Method and apparatus for optimized interferometric scattering microscopy

Abstract

A method for imaging a sample by interferometric scattering microscopy, the method comprising: illuminating the sample with at least one light source, the sample being held at a sample position including a reflective surface so as to form a reflection signal, the reflection signal including light from the light source and light scattered by the sample; detecting output light in a first time window of a first frame N1; detecting output light in a second time window of a second frame N2; calculating a ratiometric signal R, R being the ratio of N1 to N2 minus 1; estimating a ratiometric motion feature S=(S) based on frames N1 and N2. x , S y ), where S is defined as a ratiometric image that is obtained by moving a given motion vector m along x and y = (m x , m y ) measured by an invariant sample of ; estimating m as the most consistent vector so that R is approximated using S and m; and calculating a corrected ratio measurement contrast image R* based on R, S and M.

Description

Technical Field

[0001] The present invention relates to methods and apparatus for optimized interferometric scattering microscopy (referred to herein as IScat). Background Art

[0002] IScat has emerged as a powerful method for single-particle tracking with unique spatiotemporal resolution and label-free sensitivity down to the single-molecule level, and can also be used to determine single-particle masses by mass photometry and mass imaging.

[0003] IScat is disclosed, for example, in Kukura et al., “High-speed nanoscopic tracking of the position and orientation of a single virus,” Nature Methods, 2009, 6: 923-935, and in Ortega Arroyo et al., “Interferometric scattering microscopy (IScat): new frontiers in ultrafast and ultrasensitive optical microscopy,” Physical Chemistry Chemical Physics, 2012, 14: 15625-15636.

[0004] Despite its considerable potential, the widespread application of IScat is limited by the requirements for custom microscopes, unconventional cameras, and complex sample illumination, thereby restricting IScat's ability to robustly and accurately detect, image, and characterize objects as small as single molecules.

[0005] In our previous patent application WO2018 / 011591, we disclosed an interferometric scattering microscope that includes a novel, contrast-enhancing spatial mask configured to increase the relative amplitudes of the reference and scattered light fields. The microscope described therein achieves sensitivity similar to conventional IScat technology, but with significantly reduced implementation complexity and expense, enabling conventional microscopes to be configured to function as IScat through simple modification and the inclusion of a spatial mask.

[0006] However, a number of limitations on the measurement sensitivity achievable using this approach have become apparent to the inventors of the present application.

[0007] In iScat microscopy, the ratio between adjacent frames is often used to detect the signal generated by particles bound to the boundary. Therefore, a ratiometric contrast image is calculated, which shows the relative change in pixel intensity between two frames before and after a given time point.

[0008] However, this can cause problems if the sample moves between frames. Because sample motion is also associated with changes in pixel intensity, these changes are also detected in the ratiometric image and cause problems in detecting particles and determining their mass from the data. In fact, the high sensitivity of this technique means that even when the movement distance is less than one pixel, the signal from the particle can be overwhelmed by the movement of the sample carrier. Movements as small as 1% of the pixel size can produce a signal with the same amplitude as a single protein. Therefore, even very small movements can produce highly erroneous signals.

[0009] Current methods for correcting subpixel motion include iterative image registration, Fourier transforms, and optical flow. These methods cannot be directly applied to ratiometric images and are not accurate enough to perform in real time at kilohertz (kHz) frame rates.

[0010] The present invention provides methods and apparatus for optimizing IScat technology to correct for sample motion. The method disclosed herein addresses this problem by detecting motion and subtracting it from the signal. Summary of the Invention

[0011] According to one aspect of the present invention, a method of imaging a sample by interferometric scattering microscopy is provided.

