Methods and optical microscopes for high-resolution sample examination

Abstract

This invention relates to a method for optical microscopic examination of a sample (6), particularly a method for optical microscopic examination by means of laser scanning microscopy or MINFLUX microscopy, wherein drift of the sample (6) or an object in the sample (6) relative to the optical microscope (26) is detected and corrected as necessary. Specifically, this invention relates to a corresponding method for examining a sample (6) by means of laser scanning microscopy or MINFLUX microscopy. For this purpose, reference markers (8, 13) are present in the sample, and the positions of the reference markers are repeatedly determined according to the MINFLUX principle to determine the drift.

Description

[0001] Technical Field of the Invention

[0002] This invention relates to a method for optical microscopic examination of samples, wherein drift of the sample or an object within the sample relative to a measuring device is detected and corrected as necessary. In particular, this invention relates to corresponding methods for examining samples by means of laser scanning microscopy (Mikroskopie) or MINFLUX microscopy, and to an optical microscope for performing such methods.

[0003] In MINLFUX microscopy, the position of individual particles (primarily individual molecules of fluorescent dyes) is determined by scanning particles at multiple locations with a light distribution that has a local minimum intensity of excitation light. Based on the position of these individual particles, a high-resolution image or particle trajectory of the sample can be created. Because MINLFUX microscopy can achieve spatial resolution down to the single-digit nanometer scale, it is particularly dependent on the stability of the sample or the object being examined relative to the measuring apparatus of the microscope. Even the smallest displacement can lead to inaccurate positioning and cause motion blur in the image created based on the particle position.

[0004] In terms of instrumentation, slow sample drift relative to the microscope due to thermal expansion can be reduced through appropriate mechanical design and the selection of materials with low coefficients of thermal expansion. However, these measures require considerable technical work and cannot completely prevent sample drift, especially over long measurement periods. In particular, the aforementioned constructive measures are insufficient to meet the stability requirements within the nanometer range needed when using high-resolution microscopy methods. Therefore, there is particular interest in the accurate detection of drift, especially through compensation or data correction, to correct for drift. Here, drift detection should affect the actual examination of the sample as little as possible. Existing technology

[0005] References WO 2015 / 097000 A1 and Balzarotti F, Eilers Y, Gwosch KC, The publication "Nanometer resolution imaging and tracking of fluorescent molecules with minimal photon fluxes, Science 2017 Feb 10; 355(6325):606-612. doi:10.1126 / science.aak9913. Epub 2016 Dec 22. PMID: 28008086" describes a method (now abbreviated as MINFLUX) for locating spatially isolated fluorescent molecules, in which each individual molecule is illuminated at different scanning positions with an intensity distribution of excitation light having minimum intensity values. For each illumination position, the fluorescence emission excited by the excitation light is recorded, and the position of the corresponding molecule is inferred from the path of fluorescence intensity along the position of the minimum intensity. In essence, this position determination is prone to error, but this error can be reduced by iteratively applying the method. To this end, the illumination positions are adjusted before each iteration step, i.e., the illumination positions are arranged closer to the various assumed positions of the molecules. Simultaneously, the intensity of the excitation light is increased, causing the intensity gradient to increase near the intensity minimum. Alternatively, the measurement duration can be increased, which is equivalent to increasing the intensity of the excitation light in terms of effective light quantity. Using the adjusted parameters, the molecule is sequentially irradiated at each adjusted irradiation position, and the intensity of fluorescence emission is recorded. Now, based on the dependence of the fluorescence signal on the scan position of the intensity minimum, the position of the molecule can be determined with a smaller error than before. These methodological steps can be repeated until the position determination converges or until another interruption criterion is reached, such as a pre-determined maximum acceptable error. The achievable localization accuracy of the MINFLUX method according to the prior art is approximately 1 nm, making it the most accurate commercially available fluorescent molecule localization method.

[0006] Document WO 2015 / 097000 A1 further discloses a (high-resolution) image of the molecular distribution in a sample that can be obtained from the positional data of a single molecule (“MINFLUX imaging”). This method corresponds to the methods known from STORM microscopy and PALM microscopy for generating high-resolution images based on the determination of multiple positions of a single fluorescent molecule, but in the case of MINFLUX microscopy, the spatial resolution of the image is further improved by 5 nm.

[0007] A variation of the MINFLUX method is described in DE 10 2017 104 736 B3, in which the scanning of isolated fluorescent dye molecules is not performed by irradiation with an intensity distribution of excitation light having local minimum values, but rather by irradiation with two substantially complementary intensity distributions of excitation light and fluorescence suppression light. Here, the intensity distribution of the excitation light has local maximum values, while the intensity distribution of the fluorescence suppression light has local minimum values ​​at the same locations. Specifically, the fluorescence suppression light can be STED light, which prevents excited fluorescent dye molecules from emitting fluorescent photons in the edge regions of the intensity distribution of the excitation light by triggering stimulated emission. In this embodiment of the method, the excitation light and fluorescence suppression light are superimposed with such intensity distributions, as is also the case in RESOLFT microscopy and STED microscopy. A variation of the MINFLUX method utilizes the fact that the recorded fluorescence intensity for a given fluorescent dye molecule depends on the distance between the dye molecule and the local minimum intensity of the fluorescence-suppressing light, and the location of the dye molecule can be determined with high precision based on the recorded fluorescence intensities at multiple locations corresponding to the minimum intensity of the fluorescence-suppressing light. Similarly, in this variation of the MINFLUX method, the local minimum intensity can be located at several locations in the sample, and the recorded fluorescence intensity can be evaluated using the same principles as in MINFLUX microscopy. However, the difference lies in that in MINFLUX microscopy, the fluorescence intensity of the fluorescent label increases with increasing distance from the location of the local minimum intensity, while in a further implementation using fluorescence-suppressing light, the fluorescence intensity decreases with increasing distance.

[0008] As seen in STORM and PALM microscopy, the sample to be imaged moves relative to the microscope system during data acquisition, which affects positioning accuracy. During image reconstruction, this results in motion blur and a blurred image. This problem is exacerbated when positioning molecules according to the MINFLUX method, which requires high positioning accuracy. Here, acceptable relative motion between the sample and the microscope system is within 1 nm or less.

[0009] In the prior art, particularly in the prior art concerning STORM microscopy and PALM microscopy, various methods for sample stabilization are known, which involve readjusting the sample position during the duration of data acquisition or subsequently correcting the image data.

[0010] A laser scanning microscope with a temperature sensor is known from DE 199 59 228A1, in which the focal position is tracked by a correction value determined according to a calibration curve of the corresponding current temperature. However, this method is limited to the correction of axial (i.e., focusing) thermal drift.

[0011] In EP 1 548 485 A1, a method and a microscope for performing the method are proposed, wherein images of preferably stationary objects are acquired at different time points, wherein a motion estimator is used to calculate the drift between image acquisitions, and wherein drift identification is performed by comparing corresponding blocks (and, if necessary, sub-blocks) of the images. The drifts measured for each block can then be calculated as a total drift. Actual correction can be performed through post-processing of the image data or by tracking the beam position or sample position during image acquisition.

[0012] R. McGorty, D. Kamiyama, and B. Huang's "Active microscope stabilization in three dimensions using image correlation," Optical Nanooscopy 2(1),3(2013), discloses a method in which sample drift is compensated by tracking a piezoelectrically driven sample stage, with the stage drift not exceeding 10 nm in the x and y directions and not exceeding 20 nm in the z direction. Here, sample drift is determined by the Kreuzkor relation between the transmitted light image of the sample and a reference transmitted light image, where the drift in the x and y directions is determined by the location of the maximum value of the Kreuzkor relation, and the drift in the z direction is determined by the amplitude of the Kreuzkor relation. To acquire the transmitted light image, infrared light that does not affect fluorescence imaging is transmitted through the sample.

[0013] A method for lateral drift correction in PALM / STORM microscopy is known from J. Prescher's "Assembly and optimization of a super-resolution STORM microscope for nanoscopic imaging of biological structures," Dissertation, Munich (2016), which introduces fluorescent nanoparticles as reference objects into the sample. To determine the sample drift in the x and y directions, epifluorescence images of these nanoparticles are repeatedly acquired during STORM data acquisition and correlated with the initially acquired reference epifluorescence image. The drift in the x and y directions at the corresponding time points is determined based on the location of the maximum cross-correlation value and used for position correction when locating fluorescent single molecules.

