 to match any glass slide specimen. The focus detection optical system 2 is preset to a position where the focus error signal becomes zero when the focus detection beam 135 exactly converges on the surface of the coverslip 112. This flowchart (and the following flowcharts are the same) is implemented by the arithmetic unit 200. The following describes... Figure 12 The steps are explained below.

[0079] Step S12: At any representative point (XY coordinate) on the glass slide specimen, preferably at the location where the tissue section to be photographed exists, focus the objective lens 121 on the tissue section. That is, move the objective lens 121 to the position where the image of the tissue section is clearest.

[0080] Step S13: The collimating lens 132 is moved to a position where the focusing error signal is zero by the collimating lens position adjustment mechanism 133. Through this process, the focusing detection beam 135 converges on the surface of the cover glass 112 at the representative point.

[0081] Step S14: This process is the autofocus control process. Feedback control is also repeatedly implemented at coordinates other than the representative point in S12 to detect the focus error signal and move the objective lens 121 in a direction close to zero.

[0082] The above series of procedures are performed each time the glass slide specimen is changed. Therefore, even if the height of the coverslip surface varies due to the thickness deviation of the slides and coverslips in each glass slide specimen, the focusing detection beam 135 can always be focused on the surface of the coverslip 112, ensuring high focusing detection sensitivity.

[0083] <Implementation Method 3: Focus Error Signal Correction Value Mapping>

[0084] Astigmatism is a method of detecting focusing error relative to the surface of a coverslip by receiving reflected light from the coverslip surface. In autofocus systems using this astigmatism method, the position of the objective lens is controlled based on the position of the coverslip surface. However, commercially available coverslips have uneven thickness, i.e., variations in thickness caused by variations within a single coverslip. Therefore, the distance from the coverslip surface to the tissue section varies depending on the location being photographed, and this variation constitutes the focusing error. Embodiment 3 of the present invention relates to a method for suppressing focusing errors caused by uneven thickness of coverslips in glass slide specimens. The microscope structure is the same as in Embodiment 1.

[0085] <Focusing error caused by uneven cover glass thickness>

[0086] Figure 13 This is a conceptual diagram illustrating the principle that focusing errors occur due to uneven thickness of the cover glass. Figure 13 This is equivalent to a side sectional view of the coverslip. The autofocus control controls the position of the objective lens 121 to track changes in the height of the surface of the coverslip 112. However, as the thickness of the coverslip 112 varies depending on the location being photographed (XY coordinates), the distance from the surface of the coverslip 112 to the tissue section 301 being photographed also varies; therefore, this amount of variation becomes the focus error.

[0087] <Map creation and focus learning point configuration>

[0088] Figure 14 This is a top view illustrating an example of the configuration of focus learning points. The shooting range is set to include the entire tissue slice 301, which is the subject of the shooting, and focus learning points 401 are configured at predetermined intervals within the shooting range. It is desired that the configuration interval of focus learning points 401 is less than 1 / 2 of the reciprocal of the maximum spatial frequency (minimum spatial period) of the thickness variation of the coverslip.

[0089] <Interpolation>

[0090] A correction value for correcting the focus error signal can be set according to the XY coordinates of the tissue section 301 surface. The data describing this correction value is called a focus error signal correction value mapping. One method of mapping the focus error signal correction value is to store the correspondence between XY coordinates and focus error signal correction values ​​as an array in the form of a lookup table. The focus error signal correction value for the XY coordinates between focus learning points not included in the mapping is calculated using interpolation data from multiple nearby focus learning points. Alternatively, it can be set as a mathematical expression for calculating the focus error signal correction value as a function of the XY coordinates.

[0091] <Flowchart>

[0092] Figure 15This is a flowchart illustrating the process of creating a focus error signal correction value mapping and using the corrected focus error signal based on this data to implement autofocus control. The following is a further explanation... Figure 15 The steps are explained below.

[0093] Step S22: At any representative point (XY coordinate) on the glass slide specimen, preferably at the location where the tissue section to be photographed exists, focus the objective lens 121 on the tissue section. That is, move the objective lens 121 to the position where the image of the tissue section is clearest.

[0094] Step S23: The collimating lens 132 is moved to a position where the focusing error signal is zero by the collimating lens position adjustment mechanism 133. Through this process, the focusing detection beam 135 converges on the surface of the cover glass 112 at the representative point.

[0095] Step S24: Move to the position (XY coordinates) of the next focus learning point.

[0096] Step S25: At the focus learning point, align the objective lens 121 with the tissue section. That is, move the objective lens 121 to the position where the image of the tissue section is clearest.

[0097] Step S26: Detect the focus error signal and record it as the focus error signal correction value.

