阅读说明：本技术 用于对样本成像的荧光扫描显微镜和方法 (Fluorescence scanning microscope and method for imaging a sample ) 是由 约纳斯·弗林 拉尔斯·弗里德里希 于 2020-04-17 设计创作，主要内容包括：介绍了一种荧光扫描显微镜,其包括：激发光源,该激发光源被设计用于产生激发光分布,该激发光分布激发存在于样本中的荧光团,以便自发地发射荧光光子；去激发光源,该去激发光源被设计用于产生去激发光分布,该去激发光分布通过荧光光子的受激发射来使得由所述激发光分布在所述样本中激发的荧光团去激发；照明单元,该照明单元被设计用于将所述激发光分布和所述去激发光分布合并成沿着所述样本的多个照明目标点扫描的光分布,使得所述激发光分布的强度最大值和所述去激发光分布的强度最小值在相应的照明目标点中在空间上相互叠加；探测器,该探测器被设计用于检测从所述相应的照明目标点发射的荧光光子(根据这些荧光光子的到达时间)；和处理器。该处理器被设计用于,评估在所述相应的照明目标点中检测到的荧光光子的到达时间；基于所述评估生成代表相应照明目标点的第一图像点和第二图像点；把所述第一图像点合并成第一样本图像,并且把所述第二图像点合并成第二样本图像；并且,借助两个样本图像来确定在所述激发光分布的强度最大值和所述去激发光分布的强度最小值之间的空间偏移。(A fluorescence scanning microscope is presented, comprising: an excitation light source designed to generate an excitation light profile that excites fluorophores present in the sample so as to spontaneously emit fluorescent photons; a deenergizing light source designed to produce a deenergizing light distribution that deenergizes a fluorophore excited in the sample by the excitation light distribution through stimulated emission of fluorescent photons; an illumination unit designed for combining the excitation light distribution and the de-excited light distribution into a light distribution scanned along a plurality of illumination target points of the sample such that an intensity maximum of the excitation light distribution and an intensity minimum of the de-excited light distribution spatially overlap one another in the respective illumination target points; a detector designed to detect the fluorescent photons emitted from the respective illumination target points (according to their arrival times); and a processor. The processor is designed to evaluate the arrival times of the fluorescence photons detected in the respective illumination target points; generating a first image point and a second image point representing respective illumination target points based on the evaluation; merging the first image points into a first sample image and the second image points into a second sample image; and determining a spatial offset between an intensity maximum of the excitation light distribution and an intensity minimum of the de-excited light distribution by means of the two sample images.)

1. A fluorescence scanning microscope (100, 200) comprising:

an excitation light source (102) designed to generate an excitation light distribution (E) that excites fluorophores present in the sample (104) so as to spontaneously emit fluorescent photons;

a de-excitation light source (106) designed for generating a de-excited light distribution (D) that de-excites fluorophores excited in the sample (104) by the excitation light distribution (E) by stimulated emission of fluorescence photons;

an illumination unit (108) designed for combining the excitation light distribution (E) and the de-excited light distribution (D) into a light distribution scanned along a plurality of illumination target points of the sample (104) such that an intensity maximum (M) of the excitation light distribution (E) and an intensity minimum (N) of the de-excited light distribution (D) spatially overlap one another in the respective illumination target points;

a detector (110) designed to detect fluorescent photons emitted from the respective illumination target points as a function of their arrival times; and

a processor (112) designed to,

evaluating the arrival times of the fluorescence photons detected in the respective illumination target points,

generating a first image point and a second image point representing respective illumination target points based on said evaluation,

merging the first image points into a first sample image (P1) and the second image points into a second sample image (P2), and,

-determining a spatial offset (dx) between an intensity maximum (M) of the excitation light distribution (E) and an intensity minimum (N) of the de-excited light distribution (D) by means of two sample images (P1, P2).

2. The fluorescence scanning microscope (100, 200) according to claim 1, wherein the detector (110) is designed for detecting fluorescence photons emitted from the respective illumination target points by time-dependent single photon counting as a function of their arrival times.

3. The fluorescence scanning microscope (100, 200) according to claim 1 or 2, comprising an adjustment component (124) controllable by the processor (112) to influence the excitation light distribution (E) and/or the de-excitation light distribution (D) for compensating the spatial offset (dx).

4. The fluorescence scanning microscope (100, 200) according to claim 3, characterized in that the adjustment component (124) is controllable by the processor (112) to individually influence the excitation light distribution (E) and/or the de-excitation light distribution (D) for different regions of an image field for compensating the spatial offset (dx).

5. The fluorescence scanning microscope (100, 200) according to claim 3 or 4, characterized in that the adjusting means (124) is a light deflecting means arranged displaceably in the light path of the de-excitation light distribution (D) or in the light path of the excitation light distribution (E).

6. The fluorescence scanning microscope (100, 200) according to any of the preceding claims, wherein at least one of the two light sources (102, 106) is a pulsed or modulated laser light source and the detector (110) is designed for detecting the arrival time of the fluorescence photon with reference to a start time, which is determined by a light pulse or light modulation of the laser light source.

7. The fluorescence scanning microscope (100, 200) according to claim 6, wherein the processor (112) evaluates the fluorescence photons detected in the respective illumination target point in such a way that it compares the arrival time with a predetermined threshold value and assigns those fluorescence photons whose arrival time is smaller than or equal to the predetermined threshold value to the first image point and those fluorescence photons whose arrival time is greater than the predetermined threshold value to the second image point.

8. The fluorescence scanning microscope (100, 200) according to claim 6, wherein the processor (112) evaluates the fluorescence photons detected in the respective illumination target points in such a way that it determines a first fitting parameter and a second fitting parameter by adjusting a model function comprising the first fitting parameter and the second fitting parameter depending on a time distribution of the fluorescence photons given by the detected arrival time, and,

the processor (112) generates the first image point based on the first fitting parameter and generates the second image point based on the second fitting parameter.

9. The fluorescence scanning microscope (100, 200) according to claim 8, wherein the model function is given by the following function m (t):

m(t)＝a0*exp(-t/t0)+a1*exp(-t/t1)

where t represents the arrival time of the corresponding fluorescence photon,

t0 is the average lifetime of the fluorophore in the absence of said de-excited light distribution (D),

t1 is the average lifetime of the fluorophore in the presence of said de-excited light distribution (D),

a0 is the first fitting parameter, and,

a1 is the second fitting parameter.

