Methods for evaluating structured illuminated microscope images and microscopes with structured illumination

Abstract

2.1. When evaluating microscope images illuminated in a structured manner in different phases, an intermediate result image (f) is obtained. r ) with increased resolution based on the raw images (gi) using a SIM reconstruction method, especially when using a common reconstruction method (jSIM), unmodulated contributions (wide-field contributions) produce artifacts that reduce the contrast of the intermediate result image (f r ) reduce. Such artifacts remain even with jSIM using order-selective filtering. The invention aims to suppress artifacts more effectively. 2.2 By removing elements of the intermediate result image (f r ) are weighted based on elements of the cross-sectional image, and the intermediate result image weighted in this way (f r By outputting the weighted elements as a result image (f̃), artifacts can be effectively reduced or suppressed. It is particularly advantageous to generate a cross-sectional image (f) before weighting. os) of the sample using a reconstruction method and the weighting based on elements of the cross-sectional image (f os to carry out. 2.3. Fluorescence microscopy

Description

[0001] The invention relates to a method for evaluating microscope images of a sample, comprising the steps of: providing several raw digital images of the sample, which are sequentially acquired using a microscope under illumination of the sample in different phases with periodically structured illumination light; and determining an intermediate result image of the sample with increased resolution (in particular, finer than the diffraction-limited resolution of the microscope) from the raw images using a (SIM) reconstruction method. The invention also relates to a correspondingly designed microscope with structured illumination.

[0002] The resolving power of known microscopes depends on the aperture of the microscope objective and the wavelength of the light, due to the diffraction of the light received from the sample in the microscope objective. Since the usable wavelength range of visible light is finite, the resolving power of a microscope is fundamentally limited (Abbe 1873). With respect to the spatial frequencies of the sample to be imaged, this means that the support of the optical transfer function (OTF) of the microscope is limited to a finite region around the origin in the spatial frequency domain. Consequently, the microscope can only image those spatial frequencies that lie within the interval where the support does not vanish. The OTF is the point-spread function (PSF) of the microscope transformed into a spatial frequency domain. The PSF specifies how a point light source is imaged by the microscope.

[0003] By illuminating the sample in several different phases (structured illumination) and subsequently processing the raw images acquired phase by phase from the same sample plane (focal plane) (referred to as structured illumination microscopy - SIM), the lateral resolution (perpendicular to the optical axis of detection) can be improved by a factor of up to two, provided the illumination excitation intensity and the sample emission intensity are linearly related. SIM is disclosed, for example, in DE 199 08 883 A1 and in the article "Surpassing the lateral resolution limit by a factor of two using structured illumination microscopy" by M. Gustafsson (Journal of Microscopy, Vol. 198, 2000, p. 82). It is based on generating a periodic light structure on the sample under investigation, for example, by sinusoidal interference of the illumination light behind an optical grating.Due to the convolution of the sample response with the microscope's spatial frequency spectrum (PSF) in spatial space, a range of spatial frequencies of the sample structure lying outside the carrier of the optical frequency spectrum (OTF) is shifted into the central carrier interval in the spatial frequency domain. There, they superimpose on the original spatial frequency intensities, resulting in moiré effects in the raw images. The Fourier transform of each raw image contains several frequency-shifted "copies" of the sample's spatial frequency spectrum in a given frequency band. Each of these copies is referred to as a separate "order." Their number depends on the number of interfering beams from which the periodic light structure is generated.

[0004] From a set of such raw images, containing the superpositions of the shifted and original spatial frequencies, an intermediate image with increased resolution can be reconstructed using the SIM method. This intermediate image contains both the original spatial frequencies of the carrier interval and the original higher spatial frequencies that were temporarily shifted into the carrier interval by the structured illumination. As a result, the intermediate image has a higher lateral resolution than a conventional single image with uniform illumination. This resolution is referred to as superresolution when it is finer than the diffraction-limited resolving power of the microscope.

