Fast high-resolution microscopy method and fast high-resolution microscope

Abstract

To provide a fast high-resolution localization microscopy method for a sample having a fluorescence emitting body.SOLUTION: A fluorescence emitting body in a sample 2 is excited by radiation from an illumination light source 3 and radiates fluorescent radiation. The sample 2 that emits fluorescence light is imaged on a wide field camera 4 via a beam splitter 5. The operation of a microscope 1 is controlled by a control device 6. The control device 6 is connected to the wide field camera 4, the illumination light source 3 and a table where the sample 2 is arbitrarily placed via a corresponding control line, operates these elements, or receives data from these elements, especially image data from the wide field camera 4, and enables imaging having the spatial resolution exceeding the diffraction limit.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a method for high-speed and high-resolution localization microscopy of a sample containing a fluorescent phosphor. In this method, the fluorescent phosphor in the sample is excited to emit fluorescent radiation, the sample is imaged in a wide field of view at a certain spatial resolution, the excitation is brought about such that at least some of the fluorescent phosphors are separated with respect to the spatial resolution, and in localization analysis, the location of the separated fluorescent phosphors is determined with an accuracy exceeding the spatial resolution.

[0002] Furthermore, the present invention relates to a high-resolution microscope for high-speed localization microscopy of a sample containing a fluorescent phosphor, comprising an excitation device designed to excite the fluorescent phosphor in the sample to emit fluorescent radiation and an imaging device designed to image the sample at a certain spatial resolution. In this microscope, the excitation device is designed such that at least some of the fluorescent phosphors are separated within the image with respect to the spatial resolution, and the microscope comprises a localization analysis device designed to determine the location of the separated fluorescent phosphors with an accuracy exceeding the spatial resolution.

Background Art

[0003] In the state of the art, various methods have been developed in microscopy to break the diffraction limit. According to Patent Document 1 or Patent Document 2, a method abbreviated as PALM (Photo-Activated Localization Microscopy) that uses a labeling substance to image a sample is known, and the label can be activated using, for example, optical radiation. Only in the activated state can the labeling substance emit specific fluorescence radiation. Non-activated label molecules do not emit any fluorescence radiation, or at least do not emit it significantly, even after irradiation with excitation radiation. Therefore, the activation radiation is generally called a switching signal, and the molecules are called optical switches. In PALM, the switching signal is applied such that at least some of the activated labels are spaced from adjacent activated labels, so that they can be separated or substantially separated by an image processing method (in terms of the optical resolution of microscopy). This is said to separate a subset of the fluorophores. Individual images are captured for the sample thus prepared. Among them, the center of the radiation distribution caused by the resolution limit in the separated fluorophores is determined. The position of the molecules can be calculated from this center with a higher accuracy than actually allowed by the optical resolution. This procedure is called localization. Also, the increased resolution is called "super-resolution". This requires being able to distinguish a subset of the activated labels that are separated within at least a given range of optical resolution in the sample. Thereby, the location of the activated labels can be determined with higher accuracy. That is, they can be localized.

[0004] The PALM method utilizes a statistical effect to separate individual label molecules. By setting the intensity of the switching signal, it can be ensured that the probability that the labels present in a given region of the sample are activated is very low, such that there is a sufficient partial region where only the labels can emit fluorescence radiation that can be distinguished within the range of the optical resolution of the microscope. The sequence of separation and localization is repeated several times, and subsets that are differently configured each time contribute to the individual images. The localization data from these individual images are finally collected to form a high-resolution overall image.

[0005] The PALM method has been improved, for example, in Patent Document 3, with respect to the activation of the detected labels. For this purpose, fluorescent labels having specific electronic states are used. They are activated by high-intensity illumination radiation such that the majority of the molecules are put into a dark state where they are not electronically fluorescent. Thus, the remaining molecules that are fluorescing at that time can be separated with respect to optical resolution. Imaging and the capture of individual images are made as soon as a sufficient proportion of the molecules are put into the dark state, i.e., in synchronization with the excitation. The exposure time for individual images is based on the average time it takes for the excitation radiation to put the labels into the dark state by a constant irradiation. Patent Document 3 requires the manipulation of molecules in a sample to the effect that the lifetime of the dark state is extended through the addition of a chemical substance.

[0006] On the other hand, it should also be noted that modified forms of PALM, given other abbreviations such as STORM, dSTORM, etc., have been described in the technical literature. Thus, in this specification, the term "localization microscopy" is used to encompass all microscopy methods that achieve a spatial resolution exceeding the optical resolution of the device used by first separating and then localizing fluorescent molecules.

[0007] Typically, localization microscopy does not require a high spatial resolution for illumination. In many cases, simple wide-field illumination is sufficient. However, it requires that many individual images of the sample, each containing molecules to be separated, are captured. To image the sample completely, the sum of all individual images must ensure that, as far as possible, all of the molecules are separated in at least one individual image. Thus, for localization microscopy, the plurality of individual images mentioned must be captured periodically, which requires a certain duration to generate the overall image.

[0008] Also, the capture time is a problem for localization microscopy in that wide-field detection, which is advantageous in itself, necessarily requires a two-dimensional detector that must be sensitive enough for the detection of individual molecules. The sensitivity and readout noise, usually characterized by the quantum efficiency, should ideally be such that individual, isolated molecules can be detected. This requirement sets a lower limit on the integration time for the capture of individual images and thus an upper limit on the image refresh rate for cameras used in state-of-the-art localization microscopy.

[0009] Time constraints are of particular concern when microscopy of living cells is to be performed. To lead localization microscopy into a suitable time regime for this, it can be considered to extend the term "separated" in the sense that it is sufficient to include several molecules for each diffraction-limited spot in the evaluation and thus to separate the molecular population from adjacent molecular populations in terms of the resolution limit of imaging, which is the diffraction-limited spot. Thus, the density of activated molecules can already be increased during data capture, and a given number of molecules can be detected faster. However, in the case of current microscopy methods, this would come at an excessive sacrifice of localization accuracy.

[0010] Patent Document 3 attempts to avoid the problem of integration duration in that the lifetime of the dark state is increased by chemical manipulation. A similar approach is taken by Non-Patent Document 1. Here, the integration time can be set appropriately for the separation of individual molecule appearances. However, the sample then needs to be chemically manipulated.

