JP2915919B2 - Laser scanning fluorescence microscope - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a laser scanning fluorescence microscope that scans a fluorescent sample with a converged laser beam and detects fluorescence from the excited fluorescent sample to obtain a microscopic image of the sample. With regard to
The present invention relates to an improved method for acquiring an image in the thickness direction of a sample and an improvement in energy collection efficiency of fluorescence.

(Conventional technology and its problems) Fluorescence microscopes are currently generally used in many fields such as biology, medicine, and the semiconductor industry, and have high resolution, good contrast, and are suitable for quantitative measurement. For this reason, conventional transmission-type biological microscopes and electron microscopes have been replaced, or are being widely used as measuring devices to supplement these. In the illumination of a sample, a laser that overcomes the drawback of coherency of laser light, which is coherent light, with the concept of scanning, and that realizes high contrast and high resolution without the influence of scattered light by taking advantage of laser light Scanning fluorescence microscopes have recently been used for microscopic measurements.

FIG. 1 shows a principle diagram of an optical system of a commonly used uniform illumination epi-fluorescence microscope. Fluorescent microscopes often observe particularly minute structures, and therefore require higher resolution than ordinary biological microscopes. In addition, since the fluorescence (1) of the excited fluorescence is measured by weak light, the fluorescence It is desired to have high energy collection efficiency.

The resolution of the fluorescence microscope is limited by a numerical limit (Numerical Aperture; hereinafter, sometimes referred to as NA) of the objective lens (2) and a diffraction limit determined by the wavelength of the fluorescence (1). That is, in order to obtain higher spatial resolution (image a finer structure), it is necessary to increase the numerical aperture of the objective lens (2) or to use fluorescence of a shorter wavelength. However, the fluorescence wavelength is determined by the substance of the object to be measured / sample (3) and cannot be selected on the device side. On the other hand, the numerical aperture of the objective lens (2) is 1 even when an infinitely large lens is used for a dry lens.
Is the largest, even at most 1.5
It is about.

Moreover, the problem is that the traditional formula that the diffraction limit determines the resolution is limited to the case where the sample is only distributed in a certain plane.
If the sample (3) is thick, the resolution is much worse than the diffraction limit. This is because the defocus image of the structure of the sample on the surface near the focus surface is superimposed on the focus image.

For higher resolution, it is conceivable to use an objective lens with a higher numerical aperture, but this will result in a shallower depth of focus and a greater defocus effect outside of the focus plane, resulting in The resolution of the image is reduced. That is, if the numerical aperture of the objective lens is increased in order to increase the resolution, there is a contradictory problem that blur occurs from a position other than the focus surface and the resolution is reduced.

After all, in order not to be affected by defocus, it is necessary to use an objective lens having a focal depth greater than the thickness of the sample, and the resolution is determined by the numerical aperture of the objective lens.
However, in order to increase the depth of focus, it is necessary to reduce the numerical aperture of the objective lens. As the sample becomes thicker, the resolution decreases. It is well known that high resolution and long depth of focus are conflicting requirements. Only for infinitely thin samples, the above diffraction limit determines the resolution. However, since all samples are thick, the resolution is lower.

Another important point required for a fluorescence microscope is the energy collection efficiency of fluorescence. This energy collection efficiency is determined by the numerical aperture of the objective lens, and the larger it is, the higher the energy collection efficiency can be obtained, and
A good-quality fluorescent image with a good ratio can be obtained. But,
In order to observe a thick sample, as described above, a large depth of focus is required, and the numerical aperture must be reduced, which ultimately sacrifices energy collection efficiency.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above-mentioned problem of the objective lens regarding the resolution of an image in the depth direction (thickness direction, depth direction) of a thick sample and the above-mentioned problem of energy collection efficiency related to the objective lens. .

(Means for Solving the Problems) A first invention according to the present invention is a laser scanning method for scanning a fluorescent sample with a converged laser beam and detecting fluorescence from the excited fluorescent sample to obtain a microscopic image of the sample. A fundamental feature of a fluorescence microscope is that a means for converting a laser beam into an annular beam is provided in the optical path of the excitation optical system.