[0014] A drift correction method in STORM / PALM microscopy is known from C. Geisler et al., “Drift estimation for single marker switching based imaging schemes,” Optics Express 20, 7274 (2012). This method is also based on the correlation of multiple camera (raw) images. However, in this method, the sample drift as a function of time is determined by the pairwise correlation of all camera (raw) images, rather than just by the correlation with a single reference image. This significantly improves the accuracy of single-molecule localization without requiring any further assumptions about the cause or type of drift, or without modifying the data acquisition.

[0015] Invention Task

[0016] The present invention aims to provide a method and optical microscope for examining samples, wherein during sample examination, drift of the sample or an object within the sample relative to a measuring device is detected and corrected as necessary. Here, detection and correction must also meet the accuracy requirements of MINFLUX microscopy and should have as little impact as possible on the examination of the sample.

[0017] Solution

[0018] The objective of this invention is achieved by the method according to independent claim 1 and the optical microscope according to independent claim 33. Dependent claims 2 to 31 relate to preferred embodiments of the method, dependent claim 32 relates to the use of a sample carrier in the method according to the invention, and dependent claim 34 relates to a preferred embodiment of the optical microscope.

[0019] Invention Description

[0020] This invention relates to a method for optical microscopic examination of samples, wherein the sample is irradiated with a focused laser at multiple irradiation locations and optical signals from the sample are detected. Here, the examination of the sample using a focused laser can be performed according to a wide range of imaging modalities, such as using confocal laser scanning microscopy, STED microscopy, or locating individual particles in the sample according to the MINFLUX method. The method is characterized by irradiating the sample with a focused laser and detecting the optical signal based on location and, if necessary, with high spatial resolution; therefore, it is susceptible to forgery due to sample drift.

[0021] Therefore, according to the present invention, sample drift is detected by repeatedly determining the position of at least one reference marker based on the MINFLUX principle, wherein the determination of the reference marker position is integrated into the measurement process of sample examination and repeated between irradiations of the sample at consecutive irradiation positions. In particular, this method can track and compensate for sample drift over long periods (e.g., hours or days). In principle, this method (unlike some alternative methods) is not limited to two-dimensional drift detection but can be selectively applied in one, two, or three spatial directions.

[0022] The MINFLUX method for determining the position of at least one reference marker is characterized by illuminating the reference marker with an intensity distribution of scanning light having local minima at multiple scan positions around the (assumed) position of the reference marker, wherein the scanning light induces or modulates the light emission of the reference marker. Here, the scan positions are arranged within a short distance (typically within an Airy disk) around the reference marker position, which is at least approximately known (e.g., based on previous measurements). Based on the light emission recorded at different scan positions, the position of the reference particle can be determined, for example, by forming a weighted vector sum with improved accuracy. The scan positions can be illuminated progressively by the scanning light, or the scan positions can be located on a curve or define a path segment along which the sample is illuminated by continuously moving the excitation beam relative to the sample. In the second case, a time interval (dwell time) can be defined during the continuous movement of the excitation beam, by means of which photons detected by the detector are associated with a specific scan position, similar to continuous line scanning, for example, in confocal laser microscopy. The sample locations are arranged around the assumed location of the reference marker, meaning the scan locations are within a circle or sphere surrounding the assumed location (e.g., the aforementioned close-range range). This also includes, of course, just two scan locations, such as those located on opposite sides of a line intersecting the assumed location.

[0023] The specific design of the MINFLUX method can vary in many ways, and reference can be made to existing techniques of MINFLUX microscopy. Here, only the following disclosures are cited as examples: KCGwosch et al. (2019) "MINFLUX Nanoscopy Delivers Multicolor Nanometer 3D-Resolution in (Living) Cells", bioRxiv doi: 10.1101 / 734251 and M. Weber et al. (2020) "MINSTED fluorescence localization and nanoscopy", bioRxiv doi: 10.1101 / 2020.10.31.363424. Therefore, in the sense of the present specification and patent claims, the term MINFLUX method should be interpreted broadly and should include all embodiments having at least the above-described (and, if necessary, other) features.

[0024] The type of light emission used to locate a reference marker can vary considerably, and in particular can be fluorescence or phosphorescence, Rayleigh scattering or Raman scattering, CARS scattering (coherent anti-Stokes Raman scattering), or light emission generated by frequency harmonics (second harmonic, third harmonic, etc.) of the scanning light or by mixing with the scanning light. According to the invention, the scanning light is adapted to excite or induce a corresponding type of light emission from the reference marker or modulate its intensity, wherein modulation can amplify or attenuate the intensity of the light emission. If the scanning light has a modulating effect and does not itself excite or induce light emission from the reference marker, then additional excitation light is required to irradiate the reference marker. For example, a fluorescent reference marker can be scanned with scanning light that excites fluorescence. However, the scanning light may also trigger stimulated emission, thereby suppressing spontaneous (fluorescent) emission from the reference marker; in this case, additional excitation light must be used to excite the reference marker to emit light.

[0025] In particular, the reference marker is photostable. This means that the light emission of the reference marker remains essentially constant throughout the desired time period of the optical microscopy experiment. Specifically, the reference marker will not bleach or darken during the desired time period. The appropriate time period can vary depending on the experimental requirements. For example, for long-term measurements, a time period of several hours to several days may be necessary. However, for some experiments, a time period of less than one second, one millisecond, or even one microsecond may be sufficient.

[0026] In a preferred embodiment of the method according to the invention, the sample is examined by acquiring images of the sample using a laser scanning method (e.g., by means of confocal microscopy or STED microscopy). In these cases, image acquisition is typically performed by scanning the sample line by line using a focused excitation beam; in the case of STED microscopy, this is further achieved using the intensity distribution of the STED light superimposed on the focused excitation beam, which has local minima. In this case, the scanning can be interrupted periodically, for example after scanning a predetermined or pre-predictable number of scan lines, or at fixed time intervals, to determine the position of the reference marker and measure the drift according to the MINFLUX method. Then, taking into account the measured drift, the scanning of the sample continues by corrective control of the scan. To completely eliminate the effect of drift on the sample image, the accuracy of the drift correction must exceed the spatial resolution of the laser scanning method, ideally ≥2 times. Since the resolution of the MINFLUX method is inherently much higher than that of laser scanning, this requirement is usually met.

[0027] In another preferred embodiment of the method according to the invention, not only is the location of the reference marker determined, but the sample is also examined using the MINFLUX method. For this purpose, in addition to the reference marker, the sample contains other particles that can be located using the MINFLUX method. These particles also possess fluorescent or light-scattering properties, but are distinguishable from the reference marker. In many applications, these particles are single molecules of a fluorescent dye used to label structures of interest in the sample. Here, the fluorescent dye is suitable for labeling, having both a fluorescent and a non-fluorescent state, and the ratio of fluorescent to non-fluorescent dye molecules can be adjusted such that at any given time point, only a single dye molecule is in a fluorescent state per diffraction-limited volume, thereby allowing for independent localization of individual molecules. The ratio between fluorescent and non-fluorescent dye molecules can be adjusted by modifying the chemical composition of the medium, or it can be adjusted by photoinduction; for this purpose, those skilled in the art can refer to existing techniques such as PALM / STORM microscopy and MINFLUX microscopy.

[0028] In this embodiment, when examining the sample, the coordinates of the located particles are first collected, and then a high-resolution image of the sample and / or the trajectory of individual particles can be determined based on these coordinates. Here, on one hand, drift correction can be performed by tracking the sample to determine the amount of drift based on the position of a reference marker. However, the particle coordinates can also be corrected purely by computational calculation using the amount of drift, thus eliminating the need to track the sample. For example, in this case, the drift-corrected image can already be displayed as a real-time image on the display unit during measurement.

[0029] If the sample is examined using the MINFLUX method, the accuracy of drift correction must be higher than the accuracy of locating individual particles for the examination. To completely eliminate the effect of drift on localization, the accuracy should be increased to ≥2 times. Since both sample examination and drift determination are performed using the MINFLUX method, their method-related spatial resolution is the same. However, the corresponding localization accuracy in MINFLUX depends on several acquisition parameters, particularly the number of localization iterations, the corresponding arrangement of scan positions, and the intensity of the scanning light. Because the photostability of the reference marker is much higher than that of a single dye molecule, the reference marker can be localized at a higher resolution.