10. The fluorescence scanning microscope (100, 200) according to any of claims 6 to 9, wherein the excitation light source (102) is a pulsed or modulated laser light source, which determines a start time.

11. The scanning fluorescence microscope (100) according to any of claims 6 to 10, wherein the deenergized light source (106) is a continuous wave laser light source.

12. The fluorescence scanning microscope (200) according to any one of claims 6 to 10, wherein the deenergized light source (106) is a pulsed or modulated laser light source.

13. The fluorescence scanning microscope (200) according to claim 12, comprising a delay unit (228) controllable by the processor (112) to time coordinate the deenergized light source (106) with the excitation light source (102) such that the light pulses or light modulations of the deenergized light source (106) have a predetermined delay from the light pulses or light modulations of the excitation light source (102) at the location of the respective illumination target point.

14. The fluorescence scanning microscope (200) according to claim 13, wherein the threshold value corresponds to a delay that a pulse or light modulation of the deenergized light source (106) has in a position of the respective illumination target point compared to a pulse or light modulation of the excitation light source (102).

15. The fluorescence scanning microscope (200) of claim 12,

the pulse length of the deenergized light source (106) is greater than the pulse length of the excitation light source (102), and/or,

the pulse length of the deenergizing light source (106) is in the range of the mean lifetime of the excited states of the fluorophores, in particular in the range from 0.1 to 6.0 ns.

16. The fluorescence scanning microscope (100, 200) according to one of the preceding claims, wherein the processor (112) is designed for determining a further mismatch of the deenergized light distribution (D) by means of two sample images in addition to the determination of the spatial offset (dx).

17. The fluorescence scanning microscope (100, 200) according to any one of the preceding claims, wherein the intensity minimum (N) of the de-excited light distribution (D) is an intensity zero point.

18. The fluorescence scanning microscope (100, 200) according to any one of the preceding claims, wherein the processor (112) is designed for determining the spatial offset (dx) in such a way that two sample images (P1, P2) are related to each other by a cross-correlation.

19. A method of imaging a sample (104) using a fluorescence scanning microscope (100, 200), comprising the steps of:

generating an excitation light distribution (E) that excites fluorophores present in the sample (104) so as to spontaneously emit fluorescence photons;

generating a de-excited light distribution (D) that causes, by stimulated emission of fluorescence photons, de-excitation of fluorophores excited in the sample (104) by the excitation light distribution (E);

merging the excitation light distribution (E) and the de-excited light distribution (D) into a light distribution scanned along a plurality of illumination target points of the sample (104) such that an intensity maximum (M) of the excitation light distribution (E) and an intensity minimum (N) of the de-excited light distribution (D) spatially overlap each other in the respective illumination target points;

detecting fluorescent photons emitted from the respective illumination target points as a function of their arrival times;

evaluating the arrival times of the fluorescence photons detected at the respective illumination target points;

generating a first image point and a second image point representing respective illumination target points based on the evaluation;

merging the first image points into a first sample image (P1) and the second image points into a second sample image (P2); and the number of the first and second electrodes,

-determining a spatial offset (dx) between an intensity maximum (M) of the excitation light distribution (E) and an intensity minimum (N) of the de-excited light distribution (D) by means of two sample images (P1, P2).

20. Method according to claim 19, wherein the spatial offset (dx) is determined in such a way that two sample images (P1, P2) are related to each other by a cross-correlation.

Technical Field

The present invention relates to a fluorescence scanning microscope and a method of imaging a sample using a fluorescence scanning microscope.

Background

In the field of fluorescence microscopy, for high-resolution imaging of samples, the so-called STED method is generally employed, in which the sample is illuminated with a light distribution resulting from the superposition of excitation light and de-excitation light. Here, STED represents "stimulated emission depletion". Here, the excitation light is intended to excite a fluorophore present in the sample so as to spontaneously emit fluorescence. In contrast, de-excitation light (having a wavelength different from that of the excitation light) is used to de-excite the fluorophore excited by the excitation light by stimulated emission of fluorescence. In order to increase the image resolution, the excitation light is superimposed with deenergized light having a specific light distribution, which is focused in the form of a laser beam onto the corresponding illuminated target point in the sample. This deenergized light distribution has an intensity zero point and adjoining it an intensity edge which rises as steeply as possible. In order to obtain the best possible image resolution, the deenergized light distribution must be superimposed on the excitation light distribution such that the zero point of the deenergized light distribution coincides exactly with the intensity maximum of the excitation light distribution. If this is ensured, the fluorescence spontaneous emission in the outer region of the excitation light distribution is suppressed in the illumination target point, so that the spontaneously emitted fluorescence is detected only from the region around the zero point of the deexcited light distribution. If the sample region is moved through a large number of illumination target points of the sample by the scanning method, a high-resolution image of the sample can be obtained by detecting fluorescence that is not suppressed by the deexcited light distribution.

If it cannot be ensured that the zero point of the de-excited light distribution coincides with the maximum of the excited light distribution, two effects of reducing the image quality occur. First, misalignment adjustment in superposition of deenergized light and excitation light causes a decrease in image brightness. Secondly, this misalignment adjustment highlights undesirable secondary maxima in the excitation light distribution. Both effects are shown in fig. 1, where for simplicity a one-dimensional light distribution in the direction x is assumed there.

Fig. 1a shows the case of optimal superposition of the deenergized light and the excitation light. Here, the solid line represents the excitation light distribution E, and the broken line represents the deenergized light distribution D. In this optimum case, the zero point, denoted by N, of the deenergized light distribution D exactly overlaps the maximum M of the excitation light distribution E in the x direction.

Fig. 1c shows the fluorescence distribution detected in the corresponding illumination target point, which results from an optimal superposition of the deenergized light and the excitation light according to fig. 1 a. This fluorescence distribution shows a high intensity and sharp maximum, the smaller half-width of which determines the spatial resolution in the direction x.

In contrast, fig. 1b shows the case of a misalignment adjustment, in which the zero point N of the deenergized light distribution D does not coincide with the maximum value M of the excitation light distribution E. This misalignment leads to the fluorescence distribution shown in fig. 1 d. As can be seen in comparison with fig. 1c, an inaccurate superposition of the de-excited light and the excited light leads to a significant reduction of the maximum of the fluorescence distribution. Furthermore, there are significant secondary maxima in the fluorescence distribution, as indicated by the arrows in FIG. 1 d. This secondary maximum occurs at points in the x-direction where fluorescence excitation is not quenched to a sufficient extent by de-excitation of the light distribution.