A second aspect of the invention is a confocal laser scanning fluorescence microscope, which is basically characterized in that a means for changing the area of a surface to be exposed of a photoelectric detector for detecting fluorescence is provided.

(Theoretical Background of the Present Invention) When analyzing the imaging characteristics of an optical system, an optical transfer function (OTF) is used. OTF represents the spatial frequency response of the optical system and is given by a two-dimensional Fourier transform of the point spread function of the intensity. The larger this value is, the better the frequency response is.

In the following, for comparison, the three-dimensional OTF of the uniform illumination epi-illumination fluorescence microscope shown in FIG. 1 and the three-dimensional OTF of the laser scanning fluorescence microscope (confocal type) shown in FIG.
Next, a newly derived three-dimensional OTF including a detector area for detecting fluorescence in a confocal type laser scanning fluorescence microscope (FIG. 2) including parameters as parameters is examined and finally shown in FIG. The three-dimensional imaging characteristics of a laser scanning fluorescence microscope using an annular pupil as the excitation optical system and a newly derived three-dimensional OTF will be discussed.

Uniform illumination epi-illumination fluorescence microscope In this microscope, a light beam from a light source (4) having an area is reflected by a dichroic mirror (5), and a fluorescent sample (3) is uniformly illuminated by an objective lens (2) to be excited. The fluorescence emission (1) from the fluorescent sample (3) thus observed is observed by an objective lens (2) with an image detector (6) (by visual observation through a CCD camera or an eyepiece). Sample (3)
It is illuminated so that the intensity distribution at
It is not necessary to consider the spatial distribution of the illumination light, and the entire optical system may treat the fluorescent object as an incoherent luminous body and deal only with the imaging characteristics of this luminescence. 3D OTF, H (ρ,
η) is given by equation (1).

H (ρ, η) = (1 / λ ² ρA) × Re ｛[1- (λρA + 2 | η | A / ρ) ² ] ^1/2 ｝ (1) FIG. 2 shows the principle of the optical system. In this microscope, a point light source (24) forms an image on a sample (23), and fluorescence (21) from the image forming point is measured by an object lens (22) with a point detector (26). For imaging, a point light source (24) and a point detector (26) are XY scanned. When an ideal point detector is used, only the fluorescence from the focus position can be detected, and the fluorescence from a position out of focus is not detected. Therefore, the out-of-focus image is not blurred, but loses its contrast and is not observed. Only one cross section in the z direction of the three-dimensional sample (33) is sharply observed.

Here, the three-dimensional OTF of this microscope is given assuming that it has a sufficiently small detector (26).

This optical system, unlike a uniform illumination epi-illumination fluorescence microscope, does not uniformly illuminate the sample. However, the image forming optical system is equivalent to the optical system of the uniform illumination epi-fluorescence microscope.
Therefore, the three-dimensional point spread function of the intensity by this optical system And the intensity distribution by the point light source. The intensity distribution from a point light source is also , The three-dimensional point spread function of the intensity of this optical system Is given by the following equation.

Since the product in the real domain corresponds to the convolution in the frequency domain, the three-dimensional OTF and Hc (ρ · η) of the confocal laser scanning fluorescence microscope are given by Expression (3).
Note that * represents convolution, and Hc (ρ · η) is expressed by equation (1). Given by the Fourier transform of.

Hc (ρ, η) = [H (ρ, η)] * [H (ρ, η)] (3) FIG. 3 shows the result of calculating equation (3) and the three-dimensional OTF of the above (1). Shown in 1A shows a uniform illumination system, and FIG. 2B shows an optical system of a confocal laser scanning fluorescence microscope.

ρ represents the spatial frequency in the in-plane direction, and η represents the spatial frequency in the depth direction. The vertical axis is three-dimensional for each frequency
The value of OTF is shown on a log scale, and represents the response to each spatial frequency.