[0011] So-called fluctuation analysis, also known as SOFI, represents another route to high resolution. Here, individual phosphors are made to blink, and the blinking sample is imaged with wide-field detection. Thus, the increase in resolution is done through the correlation between different blinking states rather than through localization. The SOFI microscopy technique can be seen as similar to the recognition of lighthouses in navigation. There, a particular lighthouse will be different from others in its blinking behavior. When different blinking occurs at exactly the same location as the blinking of adjacent points, it is clear that a light emitter (a lighthouse in navigation, a fluorescent molecule in microscopy) must be present at this location. SOFI microscopy requires that correlated fluctuations must occur on a time scale accessible to the camera. This ultimately requires very specific dyes or labels, such as so-called quantum dots that blink very slowly. These dyes are not very suitable or not at all suitable for living samples. In addition, for a reasonable increase in resolution, a certain minimum number of individual images is required, so there is also a problem of long measurement lengths here. As in Patent Document 3, the integration period of the camera is also adapted to the duration of the bright state of the blinking emitter. Since SOFI microscopy operates without evaluating correlations, separation steps, and localization steps, it is not considered a localization microscopy. In addition, the possible increase in resolution is limited compared to localization microscopy. A further aspect in the case of microscopy of living cells is photodamage. Non-Patent Document 2 shows that irradiation with short wavelengths is many times more harmful than irradiation with longer wavelengths of light. The switching wavelength used for PALM / dSTORM etc. is 405 nm, which is another reason for being able to do without a light switch.

[0012] Therefore, there is a need for a high-resolution microscopy method that is not limited with respect to the selection of fluorescent emitters and does not require any chemical manipulation of the fluorescent emitters at all. High-speed microscopy is particularly desired. Such a method is required especially for microscopy of living cells.

Prior Art Documents

Patent Documents

[0013] [Patent Document 1] International Publication No. 2006 / 127692 [Patent Document 2] German Patent Application Publication No. 102006021317 A1 [Patent Document 3] German Patent Application Publication No. 102008024568 A1 [Non-Patent Document]

[0014] [Non-Patent Document 1] Heilemann et al., “Super-resolution imaging with small organic fluorophores.”, Angewandte Chemie International Edition, 48.37, (2009): 6903 - 6908 [Non-Patent Document 2] Waeldchen et al., Scientific Reports, 5:15348, DOI: 10.1038 / srep15348, October 2015 [Summary of the Invention] [Problems to be Solved by the Invention]

[0015] An object of the present invention is to specify a localization microscopy method and to provide a microscope designed for this method that enables high-speed imaging without chemical manipulation using conventional dyes. [Means for Solving the Problems]

[0016] The present invention provides a method for high-speed and high-resolution microscopy of a sample containing a fluorescent phosphor. A fluorescent phosphor having a first state and a second state is used. The first state is a bright state in which the fluorescent phosphor emits fluorescent radiation upon irradiation with excitation radiation. The second state is a dark state in which the fluorescent phosphor does not emit fluorescent radiation upon irradiation with excitation radiation. The fluorescent phosphor can be brought from the bright state to the dark state by irradiation with excitation radiation. The dark state has a specific lifetime, after which the fluorescent phosphor spontaneously returns to the bright state. The sample, and thus the fluorescent phosphor, is illuminated in a wide field by excitation radiation of a certain intensity. Thereby, the fluorescent phosphor is excited to flicker, and the fluorescent phosphor alternates between a natural, i.e., chemically unmodified, bright state and a dark state at a certain flicker frequency. The flickering sample is imaged in a wide field at a certain spatial resolution and detected by a camera. The intensity of the excitation radiation is set such that at least some of the fluorescent phosphors are separated in individual images based on the spatial resolution. The intensity of the excitation radiation and the refresh rate of the camera are adapted to each other such that the refresh rate at which individual images are generated does not become lower than the average flicker frequency. The individual images are subjected to a localization analysis. In this localization analysis, the location of the separated fluorescent phosphors within the individual images is determined with an accuracy exceeding the spatial resolution.

[0017] Furthermore, the present invention provides a high-resolution microscope for high-speed microscopy of a sample containing a fluorescent phosphor. The microscope comprises an excitation device designed to irradiate the sample in a wide field with excitation radiation in order to emit fluorescent radiation. The microscope further comprises an imaging device designed to image the sample in a wide field at a certain spatial resolution. The excitation device is designed such that at least some of the fluorescent phosphors are separated in the imaging based on the spatial resolution. The localization analysis device is designed to determine the location of the separated fluorescent phosphors with an accuracy exceeding the spatial resolution. The control device is designed to control the excitation device and the imaging device according to the above-mentioned microscopy method. The localization analysis device and the control device can also be combined in one device.

[0018] The present invention uses a fluorescent phosphor for confocal microscopy that is excited by illumination with excitation radiation, i.e., light of a specific wavelength, to emit fluorescent radiation and is capable of emitting fluorescent radiation upon excitation. Under illumination with excitation radiation, the fluorescent phosphor constantly passes through a cycle between the unexcited ground state and the excited state and returns to the ground state again. The transition from the ground state to the excited state is promoted by the excitation radiation. At the molecular level, this is explained by the absorption of photons of the excitation radiation. The fluorescent phosphor returns from the excited state to the ground state and emits optical fluorescent radiation during its return. At the molecular level, this change from the excited state to the ground state is related to the emission of fluorescent photons. The period during which the fluorescent phosphor remains in the excited state is on the nanosecond scale. That is, after absorption of the excitation photons, the fluorescent phosphor hardly stays in the excited state at all, but rather emits fluorescent photons almost immediately. The result of this situation is that the fluorescent phosphor illuminated by the excitation radiation emits fluorescent radiation continuously. The state of the fluorescent phosphor, which can be excited by the excitation radiation at any time and emits fluorescent radiation as long as it is illuminated, is referred to herein as the "bright state". This is contrasted with a dark state in which the fluorescent phosphor does not emit fluorescent radiation despite irradiation with the excitation radiation and is thus almost non-responsive to the excitation radiation. The dark state can be, for example, the so-called triplet state of the molecule. Such dark states and in particular triplet states have a certain lifetime. That is, since excitation and emission of fluorescent radiation using excitation radiation from the ground state are always possible, it takes a certain amount of time for the fluorescent phosphor to return from the dark state to the ground state (without emission of optical radiation) and thus to the bright state. The lifetime of the dark state is significantly longer than the lifetime of the excited state. In other words, within the framework of the present invention, the bright state includes the ground state and the excited state of the fluorescent phosphor in which fluorescent radiation is emitted. In contrast, the dark state is a state in which it is not possible to excite the fluorescent phosphor to the excited state even under illumination with the excitation radiation and thus no fluorescent radiation is emitted.