[0005] Various reconstruction methods are known, such as Wiener deconvolution or iterative methods like joint deconvolution, also known as jSIM (for example, published by Chakrova et al.: "Deconvolution methods for structured illumination microscopy" in Journal of the Optical Society of America A, Vol. 33, No. 7, 2016, p. B12), particularly using the Richardson-Lucy method. The iterative joint deconvolution methods have the particular advantage that constraints such as non-negativity can be applied, which reduces background noise amplification and thus enables higher image contrast, at least within the focal plane.

[0006] In addition to improving lateral resolution, SIM can also be used to generate an optical section through the sample perpendicular to the optical axis of detection from a set of raw images of a focal plane (so-called OS-SIM, published by Neil et al.: "Method of obtaining optical sectioning by using structured light in a conventional microscope" in Optics Letters Vol. 22, No. 24, 1997, p. 1905). The resulting cross-sectional image is, for example, quasi-confocal, and the axial resolution is also approximately equivalent to that of a confocal microscope. If a set of structured-illuminated images is acquired for a number N of focal planes, a z-stack of quasi-confocal cross-sectional images can be reconstructed.

[0007] SIM can also achieve axial super-resolution (so-called 3D-SIM). For this purpose, the illumination in the sample must also be axially periodically structured, and a separate set of raw images is acquired for each of N sample planes. From a system of equations describing the interactions in all N sample planes, a z-stack of N axially, and especially laterally, super-resolved intermediate images can be calculated. 3D-SIM is described in Gustafsson et al.: “Three-Dimensional Resolution Doubling in Wide-Field Fluorescence Microscopy by Structured Illumination”, Biophys. J., Vol. 94, 2008, p. 4957.

[0008] It is known that reconstruction methods, especially jSIM reconstruction methods, generate not only the modulated contributions but also unmodulated contributions, so-called wide-field contributions. These reduce the contrast of the SIM intermediate result image as artifacts. Wide-field contributions originate predominantly as background fluorescence from sample regions outside the focal plane of the raw images. This is especially true for thick samples.

[0009] From US 2022 / 092752 A1, it is known that order-selective filtering using bandpass filtering allows the suppression of unwanted spatial frequency ranges, especially those of the wide-field contributions, thus reducing them. Despite the filtering, wide-field contributions, especially in the axial direction, can remain. These can be particularly problematic in quantitative measurements.

[0010] The invention is based on the objective of improving a method and a microscope of the type mentioned above, so that artifacts, in particular wide-field contributions outside the focal plane, are better suppressed.

[0011] The problem is solved by a method having the features specified in claim 1 and by a microscope having the features specified in claim 10.

[0012] Advantageous embodiments of the invention are specified in the dependent claims.

[0013] In the context of the invention, the term "raw image" does not mean that the images must be free from manipulation. Rather, raw images can expressly be processed before evaluation according to the invention, for example by unfolding and / or filtering and / or alignment.

[0014] The raw images g i(The index i denotes the M different illumination phases 1...M) can be modeled as follows, for example (according to Schäfer et al.: “Structured illumination microscopy: artefact analysis and reduction utilizing a parameter optimization approach” in Journal of Microscopy, Volume 216, No. 2, p. 165): gi=Q(hdet∗(f⋅(hill∗si)))=Q(gw+gccos φi+gs sin φi), where h det , h ill the illumination or detection PSF, f the sample, s i the illumination grating and Q represent a noise function, while g w Wide-field contributions and g c , g s oscillating contributions as well as (*) the convolution operator and φ i the phase angles at which the images g i were recorded.

[0015] The raw images can be acquired sequentially before being prepared using the microscope, by illuminating the sample in different phases with periodically structured light. This illumination can be coherent or incoherent; the light structure in the sample can therefore be generated by imaging a light pattern or by the interference of multiple beams.

[0016] An interim result image f rThe weighting can, in principle, be determined using any SIM reconstruction method. However, the weighting according to the invention is particularly advantageous when a joint reconstruction method, in particular an iterative joint reconstruction method, and furthermore, in particular a method with order-selective filtering, is used to determine the intermediate result image. With these methods, the achievable suppression is particularly strong. For example, an iterative joint reconstruction can be performed using a modified Richardson-Lucy method according to Chakrova et al. (as follows): fr,k+1=fr,k∑i=1Ms^i[gi(fk⋅s^i)∗hdet]∗hdetT, where f r,k a reconstructed intermediate result image after k iterations, ŝ i = h ill * s i the imaging of the illumination grating into the sample and hdetT These are the adjoint detection PSFs. All quantities appearing in the equations above are vectors or operators in the 3D ℝ domain.

