Method for imaging a sample using a fluorescence microscope with stimulated emission depletion - Patents.com

Abstract

A method for imaging a sample (18) using a fluorescence microscope (10) with stimulated emission depletion is described. The method includes an imaging process in which the fluorescence microscope (10) is controlled by a microscope control unit (42) to generate an illumination focus by superimposing a focused de-excitation light distribution on a focused excitation light distribution. When illuminated by the illumination focus, the sample (18) is excited and emits fluorescent imaging light only within an effective excitation focus, the effective excitation focus having a reduced range compared to the range of the excitation light distribution. Multiple sample segments are sequentially scanned within a target region of the sample (18) by the illumination focus at a first spatial resolution aligned with the reduced range of the effective excitation focus, and corresponding multiple image segments are imaged, generating a raster image (58). The fluorescence microscope (10) is further controlled by the microscope control unit (42) to generate an overview image (48) of the target region at a second spatial resolution prior to the imaging process. The second spatial resolution is lower than the first spatial resolution and higher than a third resolution tailored to a range of the excitation light distribution. Further, the overview image (48) is analyzed to identify image regions without associated image information, and during the imaging process, the radiant power of at least the de-excitation light distribution is reduced when scanning sample segments assigned to the identified image regions without associated image information in the overview image (48).

Description

[Technical field]

[0001] The invention relates to a method for imaging a sample using a fluorescence microscope with stimulated emission depletion, which comprises an imaging process, in which the fluorescence microscope is controlled by a microscope controller such that an illumination focus is generated by superimposing a focused de-excitation light distribution on a focused excitation light distribution, whereby the sample, when illuminated with the illumination focus, is excited and emits fluorescence imaging light only within an effective excitation focus, the extent of which is reduced compared to the extent of the excitation light distribution, and a number of sample segments are successively scanned by the illumination focus within a target area of ​​the sample with a first spatial resolution aligned with the reduced extent of the effective excitation focus and imaged into a corresponding number of image segments, from which a raster image is generated. The invention further relates to a fluorescence microscope operating in this way. [Background technology]

[0002] Recently, fluorescence microscopy techniques have been developed that allow imaging sample structures with a spatial resolution that is not limited by the resolution limits of optical microscopy as formulated by Ernst Abbe. The so-called stimulated emission depletion, or STED for short, is one of several methods that allow such an increase in spatial resolution.

[0003] In a STED microscope, a laser beam is focused on the sample in order to excite the fluorophores present there and output a fluorescence beam. The excitation light distribution thus generated at the sample is subject to the resolution limit of the optical microscope, i.e. has an arbitrarily irreducible spatial extent. Nevertheless, to allow an improvement in the spatial resolution, a focused de-excitation light distribution is superimposed on the excitation light distribution by the emission of light that de-excites the excited fluorophores through stimulated emission. This de-excitation light distribution, hereafter also called STED light, which forms an illumination focus together with the excitation light distribution, usually has an annular intensity profile with a central minimum where the intensity drops as far as possible to zero.

[0004] To improve the spatial resolution, a de-excitation light distribution is superimposed on the excitation light distribution such that its central intensity zero point is located exactly within the region of the intensity maximum of the excitation light distribution. Therefore, in this central region, de-excitation of the fluorescent dyes excited by the excitation light distribution caused by stimulated emission does not occur. In other words, in the central region, the spontaneous fluorescence emission is not suppressed by the de-excitation light distribution.

[0005] However, the spontaneous fluorescence emission is suppressed in the outer region of the excitation light distribution following the central region. In this outer region, the intensity of the superimposed de-excitation light distribution starts from zero and initially rises sharply. As a result, the superposition of the excitation light distribution and the de-excitation light distribution causes a reduction that increases the resolution of the sample area. This sample area can be excited by the excitation light and mainly emits fluorescent light that can be used for imaging. Such a sample area, whose extent is significantly reduced relative to the extent of the excitation light distribution, is referred to hereinafter as the effective excitation focus.

[0006] The above-described situation is illustrated by way of example in Fig. 1, which shows a cross-section of a typical light distribution used in STED microscopy for improved spatial resolution (x-direction in Fig. 1).