[0019] Upon irradiation with the excitation radiation, the fluorescent phosphor remains in the bright state with a first probability. With a second probability smaller than the first probability, the fluorescent phosphor transitions to the dark state. However, since the lifetime of the dark state is much longer than that of the excited state, the probability that the fluorescent phosphor will transition to the dark state within a specific period increases with the increase in the intensity of the excitation radiation. The phosphor remains in the dark state during the lifetime of the dark state and is then unresponsive to the excitation radiation, so it is hardly switched during this lifetime. Unlike the dark state, it is not considered that the bright state itself has a definite lifetime. It is only considered that the excited state in the bright state has a lifetime. However, as already mentioned, this is many times shorter than the lifetime of the dark state. In a theoretical two-state system, a fluorescent phosphor excited by excitation radiation of a specific intensity can emit almost unlimited fluorescence radiation.

[0020] The present invention uses a general property of almost all fluorescent phosphors, namely that upon irradiation with excitation radiation, a specific proportion of the fluorescent phosphor transitions to the dark state, for example the triplet state. The proportion of the fluorescent phosphor that enters the dark state depends on the intensity of the excitation radiation. It increases as the intensity of the excitation radiation increases. For each fluorescent phosphor, there is a certain probability that it will transition from the ground state to the standard excited state through an excitation photon and then non-radiatively to the dark state. The term "non-radiative" is related to the wavelength of the fluorescence radiation. Typically, the probability of being in the dark state increases as the intensity of the excitation radiation increases. When the fluorescent phosphor is in the dark state, the fluorescent phosphor does not emit fluorescence radiation during the lifetime of the dark state. The phosphor can emit fluorescence radiation again if excited only when the fluorescent phosphor returns from the dark state to the bright state. Due to these circumstances, the fluorescent phosphor flickers, and the dark time of the flicker is determined by the lifetime of the dark state. Since the phosphor emits light while being excited in the bright state, the length of the bright time depends on how long it takes for the fluorescent phosphor to transition from the bright state to the dark state (ultimately, the average bright time depends on the intensity of the excitation radiation). The period, and thus the flicker frequency of the flicker, is formed by the sum of the bright time and the average lifetime of the dark state. Since the duration of the dark time depends on the excitation intensity, the flicker frequency is set by the selection of the excitation intensity.

[0021] In some embodiments, the dark time, i.e., the average lifetime of the dark state, is pre-determined molecule-specifically for all blinking frequencies. In some embodiments, it is not chemically manipulated. That is, the dark time is the lifetime of the natural dark state. In other words, blinking is generated in that the continuous emission of each phosphor in the bright state is interrupted by temporarily changing the phosphor to the dark state through an appropriate selection of the intensity of the excitation radiation.

[0022] The present invention blinks a dye at a desired frequency by an appropriate selection of the excitation intensity, performs localization microscopy, and adapts the integration time or the image frequency of the camera to the blinking frequency. For several reasons, this represents a systematic change away from the state of the art.

[0023] 1. The present invention deliberately causes the blinking of a phosphor and uses this blinking for localization microscopy. Image capture is designed such that the refresh rate does not become lower than the blinking frequency. For localization microscopy, the phosphor is necessarily detected several times because it is blinking. In contrast, conventional PALM techniques and their modified forms aim to illuminate the molecules only once if possible and in any case to accurately avoid blinking.

[0024] 2. For conventional phosphors, and thus for the majority of the normal applications of fluorescence microscopy, the probability of transition to the dark state depends, as mentioned, on the intensity of the excitation radiation. The on-time, and thus the blinking frequency, vary inversely with respect to each other. The stronger the excitation radiation, the more the phosphor transitions to the dark state and the shorter the on-time (the higher the blinking frequency). Thus, the blinking frequency can be appropriately set by simple optical means to match the requirements of the camera. By increasing the blinking frequency, the period during which the molecule is in the on-state is decreased. However, since the dark state is preferably the natural dark state with an invariant molecule-specific lifetime, its length remains unchanged. Thus, the ratio between the on-time and the off-time is controlled in that the dark state becomes longer compared to the on-state. As a result, this was only possible with state-of-the-art techniques involving chemical manipulation, i.e., an extension of the dark state was achieved.

[0025] 3. By referring to the blinking frequency, the present invention achieves an advantageous utilization of the interaction between the integration time / image frequency of the camera and the intensity of the excitation radiation, resulting in the ability to use conventional phosphors, which were not available for PALM, for localization microscopy without chemical influence.

[0026] 4. A camera refresh rate that does not fall below the blinking frequency can be obtained through an appropriate adaptation of the camera settings, e.g., the integration time of image capture, to the intensity of the excitation radiation. Thus, the inverse relationship between the blinking frequency defined by the excitation radiation intensity and the detection sensitivity of the camera defined by the camera integration time / refresh rate is utilized. In some embodiments, multi-emitter analysis is utilized in the localization analysis. That is, understand the term "separated" in the sense that a group of fluorescent emitters is separated from another group of fluorescent emitters, or understand the definition that at least some of the fluorescent emitters are separated taking into account the spatial resolution within an individual image. Nevertheless, since the camera operates in a preferred performance range, the resolution does not decrease as much as expected with state-of-the-art techniques.

[0027] 5. Since the refresh rate of each individual image must not be lower than the blinking frequency, the reference to the blinking frequency ensures that exactly one dark state of the blinking molecule is included in each individual image. This frequency consists of one bright state and one dark state. Therefore, the localization of the corresponding molecule is still possible with conventional methods. Thus, the localization analysis of the microscopy method according to the present invention is basically no different from the localization analysis of standard PALM or its modified forms. However, since the fluorescent emitters of the sample emit several photons during their bright times, each molecule is then detected by several photons at that time. This provides an advantageous compensation for the integration time, which is relatively short compared to the state of the art and adapted to the blinking frequency. The relatively rapid image capture surprisingly does not lead to a deterioration of the signal / noise ratio in the localization analysis.

[0028] In order to adapt the intensity of the excitation radiation and the refresh rate to each other, in some embodiments of the present invention, the lifetime of the dark state is measured in a test measurement, and the refresh rate is set accordingly. To measure the lifetime, the excitation radiation may be emitted in a pulsed manner, and the fluorescent radiation can be detected synchronously with the excitation radiation. It is also possible to perform autocorrelation analysis by exciting in continuous wave operation, as is known, for example, by fluorescence correlation spectroscopy.