From FIG. 3, the confocal laser scanning fluorescence microscope has twice the spatial frequency band in the depth direction (η direction) and the spatial frequency band in the in-plane direction (ρ direction) as compared with the uniform illumination epifluorescence microscope. It can be seen that has increased 1.7 times. Also,
In (a), the spatial frequency band in the depth direction becomes narrower toward the η axis, and only the origin has a value on the η axis. Accordingly, the uniform illumination epi-illumination fluorescence microscope has no decomposition for a sample having a structural change only in the depth direction. For example, when a thin film sample having no structural change in the plane is observed with a uniform illumination epifluorescence microscope, neither the position nor the thickness (z coordinate) of the sample can be known. On the other hand, in (b), there is no lack near the origin as in (a). Therefore, it can be seen that the laser scanning fluorescence microscope has decomposition even for a sample having a structural change only in the depth direction.

In a three-dimensional OTF confocal laser scanning fluorescence microscope in which the detector area is a parameter, as shown in FIG. 4, a pinhole (26p) is placed in front of a photomultiplier tube and a semiconductor photoelectric detector (26d). Aperture with opening (26a
p) is provided, and the pinhole (26p) limits the amount of light detected.

Here, the aperture (26ap) is a circular opening other than the pinhole (26p). The three-dimensional OTF is derived using the radius of the circular opening as a parameter.

Even if the area of the detector is changed, the excitation optical system is not affected. Therefore, only the imaging optical system needs to be considered.

Three-dimensional point spread function h of intensity due to fluorescence from the sample
(X, z) is given by the following equation using the pupil function P (μ) of the objective lens.

In the laser scanning microscope, since the image is obtained by scanning the detector as shown in FIG. 5, the three-dimensional point spread function of the intensity in the imaging optical system is obtained. Is the equation (4) and a function representing the shape of the circular opening. Given by the convolution with

 Here, a is the radius of the circular opening.

Here, it is assumed that the area of the detector is sufficiently small (a →
0) and Π are delta functions, and equation (5) is given by equation (6). FIG. 5 (b) schematically shows the state of scanning in this case.

Considering the intensity distribution of the excitation light, the three-dimensional point spread function of the intensity by the entire optical system is, as described above, the three-dimensional point spread function h ₂ of the intensity by the imaging optical system and the point spread function h ₂ It is given by the product of the intensity distribution by the light source. Therefore, the three-dimensional point spread function h ₃ of the intensity by the whole optical system is Given by

By performing a three-dimensional Fourier transform on the equation (7), a three-dimensional OTF and H ₃ (ρ, η) using a detector area as a parameter is obtained. here Is given by equation (8), Where J ₁ is a Bessel function of the first kind, The three-dimensional OTF, Ha (ρ, η) is given by the equation (9).

Ha (ρ, η) = H (ρ, η) * [H (ρ, η) · J ₁ (2πap) / πap] (9) Based on the equation (9), 3 The dimension OTF was calculated. The results are shown in FIG.
ρ represents the spatial frequency in the in-plane direction (r), and η represents the spatial frequency in the depth direction (z). The illustrated vertical axis shows the value of the three-dimensional OTF in each spatial frequency component on a log scale. The parameters used for the calculation are as follows.

Excitation wavelength: 488nm (Ar laser) Fluorescence wavelength: 520nm Lens F number: 0.5 (NA = 0.
7) Figures 6 (a), (b), (c) and (d) are based on the following conditions.

(A) ... a → 0 (confocal type) (b) ... a = radius of airy disk by excitation light (425n
m) (c)... a = radius twice as large as the radius of the air disk by the excitation light (850 nm) (d)... a → ∞ (uniform illumination epifluorescence microscope) The value of the radius a of the circular aperture is This is a value when the circular opening is projected on the sample surface, and in reality, the radius of the circular opening is (a × magnification of the microscope). The airy disk refers to the inside of the first dark ring of the airy pattern formed on the sample surface by the excitation light, as shown in FIG.

According to Fig. 6, when the area of the detector is the same as that of the air disk, or is expanded to twice the air disk, the three-dimensional OTF must have a value on the η axis and have three-dimensional resolution. I understand. Such an optical system has decomposition even for a sample having a structural change only in the z direction, for which the uniform illumination epi-fluorescence microscope has no decomposition.
In particular, when the area of the detector is the same as that of the air disk, the radius is twice as large as that of the air disk.
It has twice the depth resolution and shows a resolution close to that of a confocal type. Also, the energy collection efficiency is much higher than that of the confocal type.