[0012] The method includes a method for imaging a sample by an interferometric scattering microscope, the method comprising: illuminating the sample with at least one light source, the sample being held at a sample position including a reflective surface so as to form a reflection signal, the reflection signal including light from the light source and light scattered by the sample; detecting output light in a first time window of a first frame N1; detecting output light in a second time window of a second frame N2; calculating a ratio measurement signal R, the ratio measurement signal R being a ratio of N1 to N2 minus 1; estimating a ratio measurement motion feature S=(S) based on the local camera frames N1 and N2. x , S y ), S is a ratio measurement image, which is a ratio measurement image obtained by moving a given motion vector m along x and y = (m x , m y) measured by an invariant sample of ; estimating m as the most consistent vector so that R is approximated using S and m; and calculating a corrected ratio measurement contrast image R based on R, S and M.

[0013] By removing motion signatures from the ratiometric signal in the manner described above, rather than by conventional image alignment, real-time processing of the images at kHz rates can still be achieved.

[0014] According to the local image N, the horizontal ratio measurement motion feature S at the pixel position (x, y) is estimated x and vertical ratio measurement motion characteristics S y ,as follows:

[0015] S x [x,y]=(N[x+1,y]-N[x-1,y]) / (2N[x,y]),

[0016] S y [x,y]=(N[x,y+1]-N[x,y-1]) / (2N[x,y]).

[0017] m is estimated as the most consistent vector such that R=m x S x +m y S y .

[0018] The corrected ratiometric contrast image R* is calculated as:

[0019] R*=R-(m x S x +m y S y ).

[0020] Based on the calculated motion vector m, high-precision actuators can be used to compensate for motion, drift, or vibration to move or stabilize the sample.

[0021] Additionally, or alternatively, if the detected motion is above a predetermined level, an alarm is triggered which warns the user that the detected image may be unreliable.

[0022] The method according to claim 1, further comprising moving the sample to compensate for the motion.

[0023] The first frame may include an average of a first plurality of local frames, and the second frame may include an average of a second plurality of local frames.

[0024] The method further includes passing at least one reflected signal through a spatial filter configured to achieve a reduction in the intensity of the incident radiation, wherein the reduction in intensity is greater within a predetermined numerical aperture. Using the spatial filter eliminates most reflected light. However, because scattered light is typically present at higher apertures, it cannot be eliminated by the spatial filter. Thus, using the spatial filter achieves an image with greater contrast.

[0025] The light source may be a coherent light source.

[0026] According to the present invention, an interferometric scattering microscope is also provided, comprising: a sample holder for holding a sample at a sample position; an illumination source configured to provide illumination light; a detector; an optical system arranged to guide the illumination light to the sample position, and also arranged to collect reflected output light, the output light comprising light scattered from the sample position and illumination light reflected from the sample position, and guiding the output light to the detector; a spatial filter configured to filter the output light, the spatial filter being arranged to allow the output light to pass through but with a greater intensity reduction within a predetermined numerical aperture than at a larger numerical aperture; and a computer program device configured to instruct the apparatus to perform the steps as described above.

[0027] An interferometric scattering microscope as described above further comprises an actuator to stabilize the sample in response to the estimated motion vector m. The actuator may be a motorized actuator.

[0028] The present invention can be advantageously applied to samples comprising objects having a scattering cross section of 10 -15 Square meters (m 2 ) or smaller. Typically, the scattering cross section of such an object relative to the irradiated light can also be 10 -26 m 2 or greater, i.e., in 10 -15 m 2 to 10 -26 m 2 Examples of objects that can be studied include proteins or their small aggregates and metallic, organic or inorganic nanoparticles.

[0029] To image objects that are very weak scatterers, the spatial filter is arranged so that the intensity is reduced to 10% of the incident intensity within a predetermined numerical aperture. -2 Typically, the spatial filter can be arranged so that the intensity is reduced to 10% of the incident intensity within a predetermined numerical aperture. -4 or higher intensity (e.g. between 10 and 10 times the incident intensity) -2 to 10 -4 The output light within the range) passes through. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] To facilitate a better understanding, embodiments of the present invention will now be described by way of non-limiting examples with reference to the accompanying drawings, in which:

[0031] FIG1 is a schematic diagram of an IScat microscope of the prior art;