[0017] According to the invention, elements (2D pixels or 3D voxels represented by vectors) of the intermediate result image are weighted, and the weighted intermediate result image containing the weighted elements is output as the final result image. While all elements of the intermediate result image can be weighted, a subset thereof is sufficient. The weighting can be performed, in particular, by multiplying the respective elements.

[0018] Weighting can effectively reduce or suppress artifacts.

[0019] Before weighting, the intermediate result image can be processed, for example by unfolding and / or filtering.

[0020] It is advantageous to determine a cross-sectional image of the sample (especially without wide-field contributions) from the raw images using a reconstruction method (especially OS-SIM or a corresponding or derived method) before weighting, and to perform the weighting based on elements (2D pixels or 3D voxels represented by vectors) of the cross-sectional image. In particular, one of the elements of the intermediate result image can be weighted based on at least one of the elements of the cross-sectional image. This type of weighting leads to a significant suppression of artifacts, especially out-of-focal wide-field contributions, by enhancing the optical cross-sectional effect. Weighting is usually performed for each element of the intermediate result image. The cross-sectional image f os For example, according to Schäfer et al., it can be determined as follows, after the components g c and g s were separated: fOS=gc2+gs2

[0021] Such a cross-sectional image (and also cross-sectional images determined according to the method described by Neil et al.) has no zero-order contributions and therefore no wide-field contributions. These cross-sectional images are thus particularly well suited for weighting (and removing wide-field contributions).

[0022] Preferably, when weighting one of the elements of the intermediate result image, the element in question is combined with an element of the section image f. os multiplied. This allows for weighting with minimal computational effort. Typically, each element of the intermediate result image is multiplied by the corresponding element of the cross-sectional image with respect to its position in the sample (point-wise, i.e., pixel-wise or voxel-wise multiplication).

[0023] In a preferred embodiment, when weighting one of the elements of the intermediate result image f rThe element in question is multiplied by a factor that depends on elements of the raw images and / or elements of the intermediate result image and / or elements of the cut image. The resulting image f̃ is then obtained, for example, as follows: f˜=M⋅fr⋅fOS

[0024] This multiplication can be seen as normalization to keep the range of values ​​of the elements in the resulting image within a predefined range. However, it can also M=1 be used.

[0025] In particular, the factor can contain the mean values ​​of the raw images in the following form: M=1M⋅N∑iM∑nNgi,n(fos,fr) where M is the number of illumination phases, N is the number of pixels or voxels in an image, and <> is a scalar product operator.

[0026] Preferably, the factor depends solely on elements of the raw images. This has the advantage that the factor can be determined before the reconstruction process(s). In particular, the factor can be the inverse of an average value of the raw images: M=M⋅N∑iM∑nNgi,n

[0027] Alternatively, it can correspond to the average of the raw images: M=∑iM∑nNgi,nM⋅N

[0028] Alternatively, the factor can depend solely on elements of the intermediate result image. In particular, the factor can be the inverse of the sum of all elements, or at least some elements, of the intermediate result image. M=1∑fr

[0029] Alternatively, the factor can depend exclusively on elements of the intermediate result image and on elements of the section image: M=∑fr〈fos,fr〉

[0030] This advantageously preserves the overall intensity of the intermediate result image. Weighting redistributes the intensity.

[0031] The invention also comprises a microscope with a control unit configured to carry out the methods described above and equipped with a light source, a two-dimensional spatially resolved detector for acquiring raw images of the sample, and means for generating periodically, in particular sinusoidally, structured illumination light in the sample in different phases, wherein, in particular, no aperture optically intersecting the sample is arranged in front of the detector. The invention also comprises a computer program and a control unit, each configured to carry out the methods described above.