Various solutions are known from the prior art, which aim to ensure that the excitation maxima and the de-excitation zeros coincide locally. The first method is known in the art under the term "easy STED" and is described in publications EP 2158475A 2 and Reuss, M., "Simpler STED sessions", Ruperto-Carola University of Heidelberg,2010 (Phd. thesis). The basic idea disclosed in these publications is to direct excitation light and de-excitation light onto the sample through a common single mode fiber. This ensures that the two light distributions in the sample always exactly overlap in the desired manner. However, a problem is that a phase-shifting mask must be arranged between the exit end of the optical fiber and the objective, which phase-shifting mask is intended to influence the deexcitation light such that it has the desired light distribution at the illumination target point. In the conventional STED configuration, this phase shift mask only acts in the beam path of the deenergized light, but in contrast to this, in this method the excitation light also passes through the phase shift mask. In order to largely avoid the influence of the phase shift mask on the excitation light, the phase shift mask has to be manufactured from different materials in a complicated manner. In this case, it must be ensured that the phase shift mask produces the desired phase change at the wavelength of the deenergized light. The main disadvantage of this approach is therefore that the phase shift mask must be specifically optimized for a specific combination of excitation and de-excitation wavelengths. For this reason, the resulting microscope structure is inflexible with respect to the wavelengths that can be employed and hence the fluorophores that can be used. Another disadvantage is that the detected fluorescence also typically passes through a phase shift mask and is therefore attenuated.

Another method is described in DE 102007011305 a 1. In this method, the relative positions of the excitation light distribution and the de-excitation light distribution are measured using a calibration standard. Knowing the relative positions, the superposition can then be optimized by one or more actuators in one of the separate optical paths for the excitation light and the de-excitation light. One feature of this method is that the calibration sample is not inserted instead of the sample, but is swung into the intermediate image plane. In this way, it is not necessary to remove the sample for inspection or for optimal adjustment. However, a disadvantage of this method is that the calibration specimen has to be swung into the intermediate image plane and the microscope has to be reconfigured so that it operates in reflection mode. This slows down the method.

The adjustment of the STED microscope is also disclosed in the publication DE 102013227107 a 1. However, this method is mainly concerned with the alignment of the phase shift mask in the optical path of the deexcited light, and not with the superposition of the excitation light and the deexcited light.

All the above methods also have the following disadvantages: the effect of the sample on the relative positions of the excitation and de-excitation light cannot be taken into account. Thus, in particular, a change in the refractive index within the sample may affect the position of the light distribution. In Gould, t.j.; kromann, e.b.; burke, d.; solutions to this are described in "Auto-aligning simulated microwave using adaptive optics" of Booth, m.j. & beersdorf, j., op.lett., 2013,38, 1860-. In this method, the optimal superposition of excitation light and de-excitation light is determined by means of the high-resolution image itself. Here, a measure for evaluating the image quality is determined on the basis of the image brightness and the image sharpness. In the optimization process, the spatial light modulator, SLM for short, in the light path of the de-excited light is then controlled such that the metric is maximized. A disadvantage of this method is that for optimization a number of iteration steps are required, since the measurement value does not contain any information about the direction in which the deenergized light distribution has to be moved in order to achieve an optimal superposition.

With respect to the prior art, reference is also made to the method known in the art as "Time-Gated" STED ", which is described, for example, in Vicidomini, g.; moneron, g.; han, k.y.; westphal, v.; ta, H.; reus, m.; engelhardt, j.; eggeling, C. & Hell, S.W. "Sharper low-power STED nanoscopic by time gating" Nat Methods,2011,8,571 573. The method works with a pulsed laser light source to generate excitation light and a continuous laser light source, i.e. a continuously excited laser, for generating the de-excitation light. Here, the method takes into account that the detected arrival time of a fluorescence photon contains information about whether the photon came from a zero point region of the deenergized light distribution.

Disclosure of Invention

It is an object of the invention to provide a fluorescence scanning microscope and a method for imaging a sample using such a fluorescence scanning microscope, which allow a spatial offset between a maximum of the excitation light distribution and a minimum of the deenergized light distribution to be determined in a simple manner.

The invention achieves this object by the subject matter of the independent claims. Advantageous developments are given in the dependent claims.

A fluorescence scanning microscope, in particular a STED scanning microscope, is proposed, which comprises an excitation light source designed to generate an excitation light distribution which excites fluorophores present in the sample in order to emit fluorescent photons spontaneously, and a deenergizing light source designed to generate a deenergizing light distribution which deenergizes the fluorophores excited in the sample by the excitation light distribution by means of stimulated emission of the fluorescent photons. The fluorescence scanning microscope further comprises an illumination unit which is designed to combine the excitation light distribution and the deexcited light distribution into a light distribution which is scanned along a plurality of illumination target points of the sample, such that an intensity maximum of the excitation light distribution and an intensity minimum of the deexcited light distribution spatially overlap one another at the respective illumination target point. Furthermore, the fluorescence scanning microscope comprises a detector which is designed to detect the fluorescence photons emitted from the respective illumination target points (according to their arrival time). The fluorescence scanning microscope further comprises a processor designed to evaluate the arrival time of the fluorescence photons detected in the illumination target point. The processor is further designed for generating, based on the evaluation, a first image point and a second image point representing the respective illumination target points. The processor is further designed for merging the first image points into a first sample image and for merging the second image points into a second sample image. Finally, the processor is designed to determine a spatial offset between an intensity maximum of the excitation light distribution and an intensity minimum of the deenergized light distribution by means of the two sample images.

The solution proposed here therefore provides for the generation of two sample images which differ from one another in terms of the arrival time of the detection of the fluorescence photons which are taken into account for the generation of the respective sample image. The claimed fluorescence scanning microscope therefore makes use of the fact that the dwell time of a fluorophore in the excited state, hereinafter also referred to as the lifetime of the fluorophore, depends on whether the respective fluorophore lies in a zero region in which the light distribution is deenergized. Thus, in the zero point region of the deexcited light distribution, the lifetime is determined only by the spontaneous emissivity of the fluorophore. In contrast, in the region where the deexcited light distribution is not zero, the stimulated emission rate is also considered in addition to the spontaneous emission rate. This shortens the lifetime in the above-mentioned region. Thus, the arrival time detected by the detector also contains information about the spatial distribution of the fluorophore. This information is contained in the two sample images produced according to the invention, since these differ from one another in the arrival time of the fluorescence photons assigned to them. As a result, from this information, a spatial offset between the maximum intensity value of the excitation light distribution and the minimum intensity value of the deexcitation light can be determined by image comparison.