[0029] The integration time is preferably 90 percent, particularly preferably 50 percent, and most particularly preferably 10 percent of the lifetime of the dark state.

[0030] The present invention operates in a time regime on the order of the dark lifetimes of most standard phosphors. This range is from 1 μs to 300 μs. The excitation intensity is set such that the blinking operation is dominated by the lifetime of the dark state that does not emit fluorescence. In most cases, this is the triplet state. It is preferable to set the intensity of the excitation radiation so that the image frequency does not drop below a value of 1 / (300 μs).

[0031] In an embodiment of the microscopy method, the intensity of the excitation radiation is 1 MW / cm2 shall not exceed, preferably not exceed 50 kW / cm 2 shall not exceed, particularly preferably not exceed 2 kW / cm 2 shall not exceed.

[0032] The refresh rate preferably does not exceed 150% of the blinking frequency, more preferably does not exceed 130% of the blinking frequency, particularly preferably does not exceed 110% of the blinking frequency, extremely particularly preferably does not exceed 105% of the blinking frequency, and is selected to not exceed 100% and not reach 100%.

[0033] In a particularly preferred embodiment, the fluorescent phosphors are in a state that is not chemically affected with respect to their lifetime in the dark state. Since chemical effects on living cells are generally highly disadvantageous, this is particularly advantageous in the microscopy of living cells. The present invention adapts the output of the excitation radiation to the refresh rate. Since the photon pressure of the sample decreases at a faster refresh rate, this is particularly advantageous for the microscopy of living cells.

[0034] The microscope setup corresponds to the setup of a typical wide-field fluorescence microscope, and thus the control device is adapted to control the excitation device and the imaging device in order to match the excitation intensity and the refresh rate to each other. The excitation radiation can illuminate the sample in the form of continuous radiation (CW radiation). For fluorescence measurement, pulsed excitation and the operation of the camera synchronized therewith are preferred. This synchronous operation can be used to improve the signal-to-background radiation ratio for the emitters of individual molecules, i.e., also when background autofluorescence with a considerably long or considerably short lifetime and / or with considerably different lifetimes in the dark state and / or in the bright state is present in the sample.

[0035] The present invention adjusts the refresh rate of a camera and the flashing frequency of a sample relative to each other. There are certain advantages to setting the flashing frequency, and thus the refresh rate, relatively high. In so doing, the integration time of the camera is relatively short. In some embodiments that use a high refresh rate, an SPAD detector array within the camera is utilized. Accordingly, it is preferred that the camera be equipped with an SPAD array.

[0036] SPAD arrays having a large number of pixels are relatively difficult and expensive to obtain. In preferred embodiments, an SPAD array having a number of pixels that still does not allow diffraction-limited imaging in a wide field of view is used. In these embodiments, the technique known from German Patent Application Publication No. 102012205032A1 of Carl Zeiss Microscopy GmbH can be used. According to this technique, a sample is imaged in an image field, and the position of the image field is shifted several times across the entire sample. Partial individual images are acquired for each position of the image field, and the acquired partial individual images are assembled to form individual images.

[0037] When method steps are described below, those steps can be realized by corresponding operations of a microscope having a suitably designed control device. Accordingly, when features of a method are described, these are related to the design of the control device, for example via a suitable operating program. Features of the control device achieved, for example, via corresponding programming means are likewise related to and intended to disclose corresponding features of the operating method, and of course vice versa.

[0038] It is understood that the features specified above and further explained below can be used not only in the stated combinations but also in other combinations or alone, without departing from the spirit and scope of the present invention.

[0039] The present invention will be explained in more detail by way of example with reference to the accompanying drawings which also disclose features essential to the present invention. Localization microscopy is explained for an example of a variant form of PALM. However, as described above, this does not limit the present invention.

Brief Description of the Drawings

[0040]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Embodiments for Carrying Out the Invention

[0041] Figure 1 schematically shows a microscope 1 for imaging a sample 2 with high resolution. Here, for an imaging method having a spatial resolution increased beyond the diffraction limit of imaging, the term high resolution, which is common in the state of the art, is used. The microscope 1 serves to image a sample 2 containing a fluorescent emitter. The fluorescent emitter is excited using radiation from an illumination light source 3 and emits fluorescent radiation. The sample 2 emitting fluorescence is imaged onto a wide-field camera 4 via a beam splitter 5. The operation of the microscope 1 is controlled by a control device 6. The control device 6 is connected via corresponding control lines (not further identified) to the wide-field camera 4, the illumination light source 3, and optionally a table on which the sample 2 is placed, actuates these elements, and receives data from these elements, in particular image data from the wide-field camera 4.

[0042] In one embodiment, the wide-field camera 4 comprises a combination of a CMOS camera and an image intensifier. For example, it is realized in the form of a CMOS camera having an "enhanced high-speed CMOS camera" according to Gige-Vision, or an upstream image intensifier, for example an image intensifier "HiCATT, high-speed camera attachment" by Lambert Instruments. In this and other embodiments, the wide-field camera 4 can be adjusted with respect to its refresh rate, i.e., the exposure time or integration time. The control device 6 accepts corresponding settings via control lines. Further, in a preferred embodiment, the control device 6 establishes synchronization to the refresh rate.

[0043] The illumination light source 3 emits excitation radiation A to the sample. The intensity of this radiation is set by the control device 6 in the form of CW radiation or pulsed radiation. Further, the control device 6 optionally also performs a localization analysis on the captured individual images and generates a high-resolution overall image from the processed individual images.

[0044] Figure 3 schematically shows a sectional view of the fluorescent phosphor used in the microscope 1. The fluorescent phosphor has a ground state S0. Through the irradiation of the excitation radiation A, the fluorescent phosphor transitions to the excited state S1 and then returns to the ground state with the emission of fluorescent radiation. The ground state S0 and the excited state S1 form the bright state H of the fluorescent phosphor. In the bright state H, the fluorescent phosphor emits fluorescent radiation F at any time in response to the irradiation of the excitation radiation A. Since the lifetime of the excited state S1 is a few nanoseconds, in the bright state H, the fluorescent phosphor irradiated with the excitation radiation A continuously emits fluorescent radiation F. The fact that the emission of individual photons of the fluorescent radiation F occurs at nanosecond intervals is irrelevant in this specification and cannot be detected by some measurement techniques using the devices described in this specification.