[0098] Step S27: Determine whether steps S25 and S26 have been performed on all focus learning points. If yes, proceed to the next step S28. If no, return to step S24.

[0099] Step S28: Create a focus error signal correction value mapping that represents the relationship between the XY coordinates and the focus error signal correction value of those coordinates.

[0100] Step S29: This process is the autofocus control process. It involves repeatedly detecting the focus error signal, subtracting the current XY coordinate focus error signal correction value from the detected focus error signal, thereby correcting the focus error signal and causing the objective lens 121 to move in a direction closer to zero by feedback control.

[0101] <Implementation Method 4: Other Mapping Schemes>

[0102] <Flowchart>

[0103] In Embodiment 4 of the present invention, a collimating lens position mapping prepared from a different perspective than the focusing error signal correction value mapping described in Embodiment 3 is explained. The structure of the microscope is the same as in Embodiment 1.

[0104] Figure 16This is a flowchart illustrating the process of creating a collimating lens position map, using this data to dynamically control the position of the collimating lens 132 while simultaneously implementing autofocus control. The following is a further explanation... Figure 16 The steps are explained below.

[0105] Step S32: At any representative point (XY coordinate) on the glass slide specimen, preferably at the location where the tissue section to be photographed exists, focus the objective lens 121 on the tissue section. That is, move the objective lens 121 to the position where the image of the tissue section is clearest.

[0106] Step S33: The collimating lens 132 is moved to a position where the focusing error signal is zero by the collimating lens position adjustment mechanism 133. Through this process, the focusing detection beam 135 converges on the surface of the cover glass 112 at the representative point.

[0107] Step S34: Move to the position (XY coordinates) of the next focus learning point.

[0108] Step S35: At the focus learning point, align the objective lens 121 with the tissue section. That is, move the objective lens 121 to the position where the image of the tissue section is clearest.

[0109] Step S36: Move the collimating lens 132 to the position where the focusing error signal is zero, and record this position as the target value of the collimating lens position.

[0110] Step S37: Determine whether steps S35 and S36 have been performed on all focus learning points. If yes, proceed to the next step S38. If no, return to step S34.

[0111] Step S38: Create a collimating lens position target value mapping that represents the relationship between the XY coordinates and the collimating lens position target value at that coordinate.

[0112] Step S39: This step is the autofocus control step. The position control of the objective lens 121 is the same as in the above implementation. That is, feedback control is performed by repeatedly detecting the focus error signal and moving the objective lens 121 in a direction close to zero.

[0113] In this embodiment 4, the position of the collimating lens 132 is feedforward controlled to a target position corresponding to the XY coordinates. This target position is derived from the collimating lens position target value mapping. By adjusting the position of the collimating lens 132, the light beam can be illuminated in a more ideal state.

[0114] <Implementation Method 5: Dedicated Device for Mapping Production>

[0115] Embodiment 5 of the present invention relates to other methods of creating the focus error signal correction value mapping of Embodiment 3 and the collimating lens position target value mapping of Embodiment 4. These mappings can also be created by other devices (mapping creation apparatuses) having a structure equivalent to the microscopes of the above embodiments. For example, by creating the mapping of the slide to be observed by the microscope in advance using a mapping creation apparatus, the observation process can be performed smoothly. The mapping creation apparatus does not need to be exactly the same as the microscope of the present invention, as long as it has a structure capable of creating the mappings described in Embodiments 3 to 4.

[0116] <Implementation Method 6: Dual Astigmatism>

[0117] Figure 17 This diagram illustrates a structural example of the imaging optical system of the microscope according to Embodiment 6 of the present invention. Embodiment 6 relates to other structures of the focusing detection optical system. These other structures are the same as those in the embodiments described above.

[0118] The reflected beam 139 of s-polarized light passing through the λ / 4 plate 138 is approximately 100% reflected by the polarization beam splitter 137 and directed toward the converging lens 151. The reflected beam 139 is simultaneously converged by the converging lens 151 and directed toward the unpolarization beam splitter 154. The unpolarization beam splitter has the function of reflecting and transmitting light incident on the splitting surface at a predetermined ratio regardless of polarization direction. In this embodiment, the unpolarization beam splitter 154 reflects and transmits incident light at a 50:50 ratio. The intensity distributions of the cross-sections of the first reflected beam 140a and the reflected second reflected beam 140b passing through the unpolarization beam splitter 154 are linearly symmetrical.

[0119] The first reflected beam 140a, passing through the unpolarized beam splitter 154, is converged and incident on the first cylindrical lens 155a. The first cylindrical lens 155a is configured such that its cylindrical axis is tilted at 45 degrees relative to the Y-axis in the XY plane. The first reflected beam 140a, passing through the first cylindrical lens 155a, is thus astigmatically atomized and incident on the first photodetector 156a. The first photodetector 156a is a 4-segment photodiode, with its segmentation lines arranged along the X and Y axes. Based on the output signals of each element (A1, B1, C1, and D1) of the first photodetector 156a, a first focus error signal (FES1) is generated using Equation 3.