[0032] The invention will now be explained in more detail using exemplary embodiments.

[0033] The drawings show: Fig. 1 a multimodal microscope, Fig. 2 the well-known principle of structured lighting in several phases, Fig. 3. A flowchart for improved evaluation of structured illuminated raw images and Fig. 4 example images with and without the weighting according to the invention.

[0034] In all drawings, matching parts bear the same reference symbols.

[0035] Fig. Figure 1 shows a microscope 1 that has different operating modes. It can perform both classical microscopy techniques, i.e., microscopy techniques whose resolution is diffraction-limited, and super-resolution microscopy techniques, i.e., those whose resolution is increased beyond the diffraction limit. It is an inverted microscope. Alternatively (not shown), it can be configured as an upright microscope.

[0036] The microscope 1 captures a sample 2. For this purpose, it has an objective 3 through which the radiation passes for all microscopy procedures described below.

[0037] The objective 3, via a beam splitter 4, images the sample together with a tube lens 5 onto a two-dimensional spatially resolved area detector 6, which in this example is a CCD detector, but could also be designed as a CMOS detector, for example. In this respect, the microscope 1 has a conventional light microscope module 7, and the beam path from the sample 2 through the objective 3 and the tube lens 5 to the CCD detector 6 corresponds to a conventional wide-field detection beam path 8. The beam splitter 4 is, as indicated by the double arrow in Fig. 1 indicated, interchangeable to allow switching between beam splitters with different dichroic properties or achromatic beam splitters according to US 2008 / 0088920.

[0038] A laser scanning module 9 is also connected to the beam path to objective 3. Its LSM illumination and detection beam paths are coupled into the beam path to objective 3 via a switching mirror 11, which also functions as a beam splitter. The beam path from the switching mirror 11 to objective 3 through the beam splitter 4 is therefore a beam path in which the illumination and detection beam paths are combined. This applies both to the laser scanning module 9 and to the wide-field detection beam path 8, since, as will be explained later, illumination radiation is also coupled into the switching mirror 11. This radiation, together with the wide-field detection beam path 8, i.e., the CCD detector 6, enables microscopy techniques.

[0039] The switching mirror 11 and the beam splitter 4 are combined into a beam splitter module 12, allowing the switching mirror 11 and the beam splitter 4 to be exchanged depending on the application. This is also illustrated by double arrows. Furthermore, an emission filter 13 is provided in the beam splitter module 12, located in the wide-field detection beam path 8, and appropriately filters the spectral components that can propagate through the wide-field detection beam path 8. Naturally, the emission filter 13 in the beam splitter module 12 is also replaceable.

[0040] The laser scanning module 9 receives the laser radiation required for operation via an optical fiber 14 from a laser module 15.

[0041] In the Fig. In the construction shown in Figure 1, a collecting illumination beam path 16 is coupled to the beam splitter module 12, more precisely to the switching mirror 14, through which illumination radiation for various microscopy methods passes. Various illumination beam paths from individual illumination modules are coupled into this collecting illumination beam path 16.

[0042] For example, a wide-field illumination module 17 couples wide-field illumination radiation into the collecting illumination beam path 16 via a switching mirror 18, so that the sample 2 is illuminated in a wide field via a tube lens 27 and the objective 3. The wide-field illumination module 17 can, for example, include an HBO lamp. A further illumination module, a TIRF illumination module 19, is provided, which implements TIRF illumination when the switching mirror 18 is in a suitable position. The TIRF illumination module 19 receives radiation from laser module 15 via an optical fiber 20. The TIRF illumination module 19 has a mirror 21 that is longitudinally displaceable. The longitudinal displacement shifts the illumination beam emitted by the TIRF illumination module 19 perpendicular to the main propagation direction of the emitted illumination beam, resulting in the TIRF illumination incident on the lens 3 at an adjustable angle to the optical axis of the lens.This method easily ensures the necessary angle of total internal reflection at the coverslip. Of course, other methods are also suitable for achieving this angle adjustment.