[0030] When examining a sample according to the MINFLUX method, the measurement does not need to be interrupted when the sample is illuminated with scanning light at the irradiation location to locate the fluorophores (i.e., during the MINFLUX localization sequence). However, the fluorophores can be located as a whole or even sequentially for multiple different fluorophores before the measurement is interrupted and the location of the reference marker is determined.

[0031] A particularly advantageous embodiment of the method according to the invention is that the laser used for examining the sample is simultaneously (i.e., in the same measurement) used as scanning light for determining the position of a reference marker. This dual use of the laser is particularly useful if, for the purpose of examining the sample by means of MINFLUX microscopy or STED microscopy, an intensity distribution with local minimum intensity of the excitation light or STED light is generated anyway. In these cases, the same beam deflection device (which is configured for examining the sample by means of MINFLUX microscopy or STED microscopy) can also be appropriately controlled to guide the laser onto a reference marker in the sample, the position of which is determined according to the MINFLUX principle, and drift is detected by repeating such position determination. This drift detection (and drift correction if necessary) can be easily integrated into the sequence of sample examination without requiring any additional optical or mechanical components. Therefore, it is simpler in structure and more economical than alternative solutions that require additional beam paths, light sources, and / or detection devices. However, a more important and fundamental advantage is that the method accurately detects the drift of the reference marker relative to the optical imaging system used for examining the sample. This is the difference between drift detection and its alternatives, which typically require additional light sources, optics, separate beam paths, detectors, or other sensors. Furthermore, in alternative methods, drift between the sample, sample carrier, or microscope components and the additional elements used for drift detection cannot be detected in this way.

[0032] Specifically, in the dual use of the scanning light described above, the scanning light has the same light distribution at the focal point as the laser used to examine the sample (e.g., a ring-shaped light distribution or a locally hollow light distribution). In a variation of the MINFLUX method, the sample is illuminated by a superposition of a Gaussian excitation light distribution and a fluorescence suppression light distribution with local minima, the fluorescence suppression light being specifically used simultaneously as the scanning light to examine the sample. Thus, the light distribution at the focal point is identical when both the reference marker is positioned and the sample is examined.

[0033] Using lasers as scanning light to examine samples can be easily achieved, for example, with light-scattering nanoparticles (e.g., gold beads), because these nanoparticles scatter light over a wide range of wavelengths. Therefore, various types of excitation lasers can also be used specifically for fluorophores in the sample to generate light scattering signals.

[0034] In a preferred embodiment of the method, the drift of the reference marker relative to the optical imaging system of the microscope is calculated based on the position of at least one reference marker, and this drift is compensated for by position correction of the beam deflection unit or by tracking the sample. For example, this can be achieved by applying an offset voltage to a control voltage that controls the position of the beam deflection unit or the sample positioning stage, or by applying a corresponding offset to a digitally available position value.

[0035] As previously described, the determination of the reference marker's position is integrated into the measurement process of the examined sample and is repeated between irradiations of the sample at consecutive irradiation positions. Here, the position of at least one reference marker can be determined at regular time intervals or after a fixed number of irradiation positions. However, if necessary, the position determination can be performed not at fixed intervals or not strictly at fixed intervals, but adjusted according to the corresponding drift velocity of the sample. This process is particularly useful after the sample is placed in the microscope or after it has been (manually) moved, as this initially leads to relatively rapid sample drift. During this stage, the reference marker's position is determined at shorter intervals, while after the sample is placed, the drift velocity decreases, allowing for an increase in the time interval between position determinations. For this purpose, a drift curve is calculated based on the position determination, and the time point at which the sample has drifted by a predetermined maximum acceptable amount is determined by extrapolating this drift curve. For this purpose, the drift can be modeled, for example, using an exponentially decaying exponential function.

[0036] Adjusting the time interval for determining the reference marker's position is particularly important because position determination using the MINFLUX method has a limited capture range. This means that precise position determination of the reference marker is only possible when the distance between the scan position and the actual position of the reference marker is not too large. This capture range is typically around 200 nm, and it must be ensured that the reference marker does not leave the capture range between consecutive position determinations.

[0037] However, the frequency with which sample examination is interrupted by determining the position of the reference marker to redetermine drift must be weighed against the time required for this. The higher the frequency of drift determination, the less measurement time is available proportional to the actual sample examination time. Frequent interruptions in sample examination are often unacceptable, especially with dynamically changing samples. Therefore, drift determination preferably does not exceed 5% of the total measurement time, more preferably 2%, and particularly preferably 1%. In special cases, it is also advantageous to additionally use another external method for drift determination and / or drift correction; this method can also be based on a reference marker or other control variable located in the sample. Here, it is acceptable if the external method for drift determination itself cannot achieve long-term stability, for example, because its own drift speed is slower relative to the optical system used for sample examination. In this case, the determination of the reference marker position according to the invention is used for zero-point correction of the external method for drift determination, and the time interval for zero-point correction can be longer than the drift determination of the sample itself.

[0038] For the feasibility of the method according to the invention, the key is that the position of a given reference marker can be reliably and repeatedly determined over a long period of time, i.e., the reference marker undergoes only very little degradation or no degradation at all (e.g., in the form of photobleaching). For shorter measurements, it may be sufficient if the position determination of the reference marker can be repeated 100 or 1000 times, but for long-term measurements of several hours or even days, up to 10,000 repetitions are required.

[0039] Another requirement for the reference marker is that it must be luminescent at each position determination to allow for continuous drift tracking. If the reference marker is temporarily in a dark state (as is often the case with organic fluorescent dyes), position determination is impossible during these times. Therefore, it is possible for the sample to drift too far during periods when the reference marker is in a dark state, causing the reference marker to leave the capture range. Ideally, the reference marker should not be in a dark state, but the time spent in a dark state during measurement that cannot induce or excite light emission should not exceed 10%, preferably not more than 5%.

[0040] Due to stability requirements, individual molecules of organic fluorescent dyes (e.g., those frequently used to acquire high-resolution images or trajectories according to the MINFLUX method) are generally unsuitable as reference markers for the methods according to the invention. Instead, nanoparticles doped with highly stable fluorophores, particularly nanoparticles doped with ions or complex compounds of rare earth metals such as cerium, praseodymium, neodymium, samarium, europium, terbium, dysprosium, erbium, thulium, and ytterbium, are preferably considered as fluorescent reference markers. Such nanoparticles exhibit extremely high photostability. Alternatively, fluorescent (semiconductor) quantum dots or nanodiamonds with fluorescent defects are also suitable as fluorescent reference markers.

[0041] The reference marker does not necessarily have to be fluorescent; light emission can also be induced by the (elastic or inelastic) scattering of scanning light. Metal nanoparticles or nanorods are preferred reference markers in this regard, particularly because light scattering on these metal particles is unaffected by dark states, and these particles also possess high chemical and photophysical / photochemical stability. Here, the contrast of light emission relative to ambient light or to undesired light reflections, such as those that may occur at optical interfaces in the beam path, particularly at coverslip interfaces, can be improved by selectively detecting light emission emitted by the reference marker with only one polarization (orthogonal to the polarization of the scanning light). This utilizes the fact that depolarization or a change in polarization state of the scanning light typically occurs when light is scattered on nanoparticles. Therefore, the depolarized or polarization-changed scattered light can be detected by an analyzer that blocks the polarization of the scanning light, thus concealing the scanning light reflected at the interface. A simple implementation of this variation can be achieved by arranging a cross-linear polarizer in the beam path of the scanning light or in front of the scattering light detector used for light emission. Alternatively, a polarization beam splitter can be used to reflect the linearly polarized scanning light into the microscope beam path. Since the depolarized scattered light, scattered by the reference marker and propagating in the microscope beam path in the opposite direction to the scanning light, has a rotated polarization component relative to the scanning light, these components of the scattered light can pass through the polarization beam splitter and can be (selectively) detected by a detector positioned downstream of the polarization beam splitter in the propagation direction. The scanning light reflected at the optical interface in the beam path has the same polarization as the incident scanning light and is reflected again in the opposite direction by the polarization beam splitter.