It is particularly advantageous here that a comparison of the two sample images directly enables the direction and distance of the movement of the de-excited light distribution relative to the excited light distribution to be determined in order to compensate for the offset, so that an optimal superposition of these light distributions is achieved. The direction and distance of the offset can be determined algorithmically, so that only a few iterations are required to eliminate the offset, possibly even only one iteration. In this way, the disadvantages explained at the outset as a result of the mis-alignment adjustment, such as a reduction in the image brightness and the emergence of undesirable secondary maxima in the excitation light distribution, can be avoided.

It is also particularly advantageous that the offset between the excitation light distribution and the de-excitation light distribution is determined on the sample to be imaged itself, rather than, for example, using a calibration specimen. The influence that the sample itself has on the relative position of the excited and de-excited light distribution, e.g. due to its varying refractive index, has thus been taken into account inherently.

Since the two sample images on which the spatial offset is determined represent different arrival times of the respectively detected fluorescence photons, and the respective arrival times in turn depend on the position of the fluorophore emitting the relevant fluorescence photon, the two sample images have different resolutions. In particular, the resolution of the sample image representing early fluorescence photons (i.e., photons with relatively short arrival times) is lower than the sample image representing late fluorescence photons (i.e., photons with relatively long arrival times).

The spatial offset between the excitation light distribution and the de-excitation light distribution can be determined in particular by means of a two-dimensional sample image. However, shifts in all three spatial directions may also be detected. In this case, a three-dimensional image stack is taken, rather than a two-dimensional sample image. In all cases, the sample image or image stack representing the early fluorescence photons contains information about the position of the maximum of the excitation light distribution, whereas the sample image or image stack representing the late fluorescence photons contains information about the position of the zero point of the de-excited light distribution. The offset may be determined such that two sample images or image stacks are associated with each other, for example by cross-correlation.

In a preferred embodiment, the processor is therefore designed to determine the spatial offset between the intensity maximum of the excitation light distribution and the intensity minimum of the deexcited light distribution by means of two sample images by correlating the two sample images with one another, for example by determining a cross-correlation between the two sample images or also only between mutually corresponding partial regions of the two sample images, for example between individual or a plurality of rows and/or columns and/or portions of the sample images. The use of such a cross-correlation is particularly suitable for determining the spatial offset by means of only two sample images, without a more complex and/or less accurate method known from the prior art being required for this purpose. The latter includes, for example, the use of calibration specimens or fitting to individual fluorescent point objects. In this connection, however, it should be emphasized that, in order to determine the spatial offset between the intensity maximum of the excitation light distribution and the intensity minimum of the deenergized light distribution, it is also conceivable within the scope of the invention, for example, to determine the position of one or more individual fluorescent point objects visible in the two sample images, in particular by fitting, instead of or in addition to the use of a cross-correlation.

The solution proposed here also has the following advantages: in order to determine the spatial offset between the excitation and deenergization light distribution, the configuration of the fluorescence scanning microscope does not have to be modified, since the measurement data required for this are produced to some extent as a by-product of the actual high-resolution image recording.

The detectors are preferably designed for detecting fluorescent photons emitted from the respective illumination target points (according to their arrival time) by time-dependent single photon counting. With such a time-dependent single photon counting, a particularly accurate detection of the arrival time can be achieved. However, the detector can also detect the arrival time of the fluorescence photon in other ways, for example by adding the light intensities, i.e. the incident fluorescence photons in at least two separate time intervals following one another, wherein these at least two intervals lie within the fluorescence lifetime after a start time, to which the arrival time of the fluorescence photons is referenced. Fluorescence photons associated with one time interval are then assigned to one sample image and fluorescence photons associated with another time interval are assigned to another sample image.

The fluorescence scanning microscope preferably comprises an adjustment component which is controllable by the processor to influence the excitation light distribution and/or to de-excite the light distribution for compensating the spatial offset. By using such an adjustment member, accurate superimposition of excitation light distribution and deenergized light distribution can be automatically achieved during actual image capturing, thereby ensuring high imaging quality.

The adjustment means may preferably be controlled by the processor to individually influence the excitation light distribution and/or to de-excite the light distribution for different regions of the image field for compensating the spatial offset. In this way, the superposition of these light distributions can be adaptively optimized, as long as it is ensured that the adjusting element operates at a sufficiently high speed.

In a preferred embodiment, the adjustment member is a light deflection member movably arranged in the light path of the deenergized light distribution or in the light path of the excited light distribution. Such a light deflecting component may for example be designed in the form of a fast tilting mirror. In order to be able to adjust the axial direction even if the superposition of the light distributions is corrected in three dimensions, a displaceable lens or a displaceable lens system is conceivable as an adjustment means, which ensures that one of the two light distributions can be displaced in the axial direction relative to the other light distribution. It is also conceivable to use a Spatial Light Modulator (SLM) which takes on this task. It is thereby possible to influence, within certain limits, the lateral deflection of the light distribution and, at the same time, the axial position of the light distribution. Furthermore, the component can also be used to generate the phase information required for a spatially specific deenergized light distribution.

In a particularly preferred embodiment, at least one of the two light sources is a pulsed or modulated laser light source, wherein the detector is designed to detect the arrival time of the fluorescence photon with reference to a start time which is determined by the light pulse or light modulation of the laser light source. The use of pulsed laser light sources has certain advantages over modulated sources, since the fluorophores considered here generally have short lifetimes and require very fast light modulation.