6C and 6D, when the spatial frequency component in the in-plane direction is low, that is, the spatial frequency bandwidth in the depth direction is narrow near the η axis. That is, the resolution in the depth direction is affected by the structure in the plane. But,
In FIG. 6B, there is almost no missing portion near the origin as seen in FIGS. 6C and 6D. Therefore, when the size of the detector is the same as that of the air disk, the depth resolution is hardly affected by the in-plane structure.

Three-dimensional imaging characteristics using an annular pupil as the excitation optical system As shown in FIG.
Light of a low spatial frequency component can be cut, and light of a certain spatial frequency or higher can be transmitted.

Fig. 9 shows this annular pupil (aperture (98))
FIG. 1 shows a principle diagram of an optical system using (formed in). (94) is a point light source (laser), (97) is a lens, and (98) is an aperture that converts a laser beam into an annular beam. This annular luminous flux (99)
The light is reflected by the dichroic mirror (95), is converged by the objective lens (92), and forms an image on the three-dimensional sample (93). The fluorescence (91) from the excited sample is
Collected in 2) and detected by detector (96).

In this optical system, the intensity distribution on the sample surface by a point light source Is given by the following equation using the pupil function Pa (μ) of the annular pupil.

Here, the pupil function of the annular pupil is defined by the following equation.

The following equation (11) is obtained by subjecting the above equation (10) to three-dimensional Fourier transform and substituting the pupil function. However It is.

Hap (ρ, η) = ( 1 / λ 2 ρAmax) rect [4 | η | λA 2 max / (1- (Amax / Amin) 2)] × Re {[1- (λρAmax +2 | η | Amax / ρ ) ² ] ¹ /2-(Amax / Amin) [1- (λρAmin −2 | η | Amin / ρ) ² ] ^1/2 ｝ (11) On the other hand, a three-dimensional point image of the intensity due to light emission from the fluorescent object The three-dimensional Fourier transform of the distribution function is given by the following equation as described above.

H (ρ, η) × J ₁ (2πap) / πap (12) where J ₁ represents a Bessel function of the first kind, and H (ρ,
η) is given by equation (1).

Accordingly, the three-dimensional OTF, Hapa (ρ, η) of the entire optical system using the annular pupil as the excitation optical system is given by the following equation (13).

Hapa (ρ, η) = Hap (ρ, η) * [H (ρ, η) × J ₁ (2πaρ) / πaρ] (13) ρ represents the spatial frequency in the in-plane direction, and η represents the spatial frequency in the depth direction. Represents the spatial frequency. FIG. 10 shows 3 for each spatial frequency component.
The value of the dimension OTF is displayed on a log scale.

 The parameters used for the calculation are as follows.

Excitation wavelength: 488 nm (Ar laser) Fluorescence wavelength: 520 nm Amax: 0.5 Amin: 0.55 FIGS. 10 (a), (b), (c) and (d) are based on the following conditions.

(A) ... a → 0 (confocal type) (b) ... a = radius of airy disk by excitation light (425n
m) (c)... a = radius twice as large as the radius of the Airy disk by the excitation light (850 nm) (d)... a → ∞ Note that the actual radius of the circular aperture is a × (magnification of microscope).

Table 1 shows the in-plane resolution when the annular pupil is used for the excitation optical system and when it is not used. That is,
Shows the decomposition limit.

From Fig. 10, the optical system using the annular pupil as the excitation optical system is
It can be seen that the spatial frequency bandwidth in the η direction is uniformly narrow and has a large depth of focus. As the area of the detector increases, the depth of focus increases. On the other hand, the spatial frequency band in the in-plane direction does not decrease as much as the spatial frequency bandwidth in the depth direction even if the area of the detector is increased. Therefore, it is considered that the use of the annular pupil for the excitation optical system can increase the depth of focus while keeping the resolution in the in-plane direction high.