[0120] [Mathematical Expression 3]

[0121]

[0122] On the other hand, the second reflected beam 140b, reflected by the unpolarized beam splitter 154, is simultaneously converged and incident on the second cylindrical lens 155b. The second cylindrical lens 155b is configured such that its cylindrical axis is tilted at 45 degrees relative to the Y-axis in the YZ plane. The second reflected beam 140b passes through the second cylindrical lens 155b, thereby being astigmatically atomized and incident on the second photodetector 156b. The second photodetector 156b is a 4-segment photodiode, with its segmentation lines arranged along the Y-axis and Z-axis. Based on the output signals of each element (A2, B2, C2, and D2) of the second photodetector 156b, a second focus error signal (FES2) is generated using Equation 4.

[0123] [Mathematical Expression 4]

[0124]

[0125] In this embodiment 6, the focus error signal FES is calculated as the average of the first focus error signal FES1 and the second focus error signal FES2. That is, it is calculated using Equation 5.

[0126] [Mathematical Expression 5]

[0127]

[0128] According to the above structure, when the shape of the light spot in the photodetector surface is disturbed due to minor unevenness of the focusing object surface, the components of the disordered light spot shape in the first photodetector 156a and the second photodetector 156b are linearly symmetrical to each other. Therefore, by calculating the average of the focusing error signals FES1 and FES2 generated from the first photodetector 156a and the second photodetector 156b, the variation in the focusing error signal caused by the disorder of the light spot shape is eliminated. Therefore, according to this embodiment, even if there are minor unevenness in the focusing object surface, a stable focusing error signal can be obtained.

[0129] <Regarding variations of the present invention>

[0130] This invention is not limited to the embodiments described above, and includes various modifications. For example, without departing from the spirit of the invention, a portion of the structure of one embodiment can be replaced with the structure of another embodiment; furthermore, the structure of another embodiment can be added to the structure of one embodiment. Additionally, regarding a portion of the structure of each embodiment, other structures can be added, deleted, or replaced.

[0131] The microscope of this invention is not limited to a transmission bright-field microscope; for example, it can also be a fluorescence microscope. The illumination method can also be an epi-irradiation type. The specimen to be photographed is not limited to tissue sections. Various embodiments of this invention can also be applied to glass slide specimens containing blood or bodily fluids. The wavelength of the laser in the focusing detection optical system is not limited to 785 nm. It is preferable to select one outside the wavelength range (visible range) of the microscope's illumination source. The microscope's illumination source does not necessarily have to be an LED; it can also be a tungsten lamp, halogen lamp, xenon lamp, etc.

[0132] As described in the embodiments above, the present invention, by combining real-time focus detection based on astigmatism and correction of focus error signals based on mapping, has the advantage of being unaffected by temperature drift of the specimen height that occurs after mapping is made, compared with existing methods that use only focus mapping to control the position of the objective lens.

[0133] In the above embodiments, the arithmetic unit 200 can be constructed by hardware such as circuit devices that implement its functions, or by software that implements its functions executed by a computing device such as a CPU (Central Processing Unit).

[0134] Symbol Explanation

[0135] 1. Microscope optical system

[0136] 2. Focusing detection optical system

[0137] 3. Illumination optical system

[0138] 101 Light Emitting Diode

[0139] 102 Light-collecting lens,

[0140] 103 Condensing lens,

[0141] 104 Field stop

[0142] 105 aperture stop

[0143] 111 glass slide,

[0144] 112 Coverslip,

[0145] 121 Objective lens,

[0146] 122 Objective lens position adjustment mechanism

[0147] 123 Imaging lens

[0148] 124 camera components,

[0149] 131 Laser diode,

[0150] 132 Collimating lens,

[0151] 133 Collimating lens position adjustment mechanism

[0152] 134 aperture,

[0153] 135 Focus detection beam,

[0154] 136 1 / 2 wavelength plate

[0155] 137 Polarization beam splitter

[0156] 138 1 / 4 wavelength plate

[0157] 139 reflected beam,

[0158] 140a First reflected beam,

[0159] 140b Second reflected beam,

[0160] 141 Dichroic filter

[0161] 151 Converging Lens

[0162] 152 Cylindrical lens,

[0163] 153 Photodetector

[0164] 154 Non-polarizing beam splitter

[0165] 155a First cylindrical lens,

[0166] 155b Second cylindrical lens,

[0167] 156a First photodetector

[0168] 156b Second photodetector

[0169] 200 Computing Unit

[0170] 301 tissue sections,

[0171] 302 Sealant

[0172] 401 Focus Learning Points.