The processor preferably evaluates the fluorescence photons detected in the respective illumination target point in that it compares the arrival time with a predetermined threshold value and assigns those fluorescence photons whose arrival time is less than or equal to the predetermined threshold value to the first image point and those fluorescence photons whose arrival time is greater than the predetermined threshold value to the second image point. In this embodiment, the fluorescence photons are classified to some extent into early photons and late photons by means of a unique threshold. Here, a lower resolution sample image is generated based on the early fluorescence photons, and a higher resolution sample image spatially offset therefrom is generated based on the late fluorescence photons. The reason for the difference in resolution is that the effect of de-excitation light distribution increases with the dose of fluorophore applied. In other words, the longer the sample is illuminated by the de-energized light distribution, the greater the effect of the de-energized light distribution on the fluorophore. Thus, early fluorescence photons more correspond to the fluorescence signal obtained without using the de-excitation light, while late fluorescence photons represent the fluorescence signal corresponding to the signal detected in the conventional STED method. Thus, early fluorescence photons mainly provide information about the excitation light distribution, in particular its maxima, while late fluorescence photons mainly reflect the de-excitation light distribution, in particular its zeros.

In the above-described embodiment, the classification of fluorescence photons into early fluorescence photons and late fluorescence photons by means of a unique threshold value is a particularly simple method for determining the spatial shift of the light distribution.

However, the evaluation can be extended as follows: so that the fluorescence photons are not only classified into two classes, an early class and a late class, according to their arrival time at the detector, but the actual arrival times of all fluorescence photons are considered individually. For this purpose, the processor may, for example, evaluate the fluorescence photons detected in the respective illumination target points in such a way that it determines the first and second fitting parameters by adjusting a model function comprising the first and second fitting parameters as a function of the time distribution of the fluorescence photons given by the detected arrival times. The processor then generates a first image point based on the first fitting parameter and a second image point based on the second fitting parameter. In this way, for each image point, the model function is adjusted according to the temporal distribution of the detected fluorescence photons.

The model function is given, for example, by the following relation (1):

(1)m(t)＝a0*exp(-t/t0)+a1*exp(-t/t1)

here, t denotes the arrival time of the corresponding fluorescence photon, t0 is the average lifetime of the fluorophore in the absence of de-excited light distribution, t1 is the average lifetime of the fluorophore in the presence of de-excited light distribution, a0 is the first fitting parameter, and a1 is the second fitting parameter.

Fitting parameters a0 and a1 were determined for each illumination target point, respectively. By taking into account the actual arrival time, more information is used than in the case where the arrival times are merely divided into two categories. Thus, the spatial offset can also be determined more accurately.

The excitation light source is preferably a pulsed or modulated laser light source, which determines the start time. This has the following advantages: the conventional STED method also works with a pulsed excitation light source. Thus, in this regard, no changes need to be made to the existing STED configuration.

The deenergized light source can be designed as a continuous wave laser light source, but also as a pulsed or modulated laser light source. However, designing as a continuous wave laser light source has the following advantages: such a source is much cheaper than a pulsed source, especially if the latter is to be operated with high laser power in conventional STED applications.

A particularly preferred embodiment provides for a combination of pulsed excitation and continuous wave excitation, or a combination of pulsed excitation and pulsed excitation. However, a combination of continuous wave excitation and pulsed deexcitation may also be specified.

The fluorescence scanning microscope preferably comprises a delay unit which is controllable by the processor such that the deenergizing light source is coordinated in time with the excitation light source such that the light pulse or light modulation of the deenergizing light source has a predetermined delay in comparison with the light pulse or light modulation of the excitation light source at the location of the respective illumination target point. This embodiment is considered to be particularly advantageous when both the excitation light source and the deenergized light source are designed as pulsed or modulated laser light sources.

If the time of arrival of the fluorescence photons at the detector is divided into two categories by means of the above-mentioned threshold value, in the above-mentioned embodiment the threshold value corresponds to the delay that the light pulse or light modulation of the deenergized light source has at the location of the respective illumination target point compared to the light pulse or light modulation of the excitation light source.

In particular when both the excitation light source and the deenergizing light source are designed as pulsed laser light sources, the pulse length of the deenergizing light source is preferably greater than the pulse length of the excitation light source. Alternatively or additionally, the pulse length of the deenergizing light source may lie in the range of the mean lifetime of the excited states of the fluorophores, in particular in the range from 0.1 to 6.0 ns.

In a further embodiment, the processor can be designed to determine, in addition to the determination of the spatial offset, a further mismatch of the deenergized light distribution by means of the two sample images. In this embodiment, the spatial offset between the maximum of the excitation light distribution and the zero point of the deenergized light distribution is considered as a special case of a general mis-alignment adjustment of the deenergized light distribution, which takes into account further mismatches in addition to the aforementioned offset. An example of this is the mismatch due to spherical aberration. The latter can be reduced, for example, by the following measures: the brightness of two specimen images with low or high resolution is compared with one another and, with the aid of this comparison, for example, a Spatial Light Modulator (SLM) in the beam path of the deenergized Light or an actuator in the microscope objective is correspondingly controlled in order to reduce the spherical aberration.

This is of practical significance, especially when such spherical aberration adversely affects deenergized light distribution. The quality of the zero point, in particular of the deenergized light distribution, then has a considerable influence on the imaging quality. Due to its non-linearity, the STED method relies on the fact that the intensity of the de-energized light distribution is practically zero at its minimum and rises sharply with distance from this minimum.

Thus, the spherical aberration in the deenergized light path has a significantly more adverse effect than the spherical aberration in the excited light path. However, the sample image generated by the excitation beam alone may be used as a reference, e.g. as a reference for the image brightness. Here, by comparing the luminance of two sample images, it is not necessary that: a quality measure is used which is determined laboriously and by means of which the success of the spherical aberration correction must first be evaluated. By considering two sample images, the correction can be said to be an intrinsic reference, which is not the case, for example, if the correction quality is to be evaluated by means of a unique image.

As can be seen immediately from the above explanation, the intensity minimum of the deenergized light distribution is preferably an intensity zero point.

In another aspect, the invention features a method of imaging a sample using a fluorescence scanning microscope. The method comprises the following steps: generating an excitation light distribution that excites fluorophores present in the sample so as to spontaneously emit fluorescent photons; generating a de-excited light distribution that causes a fluorophore excited in the sample by the excitation light distribution to de-excite by stimulated emission of the fluorescence photon; combining the excitation light distribution and the de-excited light distribution into one light distribution scanned along a plurality of illumination target points of the sample such that an intensity maximum of the excitation light distribution and an intensity minimum of the de-excited light distribution spatially overlap each other in the respective illumination target point; detecting fluorescent photons emitted from the respective illumination target points (according to their arrival times) by time-dependent single photon counting; evaluating the arrival times of the fluorescence photons detected at the respective illumination target points; generating a first image point and a second image point representing the respective illumination target point on the basis of the evaluation; merging the first image points into a first sample image, and merging the second image points into a second sample image; and the spatial offset between the intensity maximum of the excitation light distribution and the intensity minimum of the deenergized light distribution is determined by means of the two sample images.