[0043] Furthermore, the illumination beam path of a manipulator module 22 is coupled to the collecting illumination beam path. This manipulator module also receives radiation from the laser module 15 via an unspecified optical fiber and scans a point- or line-shaped beam distribution over the sample 2. The manipulator module 22 thus essentially corresponds to the illumination module of a laser scanning microscope, and consequently, the manipulator module 22 can also be operated in combination with the detector of the laser scanning module 9 or the wide-field detection of the CCD detector 6.

[0044] A grating 23 is further provided in the collecting illumination beam path 16, the grating constant of which is, for example, below the cutoff frequency that can be transmitted to the sample 2 by the microscope 1. The grating 23 can, for example, be arranged in a plane of the illumination beam path 16 that is imaged onto the sample (intermediate image of the sample). For example, it can be a Ronchi grating. The grating 23 is displaceable transversely to the optical axis of the collecting illumination beam path 16, preferably in two dimensions. A corresponding displacement drive 24 is provided for this purpose.

[0045] Downstream of the grating in the collecting illumination beam path 16, an image field rotator 25 is located, which is rotated by a rotator drive 26. The image field rotator can, for example, be an Abbe-König prism. If the grating 23 is structured two-dimensionally, the image field rotator 25 can be omitted, since the resulting illumination structure does not require rotation. Instead, it can, for example, be translated in two dimensions.

[0046] The microscope 1 includes a control unit 28, for example a computer in Von Neumann architecture, which in particular has a processor as arithmetic and control unit, a random access memory as working memory and a magnetic hard disk as mass storage.

[0047] The modules, drives, and detectors of microscope 1 are all connected to the control unit 28 via unspecified lines. This connection can be made, for example, via a data and control bus. The control unit 28 controls microscope 1 in various operating modes. For example, the control unit 28 allows microscope 1 to perform classical microscopy, i.e., wide-field microscopy (WF), particularly with structured illumination (SIM), laser scanning microscopy (LSM), and also total internal reflection fluorescence microscopy (TIRF).

[0048] The microscope of Fig. The device 1 essentially comprises two modules suitable for laser scanner illumination: the laser scanning module 9 and the manipulator module 22. Of course, other combinations are also possible. These modules are coupled to the objective 3 on the sample 2 via tube lenses. The manipulator module 22 contains only the excitation part of a laser scanning module, without detection. This allows the sample to be illuminated at a single point and the illumination spot to be scanned across the sample 2.

[0049] Preferably, the manipulator module 22 also includes a switching unit, e.g., a switching lens or cylindrical lens, which enables switching between point and line illumination. This line illumination is particularly advantageous when the grating 23 is pivoted and perpendicular to the line of line illumination. Alternatively, the line illumination could be used for the dynamic (sequential) generation of structured illumination in the sample 2.

[0050] In other embodiments (not shown), a variably adjustable fringe modulator, a DMD, an SLM, or optical waveguides terminating in the pupil plane can be used as alternatives to the grating 23 to generate structured illumination in the sample 2. In these cases, the displacement drive 24 and the pivoting capability of the grating 23 are no longer required. In particular, the illumination can be achieved by generating the light structure in the sample through the interference of multiple beams, such that its period is close to the cutoff frequency of the microscope. This can also be achieved if the grating 23 is configured as a diffraction grating.

[0051] The image field rotator 25 allows the structured illumination generated by the grating 23 (or the elements replacing it) to be rotated around the optical axis of the collecting illumination beam path 16, so that the structured illumination lies at different angles in the sample 2.

[0052] To switch between individual operating modes, the switching mirrors 18 and 11, as well as the beam splitter 4, are appropriately adjusted. For this purpose, hinged or swiveling mirrors can be used in the implementation, allowing sequential switching between the operating modes. Alternatively, dichroic mirrors are also possible, which enable simultaneous operation of the different modules.

[0053] The beam splitter 4 is preferably designed as a dichroic beam splitter, the spectral properties of which are adjustable such that spectral components of the fluorescence emission from labeling molecules, which are to be detected by the CCD detector 6, enter the wide-field detection beam path 8, while the remaining spectral components are transmitted as far as possible. To increase flexibility with regard to the usability of labeling molecules with different emission characteristics, several different beam splitters 4 and emission filters 13 are arranged interchangeably in the beam splitter module 12, e.g., on a filter wheel.