[0032] FIG2 is a schematic diagram of an improved IScat microscope of the prior art;

[0033] Figure 3 shows an image from an IScat microscope with no motion correction applied; and

[0034] Figure 4 An image of an IScat microscope to which motion correction according to the present invention has been applied is shown. DETAILED DESCRIPTION

[0035] In the systems and methods described herein, the light used can be ultraviolet light (defined herein as having a wavelength in the range of 10 nanometers (nm) to 380 nm), visible light (defined herein as having a wavelength in the range of 380 nm to 740 nm), or infrared light (defined herein as having a wavelength in the range of 740 nm to 300 micrometers (μm). The light can be a mixture of multiple wavelengths. The terms "optical" and "optical" are generally used herein to refer to the light to which these methods are applied.

[0036] Figures 1 and 2 show the IScat microscope structure disclosed in WO2018 / 011591, which has many of the same structural features and functions as the device and method of the present invention.

[0037] The disclosure of WO 2018 / 011591 is incorporated herein by reference; however, for the sake of completeness, the following description will list the components and functions of the IScat microscope that are the same as those in WO 2018 / 011591 and as shown in Figures 1 and 2, and then describe the various improvements to the structure provided by the present disclosure and provide exemplary embodiments thereof.

[0038] Therefore, with reference to Figure 1, microscope 1 includes a sample holder 2 for holding sample 3 at the sample position. Sample 3 can be a liquid sample, including an object to be imaged, which will be described in more detail below. Sample holder 2 can take any form suitable for holding sample 3. Typically, sample holder 2 holds sample 3 on a surface that forms an interface between sample holder 2 and sample 3. For example, sample holder 2 can be a coverslip and / or can be made of glass. Sample 3 can be provided on sample holder 2 in a direct manner (e.g., using a micropipette).

[0039] Microscope 1 also includes an illumination source 4 and a detector 5. Illumination source 4 is configured to provide illumination light. The illumination light may be coherent light. For example, illumination source 4 may be a laser. The wavelength of the illumination light may be selected based on the properties of sample 3 and / or the characteristics to be examined. In one example, the illumination light has a wavelength of 405 nm.

[0040] Alternatively, the illumination light can be spatially modulated to eliminate speckle patterns caused by the coherent nature of the illumination and laser noise, for example, as described in detail in Kukura et al., "High-speed nanotracking of the position and orientation of single viruses," Nature Methods 2009 6: 923-935.

[0041] The output light reflected from the sample is received by the detector 5. The illumination light reaching the detector is mainly reflected from the surface of the sample (usually the interface between the sample and the sample holder), thereby causing interference to objects in the sample close to the surface.

[0042] In the example using a glass-water interface, a relatively small amount (typically only 0.5%) of the illumination light is reflected, while a significantly higher amount (typically greater than 90%) of the light scattered by nanoscale objects at the interface is scattered back in the illumination direction. Compared to a transmissive geometry, this essentially increases the ratio between the scattered and reflected light fields by more than a factor of 1000, resulting in greater interference contrast. Therefore, given a specific scatterer, illumination intensity, and exposure time, achieving the same nominal signal-to-noise ratio requires three orders of magnitude fewer detected photons than with a transmissive setup.

[0043] Typically, the microscope 1 can be operated in a widefield mode, in which case the detector 5 can be an image sensor that captures an image of the sample 3. Alternatively, the microscope 1 can be operated in a confocal mode, in which case the detector 5 can be an image sensor or a point detector, such as a photodiode, in which case a scanning device can be used to scan an area of ​​the sample 3 to create an image. Examples of image sensors that can be used as the detector 5 include a complementary metal-oxide semiconductor (CMOS) image sensor or a charge-coupled device (CCD).

[0044] The microscope 1 further comprises an optical system 10 arranged between the sample holder 2, the illumination source 4 and the detector 5. The optical system 10 is arranged as described below to direct illumination light to the sample position for illuminating the sample 3, collect output light reflected from the sample position and direct the output light to the detector 5.