[0042] Light-scattering nanoparticles in samples are sometimes used in other sample stabilization methods, such as the one described in WO 2020 / 201430A1. This may produce a synergistic effect.

[0043] Generally, the method according to the invention requires that reference markers be identified or located in the sample first, and their locations at least approximately determined. For this purpose, the sample or a portion thereof can be imaged before the examination begins, and one or more reference markers can be identified in the image, with their locations approximately determined. Alternatively, the reference markers can be located as part of the initial location determination by scanning the sample with scanning light until they are detected based on light emission. The previous scan location can then be used as a starting point for determining the location of the reference markers according to the MINFLUX method. In a preferred embodiment of the method according to the invention, the reference markers are located and / or identified in the sample based on the wavelength of their light emission. For this purpose, the wavelength of the reference markers' light emission must be different from other light signals of the sample. For example, if the sample is stained with a fluorescent dye, its fluorescence emission will undergo a redshift (Stokes shift) relative to the wavelength of the excitation light, i.e., its wavelength is greater than the wavelength of the excitation light. Conversely, when the same excitation light is (elastically) scattered at the reference marker, a scattered light signal is generated with the same wavelength as the excitation light. Therefore, even if the same excitation light is used to excite fluorescence and scatters at the reference marker, the reference marker can be distinguished from the fluorescence from the sample based on the wavelength.

[0044] Similarly, the light emission of a reference marker can be distinguished from other signals detected from the sample based on the statistical characteristics of the photon flux emitted. Light scattering on the metallic nanoparticles used as reference markers is instantaneous, while the fluorescent electronically excited states of the fluorescent dyes used to stain structures in the sample decay according to an exponential decay distribution, typically over nanoseconds. Therefore, temporal separation of scattered light and fluorescence is possible if time-resolved detection methods (e.g., time-correlated photon counting) are used for light detection, as is well known in existing techniques of fluorescence spectroscopy and fluorescence lifetime imaging (FLIM). The light emission of a fluorescent reference marker can also be temporally distinguished from other fluorescence signals from the sample, provided that the fluorescence lifetimes of the two signals differ significantly from each other.

[0045] Besides the lifetime of the excited state, other statistical parameters of photon flux can also be used for separation. As shown above, for example, almost all fluorophores have a transient dark state, into which the fluorophore can temporarily transition. Therefore, even under continuous excitation, the photon flux of fluorescence emission is repeatedly interrupted by phases of non-fluorescence emission. The duration of the through state (Ein-Zustand) with fluorescence emission, the duration of the off state (Aus-Zustand) without fluorescence emission, and the average number of fluorescence photons emitted during the through state duration depend largely on the type of the corresponding fluorophore and are therefore suitable as parameters for distinguishing different fluorophores. Accordingly, these parameters can also be used to identify reference markers in a sample and to distinguish it from other light sources in the sample.

[0046] To locate reference markers, the fact that reference markers produce substantially continuous light emission and are stationary, while fluorescence from samples can vary more significantly over time, can be utilized. Specifically, if the MINFLUX method is used to examine samples, spatially independent dye molecules from different groups are activated at different time points for localization. Therefore, fluorescence fluctuations in the sample are substantial. However, fluorescence fluctuations can also occur due to dark states and / or due to the movement of structures within the fluorescently labeled sample. In these cases, since the light emission from the reference marker accumulates while the fluorescence intensity from other sources is averaged and the signal relative to the reference marker weakens, the reference marker can be identified by the korrelation of multiple images of the sample acquired chronologically. For two (grayscale) images f1 and f2 with pixel indices i and j, the correlation amplitude r can be calculated using the following formula. ij :

[0047]

[0048] Here, and This represents the average gray values ​​of images f1 and f2, and the summation is performed over all pixels of the image. The correlation can be calculated in a similar manner for more than two images. It can be derived from the correlation amplitude r. ij Thresholds are formed to identify reference markers, and if necessary, morphological criteria such as shape and / or size can also be used to identify reference markers.

[0049] While the method according to the invention can be performed using only a single reference marker (whose position is repeatedly determined), in practice, it is often advantageous to determine the position at multiple reference markers. If the drift of multiple reference markers is detected in parallel and individually, the accuracy of the drift determination can be improved by averaging. Furthermore, the method is more robust, allowing the sample examination to continue even if one or more reference markers cannot be further determined, for example, due to photobleaching. By comparing the position determinations of multiple reference markers, it is also possible to identify when a reference marker detaches from its binding site and moves relative to the sample and other reference markers. Independent position determination of multiple reference markers also allows for the determination of the drift of (any) point in the sample based on the drift of multiple reference markers through interpolation. Thus, not only can the relevant drift motion of the entire sample or objects within the sample be detected and corrected, but also the morphological deformation of objects within the sample can be detected and corrected.

[0050] Even when drift determination is performed in parallel at multiple reference markers, it is not necessary to determine the position of all these reference markers every time sample inspection is interrupted; multiple reference markers can also be located alternately. Here, the positions of other unlocated reference markers can be inferred from the positions of the already located reference markers. In this way, unlocated reference markers can remain within the capture range of MINFLUX positioning for subsequent position determination.

[0051] When different reference markers are positioned alternately, they can be located in a fixed order or a random sequence. However, in a particularly advantageous embodiment of the method, the reference marker to be positioned is selected based on the previous irradiation position of the focused laser. Here, the closest reference marker or a reference marker whose distance from the previous irradiation position does not exceed a predetermined amount, preferably 30 μm, more preferably 20 μm, and particularly preferably 10 μm, can be selected. In this embodiment, the beam deflection oscillation is minimized, resulting in a short positioning time for the focused laser in the sample and keeping the time required for determining the position of the reference marker low.

[0052] Samples can be labeled with reference markers in various ways and methods. In a preferred embodiment of the method according to the invention, samples are labeled by attaching reference markers to or within a sample carrier (particularly a slide or coverslip). For this purpose, fluorescent or light-scattering nanoparticles are chemically bonded to the surface of the slide or coverslip or fixed by electrostatic interactions before the formulation is applied to it. Although this type of sample labeling is common in laboratory practice, samples prepared in this way do not always provide satisfactory results in terms of the adhesion between the nanoparticles and the surface. Furthermore, it is difficult to set the desired density of nanoparticles in the sample, and particularly difficult to set their reproducibility.

[0053] Therefore, slides or coverslips with reference markers pre-installed during production are particularly suitable for use in the method according to the invention. Here, nanoparticles can be firmly attached to, or even embedded in, the carrier material. However, not only nanoparticles can be used as reference markers; in principle, any mark that can be positioned in at least one spatial direction using the MINFLUX method can be used as a reference marker, i.e., in addition to dot marks, line marks and even planar marks can be used as reference markers. Here, dot marks allow for position determination in three spatial directions, line marks allow for determination of two positions in directions perpendicular to the line's extension, and planar marks also allow for position determination perpendicular to the plane.

[0054] Structures created on or within materials by mechanical engraving or laser marking, as well as structures that form scattering centers upon irradiation, are also suitable for use as reference markers. Light scattering reference markers can also be fabricated by photolithographically structuring glass surfaces. The main advantage of sample carriers prepared in this way is that the reference markers can be arranged in a reproducible manner with regular patterns and defined intervals, ensuring that there is always a reference marker with a definable position within the scanning area. In particular, laser marking can create structures not only on surfaces but also within glass, so the position of the reference marker is not necessarily limited to a plane (glass surface). Although if all reference markers are located in the same plane, the position determination of the reference marker in the z-direction (i.e., in the direction of the optical axis) is limited to the capture range of MINFLUX positioning, it is possible to use reference markers arranged at different depths to achieve position determination of the reference marker over a significantly larger range.

[0055] Labeled sample carriers are known to be usable for other purposes and are partially commercially available. In this respect, the invention includes not only the methods described and presented in various embodiments, but also the application of sample carriers with reference markers in the methods according to the invention.