[0054] The microscope described above can be used to generate a super-resolution intermediate image. The control unit 28 is appropriately designed for this purpose, for example, through suitable programming to reconstruct the intermediate image from several raw images with structured illumination at different illumination phases. Furthermore, the control unit 28 is configured to determine a cross-sectional image from the raw images, weight the intermediate image based on the cross-sectional image, and output it as the final image.

[0055] Fig. Figure 2 schematically illustrates the well-known concept of generating a super-resolution image in a single sample plane using the SIM method. The image in microscope 1 of the Fig. One microscopically examined sample is repeatedly imaged in a wide field, with different illumination conditions being set.

[0056] Fig. Figure 2 shows a set of raw images gi from a single sample plane, all depicting the same sample area but differing with respect to a light structure 29. This light structure is transferred into sample 2 during the acquisition of the raw images gi by structured illumination via the illumination beam path 16. As can be seen, the lateral, periodic light structure 29, for example, is oriented and positioned differently in the various raw images gi, but exhibits an identical period in all of them. In total, nine raw images gi (from nine different illumination and image acquisition phases) are available as examples, consisting of three different orientations of the structure 29 and three different displacement positions of the structure 29. The different orientations and displacement positions are collectively referred to as phases.Of course, other, and especially higher, numbers of different phases are also possible, as is known from the publications on the SIM principle cited above. In an alternative embodiment, it is possible to perform and / or use in a reconstruction process only those illumination and image acquisition phases in which the light structure 29 has the same orientation in each of the phases (i.e., only assumes different displacement positions), for example, to generate only the raw images g1, g4, and g7 (or to use only these in the reconstruction).

[0057] The structure shown, 29, is purely exemplary. In particular, it need not be a line structure. The schematically drawn lines along the lines can also be further structured. Likewise, instead of the line-like structure used in the SIM publications mentioned at the beginning, it is also possible to use a scanned confocal point or line illumination with confocal detection, as described in the publication "Image scanning microscopy" by C. Müller and J. Enderlein, Physical Review Letters, 104, 198101 (2010). This principle is called ISM. Of course, in this case, there are not nine orientations of structured illumination, but rather a suitable number of raw images obtained from scanning a sample 2. Each raw image gi then corresponds to a specific scan position, i.e., a specific raster state during the scanning of the image.

[0058] The control unit 28 calculates g from the recorded raw images. i for example, a super-resolved intermediate result image f can be obtained using an iterative joint SIM reconstruction procedure including order-selective filtering. r .

[0059] In Fig. Figure 3 is shown as an example of how the control unit 28 generates the raw images g. i next to the interim result image f r a cross-sectional view f os determined and the intermediate result image f r based on the cross-sectional image f os multiplication by the selected normalization factor M weighted by points.

[0060] Finally, it shows Fig. 4 example images, each in the left part in axial view (i.e. transverse section through the sample) and in the right part in lateral view (i.e. in axial section through the sample).

[0061] First, it Fig. 4A an intermediate result image obtained by iterative joint SIM reconstruction without order-selective filtering fr' and Fig. 4B a further intermediate result image f obtained by iterative joint SIM reconstruction with order-selective filtering r The suppression of extrafocal light by order-selective filtering is shown in Fig. 4B is clearly visible. However, artifacts are still visible in the axial section. Fig. 4C represents the cross-sectional image obtained from the same raw images. os in Fig. 4D is the resulting image f̃ in the form of the cross-sectional image f os weighted interim result image fr' The weighting is shown here, for example, according to: f˜=∑fr〈fos,fr〉⋅fr⋅fos

[0062] This means that the overall intensity is maintained in this case: ∑f˜≡∑fr

[0063] In an alternative embodiment (not shown), the resulting image can be composed of a Fig. 4A intermediate result image determined without order-selective filtering fr' to be determined: f˜=∑fr'〈fos,fr'〉⋅fr'⋅fos