It can be seen from Table 1 that the optical system using the annular pupil as the excitation optical system has a higher in-plane resolution regardless of the area of the detector than the optical system not using the annular pupil. However, if an annular pupil is used for the excitation optical system, the spatial frequency band in the depth direction is narrowed, so that it should be correct to have a high in-plane resolution only near η = 0. In other words, it can be seen that high resolution can be obtained by performing excitation using a ring pupil on a sample having no structural change in the depth direction. That is, when a sample having no structure in the depth direction (the structure does not change in the z direction) is excited using a circular pupil, the resolution of the focus plane is high, but an image other than the focus plane remains blurred. After all, the fine structure of the focus plane cannot be observed in detail. On the other hand, when the same sample is excited by using the annular pupil, the resolution of the focus plane is slightly lower than that in the case of being excited by using the circular pupil, but no defocus occurs because the depth of focus is deep. Therefore, for observation of a sample having no structure in the depth direction, higher resolution can be obtained by excitation using the annular pupil.

Comparing Fig. 6 (excitation method with circular pupil) and Fig. 10 (excitation method with annular pupil), in the circular pupil excitation method, if the aperture in front of the detector is widened, the three-dimensional OTF becomes a figure-eight. Becoming (corresponding to the out-of-focus image blurring and spreading)
In the annular pupil excitation method, since the three-dimensional OTF is rod-shaped and the spatial frequency band in the z-axis direction is narrower than in the circular pupil excitation method, the depth of focus is deep, and the out-of-focus image is not blurred but intensity. It can be confirmed that is lowered (disappears). In a uniform illumination epi-illumination fluorescence microscope (corresponding to FIG. 6 (d)) or a confocal laser scanning fluorescence microscope (corresponding to FIG. 6 (a)), the depth of focus is increased (3 in the η-axis direction). Dimension OT
In order to reduce the width of F, the numerical aperture of the objective lens is reduced, but this also narrows the band in the ρ-axis direction as an example, resulting in a reduction in resolution. In addition, when the numerical aperture of the objective lens is reduced, the amount of fluorescent light that can pass through the objective lens decreases, so this method is fatal in the measurement of weak light, but there is no other method.

However, when an annular pupil is provided in the excitation optical system of the laser scanning fluorescence microscope, as shown in FIG. 10 (d), although the three-dimensional OTF is sufficiently narrow in the η direction, Rather than the circular pupil excitation method (FIG. 6 (d)), it has a wide band imaging characteristic, and can simultaneously realize high resolution and a long depth of focus.

(Example) FIG. 11 shows an example. An epi-fluorescence microscope (30) was used. The light source is an Ar laser with a wavelength of 488 nm (3
1) Spectra-Physics 161B. Fluorescence is detected by a CCD camera (32) of TI-23A manufactured by NEC Corporation. To be able to connect the CCD camera (32)
A relay lens (33) that can change the magnification from 1 to 2.25 is used. The sample (34) is placed on a vertically movable stage table (35) having an opening at the center. As the sample (34),
A glass slide coated with a fluorescent paint uniformly is used.

(38) is a beam expander for expanding the diameter of the laser beam, and the converging lens (38) converges this laser beam on the position of the field stop of the microscope (30). The convergent light goes to the dichroic mirror (39), where it is reflected,
It faces the objective lens (40) and forms an image on the sample (34) by the objective lens (40). The sample (34) is excited by the imaging spot of the light source and emits fluorescence. Sample (34)
Fluorescence from the object passes through the objective lens (40) and is passed through a 510 nm dichroic mirror (39) and an absorption filter (41).
After the light wave component having a wavelength of 520 nm or less is cut, the light passes through a relay lens (33) and is imaged on a light receiving surface of a CCD camera (32) attached to an image forming position above the lens barrel. In addition,
The objective lens (40) has a magnification of 20 and an NA of 0.5. The focal length of the lens (38) is 250 mm.

The fluorescent pattern received by the light receiving surface of the CCD camera (32) is read out by the CCD camera (32) and input as an analog video signal to the image capturing device (51), where it is converted into a 6-bit digital signal. After being converted,
Stored in frame memory. The image data is directly transmitted between the image capture device (51) and the computer (52).
The data is transferred to the computer (52) via the A bus. The image data processed by the computer (52) is displayed on a display monitor (53). In the embodiment, the image capturing device (51) uses a VM01B1 frame grabber manufactured by Shibasoku Co., Ltd., and the computer (52) uses DE (Ltd.).
Uses a C minicomputer or Micro VAX II.