[0007] In FIG. 1, A shows the excitation light distribution, and B shows the de-excitation light distribution, i.e., STED light. The part of the excitation light distribution A that has only the excitation effect is located in the central region between the two vertical lines V and changes according to the course of the excitation light distribution in the central region, the excitation effect of this part is implied in FIG. 1 by the curve C. The curve C therefore also reflects the effective excitation focus mentioned above. The saturation intensity I sat is also shown in Figure 1. This shows the intensity required to completely de-excite the fluorophores contained in the sample through stimulated emission using STED light B.

[0008] The problem with the light distribution shown in FIG. 1 is that in order to restrict the effective excitation focus to as small an area of ​​the sample as possible, the STED light B must have a zero point in the center where the excitation light distribution has its maximum intensity, and at the same time must already have a saturation intensity I as close as possible to this zero point. sat Such a limited sample area defines the effective point spread function, or PSF for short, and therefore the spatial resolution. However, the required value I at zero and in the immediate vicinity of zero is sat Such intensity distributions of the STED light, with an intensity rising as strongly as possible to the zero point, are currently only produced with relatively high radiation powers. This entails that the intensity distribution of the STED light B first varies with increasing distance from the zero point, from I sat However, such a light intensity, which entails a high radiation power not necessary for fluorescence suppression per se, exposes the sample area to which it is applied to a particularly high light load, which can lead to strong photobleaching of the fluorochromes and even damage to the examined tissue. Therefore, during a conventional STED imaging process, all sample segments that are to be imaged with extremely high resolution are scanned, i.e. scanned, with the overall light distribution shown in Fig. 1, i.e. also scanned with an intensity overshoot of the STED light B, in particular far from the zero point. In the intensity overshoot, the light intensity reaches the saturation intensity I sat Higher than.

[0009] Photobleaching, which occurs in conventional STED microscopy, often precludes multiple imaging of a sample, for example to detect changes in the sample over time. Often, photobleaching is so strong that it is not possible to record one raster image at a time, or the recorded images contain very little signal and are difficult to evaluate. This is particularly problematic in three-dimensional image recording processes, where each sample segment is hit by the light distribution very frequently and is therefore bleached particularly strongly.

[0010] From the prior art, several solutions are known that aim to reduce photobleaching even in STED applications. One of these solutions proposes to measure, during the raster process, at each pixel or sample segment whether the fluorescence signal is above a predefined boundary value during a short diagnostic period. If the fluorescence signal is above a predefined boundary value, the illumination focus, which is combined from the excitation light distribution and the STED light, remains switched on and the detected fluorescence signal for this sample segment is stored. On the other hand, if the detected fluorescence signal during the diagnostic period is below the boundary value, it is assumed that no fluorochromes are present in the STED detection area, and therefore the illumination focus is switched off. This leaves fluorochromes outside the STED detection area unharmed by the illumination focus.

[0011] In "Strong signal increase in STED fluorescence microscopy by imaging regions of subdiffraction extent" (PNAS 2017, vol. 114 no. 9, 2125-2130) by Goettfert et al., only very small subregions of the sample are scanned with the illumination focus. Such subregions are small, and the region of interest is much smaller than the confocal PSF. In this way, it is ensured that the sample structures of interest are only contacted with the STED light distribution in the region directly following the zero point, but not in its outer regions, where the STED light distribution has a particularly high intensity. Therefore, the method can only be used for very small image regions, typically below 150 nm. Furthermore, special electro-optical scanners are required to implement the method, since conventional mechanical scanners are not suitable for such small image regions.

[0012] Heine et al., "Adaptive-illumination STED nanoscopy" (PNAS 2017, vol. 114 no. 37, 9797-9802) also presents a method to analyze the fluorescent signal coming from each individual sample segment during the STED imaging process. However, here the signal is not analyzed at the full intensity of the illumination focus. Rather, the radiation power of the illumination focus is gradually increased.

[0013] In such a process, which is carried out pixel by pixel, the detection of the STED light intensity, which is carried out for analytical purposes, already takes some time. Only afterwards does the actual data recording for this pixel take place, which also takes time. This makes the overall measurement time extremely long. The dwell time of the illumination focus is therefore currently about 80-100 μs per pixel. Typically, when recording a two-dimensional image with a size of 10 μm×10 μm for a pixel edge length of 10 nm, this results in an overall measurement time of 80-100 seconds. If a three-dimensional image is recorded, such a measurement time occurs for each individual image plane, which significantly increases the time required.