[0045] The optical system 10 includes an objective lens 11, which is a lens system disposed in front of the sample holder 2. The optical system 10 also includes a condenser lens 12 and a tube lens 13. The condenser lens 12 condenses the irradiation light from the light source 4 (indicated by a solid line in FIG. 1 ) onto the sample 3 at the sample position through the objective lens 11.

[0046] Objective lens 11 collects output light, which includes: (a) illumination light reflected from the sample position (shown by a solid line in FIG1 ) and (b) light scattered from sample 3 at the sample position (shown by a dotted line in FIG1 ). The reflected light is mainly reflected from the interface between sample holder 2 and sample 3.

[0047] Typically, this is a relatively weak reflection, such as a glass-water reflection. For example, the intensity of the reflected illumination light may be 0.5% of the intensity of the incident illumination light. The scattered light is scattered by objects in the sample 3. The scattered light from objects at or near the sample surface constructively interferes with the reflected light and is therefore visible in the image captured by the detector 5.

[0048] As shown in Figure 1, the reflected illumination light and the scattered light have different directions. In particular, the reflected illumination light has a numerical aperture resulting from the geometry of the light beam output by the light source 4 and the optical system. The scattered light scatters over a wide range of angles and therefore fills a larger numerical aperture than the reflected illumination light. Tube lens 13 focuses the output light from objective lens 11 onto detector 5.

[0049] The optical system 10 further comprises a beam splitter 14 arranged to split the optical paths of the illumination light from the light source 4 and the output light directed to the detector 5. The beam splitter 14 may have a conventional structure that provides partial reflection and partial transmission of light incident thereon.

[0050] In the example of the present disclosure, the light source 4 deviates from the optical path of the objective lens 11, so that the illumination light from the light source 4 is reflected by the beam splitter 14 to the objective lens 11. Conversely, the detector 5 is aligned with the optical path of the objective lens 11, so that the output light from the sample position is transmitted through the beam splitter 14 to the detector 5.

[0051] In addition to the components described above, which may have conventional structures, the microscope 1 also includes a spatial mask or filter 20. In the example of FIG1 , the spatial filter 20 is formed on the beam splitter 14 and is therefore located behind the rear aperture of the objective 11 and, in turn, directly behind the rear focal plane 15 of the objective 11. However, the spatial filter 20 may also be placed at other points along the optical path of the IScat microscope to achieve the same effect as described below.

[0052] The spatial filter 20 is arranged to filter the back-propagating output light from the sample holder interface to the detector 5. Thus, in the disclosed example where the detector 5 is aligned with the optical path of the objective 11, the spatial filter 20 is transmissive.

[0053] Spatial filter 20 can be partially transmissive and thus allow output light to pass, including the reflected illumination light, but at a reduced intensity. Spatial filter 20 is also aligned with the optical axis and has a predetermined aperture such that it provides intensity reduction within a predetermined numerical aperture. Numerical aperture is defined in its conventional manner as a dimensionless quantity that characterizes the angular range relative to the sample position from which the output light originates.

[0054] Specifically, the numerical aperture NA may be defined by the following equation: NA=n·sin(θ), where θ is the collection half-angle and n is the refractive index of the material (eg, the material of the components of the optical system 10 ) through which the output light passes.

[0055] The spatial filter 20 can be formed in any suitable manner, typically comprising layers of deposited material. The material can be, for example, a metal such as silver. In some embodiments, the spatial filter can include one or more dielectric coatings. In some embodiments, the spatial filter can be formed to partially reflect incident radiation within a given angular range. Deposition can be performed using any suitable technique.

[0056] Because sub-diffraction-sized objects near the interface preferentially scatter light into a larger numerical aperture than the reflected illumination light, the intensity reduction provided by the spatial filter 20 preferentially reduces the detected intensity of the reflected illumination light compared to the scattered light. Thus, at low numerical apertures, the intensity reduction provided by the spatial filter 20 primarily affects the reflected illumination light and minimizes the effect on the scattered light, thereby maximizing contrast in the captured image. The enhanced imaging contrast enables high-contrast detection of objects that are weak scatterers.