[0056] As an alternative to attaching reference markers to or within a sample carrier, reference markers can also be coupled to objects within the sample, such as cells. In this case, the reference markers are no longer necessarily fixed in position relative to the entire sample, but rather relative to the objects they are coupled to. In this implementation, the determination of the reference marker position no longer involves detecting only the drift of the entire sample, but alternatively or additionally, detecting the drift or active movement of the objects or cells coupled to the reference markers. This is a preferred implementation of the method in many applications examining intracellular structures, as passive and active cell movement can be as disruptive to the examination as drift of the entire sample. Furthermore, this implementation of the method allows not only the detection of coordinated movement of the sample or cells as drifting, but also the detection of morphological changes (e.g., stretching or compression) that occur during the examination due to growth or cell division processes. For this purpose, drift at any location within the sample can be determined by interpolation based on the drift of multiple reference markers located around the corresponding location. Then, the reference markers form a (time-varying) two-dimensional or three-dimensional mesh, on which the coordinates of points between mesh points can be interpolated using the warping algorithm.

[0057] In special cases, objects in a sample can also possess inherent particles suitable as reference markers. An example of this is the heme pigment produced by the malaria pathogen Plasmodium falciparum during hemoglobin degradation, which is present in infected red blood cells. These pigments are birefringent, thus polarization contrast is particularly suitable for determining their location.

[0058] In addition to the methods according to the invention described in the various embodiments, the invention also includes an optical microscope having an objective lens, a light source for scanning light, a beam shaping device for forming the scanning light in a sample with an intensity distribution having local intensity minima, a scanning device for positioning the scanning light in the sample, and a detector for detecting light emission from a reference marker in the sample. The optical microscope is characterized by having a control unit configured to perform the methods according to the invention. To correct for sample drift determined at the reference marker, the optical microscope preferably has a sample stage with a driver, which can be used to compensate for sample drift through tracking.

[0059] Advantageous improvements of the invention are derived from the patent claims, specification, and drawings, as well as the related descriptions of the drawings. The advantages described in the features and / or combinations of features of the invention are merely exemplary and may have alternative or cumulative effects.

[0060] The following provisions apply to the disclosure (but not the scope of protection) of the original application documents and patents: Further features can be found in the accompanying drawings, particularly the relevant arrangements and effective connections shown. Combinations of features from different embodiments of the invention or from different patent claims may also deviate from the selected reference relationships in the patent claims, and are hereby set forth. This also applies to features shown in separate drawings or mentioned in their descriptions. These features may also be combined with features from different patent claims. Similarly, in further embodiments of the invention, features listed in the patent claims may be omitted, but this does not apply to the independent patent claims of granted patents.

[0061] Reference marks included in patent claims do not constitute a limitation on the scope of protection of the patent claims. They are only used to make the patent claims easier to understand. Brief description of the attached diagram

[0063] Figure 1 An embodiment of the method according to the present invention is illustrated in a flowchart.

[0064] Figure 2 Another embodiment of the method according to the invention is illustrated in a flowchart.

[0065] Figure 3 Another embodiment of the method according to the invention is illustrated in a flowchart.

[0066] Figure 4 A schematic diagram of the method according to the present invention is shown.

[0067] Figure 5 The diagram illustrates how drift is determined by interpolation from the positions of multiple reference markers.

[0068] Figure 6 A sample carrier for use in the method according to the invention is shown.

[0069] Figure 7 An optical microscope according to the invention, based on a first embodiment, is shown.

[0070] Figure 8 An optical microscope according to the invention, based on a second embodiment, is shown.

[0071] Attached Figure Description

[0072] exist Figure 1 The flowchart illustrating an embodiment of the method according to the present invention is shown in the figure. Figure 1 The sample is examined using laser scanning methods, such as confocal microscopy or STED microscopy. For this purpose, the sample or structures of interest within the sample are stained with fluorescent markers. In the procedure... Figure 1In the diagram, the dashed line represents the data flow.

[0073] First, reference markers are identified in the sample, and drift is determined by repeated position determination during sample examination. For example, reference markers can be identified in an overview image of the sample based on characteristic shapes or specific light emission, and the position of the reference markers in the sample can be determined at least approximately. For example, metallic nanoparticles can be identified in a reflected light image of the sample based on their bright, point-like scattered light signals.

[0074] For example, such a reflected light image can be overlaid with a raster fluorescence image of the sample, allowing the position of reference markers relative to the sample structure to be displayed based on the overlay. For instance, a user can select one or more reference markers from the image to determine drift. This selection can be made, for example, by marking areas on a graphical user interface, such as by defining and moving frames using a mouse or touchscreen.

[0075] When a reference marker is selected, its position is first determined using the MINFLUX method to establish a starting position r02. This starting position serves as a reference position for drift measurement in subsequent reference marker position determinations. In the illustrated embodiment, the sample inspection is then performed row by row; initially, a row counter i is set on the first scan row, and the row is scanned point by point, probing the fluorescence of each pixel in the row and storing the fluorescence intensity in the pixel value memory 3. In the next step, it is checked whether all scan rows have been scanned, i.e., whether the sample inspection is complete. If so, the method ends; otherwise, the row counter i is incremented.

[0076] In the next step, it is checked whether reference marker positioning and drift correction should be performed, which requires determining the time elapsed since the last positioning of the reference marker. If the (predetermined) drift measurement interval has been reached, the reference marker position is determined again according to the MINFLUX method. Alternatively, the reference marker position can be determined repeatedly after a predetermined number of scan lines have been scanned.

[0077] The drift correction is calculated based on the difference between the current position r4 and the starting position r02 of the reference marker. This drift correction is used to track the sample and correct the drift. The sample inspection is then continued by scanning the next scan line i.

[0078] exist Figure 2 The flowchart illustrating an embodiment of the method according to the present invention is shown in the figure. Figure 1Unlike the embodiments described above, the MINFLUX method is used to examine the samples. For this purpose, the sample or the structure of interest within the sample is labeled with a fluorescent dye that can photoinducibly switch between a fluorescent state and a non-fluorescent dark state in at least one direction.

[0079] As before, at the start of the examination, a reference marker is identified in the sample. Drift is determined by repeated position determination, and the position of the reference marker is first determined using the MINFLUX method to determine the starting position r02. The drift correction Δr is initialized to a value of 0.

[0080] In the next step, only a small number of fluorescent dye molecules used to label the sample are converted to a fluorescent state through photoactivation. Here, the density of fluorescent dye molecules is adjusted so that at most one fluorescent dye molecule exists within the diffraction-limited volume.

[0081] In the next step, the previously photoactivated dye molecules are localized according to the MINFLUX method. Here, localization involves initially determining the approximate location of the dye molecules as a starting position value, which can be determined by systematically scanning the sample (or a portion of the sample) or by acquiring epifluorescence images of the sample. Starting from the corresponding initial localization value of the dye molecules, the positions of the dye molecules are progressively localized with increasing precision according to the MINFLUX method. For this purpose, the dye molecules are scanned at multiple scanning locations using an intensity distribution of excitation light with local minima, and the improved position r of the dye molecules is determined based on the fluorescence intensity detected at these scanning locations. In this embodiment of the method according to the invention, the position of the dye molecules is then corrected by a drift correction value Δr, and this position is then stored in the coordinate memory 51.

[0082] If a sufficient number of monomolecule localizations are achieved, the measurement can be interrupted and terminated at this point. Various criteria can be used as termination criteria; for example, the measurement can be terminated if a predetermined or determinable number of monomolecule localizations are observed or a predetermined signal-to-noise ratio is reached. The measurement can also be terminated interactively by the operator.

[0083] As in the previously described embodiment, it is now checked whether reference marker positioning and drift correction should be performed. To do this, the time elapsed since the last positioning of the reference marker is determined. If the (predetermined) drift measurement interval has been reached, the position of the reference marker is determined again according to the MINFLUX method. An updated drift correction Δr is calculated based on the difference between the current position r4 and the initial position r02 of the reference marker. The sample is then examined further by re-photoactivating and repositioning a small number of dye molecules.

[0084] exist Figure 3The diagram shows a flowchart of another embodiment of the method according to the present invention. Figure 1 The illustrated implementation also uses laser scanning methods, such as confocal microscopy or STED microscopy, to examine the samples. The implementation of the method shown here is suitable for repeatedly examining samples over longer periods of time, such as hours or even days. Here, drift determination is again integrated into the sample examination, but instead of being performed between consecutive image rows, it is performed during pauses between consecutively acquired images of the sample or image stacks. Therefore, the drift measurement interval is checked after each image acquisition to see if it has expired. If so, as... Figure 1 As shown, drift is determined and samples are tracked at reference markers. The drift measurement interval can be predetermined, dynamically determined based on previously determined drift values, or determined in advance. Then, wait loop 5 is executed until the image acquisition interval expires; then a new image acquisition begins. During the repeated execution of the wait loop, the drift measurement interval is further checked for expiration, and further drift determination and sample tracking are performed as necessary.