[0045] The excitation radiation A can also cause the fluorescent phosphor to reach the triplet state T1 via the excited state S1. This has a lifetime in the range of several microseconds, for example, 1 μs to 300 μs. While the fluorescent phosphor is in the triplet state T1, the fluorescent phosphor does not respond to further irradiation with the excitation radiation A. The transitions to and from the triplet state are radiationless (at least with respect to the wavelength of the fluorescent radiation F), and thus are depicted by dashed lines in Figure 3. Since the fluorescent phosphor does not emit fluorescent radiation F despite illumination with the excitation radiation A at this time, the triplet state T1 is thus a dark state D. Only when the fluorescent phosphor returns to the ground state S0, and thus exits the dark state D and is again in the bright state H, does the fluorescent phosphor continuously emit fluorescent radiation F in response to illumination with the excitation radiation A.

[0046] The probability of the transition from the excited state S1 to the ground state S0 is much higher than the probability of the transition from the excited state S1 to the triplet state T1. Therefore, in response to the irradiation of photons of the excitation radiation A, the fluorescent phosphor is much more likely to remain in the bright state H than to change to the dark state D. However, the probability is not zero. As the intensity of the excitation radiation A increases, that is, along with an increase in the impact by the photons of the excitation radiation so to speak, the probability that the fluorescent phosphor will transition to the dark state D within a certain period increases.

[0047] The item diagram shown in FIG. 3 should be construed as merely an example, and is only intended to characterize the substance used herein as a fluorescent emitter in the sense that there are bright state H and dark state D both reachable by simply irradiating with excitation radiation A. In dark state D, it is not absolutely necessary for a single state to exist in the form of triplet state T1, and a multi-state system is also possible. However, it is important that the fluorescent emitter transfers to dark state D through the incidence of excitation radiation A with a non-zero probability of transition.

[0048] This results in a significant difference for PALM variants that use radiation of a wavelength different from that of excitation radiation A. A PALM variant known as dSTORM, which operates with specific switching radiation, is just an example. Such distinct switching radiation is not used in the embodiments of the present invention. Rather, the fluorescent emitter transfers to the dark state solely (with a certain probability) through the irradiation of excitation radiation A.

[0049] The effects of this situation are shown in FIG. 4. This figure represents the time sequence of the light emission of the fluorescent emitter. Among them, the lower level 8 indicates a dark fluorescent emitter, and the upper level 9 indicates a fluorescent emitter that is emitting fluorescence. As explained, the fluorescent emitter is made to enter the dark state D with a certain probability by the excitation radiation A. Therefore, through the irradiation of the excitation radiation A, the fluorescent emitter will sooner or later transition from the bright state (upper level 9) that emits light to the dark state (lower level 8). During this dark state D, the emitter remains dark. The duration of this dark state cannot be optically affected on the fluorescent emitter in the above case. The dark time (the lower level 8 of the curve in FIG. 4) corresponds to the lifetime td of the dark state D. Thereafter, the fluorescent emitter exits the dark state D again and emits light (upper level 9 in FIG. 4). Purely statistically, the fluorescent emitter will return to the dark state D again after a certain period. The period during which the fluorescent emitter emits light represents the bright time th. The bright time th depends on the probability of transition to the dark state D and thus on the intensity of the excitation radiation A. As a result, the fluorescent emitter performs the blinking 7 formed by the succession of non-emission (lower level 8) and emission (upper level 9). The period PB of this blinking has a reciprocal that is the blinking frequency fB and is formed by the sum of the lifetime td of the dark state D and the bright time th.

[0050] When the intensity of the excitation radiation A is increased, the probability that the emitter transitions to the dark state D increases, thereby reducing the bright time th. This situation is depicted by the dashed line in FIG. 4. By increasing the intensity of the excitation radiation A, statistically on average, the dark state D is reached earlier, so the average bright time th is shortened. In FIG. 4, this is plotted with a dashed line by the displacement 10 in front of the second dark state as seen from the left. It can be immediately seen that the blinking frequency fB increases (the blinking period PB decreases), and the lifetime td of the dark state D remains unchanged. It preferably corresponds to the natural lifetime of the dark state D, for example, the natural lifetime of the triplet state T1.

[0051] The control device 6 is designed as a whole such that the microscope 1 performs a microscopy method based on localization, for example, the PALM method.

[0052] The method steps to be executed are represented as a flow chart in FIG. 2. After the start of the method in step S1, in the excitation step S2, the sample 2 is illuminated by the excitation radiation A from the illumination light source 3. This is done so that the individual fluorescent emitters are separated with respect to the spatial resolution of the imaging. Imaging is performed in the subsequent image capture step S3, in which individual images of the sample are supplied in a wide field of view in such a way that the distance between adjacent fluorescent emitters of at least the individual fluorescent emitters is greater than the spatial resolution of the wide field of view imaging. This state is schematically shown in FIG. 5, which shows the individual images 14 of the sample 2. In the individual images, the individual fluorescent emitters 11 are separated. Here, they are plotted as asterisks. The non-emitting fluorescent emitters, which are in the dark state D at this point in time, are symbolically represented by rings in FIG. 5. Most of the fluorescent emitters, especially those in region 13, are spaced so that they can be separated from each other. However, for example, in region 12, the individual fluorescent emitters are very close to each other and adjacent, so that they cannot be distinguished optically and in the localization analysis. The fact that 100% separation cannot be achieved (and this is not even necessary at all) is well known to those skilled in the art regarding localization analysis.

[0053] In the subsequent localization step S4, for these separated fluorescent emitters, the location in the image is determined with a spatial resolution greater than what is actually allowed by the diffraction limit. Fluorescent emitters that cannot be separated, such as those within region 12, are discarded in localization step S4. This can occur in that the radiation captured from the separated fluorescent emitters is utilized in localization step S4 to determine whether it emanates from one fluorescent emitter or from a small number of fluorescent emitters. The location is determined from this knowledge, and the fluorescent emitters are positioned with a higher accuracy than what is allowed by the diffraction limit. The localization analysis performed in localization step S4 can be envisioned in a simplified manner such that the center of the airy disk of the separated fluorescent emitter is determined, and that center represents the location at that time. The location of the fluorescent emitter can be determined more accurately than simply determining the center by fitting to a point spread function. This function can be approximated, for example, by a 2D Gaussian distribution, or an experimentally determined point spread function of the optical system can also be used. For performing the localization analysis, many different methods are known in the state of the art, for example, from the publications mentioned in the introduction section of this specification.