[0057] Contrast enhancement can be understood with reference to the following. When the spatial filter 20 passes a portion of the output light within a predetermined numerical aperture (i.e., partially transmitted in this example), a portion of the illumination light field and a portion of the scattered light field reach the detector and interfere to obtain a sufficiently coherent illumination source. det The light intensity is given by the formula:

[0058] I det =|E inc | 2 {r 2 t 2 +|s| 2 +2rt|s|cosΦ},

[0059] Among them, Einc is the incident light field, r 2 is the reflectivity of the interface, t 2 is the transmittance of the spatial filter 20, s is the scattering amplitude of the object, and Φ is the phase difference between the transmitted irradiation light and the scattered light.

[0060] The additional filtering provided by the spatial filter 20 enables direct selection of the transmittance t of the spatial filter 20. 2 The amplitude of the reference field is tuned rather than being fixed by the reflectivity of the glass-water interface as in standard IScat. In the case where the spatial filter 20 is a layer of deposited material, the transmittance t 2 The selection can be made by the choice of the material and / or thickness of the layer. For example, such tuning can be performed depending on the scattering object of interest, the camera full well capacity and the magnification.

[0061] Brightfield illumination ensures that the strongest unwanted back reflections (typically from the objective lens) are directed away from the detector 5, minimizing the imaging background and enabling a large field of view without the need for complex scanning of the illumination beam.

[0062] In order to image an object with relatively weak scatterers, the spatial filter 20 may be arranged to reduce the intensity of the reflected illumination light within a predetermined numerical aperture to 10 times the incident intensity (herein, the intensity of the output light incident on the spatial filter 20). -2 to 10 -4 Strength within the range.

[0063] For example, a sample comprising objects having a mass of 5000 kDa or less can be imaged. In general, the disclosed techniques are applicable to samples comprising objects having a mass of 10 kDa or greater, such as objects having a mass in the range of 10 kDa to 5000 kDa, and / or to samples comprising objects having a scattering cross section of 10 with respect to the illuminating light. -12 m 2 or less, or more preferably 10 -17 m 2 Typically, the scattering cross section of such objects relative to the irradiation light can also be, for example, in the range of 10 -17 m 2 to 10 -26 m 2 Examples of objects to which the disclosed technology can be applied for imaging include proteins or small aggregates thereof, or their binding partners.

[0064] To image stronger scatterers simultaneously, the transmittance of the second filter can be set to 1 to 10 -2 Any value in between, depending on the desired detection range.

[0065] Referring to Figure 2, Figure 2 shows a second exemplary structure of the microscope 1. The structure of Figure 2 is also disclosed in WO 2018 / 011591 and is also suitable for optimization by applying the technology of the present invention.

[0066] 2 positions the spatial filter 20 at a conjugate focal plane 21 of the rear focal plane of the objective lens 11, rather than behind the rear aperture of the objective lens 11. The conjugate focal plane 21 of the rear focal plane 15 of the objective lens 11 is formed between a pair of telescope lenses 22 and 23 located behind the tube lens 13.

[0067] An acousto-optic deflector 32 is arranged behind the light source 4 to provide scanning of the illumination light. The acousto-optic deflector 32 can be operated to scan an area of ​​the sample 3 to create an image and / or provide spatial modulation for eliminating speckle patterns caused by the coherent nature of the illumination and laser noise as described above.

[0068] The condenser lens 12 is replaced by a pair of telecentric lenses 30 and 31 which perform the function of imaging any modifications to the beam path at the acousto-optic deflector 32 to the back focal plane of the imaging objective.

[0069] The positions of the light source 4 and the detector 5 are opposite to those of the structure of FIG. 1 , so that the illumination light from the light source 4 is transmitted to the objective lens 11 through the beam splitter 14 , and conversely, the output light from the sample position is reflected by the beam splitter 14 to the detector 5 .