[0085] exist Figure 4 The method according to the invention is illustrated schematically in the figure. Sample 6 comprises cells 7 to be examined and reference markers 8, wherein the reference markers are in the form of light-scattering metal nanoparticles 9. Cells 7 are examined by means of confocal laser scanning microscopy, i.e., by scanning the cells 7 in a meandering shape line by line with a focused laser. Sample 6 is positioned at a starting position 10 on a sample stage 11 having a actuator 12, which allows sample 6 to be moved in the x and y directions. If drift correction is required in the three spatial directions, additional adjustments are needed in the z-direction perpendicular to the drawing plane.

[0086] First, reference markers 8 are identified in sample 6, and one of these reference markers 13 is selected to perform drift determination. Preferably, the selection is made such that the reference marker 13 is located near, but not within, the region of the sample 6 to be examined. A first MINFLUX localization is then performed on the reference marker 13. For this purpose, the marker is illuminated with an intensity distribution of scanning light having local minima at a plurality of scan positions 14 (two are used as an example here) densely arranged around the marker, and the reference marker 13 is precisely located based on the intensity of the scattered light recorded at these scan positions 14. This first localization defines the starting position 10 of the reference marker 13, and thus defines the starting position 10 of sample 6 or cells 7 in sample 6. Then, images of cells 7 are acquired by means of raster scanning, for which sample 6 is scanned line by line (here, meandering) starting from the first scan line 15. Once the time elapsed since the first or last position determination reaches the drift measurement interval, the grid scan is interrupted 16, and the reference marker 13 is repositioned using MINFLUX, thus providing the drift position 17 of the reference marker 13 relative to the starting position 10. For correction, the sample is now tracked based on the difference between the drift position 17 and the starting position 10 using the driver 12 of the sample stage 11. Subsequently, image acquisition continues using the grid scan.

[0087] exist Figure 5 The diagram outlines how the drift of any point 18 can be determined by interpolation from the positions of multiple reference markers 8, the positions of which cannot be determined by MINFLUX localization alone. For this purpose, cells 7 are labeled with multiple reference markers 8, arranged along an approximately regular grid 19 in the example shown. (For clarity, only two of the reference markers are shown as examples in the figure.) At the start of the inspection, i.e., at time t = 0, the reference markers 8 are in initial position 10, determined by a first MINFLUX localization of all reference markers 8 prior to the start of the inspection. The initial position 20 of point 18 in the sample (whose drift will be corrected during the inspection period) is determined by interpolation relative to the initial position 10 of the reference markers 8 surrounding point 18.

[0088] At a subsequent time point t>0 during sample examination, for example, cell 7 deforms due to cell growth. Therefore, the position 21 of reference marker 8 changes, where the change in position is not uniform but may vary depending on the reference marker 8. Now, the new MINFLUX localization of reference marker 8 (or at least the reference markers surrounding point 18) provides a set of positions from which the changed position 22 of point 18 can be interpolated.

[0089] The MINFLUX method performed for locating a reference marker may include, if necessary, a pre-location step after selecting the reference marker from an overview image (e.g., a superposition of an incident light image showing the rough location of the reference marker and a grid fluorescence image of the sample structure marked with a fluorescent marker), in which an initial position estimate of the reference marker is determined. For this purpose, for example, the projection of the detector pinhole can be moved circularly in the sample plane while the excitation light distribution remains stationary relative to the sample (a so-called pinhole-orbit). In this way, an initial position estimate of the reference marker can be determined in the lateral direction (i.e., in the sample plane perpendicular to the optical axis of the excitation beam). Subsequently, a so-called 2D MINFLUX method can be performed to precisely locate the reference marker.

[0090] If 3D localization of a reference marker is required (which is often the case if structures in the sample labeled with fluorescent dyes also need 3D imaging), then alternative pre-localization methods may have to be used, as the pinhole track method does not provide any information about the axial position of the reference marker. For example, convergence of position estimates based on the initial position of the excitation light distribution can be used for 3D pre-localization, particularly for light-scattering reference markers. For this purpose, for example, multiple iterative MINFLUX localization sequences can be performed on the point cloud surrounding the location of the reference marker estimated from an overview image. In the localization sequence, the central minimum of the excitation light distribution is first placed sequentially at points of the illumination pattern, which are arranged around random points in the point cloud, and the photon count at each point is determined. A new position estimate is then calculated, and in the next step, the illumination pattern is arranged around the new position estimate. This sequence is performed for each point until the position estimate diverges (if the starting point is outside the capture range of the reference marker) or converges (if the starting point is within the capture range). The closer the starting point is to the actual location of the reference marker, the faster the position estimate converges. Based on the route of the localization sequence, reference markers can be pre-positioned using 3D illumination patterns and 3D point clouds of lateral and axial starting points. Subsequently, the 3DMINFLUX method can be applied to determine the position of the reference markers at high resolution.

[0091] In particular, due to its lower efficiency, the described pre-positioning method requires more positioning steps to locate the sample structure than the MINFLUX method. Here, the reference markers, due to their high photostability, can withstand stronger exposures. Because of the additional time required, the 3D pre-positioning method is specifically used only for initial position determination of the reference markers; for further position determination in drift measurements, pre-positioning is not used, and the 3D MINFLUX method is used directly instead. For subsequent measurements using the 3D MINFLUX method, the positions of the reference markers determined in previous steps can be used as starting values, without pre-positioning.

[0092] Specifically, compared to the conventional MINFLUX method, the pre-positioning based on the convergence of the position estimate does not require prior information about the background signal. However, to improve robustness and convergence speed, the illumination pattern used can specifically include a center position that corresponds precisely to the previous position estimate, where the background signal is estimated by the number of photons detected at that center position (especially by factor scaling).

[0093] Figure 6 A sample carrier 23 made of glass is shown, having a preparation region 24 within which the sample carrier 23 has reference markers 8 suitable for use in the method according to the invention. Such reference markers 8 can be, for example, fixed metal nanoparticles 9 that scatter light; however, they can also be produced, for example, by laser engraving in the form of point-like scattered light centers on the surface or in the glass volume. In the example shown, the sample carrier has reference markers 8 arranged in a regular hexagonal pattern in the preparation region. The sample carrier 23 may have other features, such as a marking region 25.

[0094] In preparation area 24, the sample to be examined can be applied to sample carrier 23. Preferably, the sample is fixed on the surface of sample carrier 23 to ensure a fixed positional relationship with reference marker 8. For example, cells can be cultured on the surface of sample carrier 23, or cells can be chemically immobilized on a glass surface.

[0095] Figure 7 The structure of an optical microscope 26 for performing the method according to the invention is schematically shown. A light source 27 provides scanning light 28, which on one hand excites fluorescent particles located in the sample 6, and on the other hand is scattered by light-scattering reference particles located in the sample 6.

[0096] The scanning light 28 beam 29 passes through two beam deflection devices 30 connected in series (designed here as electro-optic deflectors (EOD) 31), which deflect the beam 29 in either the horizontal or vertical direction. A portion of the scanning light 28 is reflected off the beam path at a partially transparent mirror 44; this portion of the beam is captured in a beam trap 49. The wavefront of the portion of the scanning light 28 transmitted at the partially transparent mirror 44 is shaped by a liquid crystal modulator (spatial light modulator, SLM) 32 such that, when subsequently focused by the objective lens 33, an intensity distribution of the scanning light 28 with local minimum intensity is produced in the sample 6. The beam reflected by the liquid crystal modulator 32 is coupled to the main beam path 35 of the optical microscope 26 via a beam coupler 34. The scanning light 28 is deflected via a scanner 36 and a scanning lens 37 into the rear aperture of the objective lens 33, which focuses the scanning light 28 onto the sample 6, which is held on a sample stage 11 with a driver 12. The sample stage 11 with driver 12 constitutes the sample positioning unit 50.