[0054] One single execution of the localization step S4 in the individual images 14 gives the location only for the fluorescent emitters that emit light, i.e., are not in the dark state D, within the corresponding wide-field image. Further, only the isolated fluorescent emitters are localized (thus, for example, the emitters in region 12 are not localized). If possible, in order to image the sample for all fluorescent emitters, steps S2 - S4 are repeated several times such that, in the best case, a subset of the isolated fluorescent emitters has all the fluorescent emitters 11 at least once. Thus, a question step S5 follows the localization step S4, where it is checked whether the sample 2 has been imaged sufficiently. For example, this step can access a large number of locations obtained from all previous cycles before the localization step S4. Such means are also used in the PALM method or other localization-based high-resolution methods that operate without determining the location tolerance, and thus such questions are well-known to those skilled in the art of the state of the art. If the sample has been detected comprehensively enough (J branch), the overall image is generated in step S6 and the image is displayed. Otherwise (N branch), steps S2 - S4 are executed again, and care is taken to separate other fluorescent emitters as much as possible during the repetition cycle.

[0055] (Since it is the simplest) The most advantageous option is to first capture all the individual images 14 and save them, and then perform a localization analysis on each of the saved individual images 14. Alternatively, in particular when there is sufficient processing power, on-line analysis is performed. After the generation and display of the overall image in the image display step S6, step S7 ends the method. The sample 2 has been imaged with a spatial resolution better than that allowed by the optical diffraction limit.

[0056] The excitation radiation A excites the fluorescent phosphor in the sample to blink, and the fluorescent phosphor alternates between the bright state H and the dark state D at the blink frequency fB. The excitation step S2 provides a blinking sample via excitation radiation A of appropriate intensity. The blinking sample 2 is imaged in the wide-field imaging device configuration in the image capture step S3 and detected by the wide-field camera 4. The camera has a refresh rate, and thus defines an image capture time for each of the individual images 14. The image capture time is determined by the integration time of the individual image generation process. The control device 6 sets this integration time with the wide-field camera 4. Furthermore, the control device 6 sets the intensity of the excitation radiation A with the illumination light source 3.

[0057] The intensity of the excitation radiation A and the blink frequency fB at which the phosphor 11 blinks are related to each other. When the excitation intensity A increases significantly, the blink frequency fB increases. At the same time, an increase in the refresh rate decreases the detection sensitivity of the wide-field camera 4. The corresponding sensitivity curve or other data describing the sensitivity dependence are stored in the control device 6. Similarly, the relationship between the excitation intensity A and the blink frequency fB, or the relationship between the excitation intensity and the average bright time th, are stored in the control device 6. This relationship can be stored as a curve or in the form of data (e.g., as a table). It is either known in advance for the phosphor or determined in a special test measurement mode of the microscope. The test measurement mode is executed by the control device 6 upon request or automatically. In this test measurement mode, the control device 6 detects the lifetime td of the dark state D and / or the bright time th and / or the blink frequency fB (as one option as a function of the excitation intensity) through the pulsed operation of the synchronized illumination light source 3 and the wide-field camera 4. Alternatively, the lifetime measurement can also be performed using CW excitation radiation (see above).

[0058] The control device 6 sets the intensity of the excitation radiation A emitted by the illumination light source 3 and the refresh rate of the wide-field camera 4 according to the requirements stored in the control device 6 or the requirements input into the control device 6.

[0059] The requirement can be asserted, for example, that image capture is required as fast as possible. The control device 6 determines the minimum refresh rate required to image the sample 2 from the stored relationship between the refresh rate of the wide - field camera 4 and the optical detection limit, and sets the intensity of the excitation radiation A in the illumination light source 3 such that the blinking frequency fB is lower than the refresh rate. Since the interaction between the intensity of the excitation radiation A and the refresh rate of each individual image capture is adapted to minimize the generation of the overall image, this mode of operation can be interpreted as a high - speed mode of operation. When determining the maximum allowable refresh rate of each individual image capture, the requirement for the separation of a part of the fluorescent emitter 11 can be further reduced by performing a localization analysis in step S4, as a multi - emitter analysis, that is, allowing several molecules to enter the evaluation for each diffraction - limited point. Accordingly, the performance requirements for the wide - field camera 4 are further reduced and the possible refresh rate is increased (however, this is at the expense of a somewhat but constant resolution acceptable for some applications).

[0060] The opposite extreme of image capture as fast as possible is the mutual adaptation of the intensity of the excitation radiation A and the refresh rate, aiming at the maximum separation of the fluorescent emitter 11. In this case, the control device 6 increases the intensity of the excitation radiation A in the illumination light source 3, thereby increasing the blinking frequency. At the same time, the refresh rate is also set to the maximum, thereby ultimately increasing the degree of separation. The shorter the exposure time compared to the lifetime of the dark state, the greater the separation can be. In this way, the localization accuracy, and thus the high resolution, is increased (however, at the expense of the photon load or pressure of the sample).

[0061] In a preferred embodiment, the control device 6 has an input interface for specifying parameters regarding the mutual adaptation of the refresh rate and the excitation radiation intensity. The interface can be, for example, an input interface for the user or a data interface for supplying corresponding parameter data. In one embodiment, the input via parameter data or the user can preferably include a value specifying the image capture rate on a relative scale, for example from 0 to 100. In another embodiment, instead or in addition, here preferably on a dimensionless scale, parameters of the localization accuracy and / or the photon pressure of the sample can be specified. From the parameter value or several parameter values, the control device 6 determines, according to a stored relationship or table, the refresh rate or the integration time or the exposure time set in the wide-field camera 4, and the intensity of the excitation radiation A set in the illumination light source 3 (as CW or pulsed radiation).

[0062] As described in the summary part of this specification, in some embodiments, it includes a control device 6 that further measures the characteristic variables of the emitter. This can be, for example, the measurement of the blink frequency fB or the lifetime td of the dark state D. The microscope 1 uses a wide-field camera 4, and the refresh rate of the wide-field camera 4 is set or adapted to the blink frequency fB. To enable the realization of a high refresh rate, in some embodiments, a wide-field camera 4 equipped with a SPAD array is used.