[0070] The beam splitter 14 is a polarization beam splitter, and a quarter wave plate 33 is provided between the beam splitter 14 and the sample 3 so that the beam splitter 14 splits the light. The mirror 34 is arranged to deflect the output light reflected by the beam splitter 14.

[0071] The ratiometric frame R is calculated from two native camera frames N1 and N2 such that for each pixel at position (x, y), the ratiometric signal is calculated as:

[0072] R[x,y]=N2[x,y] / N1[x,y]-1,

[0073] However, if the sample moves between frames N1 and N2, there are contributions from neighboring pixels. For example, with sub-pixel movement, a pixel at position (x, y) may have contributions from any of its neighboring pixels, including (x+1, y), (x-1, y), (x, y+1), and (x, y-1).

[0074] Considering only the motion in the x direction, when obtaining N2 based on the pixel N1[x+1, y] at frame N1, there may be a motion contribution m to the pixel N2[x, y] x. First, N2[x, y] can be expressed as:

[0075] N2[x,y]=m x N1[x+1,y]+(1-m x )N1[x,y],

[0076] The contribution to the ratio measurement image is:

[0077]

[0078] Therefore, if the ratio of neighboring pixels can be calculated, the motion contribution can be calculated as the best fit m for all pixels in the frame. x value.

[0079] Similarly, for motion in the y direction, when obtaining N2 from pixel N1[x,y+1] at frame N1, there may be a motion contribution m to pixel N2[x,y]. y This will contribute to the ratiometric image as follows:

[0080]

[0081] Check the motion vector m between the two-dimensional frames = (m x , m y ), m is the most consistent vector estimated based on R and S, such that

[0082] R=m x S x +m y S y

[0083] The horizontal ratio at the pixel index (x, y) measures the motion feature S x and vertical ratio measurement motion characteristics S y N can be estimated from the native image as:

[0084] S x [x,y]=(N[x-1,y]-N[x+1,y]) / (2N[x,y]),

[0085] S y [x,y]=(N[x,y-1]-N[x,y+1]) / (2N[x,y]).

[0086] Then, the motion-corrected ratio image R* can be calculated as:

[0087] R*=R-(m x S x +m y S y ).

[0088] Although this uses linear interpolation between pixels, more advanced interpolation (such as Lanczos resampling) can also be used to generate finer motion estimates. In addition, by calculating the gradient of pixels across frames, the method can be applied to the motion of multiple pixels.

[0089] Other applications of this method include detecting motion parallel to the viewing direction. This will be detected as nonlinear motion (rather than linear motion from translation as described above). Blurs in the image need to be calibrated using the optical properties of the microscope.

[0090] Although the description uses separate local frames N1 and N2, average frames may be used instead of these local frames.

[0091] The IScat microscope may include an alarm or alert, wherein if the detected motion is above a predetermined value, an alarm signal or alert signal is sounded. The alarm may be an audio alarm or a visual alarm.

[0092] The IScat microscope may include an actuator for correcting or compensating for motion, drift, or vibration. The actuator may be a motorized actuator capable of moving the sample in three orthogonal directions.

[0093] Figure 3 and Figure 4 Shown are images obtained under an IScat microscope from a sample with three different molecules: a monomer, a dimer, and a trimer.

[0094] Figure 3 Depicted are (ratiometric) frames captured on an iScat microscope without motion suppression according to the present invention. Due to motion, individual particles cannot be distinguished. Motion is detected by the program, as indicated by the exclamation mark in the upper right corner of the application window. In the lower left panel, the amplitude of the motion is plotted against time. Figure 3 The right side of the image shows a histogram of the detected particle masses falling on the observation slide. There are no distinct peaks for the three different types of molecules because the noise caused by motion interferes with the molecule selection and identification process.