In the embodiment, the effects of the detector area on the depth resolution and the energy collection efficiency are verified based on image processing. While shifting the sample (34) by 10 μm in the z direction, seven images (total image 256 × 256) were collected. A detector having each area was simulated on the 256 × 256 two-dimensional data thus obtained, and the output intensity from each detector having a different area was calculated. For example, if it is a pinhole,
The intensity of one point (one pixel) having the highest intensity in the image is taken as the output intensity (FIG. 12 (a)), and the output intensity from the detector having a certain area is determined by the pixel in the corresponding area in the image. (Fig. 12 (b), (c),
(D)). This calculation and the like were performed by the computer (52). FIG. 12 (a) shows a case of a pinhole.
(B) schematically shows the case where the radius is twice the radius of the airy disk, and (d) schematically shows the case where the area is large.

FIG. 13 is a plot of the above experimental results, showing the change in the output intensity from the detector with respect to the distance in the z direction.

FIG. 13 (a) plots the z coordinate when the focus position is set to 0 on the horizontal axis and the output intensity on the vertical axis. It can be seen that the intensity decreases as the area of the detector decreases. For example, when a pinhole is used, the detected fluorescence energy is very small.

On the other hand, FIG. 13 (b) plots the coordinates of the focus position on the horizontal axis and the area of each detector on the vertical axis, normalized by the output intensity at the focus position. As the area of the detector becomes smaller, the output intensity immediately decreases. Assuming a sufficiently large detector, the output intensity hardly changes.

For a sample having a structural change only in the depth direction, such as the sample used in this example, the uniform illumination epifluorescence microscope did not have any decomposition. From the above results, if the detector area is sufficiently large, such decomposition of the sample will be lost.However, make the detector the same size as the Airy disk or change the detector radius to the radius of the Airy disk. 2
When it is doubled, it can be seen that the sample having the structural change only in the z direction is decomposed and has a three-dimensional resolution. Approximately 30% when using a detector of the same size as an Airy disk, and approximately 70% when using a detector whose radius is twice the radius of an Airy disk, compared to the case using a pinhole. Has a reduced depth resolution. From this result, even if the area of the detector is increased from the pinhole to the same size as the air disk, the depth resolution only decreases by about 30%, and the energy collection efficiency is considerably higher than when the pinhole is used. can do.

In addition, in order to limit the light receiving area of the detector, a CCD camera (32) is used as in the embodiment, and image processing is performed. The detector may be installed, and an aperture having a predetermined opening may be provided in front of the detector. If you want to change the size of the opening, it is convenient to use a turret disk aperture (60) in which openings whose diameters are slightly changed are arranged along the circumference of the disk.
Is provided.

Although the scanning mechanism is omitted in FIG. 11, the scanning may be performed by any of the well-known techniques of beam scanning or stage scanning. In the case of beam scanning, one of a galvanometer mirror, a resonance galvanometer mirror, a polygon mirror, a turbine drive polygon mirror, an AO deflector, and a hologram scanner is used. In the stage scanning, a method using any of a stepping motor, a piezo element, a voice coil, and the like is employed.

In the embodiment shown in FIG. 11, when an annular pupil is formed in the excitation optical system, simply, between the laser (31) and the dichroic mirror (39), an annular aperture aperture for converting the laser beam into an annular beam is provided. (70) is installed. Either after the beam expander (36) or after the lens (38). Further, the reflection surface of the dichroic mirror (39) may be formed in an annular shape. More simply, a light-shielding circular mask whose center is coincident with the optical axis may be attached to the lenses (36) and (38).

FIGS. 14 and 15 schematically show how a sample is observed when the annular opening aperture (70) is provided. It is assumed that there is a circular object, a square object, and a triangular object within the depth of focus (z direction) of the sample.