[0063] In the improvement of this method, a SPAD array that should theoretically be insufficient for the required resolution in the desired sample field is used. FIG. 6 schematically shows this embodiment of the microscope 1. The microscope 1 images a flat sample 2. A sample region P of the sample 2 is imaged. The microscope 1 includes an objective lens 23 and a further lens 24 for imaging the sample field P. A part of the sample field P is imaged by spatial resolution, i.e., by the pixelated detector 25. Thereby, in contrast to the design of FIG. 1, the image field of the sample 2 is imaged onto the detector 25 that is smaller than the detected sample field. The sample 2 is exposed to the excitation radiation A from a laser, which is an example of the illumination source 3. The excitation radiation A excites the fluorescent phosphor 11 in the sample 2 to emit fluorescent radiation F. The excitation radiation A is coupled into the beam path to the objective lens 23 via an optical system and a beam splitter 5 that are not described in more detail. The radiation travels from the beam splitter 5 to the deflection mirror 29 and the adaptive mirror 30. The deflection mirror 29 helps to keep the beam path compact and has no further significance in other respects. The adaptive mirror 30 includes mirror segments that can be actuated individually. The adaptive mirror 30 is arranged in the intermediate image plane ZB of the imaging beam path and the illumination beam path. The adaptive mirror 30 is an example of an image field shift device for shifting the image field over the entire sample region P, and this image field is imaged by the detector 25. Another option is sample shifting by a mirror scanner or a sample holder. The aperture 32, which is also arranged in the intermediate image plane ZB, also contributes to image field selection. The image of the sample 2 reaches the detector 25, which is also in the intermediate image plane ZB, via the lens 33 and the pivoting mirror 34. The pivoting mirror 34 alternatively enables viewing the sample with a normal wide-field camera 35. The pivoting mirror 34 and the camera 35 are optional. During the operation of the camera 35, the aperture 32 is set to fully open in a convenient manner or removed from the beam path, resulting in the entire sample field of the sample 2 being illuminated. A lens 31 is also provided in the imaging beam path, which provides intermediate imaging like the lens 33. The lens 31 defines the intermediate image plane ZB in which the aperture 32 is arranged.The lens 33 defines an intermediate image plane ZB in which the detector 25 is arranged.

[0064] The function of the image field displacement mechanism realized by the adaptive mirror 30 in the embodiment of FIG. 6 is apparent from FIG. 7 which shows a plan view of the sample field P of the sample 2 to be detected. An image field considerably smaller than the sample field P is labeled 37. The image field 37 is shifted over the entire sample field P. The corresponding shift positions and the positions of the image field 37 resulting from the shift positions are schematically shown in FIG. 7.

[0065] FIG. 8 shows that the image displacement device positions the image field 37 at various positions 38 such that an overlapping region 39 occurs between adjacent positions 38. This is optional but makes it easier to later assemble the individual partial images to form one image reproducing the sample field P.

[0066] In the embodiment using the adaptive mirror 30, the adaptive mirror 30 includes one or more grouped elements that shift the image field 37 over the entire sample field P. The size of the image field 37 imaged by the detector 25 is determined in advance, on the one hand, by the surface area of the detector 25 itself and, on the other hand, by the size of the aperture 32. The aperture 32 improves imaging but is not essential. If the aperture 32 is omitted, the corresponding intermediate image plane ZB and the optical means for generating it can be dispensed with.

[0067] Detector 25 is a detector array that, when the entire sample field P is imaged, will ultimately have several pixels that far exceed the optical resolution limit defined by the objective lens 23. Therefore, detector 25 is not suitable by itself for confocal microscopy. Therefore, the size of the image field 37 and the magnification of microscope 1 are selected such that the structural length in sample 2 corresponding to the resolution limit is equal to or larger than the size of the pixels on the detector. Nevertheless, the sample field P can be scanned by shifting the image field 37 to several positions. Therefore, a detector with a high measurement speed can be used. In the case of an electronic detector, the refresh rate of individual images is limited by the integration time and readout time of the detector. The detector 25 of microscope 1 can be, for example, an array of avalanche photodiodes, photomultiplier tubes or GaAsP hybrid detectors that are much smaller in size but much faster than relatively large conventional CMOS-based or CCD-based two-dimensional detectors. Therefore, the sample field P is imaged by the spatial resolution detector 25. Here, optical imaging is performed such that the maximum resolution is the diffraction limit and the structural length corresponds to the detector pixels on the scale of that structural length. The detected image field 37 is much smaller than the sample field P. For the detection of the entire sample field P, the image field is brought to various positions 38. For each position, the individual image 14 below sample 2 is captured corresponding to the current position 38. Subsequently, the individual images below are assembled to form the individual image 14 of the sample field P. This further development allows the use of a high-speed detector array with a pixel number that is not sufficient for standard wide-field imaging. For example, when an SPAD array with 128×128 pixels is used, as known by Niclass et al., IEEE Journal of Solid State Circuits 43, page 2977, 2008, a total of 5×5 different image field positions can cover a typical sample field with a size of 50 μm.

[0068] Figure 9 shows two images generated using the described microscope embodiments. An Elyra type microscope by Carl Zeiss Microscopy GmbH was used. As a detector for image capture, an Ultracam7 by Video Scope International, Ltd. of Dulles, USA was used. The phosphor is the dye Alexa488. This dye marks phalloidin, thereby coloring a part of the cytoskeleton (actin) of the cell. The intensity of the excitation radiation was 0.1 kW / cm at a wavelength of 488 nm. 2 was.

[0069] In the upper image, the standard wide-field image 41 of FIG. 9 is shown, while the lower image shows the high-resolution overall image 42 of the same sample. 30,000 individual images were generated, and those images were captured at a frame rate of 32 kHz corresponding to an exposure time of 31.25 μs including the return time. A total measurement time of 1 second was required. For the localization analysis, a 2D Gaussian function was applied. With the conventional PALM microscopy method using an EMCCD detector, 10 to 17 minutes would be required for the same measurement. The wide-field image 41 was obtained by simply summing 30,000 individual frames without any processing. Therefore, the wide-field image 41 corresponds to an image with an exposure time of 1 second.

[0070] Figure 10 shows two marked sections "1" and "2" in the wide-field image 41 and the overall image 42 of FIG. 9. The upper representation in FIG. 10 shows two curves for section "1", namely, curve 43 for the wide-field image 41 and curve 44 for the high-resolution overall image 42.

[0071] As can be seen, the spatial resolution in the high-resolution overall image 42 is very good. The lower representation in FIG. 10 shows that, referring to interval "2", inaccurate information may indeed occur in a wide field of view for some intervals, thereby, for example, over-illuminating adjacent bright regions within the curve. This is avoided by the high-resolution overall image 42. This difference can be clearly recognized in the lower representation in FIG. 10 (refer, for example, to the region from 0.8 μm to 0.9 μm). The curve 45 for the wide-field image 41 exhibits a maximum at positions where, according to the curve 46 of the high-resolution overall image 42, there is actually no sample structure or only a very slight one.