Drawings

The invention is explained below by means of embodiments with reference to the drawings. Wherein:

FIG. 1 is a graph illustrating how spatial superposition of de-excited and excited light distributions affects the detection of fluorescent signals;

FIG. 2 shows a schematic view of a fluorescence scanning microscope according to an embodiment;

FIG. 3 is a graph illustrating spatial distributions of fluorescence signals representing early fluorescence photons and late fluorescence photons in accordance with an embodiment;

FIG. 4 is a graph showing the spatial distribution of fitting parameters used for evaluation according to another embodiment;

FIG. 5 shows a fluorescence scanning microscope according to another embodiment; and

fig. 6 shows an exemplary graph illustrating how pulsed excitation light and pulsed de-excitation light are coordinated in time with each other.

Detailed Description

Fig. 2 shows a schematic view of a fluorescence scanning microscope 100 according to an embodiment. In the following, the basic structure and the basic operation of the fluorescence scanning microscope 100 will first be briefly outlined, and then a specific implementation according to the illustrated embodiment will be explained in more detail.

The fluorescence scanning microscope 100 comprises an excitation light source 102 designed to produce an excitation light profile E of the type shown in fig. 1 that excites fluorophores present in the sample 104 so as to spontaneously emit fluorescence light. The wavelength of the excitation light profile E produced by the excitation light source 102 is therefore designed for the fluorophore used in the particular application.

The fluorescence scanning microscope 100 further comprises a de-excitation light source 106, which is designed for generating a de-excited light distribution D of the type shown in fig. 1, which de-excites fluorophores excited in the sample 104 by the excitation light distribution through stimulated emission of fluorescence light. The wavelength of the de-excited light distribution D produced by the de-excited light source 106 is also matched to the fluorophore used in the particular application. The wavelength of the deenergized light distribution D is specifically selected such that fluorophores present in the sample 104 are reliably caused to return from their excited states to the ground state when illuminated with the deenergized light distribution D by stimulated emission. For this purpose, the deenergized light distribution D preferably has a wavelength approximately equal to the wavelength of fluorescence emitted by the fluorophore during transition from the excited state to the ground state.

The fluorescence scanning microscope 100 also has an illumination unit, generally indicated at 108 in fig. 2. The illumination unit is designed to combine the excitation light distribution E and the deenergized light distribution D into a superimposed light distribution and to scan a plurality of illuminated target points of the sample 104 with the light distribution thus produced. The combination proceeds as follows: so that the intensity maxima M of the excitation light distribution E and the intensity minima N of the deenergized light distribution D spatially overlap one another in the respective illumination target points. The spatial superimposition illustrated in fig. 1a, as described exemplarily in the opening paragraph, is intended here. However, since a certain misalignment adjustment cannot generally be avoided, a non-negligible spatial shift between the intensity maximum M of the excitation light distribution E and the intensity minimum N of the deenergized light distribution D often occurs, as exemplarily shown in fig. 1 b.

The fluorescence scanning microscope 100 also has a detector 110 that detects fluorescence photons emitted from the corresponding illumination target point. The detector 110 is designed to measure light intensity that varies rapidly with time. The detector is thus capable of detecting fluorescent photons emitted from the corresponding illumination target point by time-dependent single photon counting.

Finally, the fluorescence scanning microscope 100 comprises a processor 112 which enables the determination of the spatial offset between the excitation light distribution E and the de-excited light distribution D. To this end, the processor 112 evaluates the arrival times of the fluorescence photons detected in the respective illumination target points, which are determined by the detector 110. Based on this evaluation, the processor 112 then generates a first image point and a second image point, both representing the same illumination target point. The processor 112 applies the same to all illumination target points scanned with a light distribution consisting of an excitation light distribution E and a deenergized light distribution D. The processor 112 thus generates a plurality of first image points combined into a first sample image and a plurality of second image points combined into a second sample image. In this way, two sample images are generated, by means of which the processor 112 then determines the spatial offset between the intensity maximum M of the excitation light distribution E and the intensity minimum N of the deenergized light distribution D.

The structure specifically shown in fig. 2 represents only an exemplary embodiment for realizing the above-described functional principle, and in particular, the fluorescence scanning microscope 100 should not be limited to this specific embodiment. For example, in the embodiment according to fig. 2, the excitation light source 102 is designed as a pulsed laser light source, while the excitation deactivation light source 106 is a continuous wave laser light source. The two light sources 102, 106 each feed their light to an illumination unit 108 to which all microscope components are assigned, except for the two light sources 102, 106, the detector 110 and the processor 112 in the embodiment according to fig. 2.

Specifically, the excitation light source 102 emits excitation light L1, which is reflected by the fixed mirror 114 onto the first wavelength selective beam splitter 116. The first wavelength selective beam splitter 116 reflects the excitation light L1 onto a second wavelength selective beam splitter 118, which transmits the excitation light L1 in the direction of the grating device 120. In contrast, deenergized light source 106 emits deenergized light L2 onto phase-shift mask 122, which affects deenergized light L2 such that deenergized light distribution D generated from excitation light L2 in sample 104 has the desired zero point N. After passing through the phase shift mask 122, the deenergized light L2 is reflected onto the second wavelength selective beam splitter 118 at an adjustment component 124, such as a movable tilting mirror, that is controllable by the processor 112. Alternatively, for example, an SLM may also be used as the adjusting means 124. Further, the phase shift mask 112 and the adjusting section 124 may be formed of the same member such as an SLM. Wavelength selective beam splitter 118 reflects the de-excitation light L2 in the direction of grating device 120, which is also controlled by processor 112.

The excitation light L1 and the deenergized light L2 are then superimposed on each other by the second wavelength selective beam splitter 118 and fed to the grating device 120. From this light barrier arrangement, the superimposed light distribution is focused by the objective 126 onto the respective illumination target point, whereby a light distribution is generated in the illumination target point in the desired shape, which light distribution specifies a spatial superposition of the intensity maxima M of the excitation light distribution E and the intensity minima N of the deenergized light distribution D. The light barrier arrangement 120 ensures that the superimposed light distribution is moved along the sample 104, so that a large number of illumination target points of the sample 104 are scanned by the light distribution.