[0097] In the configuration shown, a galvanometer-based scanner 36 is used for relatively slow but coarse localization of the focused scanning light 28 in sample 6, which can be achieved within a large image field, while an EOD 31 is used for rapid localization of the minimum intensity of the scanned reference marker. Here, the EOD 31 allows for high-speed localization, but the localization range is limited to a few micrometers. The fluorescence 38 of the fluorescent particles and the scattered light 39 of the reference particles received by the objective lens 33 from sample 6 propagate along the main beam path 35 in the opposite direction to the scanning light 28. The fluorescence 38, which is wavelength-shifted relative to the scanning light 28, passes through the beam coupler 34, is purified by the filter 40, focused by the lens 41 through the confocal pinhole 42, and detected by the fluorescence detector 43 disposed behind the pinhole 42. Conversely, the scattered light 39 of the reference particles received by the objective lens 33 from sample 6 has the same wavelength as the scanning light 28 and is therefore reflected out of the main beam path 35 by the beam coupler 34. It continues to move in the opposite direction to the scanning light 28 and is deflected onto the scattered light detector 45 by the partially transparent reflector 44.

[0098] According to the present invention, the optical microscope 26 has a control unit 46 that generates control signals 47 for controlling the scanner 36, EOD 31, and driver 12, and receives detector signals 48 from the fluorescence detector 43 and the scattered light detector 45. The control unit 46 is configured to position a scanning light 28 onto a single fluorescent particle via the scanner 36, and to position a single particle in the sample 6 by repeatedly scanning the single particle in the sample 6 with the scanning light 28 within a close range and determining the position of the single particle based on the amount of fluorescence 38 detected by the fluorescence detector 43 at the scanning position. Here, the control unit 46 periodically interrupts the positioning of the single fluorescent particle to determine the position of at least one reference marker according to the MINFLUX method. For this purpose, the reference marker is scanned with the scanning light at multiple scanning positions within a close range, and the position of the reference marker is calculated based on the amount of scattered light 39 detected by the scattered light detector 45 at the scanning position. Drift calculated based on the difference from the initial position of the reference marker is compensated by controlling the driver 12 of the sample stage 11 to track the sample 6 in the opposite direction.

[0099] Figure 8 The embodiment of the optical microscope 26 shown is similar to Figure 7 The difference in the optical microscope 26 shown lies in the position of the scattered light detector 45 in the beam path, and particularly in the way the scattered light 39 is coupled out. According to Figure 8 Beam splitter 52 is arranged in the main beam path 35 between beam coupler 34 and scanner 36. Beam splitter 52 has a reflective surface, the surface normal of which forms a small angle (i.e., particularly 10° or less, preferably 5° or less, and more preferably 1° or less) with the beam propagation direction in the main beam path 35. For greater clarity, Figure 8 The angle shown is larger than that of a typical embodiment of the optical microscope 26. A small portion (e.g., about 5%) of the scattered light 39 is coupled out of the main beam path by the beam splitter 52 and deflected onto the scattered light detector 45, which is specifically designed as a photomultiplier. Because the beam splitter 52 is located on the objective side of the beam coupler 34, and because the beam splitter 52 specifically reflects wavelength-independent light, a particular portion of fluorescence is also coupled out. However, because the angle between the beam splitter 52 and the main beam path 35 is small, this portion of light is also very small.

[0100] Beam splitter 52 is specifically designed as a so-called beam sampler. Here, it is a component made of a material with a higher refractive index than air (e.g., glass) having a flat reflective surface (specifically, whose reflection is independent of wavelength and polarization). In particular, a second surface opposite the reflective surface is provided with an anti-reflective coating to prevent beam multiplication. Additionally, the second surface can be arranged at a small angle to the reflective surface to avoid interference within the component. As an alternative to the beam sampler, any reflective surface (e.g., a reflective surface of a component already present in the beam path), such as a window, can be used to couple scattered light out.

[0101] It is advantageous to place the beam splitter 52 near the objective lens 33 because the less the wavefront of the detected scattered light 39 is distorted by the optical elements (e.g., liquid crystal modulator 32) in the beam path, the better, which improves the determination of the position of the reference marker.

[0102] To achieve a particularly space-saving and low-cost arrangement, a lens already present in the main beam path 35 can be used. Figure 8 (Not shown in the image) to focus the scattered light 39 onto the scattered light detector 45. That is, the beam splitter 52 is not arranged in the collimated beam path. For example, such a lens could be part of an optical repeater or telescope that mirrors the active surface of the liquid crystal modulator 32 into the pupil of the objective lens 33.

[0103] List of reference numerals

[0104] 1. Flowchart

[0105] 2. Starting position

[0106] 3-pixel value memory

[0107] 4 positions

[0108] 5. Waiting loop

[0109] 6 samples

[0110] 7 cells

[0111] 8 Reference Markers

[0112] 9 nanoparticles

[0113] 10 Starting position

[0114] 11 Sample Stage

[0115] 12 drives

[0116] 13 Reference Markers

[0117] 14 Scanning Position

[0118] 15 scan lines

[0119] 16 Interruption

[0120] 17 Drift Position

[0121] 6 PM

[0122] 19 grids

[0123] 20 Starting position

[0124] 21 positions

[0125] 22 positions

[0126] 23 Sample carriers

[0127] 24 Preparation Zone

[0128] 25 Marked Area

[0129] 26 Optical Microscope

[0130] 27 Light Source

[0131] 28 Scanning light

[0132] 29 beams

[0133] 30 Beam deflection device

[0134] 31 Electro-optic deflector (EOD)

[0135] 32 Liquid Crystal Modulator

[0136] 33 Objective lens

[0137] 34 Beam Coupler

[0138] 35. Main Beam Path

[0139] 36 Scanners

[0140] 37 Scanning Lens

[0141] 38 Fluorescence

[0142] 39 Scattered light

[0143] 40 Filter

[0144] 41 Lens

[0145] 42 pinholes

[0146] 43 Fluorescence Detector

[0147] 44 Partially transparent reflectors

[0148] 45 Scattered Light Detector

[0149] 46 Control Unit

[0150] 47 Control Signals

[0151] 48 Detector Signal

[0152] 49 Beam Trap

[0153] 50 sample localization units

[0154] 51 Coordinate Memory

[0155] 52 beam splitter.

Claims

1. A method for optical microscopic examination of a sample (6) marked with reference markers (8, 13), the reference markers being photostable, wherein the examination comprises sequentially irradiating the sample (6) with a focused laser at multiple irradiation locations. Its features are, The examination of the sample (6) was repeatedly interrupted between irradiations at the irradiation location, and the position (21, 22) of at least one reference marker (13) was determined according to the MINFLUX method. In this process, at least one reference marker (13) is illuminated with an intensity distribution of scanning light (28) having local minimum values ​​at a plurality of scanning positions (14) arranged in a close range around the reference marker (13) to determine its position. The scanning light (28) induces or modulates the light emission of the reference marker (13), and the position (21, 22) of the at least one reference marker (13) is calculated based on the light emission of the reference marker (13) detected by the plurality of scanning positions (14) of the scanning light (28).

2. The method according to claim 1, characterized in that, The light emission of the reference marker (13) is fluorescence or phosphorescence, Rayleigh scattering or Raman scattering, coherent anti-Stokes Raman scattering (CARS) light, or light emission generated by frequency doubling of the scanning light (28) or by frequency mixing with the scanning light (28).

3. The method according to claim 1 or 2, characterized in that, Selectively detect polarizations orthogonal to the polarization of the scanning light (28) from the light emission of the reference marker (13).

4. The method according to claim 1, characterized in that, The laser used to examine the sample is also a scanning light (28) used to determine the position (21, 22) of the at least one reference marker (8, 13).

5. The method according to claim 4, characterized in that, The examination of the sample (6) includes using the MINFLUX method to determine the positions of the individual particles in the sample (6) that are spatially separated from each other.

6. The method according to any one of claims 1-2 and 4-5, characterized in that, The drift of the reference marker (8, 13) relative to the optical imaging system of the optical microscope (26) is calculated based on the position (21, 22) of the at least one reference marker (13), and the drift is compensated by position correction by the beam deflection unit or by tracking the sample (6) by the sample positioning unit (50).

7. The method according to any one of claims 1-2 and 4-5, characterized in that, The positions (21, 22) of the at least one reference marker (13) are determined at regular time intervals.