[0072] Although the triplet state has been described as an example of a dark state, this should be understood merely as an example. There are fluorescent dyes well-known to those skilled in the art of fluorescence microscopy that have dark states other than the triplet state.

[0073] Performing the microscopy method and the design of the corresponding microscope 1 is particularly advantageous in the following time regime, namely when the dark lifetime td is in the range of 3 μs to 320 μs. The integration time of the camera is preferably shorter than this lifetime. Thus, possible extreme values are th = 0.1td, td = 3 μs, PB = 3.3 μs, and fB = approximately 300 kHz. In another extreme, th is approximately equal to td = 300 μs, PB = 600 μs, and fB = 1.6 kHz.

Claims

1. A method for high-speed and high-resolution microscopic examination of a sample (2) containing a fluorescent phosphor (11), comprising: - A fluorescent phosphor (11) having a first state and a second state is used. The first state is a bright state (H) in which the fluorescent phosphor (11) emits fluorescent radiation (f) upon irradiation with excitation radiation (a). The second state is a dark state (D) in which the fluorescent phosphor (11) does not emit fluorescent radiation (f) upon irradiation with excitation radiation (a). The fluorescent phosphor (11) can be changed from the bright state (H) to the dark state (D) by irradiation with excitation radiation (a). The bright state (H) has a specific bright time (th) during which the fluorescent phosphor (11) remains in the bright state (H). The dark state (D) has a specific lifetime (td), after which the fluorescent phosphor (11) spontaneously returns to the bright state (H). The lifetime (td) of the dark state (D) is from 1 to 300 μs. - The sample (2), and thus the fluorescent phosphor (11), is illuminated in a wide field by excitation radiation (a) of a certain intensity, whereby the fluorescent phosphor (11) is excited to flash (B). As a result, the fluorescent phosphor (11) alternates between the bright state (H) and the dark state (D) at a flashing frequency (fB) formed by the sum of the lifetime (td) of the dark state (D) and the bright time (th). - The flashing sample (2) is imaged in a wide field at a certain spatial resolution and detected by a camera (4). - The intensity of the excitation radiation (a) is set such that at least some of the fluorescent phosphors (11) are separated into individual images (14) with respect to the spatial resolution. The intensity of the excitation radiation (a) and the refresh rate of the camera (4) are adapted to each other such that the refresh rate at which the individual images (14) are generated is not lower than the flashing frequency (fB). - The individual images (14) are subjected to a localization analysis (S4), and the location of the separated fluorescent phosphors (11) within the individual images (14) is determined with an accuracy exceeding the spatial resolution. A high-speed and high-resolution microscopic examination method.

2. The microscopic examination method according to claim 1, wherein the integration time for generating the individual images is not longer than the lifetime (td) of the dark state (D).

3. In the test measurement, the excitation radiation (a) is irradiated in a pulsed manner, and the lifetime (td) of the dark state (D) is measured. The refresh rate is set accordingly with the result that the integration time is 90% of the lifetime (td) of the dark state (D), the microscopy method according to claim 1 or 2.

4. The refresh rate is set such that the integration time is 50% of the lifetime (td) of the dark state (D), the microscopy method according to claim 3.

5. The refresh rate is set such that the integration time is 10% of the lifetime (td) of the dark state (D), the microscopy method according to claim 3.

6. The integration time is not longer than 300 μs, the microscopy method according to any one of claims 2 to 5.

7. The integration time is not longer than 100 μs, the microscopy method according to claim 6.

8. The integration time is not longer than 50 μs, the microscopy method according to claim 6.

9. The integration time is from 1 μs to 30 μs, the microscopy method according to claim 6.

10. The fluorescent phosphor (11) is in a state that is not chemically affected by the lifetime (td) of the dark state (D). The microscopy method according to any one of claims 1 to 9.

11. The intensity of the excitation radiation (a) does not exceed 1 MW / cm ２ The microscopy method according to any one of claims 1 to 10.

12. The excitation radiation does not exceed 50 kW / cm ２ The microscopy method according to claim 11.

13. The excitation radiation does not exceed 2 kW / cm ２ The microscopy method according to claim 11.

14. The intensity of the excitation radiation (a) is set such that the on-off frequency (fB) has a value not lower than 1 / (300 μs), the microscopy method according to any one of claims 1 to 13.

15. The on-off frequency has a value not lower than 1 / (50 μs), the microscopy method according to claim 14.

16. The on-off frequency has a value from 1 / (30 μs) to 1 / (1 μs), the microscopy method according to claim 14.

17. In the generation of the individual images (14), the excitation radiation (a) is irradiated as CW radiation, the microscopy method according to any one of claims 1 to 16.

18. A high-speed and high-resolution microscope for high-speed microscopy of a sample (2) containing a fluorescent phosphor (11), - An excitation device (3) designed to excite the sample (2) in a wide field of view by excitation radiation (a) in order to emit fluorescence radiation (f), and an imaging device (4) designed to image the sample (2) in a wide field of view with a certain spatial resolution, wherein the excitation device (3) is designed such that at least some of the fluorescent phosphors (11) are separated in the imaging with respect to the spatial resolution, the excitation device (3) and the imaging device (4); - A localization analysis device (6) designed to determine the location of the separated fluorescent phosphors (11) with an accuracy exceeding the spatial resolution; - A control device (6) designed to control the excitation device (3) and the imaging device (4) according to the method according to any one of claims 1 to 17; A high-speed high-resolution microscope comprising the same.

19. The microscope according to claim 18, wherein the imaging device (4) includes a CMOS 2D detector (4a) provided with an image intensifier (4b) connected upstream of the beam path.

20. The microscope according to claim 18, wherein the refresh rate has a value not lower than 1 / (300 μs).

21. The microscope according to claim 20, wherein the value is not lower than 1 / (100 μs).

22. The microscope according to claim 20, wherein the value is not lower than 1 / (50 μs).

23. The microscope according to claim 20, wherein the value is from 1 / (30 μs) to 1 / (1 μs).

24. The microscope according to any one of claims 18 to 20, wherein the imaging device (4) includes a SPAD array (25) as a 2D detector.

25. The microscope according to claim 24, comprising an image field displacement device (30) for imaging a sample field (P) on the SPAD array (25) by displacing an image field (37) and assembling each individual image (14) from several partial individual images.

Images