The sample 104 illuminated with the superimposed light distribution emits fluorescence L3, which is returned to the grating device 120 through the objective lens 126. In the exemplary embodiment according to fig. 2, a so-called descan detection of the fluorescence L3 is thus provided. The fluorescence L3 then passes through two wavelength selective beam splitters 118, 116 in sequence and falls onto detector 110, which detects fluorescence L3 and outputs a corresponding output signal S to processor 112.

As already explained, detector 110 detects fluorescence photons representative of fluorescence L3 by time-dependent single photon counting. The detector 110 detects the arrival time of the fluorescence photons with reference to a start time, which is determined in the exemplary embodiment according to fig. 2 by a light pulse emitted by the excitation light source 102, which is designed as a pulsed laser light source. To this end, the excitation light source 102 outputs an electrical trigger signal T to the processor 112, from which the time point of the respective laser pulse and thus the above-mentioned start time can be determined.

Based on the fluorescence photon arrival times evaluated by the processor 112, the processor 112 controls the adjustment component 124 in order to influence the de-excited light distribution D such that a spatial shift between an intensity maximum M of the excited light distribution E and an intensity minimum N of the de-excited light distribution D is compensated. In the exemplary embodiment according to fig. 2, the adjusting element 124 is located in the beam path of the deactivation light L2. However, it is also conceivable to provide corresponding adjusting means in the beam path of the excitation light E. The adjusting member 124, which in the embodiment according to fig. 2 is designed as a tilting mirror, also offers the following possibilities if it can be controlled fast enough by the processor 112: the deenergized light distribution D is influenced individually for different regions of the acquired image field in the adaptation process.

As explained further above, the processor 112 generates two sample images based on the evaluation of the fluorescence photons, by means of which sample images a spatial offset between the excitation light distribution and the de-excited light distribution can be determined. In the exemplary embodiment according to fig. 2, the processor 112 evaluates the fluorescence photons detected in each illumination target point for this purpose in that it compares the arrival time with a predetermined threshold value. The processor 112 assigns those fluorescence photons whose arrival time is smaller than or equal to the predetermined threshold value to the first image point corresponding to the respective illumination target point. Instead, the processor 112 assigns those fluorescence photons whose arrival time is greater than the predetermined threshold value to a second image point corresponding to the same illumination target point. In this manner, the processor 112 generates two sample images, one of which is assigned to early fluorescence photons and the other of which is assigned to late fluorescence photons. This way of evaluation is exemplarily shown in fig. 3 for the one-dimensional case.

In fig. 3, the dashed line P1 shows the curve of the change in the fluorescence signal of a point-like object generated by early fluorescence photons along the direction x. In contrast, the solid line P2 shows the relevant fluorescence signal obtained from the late fluorescence photon. The fluorescence signal P2 has a small half-width and is therefore suitable for producing high-resolution images of the sample.

The half-widths of the fluorescence signals P1, P2 are due to the fact that the fluorescence signal P1, represented by the early fluorescence photons, is still slightly influenced by the de-excitation light distribution D. In contrast, the de-excited light distribution D is fully effective for late photons, thereby reducing the half-width of the fluorescence signal P2 and improving resolution.

In the example according to fig. 3, the zero point N of the deenergized light distribution D is offset by a spatial offset dx with respect to the maximum M of the excitation light distribution E. The absolute distance and direction of the spatial offset dx can be determined from the fluorescence signals P1, P2 shown in fig. 3.

With regard to the view according to fig. 3, it should be noted that the fluorescence signals P1, P2 shown there are normalized. By this normalization, the intensity drop caused by the spatial offset dx and explained at the outset with reference to fig. 1d is calculated.

The example according to fig. 3 represents a particularly simple way of evaluating fluorescence photons, the comparatively expanded evaluation being shown in the view according to fig. 4. In this example, the processor 112 evaluates the fluorescence photons a1 detected in the respective illumination target points in that it adjusts the model function given above according to relation (1), for example, as a function of the temporal distribution of the detected fluorescence photons and determines therefrom two fitting parameters a0, a 1. Based on these two fitting parameters a0, a1, the processor 112 then generates two image points associated with the same illumination target point, and subsequently uses a large number of such image points to generate two sample images from which the spatial offset dx is determined.

Fig. 4 exemplarily shows the course of two fitting parameters a0 and a1 along the axis x, assuming that there is a spatial offset dx between the maximum M of the excitation light distribution E and the zero point N of the de-excited light distribution D. Here, the solid line shows a variation curve of the fitting parameter a1, and the dotted line shows a variation curve of the fitting parameter a 0. The curve of a1 is similar to the fluorescence signal P2 represented by late photons in FIG. 3. This is because the fitting parameter a1 is a measure of the number of long-lived fluorophores, only those which produce late fluorescence photons. In contrast, the curve fitted to the parameter a0 is similar to the fluorescence signal P1 represented by the early photons in fig. 3. In contrast thereto, however, the fitting parameter a0 has a minimum value at the zero point N position of the deenergized light distribution D. This is because the fitting parameter a0 is a measure of the short-lived fluorophore, and the short-lived fluorophore is not present at the zero-point N position of the deenergized light distribution D.

In fig. 5, a fluorescence scanning microscope 200 is shown, which is a modified embodiment compared to fig. 2. In this modified embodiment, not only the excitation light source 102 but also the deenergized light source 106 is designed as a pulsed laser light source.

The embodiment according to fig. 5 also has a delay unit 228 that can be controlled by the processor 112. Here, the processor 112 suitably controls the delay unit 228 such that the deenergized light source 106 is coordinated in time with the excitation light source 102. In particular, the delay unit 228 ensures that the respective light pulse of the deenergized light source 106 has a predetermined delay compared to the light pulse of the excitation light source 102 at the location of the illumination target point under consideration.

To achieve the desired time coordination, the excitation light source 102 outputs a trigger signal T to both the processor 112 and the delay unit 228. The delay unit 228 generates a trigger signal T' delayed with respect to the trigger signal T. Here, the delay may be adjusted by the processor 112. The delayed trigger signal T' is used to synchronize the pulses of light of the deenergizing light source 106 with the pulses of light of the excitation light source 102.