[0095] Figure 4 The same (ratiometric) frames captured on an IScat microscope are depicted, but this time the motion has been corrected according to the present invention. The two particles, seen as black circles, can be clearly seen, and the motion amplitude map in the lower left panel shows values ​​well below Figure 3 In the histogram calculated from the ratio measurement frames, three distinct peaks can be seen, corresponding to the masses of the three types of particles in the sample.

[0096] Microscope 1 can be used to perform IScat technology for a wide range of applications, including single-molecule detection. A particular application is label-free imaging of weak scatterers, where the object of interest must be consistently detected against a large background, which reduces image contrast. Microscope 1 can be used for a wide range of studies and measurements, such as measuring any change in refractive index, including single-molecule binding / unbinding, phase transitions, clustering, assembly / disassembly, aggregation, protein / protein interactions, protein / small molecule interactions, and high-sensitivity label-free imaging.

[0097] Therefore, microscope 1 has numerous applications, ranging from basic research to industrial applications, such as in the pharmaceutical industry. For example, IScat is currently the world's most sensitive label-free single-molecule imaging biosensor, which could have a significant impact on the surface plasmon resonance sensing market. Furthermore, as described above, microscope 1 can be used for mass measurement, serving as a precise, accurate, and high-resolution single-molecule mass spectrometer in solution, with numerous applications in research and industry.

Claims

1. A method for imaging a sample by interferometric scattering microscopy, the method comprising: illuminating a sample using at least one light source, the sample being held at a sample location comprising a reflective surface such that an output light is formed, the output light comprising illumination light reflected from the sample location and light scattered from the sample location; detecting the output light in a first time window of a first frame N1; detecting the output light in a second time window of a second frame N2; Calculate the ratio measurement signal R, which is the ratio of N1 to N2 minus 1; The motion feature S is measured by estimating the ratio of the first frame N1 and the second frame N2 = (S x , S y ), S is a ratio measurement image, which is a ratio measurement image obtained by moving a given motion vector m=(m along x and y. x , m y ) measured by an unchanging sample; Estimate m as the most consistent vector so that R is approximated using S and m; and Calculate the corrected ratio measurement contrast image R* based on R, S and M; Wherein, the corrected ratio measurement contrast image R* is calculated as: R* = R –(m x S x + m y S y )。 2. The method according to claim 1, wherein Based on the native image N, estimate the horizontal ratio measurement motion feature S at the pixel position (x, y) x and vertical ratio measurement motion characteristics S y ,as follows: S x [x, y] =(N[x+1, y] - N[x-1, y]) / (2 N[x, y]), S y [x, y] =(N[x, y+1] - N[x, y-1]) / (2 N[x, y])。 3. The method according to claim 1 or 2, wherein: m is estimated as the most consistent vector such that: R = m x S x + m y S y 。 4. The method according to claim 1 or 2, further comprising: The sample was moved to compensate for motion.

5. The method according to claim 1 or 2, wherein: If the detected motion is above a predetermined level, an alarm is triggered.

6. The method according to claim 1 or 2, further comprising: The output light is passed through a spatial filter configured to achieve an intensity reduction of incident radiation, the intensity reduction being greater within a predetermined numerical aperture.

7. The method according to claim 1 or 2, wherein: The light source is a coherent light source.

8. An interference scattering microscope, comprising: a sample holder for holding the sample at a sample position; an illumination source arranged to provide illumination light; detector; an optical system arranged to direct the illumination light to the sample location and further arranged to collect reflected output light, the output light comprising light scattered from the sample location and illumination light reflected from the sample location, and direct the output light to the detector; a spatial filter configured to filter the output light, the spatial filter being arranged to pass the output light but with a greater intensity reduction within a predetermined numerical aperture than at numerical apertures greater than the predetermined numerical aperture; and Computer program device configured to instruct an apparatus to perform a method according to any one of the preceding claims.

9. The interferometric scattering microscope of claim 8, further comprising an actuator for stabilizing the sample in response to the estimated motion vector m.

Images