8. The method according to any one of claims 1-2 and 4-5, characterized in that, If the position (21, 22) of the reference marker (13), which is determined by time extrapolation of the drift curve created based on the previous position of the reference marker (13), changes by a predetermined or predictable amount, the position (21, 22) of the at least one reference marker (13) is re-determined.

9. The method according to any one of claims 1-2 and 4-5, characterized in that, The measurement duration for determining the position of the at least one reference marker (8, 13) shall not exceed 5% of the total measurement duration for examining the sample (6).

10. The method according to any one of claims 1-2 and 4-5, characterized in that, The position (21, 22) of the at least one reference marker (13) can be determined multiple times before further position determination is no longer possible due to the degradation of the reference marker (13).

11. The method according to any one of claims 1-2 and 4-5, characterized in that, During the examination of the sample (6), the reference markers (8, 13) are in a dark state that cannot be induced or excited to emit light for no more than 10% of the time.

12. The method according to any one of claims 1-2 and 4-5, characterized in that, The reference markers (8, 13) are light-scattering nanoparticles or fluorescent nanoparticles.

13. The method according to any one of claims 1-2 and 4-5, characterized in that, Based on the wavelength of light emitted by the reference markers (8, 13), the reference markers (8, 13) are found and / or identified as reference markers (8, 13) in the sample (6).

14. The method according to any one of claims 1-2 and 4-5, characterized in that, Based on the statistical characteristics of the photon flux emitted by the reference markers (8, 13), the reference markers (8, 13) are found and / or identified as reference markers (8, 13) in the sample.

15. The method according to any one of claims 1-2 and 4-5, characterized in that, Based on the lifetime of light emission of the reference markers (8, 13), the reference markers (8, 13) are found and / or identified as reference markers (8, 13) in the sample (6).

16. The method according to any one of claims 1-2 and 4-5, characterized in that, The reference markers (8, 13) are found and / or identified as reference markers (8, 13) in the sample (6) by the correlation of multiple images of the sample (6).

17. The method according to any one of claims 1-2 and 4-5, characterized in that, Position determination is performed alternately or at different reference markers (8, 13) in the same positioning step.

18. The method according to claim 17, characterized in that, In order to determine the position of the reference markers (8, 13), in each case a reference marker (13) is selected from the sample (6), the distance between the reference marker (8, 13) and the last irradiation position of the focused laser not exceeding 30 μm.

19. The method according to claim 17, characterized in that, The drift of point (18) in the sample (6) is determined by interpolation based on the drift of multiple reference markers (8, 13).

20. The method according to any one of claims 1-2, 4-5, and 18-19, characterized in that, The sample (6) is examined by acquiring images of the sample (6) using a laser scanning method.

21. The method according to claim 20, characterized in that, The position of the at least one reference marker (8, 13) is determined with an accuracy at least corresponding to the spatial resolution of the laser scanning method, and with an accuracy exceeding the spatial resolution of the laser scanning method by at least twice the spatial resolution of the laser scanning method.

22. The method according to claim 5, characterized in that, The position (21, 22) of the at least one reference marker (8, 13) is determined with an accuracy at least corresponding to the position of the particle, and with an accuracy exceeding the position determination accuracy of the particle by at least twice the accuracy of the position determination accuracy of the particle.

23. The method according to claim 5, characterized in that, The examination of the sample (6) includes calculating a high-resolution image of the sample (6) and / or the trajectory of the individual particles based on their positions.

24. The method according to claim 5, characterized in that, The drift of the reference marker (8, 13) relative to the optical imaging system of the optical microscope (26) is calculated based on the position (21, 22) of the at least one reference marker (13), and the position (21, 22) of the individual particle determined by the MINFLUX method is corrected by the amount of the drift.

25. The method according to any one of claims 1-2, 4-5, 18-19, and 21-24, characterized in that, The reference markers (8, 13) are fixed in position relative to the sample carrier (23).

26. The method according to claim 25, characterized in that, The reference markers (8, 13) are designed as mechanical engravings or laser markings on or within the sample carrier (23).

27. The method according to claim 25, characterized in that, The reference markers (8, 13) are formed by photolithographic surface structuring of the sample carrier (23).

28. The method according to any one of claims 1-2, 4-5, 18-19 and 21-24, characterized in that, The reference markers (8, 13) are fixed in position relative to the objects in the sample (6).

29. The method according to claim 28, characterized in that, The reference markers (8, 13) are inherent components of the object.

30. The method according to claim 29, characterized in that, The inherent component of the object is a colored pigment.

31. The method according to any one of claims 1-2, 4-5, 18-19, 21-24, and 29-30, characterized in that, The measured drift is determined by the position of the at least one reference marker (8, 13) for zero-point correction of the drift determination supplementary method.

32. The method according to claim 9, characterized in that, The measurement duration for determining the position of the at least one reference marker (8, 13) shall not exceed 2% of the total measurement duration for examining the sample (6).

33. The method according to claim 32, characterized in that, The measurement duration for determining the position of the at least one reference marker (8, 13) shall not exceed 1% of the total measurement duration for examining the sample (6).

34. The method according to claim 10, characterized in that, The position (21, 22) of the at least one reference marker (13) can be determined 100 times before further position determination is no longer possible due to the degradation of the reference marker (13).

35. The method according to claim 34, characterized in that, The position (21, 22) of the at least one reference marker (13) can be determined 1,000 times before further position determination is no longer possible due to the degradation of the reference marker (13).

36. The method according to claim 35, characterized in that, The position (21, 22) of the at least one reference marker (13) can be determined 10,000 times before further position determination is no longer possible due to the degradation of the reference marker (13).

37. The method according to claim 11, characterized in that, During the examination of the sample (6), the reference markers (8, 13) are in a dark state that cannot be induced or excited to emit light for no more than 5% of the time.

38. The method according to claim 37, characterized in that, During the examination of the sample (6), the reference markers (8, 13) are never in a dark state that cannot be induced or excited to emit light.

39. The method according to claim 12, characterized in that, The light-scattering nanoparticles are metallic nanoparticles.

40. The method according to claim 12, characterized in that, The fluorescent nanoparticles are nanoparticles doped with rare earth metal ions or composite compounds, fluorescent quantum dots, or nanodiamond nanoparticles with fluorescent defects.

41. The method according to claim 18, characterized in that, In order to determine the position of the reference markers (8, 13), in each case a reference marker (13) is selected from the sample (6), the distance between the reference marker (8, 13) and the last irradiation position of the focused laser not exceeding 20 μm.

42. The method according to claim 41, characterized in that, In order to determine the position of the reference markers (8, 13), in each case a reference marker (13) is selected from the sample (6), the distance between the reference marker (8, 13) and the last irradiation position of the focused laser not exceeding 10 μm.

43. The method according to any one of claims 1-2, 4-5, 18-19, and 21-24, characterized in that, The reference markers (8, 13) are combined with the sample carrier (23).

44. The method according to any one of claims 1-2, 4-5, 18-19, and 21-24, characterized in that, The reference markers (8, 13) are embedded in the sample carrier (23).

45. The method according to claim 28, characterized in that, The reference markers (8, 13) are combined with the objects in the sample (6).

46. ​​The use of a sample carrier (23) in the method according to any one of claims 1 to 45, wherein the sample carrier (23) is attached with reference markers (8, 13).

47. The application according to claim 46, wherein, The sample carrier (23) is engraved with reference markers (8, 13).

48. The application according to claim 46, wherein, Reference markers (8, 13) are embedded on the sample carrier (23).

49. An optical microscope (26), comprising: Objective lens (33). The light source for scanning light (27). A beam shaping device for forming an intensity distribution with local minimum intensity values ​​of the scanning light (28) in the sample (6). A scanning device for locating the scanning light (28) in the sample (6). A detector is used to detect the light emission of reference markers (8, 13) in the sample (6). Control unit (46) for controlling the scanning device and processing the detector signal (48) output by the detector. Its features are, The control unit (46) is configured to perform the method according to any one of claims 1 to 42.

50. The optical microscope (26) according to claim 49, characterized in that, The optical microscope (26) includes a sample stage (11) with a driver (12), and the control unit (46) controls the driver (12) to compensate for drift determined by the position determination of at least one reference marker (8, 13).

Images