[0014] It should be noted that when using conventional mechanical scanners, it is usually easy to quickly jump from a pixel where no signal was detected to the next pixel, thereby shortening the dwell time at the former pixel by at least the time required to read out the image signal. Such jumps introduce irregularities into the scanning process due to the inertia of the scanner, which causes image drops. Therefore, in the prior art, all pixels must always be scanned at the same slow speed. Summary of the Invention [Problem to be solved by the invention]

[0015] It is an object of the present invention to provide a method for imaging a sample using a fluorescence microscope with stimulated emission suppression that is fast and sample-friendly. Furthermore, it is an object of the present invention to provide a fluorescence microscope that operates accordingly. [Means for solving the problem]

[0016] The invention achieves the above-mentioned object by means of the features of the independent claims. Advantageous developments are given in the dependent claims.

[0017] The present invention provides a method for imaging a sample using a fluorescence microscope with stimulated emission suppression, the method comprising an imaging process, in which the fluorescence microscope is controlled by the microscope controller in such a way that an illumination focus is generated by superimposing a focused de-excitation light distribution for stimulated emission suppression on a focused excitation light distribution, whereby the sample, when illuminated with the illumination focus, is excited and emits fluorescence imaging light only within an effective excitation focus, the extent of which is reduced compared to the extent of the excitation light distribution, and a number of sample segments are successively scanned by the illumination focus in a target area of ​​the sample with a first spatial resolution adapted to the reduced extent of the effective excitation focus and imaged into a corresponding number of image segments, from which a raster image is generated. In the present invention, the fluorescence microscope is further controlled by the microscope controller in such a way that, prior to the imaging process, an overview image of the target area is generated with a second spatial resolution, where the second spatial resolution is lower than the first spatial resolution and higher than a third spatial resolution adapted to the extent of the excitation light distribution. Further, the overview image is analyzed to identify image regions without associated image information, and during the imaging process, the radiant power of at least the de-excitation light distribution is reduced when scanning sample segments that are assigned to the image regions without associated image information identified in the overview image.

[0018] In order to protect the sample from photobleaching, the invention provides a step that is performed before the original imaging process, in which an overview image of the target area is recorded in order to identify image areas therein that do not carry relevant image information. In such image areas, there are no sample structures of interest that must be imaged with very high spatial resolution in the subsequent STED imaging process. For sample segments that are assigned to such image areas that are not relevant for imaging, the illumination focus can therefore be adjusted so that the resulting light load on the sample is reduced. In the present invention, this is done by reducing at least the radiation power of the de-excitation light distribution, which is relatively high in the normal STED imaging process in order to minimize the effective excitation focus. In order to be able to analyze the overview image in order to identify image areas that do not carry relevant image information, the overview image is recorded with a spatial resolution that is, on the one hand, higher than the resolution obtained using only the excitation light distribution, and on the other hand, lower than that used in the original STED imaging in order to generate a raster image resolved with very high resolution. This allows overview images to be recorded much faster than STED raster images.

[0019] Thus, the methods of the present invention offer a significant speed advantage over conventional methods, which are designed to avoid bleaching effects and which determine whether the STED light should be turned on or off pixel by pixel during the scanning process, i.e. for each sample segment imaged. Thus, the generation and analysis of overview images according to the present invention saves the time required for the analysis process, which in the prior art is performed pixel by pixel.

[0020] The present invention uses a saturation intensity I satThis allows for an effective bleaching reduction by adjusting at least the emission power of the STED light based on the analysis of the overview image in order to expose as few sample structures as possible to a STED intensity above I. This can be done, for example, in that the full STED intensity is only activated if the sample structure to be imaged or as large as possible the process structure part is located in a region closely distributed around the central zero point of the STED light, i.e., referring to FIG. 1 , between the two vertical lines V. If the sample structure to be recorded is located in a region in which the STED intensity would be higher than the saturation intensity for the full STED emission power or in a region where the excitation light does not reach at all, the STED emission power can be set according to the method of the invention so that the sample structure is exposed to at most the saturation intensity I. sat It can be set so that only the excitation light is applied. Alternatively, the STED light can be switched off completely. The full STED radiation power should be set only if the sample structure to be recorded is located within a close region around the zero point of the STED light distribution. Such a region typically extends for a distance of 50-100 nm from the zero point. In either case, the range is usually smaller than the range of the excitation light distribution, which specifies the spatial resolution of the confocal microscope used for STED imaging.