[0064] The overall intensity remains the same here as well. Reference symbol list 1 microscope 2 Sample 3 lenses 4 beam splitters 5 Tube lens 6 CCD detector 7 Light microscope module 8 Wide-field detection beam path 9 Laser scanning module 11 shift mirrors 12 Beam splitter module 13 emission filters 14 optical fibers 15 laser modules 16 Collective lighting beam path 17 Wide-field lighting module 18 shift mirrors 18 19 TIRF lighting module 20 optical fibers 21 mirrors 22 Manipulator module 23 grids 24 Displacement drive 25 Image field rotator 26 Rotator drive 27 Tube lens 28 Control unit 29 Lighting structure G l Interim result image f r Interim result image f os Cross-sectional view f̃ Result image QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 199 08 883 A1

[0003] US 2022 / 092752 A1

[0009] US 2008 / 0088920

[0037] Zitierte Nicht-Patentliteratur

[0000] Chakrova et al.: „Deconvolution methods for structured illumination microscopy“ in Journal of the Optical Society of America A, Band 33, Nr. 7, 2016, S. B12

[0005] Neil et al.: „Method of obtaining optical sectioning by using structured light in a conventional microscope“ in Optics Letters Bd. 22, Nr. 24, 1997, S. 1905

[0006] Gustafsson et al.: „Three-Dimensional Resolution Doubling in Wide-Field Fluorescence Microscopy by Structured Illumination“, Biophys. J., Bd. 94, 2008, S. 4957

[0007] Schäfer et al.: „Structured illumination microscopy: artefact analysis and reduction utilizing a parameter optimization approach“ in Journal of Microscopy, Band 216, Nr. 2, S. 165

[0014] Image scanning microscopy“ von C. Müller und J. Enderlein, Physical Review Letters, 104, 198101 (2010

[0057]

Claims

[1] Method for evaluating microscopic images of a sample (2), comprising the steps: - Providing several digital raw images (gi) of the sample (2) taken sequentially using a microscope (1) under illumination of the sample (2) in different phases with periodically structured illumination light (29) and - Determining an intermediate result image (f r ) of the sample (2) with increased resolution using the raw images (g i ) using a reconstruction method, characterized by - Weighting of elements of the intermediate result image (f r ) and - Output of the intermediate result image (f r ) with the weighted elements as the result image (f). [2] Method according to claim 1, wherein a cross-sectional image (f) is taken before weighing. os ) the sample is determined from the raw images using a reconstruction method and the weighting is based on elements of the cross-sectional image (fos ). [3] Method according to claim 2, wherein when weighting one of the elements of the intermediate result image (f r ) the element in question with an element of the cross-sectional image (f os ) is multiplied. [4] Method according to one of the preceding claims, wherein to determine the intermediate result image (f r ) a common reconstruction method, in particular an iterative common reconstruction method, is used. [5] Method according to one of the preceding claims, wherein when weighting one of the elements of the intermediate result image (f r ) the element in question is multiplied by a factor which is determined by elements of the raw images (g i ) and / or elements of the intermediate result image (f r ) and / or elements of the cross-sectional image (f os depends. [6] Method according to claim 5, wherein the factor consists exclusively of elements of the raw images (gi depends. [7] Method according to claim 5, wherein the factor consists exclusively of elements of the intermediate result image (f r depends. [8] Method according to claim 5, wherein the factor consists exclusively of elements of the intermediate result image (f r ) and elements of the cross-sectional image (f os depends. [9] Method according to any one of the preceding claims, wherein the raw images (g i ) before being provided by the microscope (1) sequentially under illumination of the sample (2) in different phases with periodically structured illumination light (29). [10] Microscope (1) with a control unit (28) configured to carry out a method according to one of the preceding claims, and with a light source (15), a two-dimensional spatially resolving detector (6) for recording raw images (g i) the sample (2) and means for generating periodically structured illumination light (29) in the sample (2) in different phases, wherein in particular no aperture optically cutting the sample (2) is arranged in front of the detector (6). [11] Computer program designed to carry out a method according to one of the method claims. [12] Control unit (28) configured to carry out a procedure according to one of the procedure claims.

Images