Claims

1. A microscope for observing samples covered by coverslips, characterized in that, The microscope has the following features: A collimating lens makes the light emitted from a light source approximately parallel; An objective lens that converges light transmitted through the collimating lens relative to the sample; A position adjustment mechanism that moves the position of the collimating lens; as well as A detector that detects reflected light from the sample. The position adjustment mechanism moves the collimating lens so that the light passing through the objective lens converges onto the surface of the coverslip. The detector detects the reflected light as a focus error signal representing the offset of the convergence point of the light passing through the objective lens relative to the focusing surface.

2. The microscope according to claim 1, characterized in that, With the objective lens pre-adjusted to focus on the sample, the position adjustment mechanism moves the position of the collimating lens so that the light passing through the objective lens converges onto the surface of the coverslip.

3. The microscope according to claim 1, characterized in that, The position adjustment mechanism moves the position of the collimating lens toward a direction where the focusing error signal approaches zero.

4. The microscope according to claim 1, characterized in that, The microscope further includes a computation unit that acquires a correction value map, which describes, according to the planar position on the surface of the coverslip, the correction values ​​used to correct for variations in the focusing error signal caused by the thickness of the coverslip. The arithmetic unit uses the correction value to correct the focus error signal at each of the plane positions described by the correction value mapping. The arithmetic unit controls the position of the objective lens in a direction that brings the corrected focus error signal closer to 0.

5. The microscope according to claim 4, characterized in that, The position adjustment mechanism moves the collimating lens to the position where the focusing error signal is 0. The computing unit obtains the focusing error signal based on aligning the focal point of the objective lens with the sample according to the planar position. The calculation unit records the focus error signal obtained according to the plane position as the correction value.

6. The microscope according to claim 1, characterized in that, The microscope further includes a computation unit that acquires a position map that describes the target position of the collimating lens where the focusing error signal is 0, according to the planar position on the surface of the coverslip. The arithmetic unit controls the position of the collimating lens toward the target position by mapping each of the plane positions described in the position map.

7. The microscope according to claim 6, characterized in that, The position adjustment mechanism moves the collimating lens to the position where the focusing error signal is 0. Based on aligning the objective lens with the sample according to the planar position, the arithmetic unit moves the collimating lens to the position where the focusing error signal is 0. The calculation unit records the position of the collimating lens where the focusing error signal is 0 at each of the planar positions as the target position.

8. The microscope according to claim 4, characterized in that, The microscope further comprises: a mapping device that creates a correction value mapping, which records, according to the planar position on the surface of the coverslip, correction values ​​used to correct for variations in the focusing error signal caused by the thickness of the coverslip. The microscope further includes a processing unit that uses the correction value mapping to correct the focusing error signal. The arithmetic unit uses the correction value at each of the planar positions described by the correction value mapping to correct the focus error signal. The arithmetic unit controls the position of the objective lens in a direction that brings the corrected focusing error signal closer to 0.

9. The microscope according to claim 6, characterized in that, The microscope also includes a mapping device for creating a position map that records the target position of the collimating lens where the focusing error signal is 0 according to the planar position on the surface of the coverslip. The microscope also includes a computing unit that uses the position mapping to control the position of the collimating lens. The arithmetic unit controls the position of the collimating lens toward the target position by mapping each of the plane positions described in the position map.

10. The microscope according to claim 1, characterized in that, The microscope also includes a beam-splitting element that guides the reflected light to the first detection optical system and the second detection optical system, respectively. The beam splitter guides the reflected light transmitted through it to the first detection optical system, and guides the reflected light from the beam splitter to the second detection optical system. The intensity distribution of the reflected light guided to the first detection optical system and the intensity distribution of the reflected light guided to the second detection optical system are linearly symmetrically reversed.

11. The microscope according to claim 1, characterized in that, The detector detects the focus error signal based on astigmatism.

12. The microscope according to claim 1, characterized in that, The detector has: a detection element, which is divided into at least four parts. Each of the aforementioned detection elements detects the reflected light and outputs its detection result.

13. A method for observing a sample covered by a coverslip, characterized in that, have: The step of using a collimating lens to make the light emitted from the light source approximately parallel; The step of using an objective lens to converge the light transmitted through the collimating lens relative to the sample; The step of moving the position of the collimating lens; as well as The step of detecting reflected light from the sample, In the step of moving the position, the position of the collimating lens is moved so that the light passing through the objective lens is focused onto the surface of the coverslip. In the detection step, the reflected light is detected as a focus error signal representing the offset of the convergence point of the light passing through the objective lens relative to the focusing surface.

Images