[0021] Advantageously, when scanning sample segments that are assigned to image areas identified in the overview image and that do not have associated image information, the radiant power of the excitation light distribution is additionally reduced, which makes it possible to reduce photobleaching more effectively.

[0022] In an advantageous configuration, the radiation power is reduced to zero, which means that the excitation light or the de-excitation light used for stimulated emission suppression is switched off during the STED imaging process for sample segments that do not have the sample structure of interest.

[0023] The overview image is advantageously generated at the second spatial resolution by first recording a raw image at a spatial resolution lower than the second spatial resolution and then performing image processing on this raw image to increase the spatial resolution. Such an embodiment provides a two-stage process for generating the overview image, where in a first, substantially hardware-based stage, image data is provided in the form of a relatively low-resolution resolved raw image, which is then further processed in a second, software-based stage into a higher-resolution resolved overview image.

[0024] In a particularly advantageous configuration, the raw image is generated by scanning the target area with the excitation light distribution at the third spatial resolution, which has the advantage that for the generation of the raw image substantially the same imaging process can be used as for the subsequent recording of the STED image, with the difference that for recording the raw image the sample is illuminated only with excitation light and not with de-excitation light.

[0025] The above-mentioned configuration of the method is advantageously developed in that the raw images are generated by confocal raster imaging. The confocal raw images, typically with a spatial resolution in the range of 250 nm, form a suitable basis for subsequent image processing, whereby a more highly resolved overview image is generated.

[0026] The generation of the raw images does not necessarily have to be based on a scanning microscope process: the raw images may for example be generated using wide-field illumination.

[0027] In a particularly advantageous embodiment, the image processing provides for deconvolution of the raw image to generate an overview image. For example, in the prior art, highly efficient deconvolution algorithms are available, which intelligently utilize the computing power of modern workstation computers, allowing extremely short computing times of less than a few tens of milliseconds. This allows overview images to be generated in an extremely short time. Furthermore, deconvolution is particularly well suited to improving image resolution when the sample is stained with a rather low density of fluorescent dyes, i.e. when the fine sample structures examined in the original imaging process are not spatially too closely adjacent. In such circumstances, the method of the invention is particularly well suited to reducing bleaching.

[0028] In an alternative embodiment, the overview image is generated at the second spatial resolution by direct imaging of the target area, in which case no additional image processing is required.

[0029] For example, an overview image can be generated using patterned illumination.

[0030] It is also possible to generate an overview image by scanning the target area with the illumination focus, in which case the radiation power of the de-excitation light distribution is reduced compared to the radiation power used in the imaging process. On the one hand, the reduced radiation power is weak enough that it does not bleach the sample or only bleaches it very slightly. On the other hand, the reduced radiation power is at least strong enough that the desired spatial resolution of the overview image is obtained. In such a case, it is also possible to record several overview images with weaker but different STED radiation powers.

[0031] However, other methods for directly generating the overview image can also be used, such as the method referred to in the literature as "Pixel Reassignment", which is described, for example, in "Optik" by CJR Sheppard (1988) and "Optics Letters" by CJR Sheppard et al. (2013).

[0032] Advantageously, based on the analysis of the overview image, a radiant power table is created and stored, which assigns at least one value of radiant power of the de-excitation light distribution to each sample segment imaged during the imaging process.

[0033] In a particularly advantageous embodiment, exclusion regions are identified from image regions without associated image information identified in the overview image, and sample segments assigned to these exclusion regions are skipped when the target region is scanned at the first resolution. This allows, for example, in a high-resolution STED imaging process in which one or more mutually consecutive rows of image segments are assigned, to skip relatively large sample regions, which further accelerates the imaging. Skipping entire rows is particularly advantageous, since the scanning mirrors used in conventional STED scanning microscopes that direct the illumination focus to the next row are moved relatively slowly, and image loss is not noticeable or is much less noticeable when the scanning mirror has to move different distances in various regions. This is the case, for example, when the scanning mirror continues to jump row by row in rows with image information and skips one or several rows in rows without image information. Here, processing can be carried out with conventional mechanical scanners that allow scanning of large image regions. In particular, no special electro-optical scanners are used that can only scan very small image regions.