In FIG. 14, when focused on a square object surface of 2, a confocal laser scanning fluorescence microscope (without aperture)
As shown in FIG. 15 (a), a clear image can be obtained only for the rectangular object, and the information is completely lost for the one circular object and the three triangular objects. However, when the annular aperture (70) is used, as shown in FIG. 15 (b), images of the object in all 1, 2, and 3 cross sections, and a clear image with good contrast are simultaneously observed. All cross-sectional images within a deep depth of focus can be observed in a list (in a perspective plan view) sharply without blurring.

The above description relates to a laser scanning fluorescence microscope and is directed to a fluorescent sample. However, the various methods according to the present invention disclosed herein are exactly the same for a microscope using Raman scattering in laser scanning. The method can be applied and the same effect can be expected, and the method of the present invention can be applied to a microscope using Raman scattering.

(Effects of the Invention) As described above, according to the first aspect of the invention in which the excitation is performed by the annular luminous flux, it is possible to increase the depth of focus while maintaining a high in-plane resolution. Therefore, it is possible to increase the NA of the objective lens, that is, to increase the diameter, thereby improving the energy collection efficiency of the fluorescent light. In addition, a bright image within the depth of focus range in the depth direction of the sample can be obtained in a list (in a perspective plan view). In addition, the confocal type eliminates the influence of the defocused image,
A sample image in a deep depth of focus range can be obtained in a list (in a perspective plan view) with high resolution and high contrast without a background.

According to the second invention in which the area of the detector for detecting fluorescence can be set to a predetermined area other than the pinhole, a high energy collection efficiency can be obtained in combination with a high depth resolution,
A high-performance fluorescence microscope having a high SN ratio and excellent three-dimensional resolution can be realized.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a principle diagram of an optical system of a uniform illumination epi-illumination type fluorescence microscope,
FIG. 2 is a principle diagram of an optical system of a confocal laser scanning fluorescence microscope, FIGS. 3 (a) and 3 (b) each show a three-dimensional OTF, FIG. 4 is an explanatory diagram of point detection, and FIG. Figures (a) and (b)
Fig. 6 (a) is an explanatory view of the scanning of the detector.
(B), (c), and (d) show three-dimensional OTFs with different detector areas, FIG. 7 is an explanatory diagram of an air disk, FIG. 8 is an explanatory diagram of an annular pupil, and FIG. Principle of the optical system of a confocal laser scanning fluorescence microscope using an annular pupil.
10 (a), (b), (c) and (d) show three-dimensional OTFs with different detector areas, FIG. 11 shows an embodiment, and FIGS. 12 (a) and (b) ), (C) and (d) schematically show the simulated detector area, respectively.
13 (a) and 13 (b) are graphs of fluorescence intensity plotting the results obtained using the examples, FIG. 14 is a diagram showing a schematic cross section of the sample, and FIGS. 15 (a) and (b). () Is an explanatory diagram schematically showing a difference between images observed by the confocal laser scanning fluorescence microscope and the microscope of the example. 31 ... Laser, 32 ... CCD camera, 34 ... Fluorescent sample, 40
…… Objective lens, 60… Aperture with variable aperture area, 70 …… Aperture with annular aperture.

Continuation of the front page (56) References JP-A-56-12615 (JP, A) JP-A-61-219920 (JP, A) JP-A-63-292111 (JP, A) JP-A-59-4954 (JP) JP-A-61-105521 (JP, A) JP-A-2-267512 (JP, A) JP-B-7-3508 (JP, B2) Journal of Microscopy, Vol. 149 (1988), pp. 51-66 OPTICS LETTERS, Vol. 10 (2) (1985), p. 53-55 (58) Field surveyed (Int. Cl. ⁶ , DB name) G02B 21/00-21/36

Claims (1)

(57) [Claims] 1. A confocal laser scanning fluorescence microscope, comprising: means for converting a laser beam into an annular beam in an optical path of an excitation optical system for exciting a fluorescent sample; and a photoelectric device for detecting fluorescence generated by the excitation. A laser scanning fluorescence microscope characterized in that a means is provided in proximity to a surface to be exposed of a detector and capable of variably setting the size of a circular opening including a pinhole.