The evaluation of the fluorescence photons detected in the embodiment according to fig. 5 can be performed according to the methods shown in fig. 3 and 4. In the case of a simpler evaluation, in which the fluorescence photons are only classified into early and late photons, the delay T' can be set such that it corresponds to the threshold value for this classification.

With regard to the time coordination of the light sources 102, 106 in the embodiment according to fig. 5, the light pulses to avoid deenergizing the light source 106 are much shorter than the lifetime of the fluorophores. Furthermore, it is possible in principle, but disadvantageous, for the two light pulses to coincide exactly in time. Since in this case the stimulated emission of fluorescence photons occurs substantially simultaneously with the excitation. However, after stimulated emission, all fluorophores that are still excited have the same lifetime, and therefore are difficult to distinguish by the time they've fluorescence photons reach the detector 110. This situation is shown in fig. 6 a.

To avoid this, the delay unit 228 in fig. 5 may, for example, be configured such that the light pulses emitted by the deenergized light source 106 arrive at the sample 104 with a delay time interval dt from the light pulses emitted by the excitation light source 102. This solution is shown in fig. 6 b. Within a time span dt after the excitation light pulse, fluorescence photons are emitted according to the excitation light distribution E. Stimulated emission is then performed by outputting a deenergized light pulse. Here, a portion of the fluorescence is quenched, so that the fluorophore emits a photon preferably in the zero point N region of the deenergized light distribution D. In this way, the sample image generated by the early fluorescence photons provides information about the position of the maximum M of the excitation light distribution E, while the sample image generated by the late fluorescence photons provides information about the position of the zero point N of the deenergized light distribution D.

Another possible time coordination is shown in fig. 6 c. The pulse length of the deenergized light pulse can then be adjusted, for example, such that it lies within the mean lifetime range of the excited states of the fluorophores, for example in the range from 0.1 to 6.0 ns. In this case, the stimulated emissions are distributed along a time range of the pulse length of the de-energized light pulse, or along a portion of the de-energized light pulse that arrives at the sample 104 after the excitation light pulse in time. Due to this time stretching of the stimulated emission, the sample image built up from the excitation light distribution E can be obtained from the early photons, whereas the sample image consisting mainly of the region of the zero point N of the deenergized light distribution D can be generated from the late fluorescence photons.

The embodiments explained above are to be understood purely exemplary. . In particular, the embodiments are also not limited to the specifically described combination of pulsed laser light sources and continuous wave laser light sources. For example, the fluorescence scanning microscope 100, 200 can also operate in the following operating modes: excitation is performed by a continuous wave laser light source, and deenergization is performed by a pulsed laser light source. In this embodiment, the time at which the fluorescence photon reaches the detector 110 is related to the point in time of the pulse of de-excitation light D. The fluorescence photons detected shortly after the deenergizing pulse then contain information about the position of the zero point N of the deenergized light distribution D. In contrast, fluorescence photons detected shortly before or longer after the deenergizing pulse provide information about the position of the maximum M of the excitation light distribution E.

Although some aspects within the scope of the described apparatus have been described, it is clear that these aspects are also a description of a corresponding method, wherein a block or an apparatus corresponds to a method step or a function of a method step. Similarly, aspects described in the context of method steps are also a description of the respective block or component or of a characteristic of the respective device. Some or all of the method steps may be performed by (or using) hardware devices, such as processors, microprocessors, programmable computers, or electronic circuits. In some embodiments, one or more of the most important method steps may be performed by such an apparatus.

Embodiments of the invention may be implemented in hardware or software, depending on certain implementation requirements. Implementation can be via a non-volatile storage medium, such as a floppy disk, DVD, blu-ray, CD, ROM, PROM and EPROM, EEPROM or flash memory, on which electronically readable control signals are stored, which cooperate (or can cooperate) with a programmable computer system such that the respective method is performed. Accordingly, the digital storage medium may be computer-readable.

Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which can cooperate with a programmable computer system so as to carry out one of the methods described herein.

Generally, embodiments of the invention can be implemented as a computer program product having a program code, which is effective to perform one of the methods when the computer program product runs on a computer. The program code may be stored, for example, on a machine-readable carrier.

Other embodiments include a computer program stored on a machine-readable carrier for performing one of the methods described herein.

In other words, an embodiment of the invention is thus a computer program having a program code for performing one of the methods described herein, when the computer program runs on a computer.

Thus, another embodiment of the invention is a storage medium (or data carrier or computer-readable medium) comprising a computer program stored thereon for performing one of the methods described herein when it is executed by a processor. The data carrier, the digital storage medium or the recording medium is typically tangible and/or not seamless. Another embodiment of the present invention is an apparatus as described herein that includes a processor and a storage medium.

Thus, another embodiment of the invention is a data stream or a signal sequence forming a computer program for performing one of the methods described herein. The data streams or signal sequences can be configured, for example, in such a way that they are transmitted via a data communication connection, for example via the internet.

Another embodiment includes a processing mechanism, such as a computer or programmable logic device, configured or adapted to perform one of the methods described herein.

Another embodiment comprises a computer on which a computer program for performing one of the methods described herein is installed.

Another embodiment according to the present invention comprises an apparatus or system configured to transmit a computer program (e.g., electronically or optically) for performing one of the methods described herein to a receiver. The receiver may be, for example, a computer, a mobile device, a storage device, etc. The apparatus or system may comprise, for example, a file server for transmitting the computer program to the receiver.

In some embodiments, programmable logic devices (e.g., field programmable gate arrays, FPGAs) may be used to perform some or all of the functions of the methods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor to perform one of the methods described herein. In general, the methods are preferably performed by any hardware device.

List of reference numerals

100 fluorescent scanning microscope

102 excitation light source

104 sample

106 de-excitation light source

108 lighting unit

110 detector

112 processor

114 mirror

116 wavelength selective optical splitter

118 wavelength selective optical splitter

120 grating device

122 phase shift mask

124 adjustment member

126 objective lens

228 delay unit

De-excitation light distribution

E excitation light distribution

Intensity minimum of N deenergized light distribution

Intensity maximum of M excitation light distribution

P1 fluorescent Signal

P2 fluorescent Signal

a0 fitting parameters

a1 fitting parameters

dx offset

dt time interval

T trigger signal

T' trigger signal

L1 excitation light

L2 deenergizes light

L3 fluorescence