[0034] Advantageously, the second spatial resolution is 1.5 to 2 times higher compared to the third spatial resolution.

[0035] The invention will be explained in more detail below on the basis of the drawings. [Brief description of the drawings]

[0036] [Figure 1] FIG. 1 shows a cross-section of the light distribution used in STED microscopy to improve spatial resolution. [Diagram 2] 1 shows a schematic diagram of a STED microscope of the present invention. [Diagram 3] 2 shows a schematic representation of a raw image produced in accordance with the present invention, in which sample structures are visible; [Figure 4] 2 shows a schematic diagram of an overview image produced in accordance with the present invention, in which a sample structure is visible; [Diagram 5] 1 shows a schematic diagram illustrating the function of a table created on the basis of an overview image for controlling the radiant power of an illumination focus. [Figure 6] FIG. 1 shows a schematic diagram illustrating the resulting STED raster image, in which the sample structures are visible. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0037] FIG. 2 shows a STED microscope 10 according to the invention, which is designed to image fluorescing sample structures with high spatial resolution using the principle of stimulated emission depletion, known per se.

[0038] The STED microscope 10 comprises an excitation light source 12 and a de-excitation light source 14, each configured as a laser light source. The excitation light source 12 emits an excitation light beam 16, the wavelength of which is selected such that it excites fluorochromes contained in the sample 18 to be imaged for spontaneous emission of fluorescent light. In contrast, the de-excitation light source 14 emits a de-excitation light beam 20, the wavelength of which is designed to de-excite fluorochromes contained in the sample 18 according to the principle of stimulated emission depletion.

[0039] The excitation light beam 16 emitted by the excitation light source 12 is reflected by the first dichroic beam splitter 23 towards the second dichroic beam splitter 24, where the excitation light beam 16 merges with the de-excitation light beam 20 that has previously passed through the phase modulator 22. To this end, the second beam splitter 24 is configured such that it passes the excitation light beam 16 while reflecting the de-excitation light beam 20.

[0040] The light beam 16 and the light beam 20, superimposed on each other by the second beam splitter 24, further propagate to a raster module 26 that includes a raster mirror 28. After reflection at the raster mirror 28, the excitation light beam 16 and the de-excitation light beam 20 pass through a scan optics 30 as well as a tube optics 32, and are then focused onto the sample 18 by an objective lens 34.

[0041] The light beams 16, 20 focused by the objective lens 34 are superimposed in the sample 18 to form an illumination focal point. The raster module 26 moves the illumination focal point over the sample 18 in a scanning motion by tilting the raster mirror 28. In this way, a target area of ​​the sample 18 is scanned by the illumination focal point during the STED imaging process.

[0042] The phase modulator 22 influences the de-excitation light beam 20 emitted by the de-excitation light source 14 in such a way that the de-excitation light distribution generated in the sample 18 by the objective lens 34 forms a ring focus. The ring focus is located on the outer region of the excitation light distribution focused on the sample 18 by the objective lens 34. As a result, the sample area illuminated at the de-excitation focus is de-excited through stimulated emission depletion and is not further excited by the excitation light distribution to emit fluorescence imaging light. Thus, an effective excitation focus is formed in the central region of the illumination focus, the extent of which is smaller than the extent of the excitation light distribution focused on the sample 18. Only within such an effective excitation focus, the fluorescent dyes contained in the sample are excited to emit fluorescence light.

[0043] The spatial resolution with which the sample 18 is scanned with the illumination focus during the STED imaging process is adapted to the range of the above-mentioned effective excitation focus, which is significantly reduced by the superposition of the de-excitation light distribution through stimulated emission suppression in comparison with the range of the excitation light distribution radiated to the sample 18, resulting in a correspondingly high spatial resolution with which a raster image is generated during the STED imaging process.

[0044] 2 passes through the objective lens 34, the tube optics 32 and the scan optics 30 before returning to the raster mirror 28. After reflection from the raster mirror 28, the fluorescent light 36 passes in turn through two dichroic beam splitters 24, 23, the detection pinhole aperture 26 and a filter 38 before being detected by a detector 40.

[0045] The STED microscope 10 further comprises a microscope controller 42 which controls the overall operation of the STED microscope 10. In particular, the microscope controller 42 controls the methods for reducing photobleaching effects, which are described hereinafter.

[0046] 3 to 6 show, purely by way of example, how the method according to the invention can be implemented.

[0047] In a first step, the microscope control 42 drives the STED microscope 10 in such a way that, prior to the actual STED imaging process, a raster image representing a confocal raw image 44 to be further processed according to the invention is recorded in a manner known per se in confocal mode. For this purpose, the sample 18 is illuminated only by the excitation light beam 16 and not by the de-excitation light beam 20. The excitation light beam 16 is now moved in a scanning motion over the sample 18 by the raster mirror 28. The spatial resolution with which the confocal raw image 44 is generated in such a first step is adapted to the spatial range of the excitation light distribution that the objective lens 34 focuses on the sample 18. Such a range is much larger than the range of the effective excitation focus utilized in STED imaging, so that the spatial resolution is much reduced compared to the STED resolution. As a result, the confocal raw image 44 can be generated much faster and more gently on the sample than the STED image to be recorded later. As shown in FIG. 3, a sample structure of interest 46 is imaged in a raw confocal image 44 at a relatively low spatial resolution.

[0048] 4 shows a second step of the method, in which an overview image 48 is generated on the basis of the confocal raw image 44 in order to increase the spatial resolution by image processing, for example by using a deconvolution algorithm. In the illustrated embodiment, the spatial resolution of the overview image 48 is about 1.5 to 2 times higher compared to the spatial resolution of the confocal raw image 44. Correspondingly, the sample structure, labeled with reference number 50 in FIG. 4, is imaged much sharper than in the confocal raw image 44 shown in FIG.

[0049] In FIG. 5, a third step of the method is shown, in which the overview image is first analyzed in order to identify image regions without associated image information. In such a third step, a radiation power table is then created and stored based on the analysis of the overview image 48. It assigns to each sample segment that is to be imaged with a relatively high spatial resolution in the subsequently performed STED imaging process control information regarding the radiation power of the illumination focus that should be used for such a sample segment. In this embodiment, such control information indicates whether the entire illumination focus should be switched on or off for the sample segment under observation. FIG. 5 shows a map 52 representing the assignment of the above-described control information to the individual sample segments in the form of a radiation power table. Here in FIG. 5, the image regions outside the region marked with the reference number 56 are assigned to sample regions for which the illumination focus should be switched off in the STED imaging process. In contrast, the above-mentioned region 56 indicates sample segments for which the illumination focus should be switched on in the STED imaging process.

[0050] Figure 6 finally shows the fourth step of the method in which the STED imaging process is performed, taking into account the control information generated in step 3. Figure 6 now shows the resulting STED raster image 58 with sample structures 60 visible in high resolution therein.

[0051] It should be pointed out again here that Figures 3 to 6 show only a special and particularly simply constructed realization of the method according to the invention. In such a special embodiment, the map 52 shown in Figure 5 relates to the illumination focus as a whole, without distinguishing between STED light and excitation light distribution. That is to say, in such a special example, the control information contained in the map 52 consists, so to speak, only of 0s and 1s, respectively, the illumination focus for sample segments to which 0 is assigned is completely switched off, whereas the illumination focus for sample segments to which 1 is assigned is switched on. However, it is also possible to provide a more complex radiation power adjustment, in particular with regard to the STED light, which also covers intermediate values. This results in a correspondingly more complex radiation power map. This map realizes an assignment between the image information contained in the overview image 48 and the illumination focus, in particular the STED light.

Claims

1. A method for imaging a sample (18) using a fluorescence microscope (10) with stimulated emission depletion, the method comprising an imaging process in which the fluorescence microscope (10) is controlled by a microscope controller (42) as follows: A focused de-excitation light distribution is superimposed on the focused excitation light distribution for stimulated emission suppression to generate an illumination focus, whereby the sample (18) is excited and emits fluorescence imaging light only within an effective excitation focus, the range of which is reduced compared to the range of the excitation light distribution, when illuminated with the illumination focus; 1. A method for scanning a number of sample segments in succession within a target area of ​​the sample (18) with the illumination focus at a first spatial resolution aligned with the reduced range of the effective excitation focus, imaging a number of corresponding image segments and generating a raster image (58) therefrom, comprising: The fluorescence microscope (10) is further controlled by the microscope control unit (42) as follows: Prior to the imaging process, an overview image (48) of the target area is controlled to be generated at a second spatial resolution, the second spatial resolution being lower than the first spatial resolution and higher than a third spatial resolution tailored to the range of the excitation light distribution, and further analysing the overview image (48) to identify image areas without associated image information; reducing, during the imaging process, at least the radiant power of the de-excitation light distribution when scanning a sample segment that is assigned to an image region identified in the overview image (48) that does not have associated image information. method.

2. additionally reducing the radiation power of the excitation light distribution when scanning sample segments that are assigned to image regions identified in the overview image (48) that do not have associated image information. The method of claim 1.

3. reducing said radiation power to zero; 3. The method according to claim 1 or 2.

4. first recording a raw image (44) at a spatial resolution lower than the second spatial resolution, and then performing image processing on the raw image (44) to increase the spatial resolution, thereby generating the overview image (48) at the second spatial resolution; 4. The method according to any one of claims 1 to 3.

5. generating the raw image (44) by scanning the target area with the excitation light distribution at the third spatial resolution; The method of claim 4.

6. The raw image (44) is generated by confocal raster imaging. The method of claim 5.

7. The raw image (44) is generated using wide field illumination. The method of claim 4.

8. The image processing provides a deconvolution of the raw image (44).

8. The method according to any one of claims 4 to 7.

9. The overview image (48) is generated at the second spatial resolution by direct imaging of the target area.

4. The method according to any one of claims 1 to 3.

10. generating said overview image (48) using patterned illumination; 10. The method of claim 9.

11. generating the overview image (48) by scanning the target area with the illumination focus, wherein the radiant power of the de-excitation light distribution is reduced compared to the radiant power used in the imaging process; 10. The method of claim 9.

12. generating and storing a radiant power table based on said analysis of said overview image (48), said radiant power table assigning at least one value of said radiant power of said de-excitation light distribution to each sample segment imaged during said imaging process; 12. The method according to any one of claims 1 to 11.

13. identifying exclusion regions from the image regions identified in the overview image (48) that have no associated image information, and skipping sample segments assigned to the exclusion regions when the target region is scanned at the first spatial resolution; 13. The method according to any one of claims 1 to 12.

14. The second spatial resolution is 1.5 to 2 times higher than the third spatial resolution.

14. The method according to any one of claims 1 to 13.

15. A fluorescence microscope (10) for imaging a sample (18) by stimulated emission depletion, said fluorescence microscope (10) comprising a microscope control (42) configured to control said fluorescence microscope (10) in an imaging process, said microscope control (42) being configured to: a focused de-excitation light distribution is superimposed on the focused excitation light distribution for stimulated emission suppression to generate an illumination focus, whereby the sample (18) is excited and emits fluorescence imaging light only within an effective excitation focus, the range of which is reduced compared to the range of the excitation light distribution, when illuminated with the illumination focus; In a fluorescence microscope (10), a number of sample segments are successively scanned within a target area of ​​the sample (18) by the illumination focus at a first spatial resolution adapted to the reduced range of the effective excitation focus and imaged into a corresponding number of image segments, from which a raster image is generated, The fluorescence microscope (10) is further controlled by the microscope control unit (42) as follows: Prior to the imaging process, an overview image (48) of the target area is controlled to be generated at a second spatial resolution, the second spatial resolution being lower than the first spatial resolution and higher than a third spatial resolution tailored to the range of the excitation light distribution, and the overview image (48) is analyzed to identify image areas without associated image information; during the imaging process, at least the radiation power of the de-excitation light distribution is reduced when scanning a sample segment that is assigned to an image area identified in the overview image (48) that does not have associated image information. Fluorescence microscope (10).

Images