Method and system for compensating for substrate thickness errors

Abstract

A system and method for imaging a sample on a top surface of a sample coverslip that uses an objective lens positioned beneath the sample coverslip to capture a sample image of the sample on the top surface of the sample coverslip, capture reference image data obtained from light reflected from the top surface of a calibration coverslip, capture test image data obtained from light reflected from the top surface of the sample coverslip, process the reference image data and the test image data to generate a calculated point spread function associated with the objective lens and the coverslip in use, and deconvolve the sample image with the calculated point spread function, thereby reducing artifacts from the sample image.

Description

[Background technology]

[0001] (CROSS-REFERENCE TO RELATED APPLICATIONS) This application claims priority to and benefit of U.S. Provisional Application No. 63 / 134,033, filed January 5, 2021, which was filed as a PCT International Patent Application on January 5, 2022, the disclosure of which is incorporated herein by reference in its entirety.

[0002] Technicians often use optical microscopy imaging systems during high-content screening (HCS) to capture image quality of microscopy samples. A sample holder, such as a microtiter plate, slide, dish, etc., can support the microscopy sample during the screening process. Automated microscopy imaging systems can include an objective lens coupled to an electronic imaging device, such as a charge-coupled device (CCD) or a complementary metal-oxide semiconductor (CMOS) chip, to generate an image of the microscopy sample. The position of the objective lens relative to the sample holder can be adjusted to bring the microscopy sample into focus on the imaging device.

[0003] To improve imaging efficiency, multiple imaging devices can be used to image multiple wells in parallel (i.e., simultaneously). However, the time required to focus the objective lenses of each of the multiple imaging devices can eliminate any efficiency that could be gained from parallel imaging. Furthermore, focusing each objective lens individually can also increase the complexity of the imaging system. In various imaging configurations, such as inverted microscopes or epifluorescence microscopes, samples are viewed through glass coverslips, glass slides, and the bottom plate of the sample holder (e.g., Petri dishes, microtiter plates). However, variations in the thickness and / or curvature of the sample holder base can prevent accurate focus over a range of measurement locations. As a result, the focal position of the objective lens needs to be corrected at each measurement location with thickness and / or curvature variations to obtain individual in-focus images for all measurement locations. Because high-content screening may involve imaging hundreds or thousands of measurement samples, some microscopy imaging systems can be configured to automatically maintain focus at each measurement location.

[0004] Even with autofocus equipment, other errors in the optical system can prevent accurate focus over a range of measurement locations. Generally, high-magnification objective lenses are designed to obtain clear images when observing fixed samples using a cover glass piece whose thickness and refractive index are predetermined. Observing using a cover glass piece whose thickness and refractive index may deviate from the standard introduces aberrations that can obscure clear images. Lenses with larger numerical apertures can visualize finer details than lenses with smaller numerical apertures. Furthermore, lenses with larger numerical apertures collect more light and generally provide brighter images, but at the expense of a shallower depth of field. However, the larger the numerical aperture of the objective lens, the more noticeable distortions (such as spherical aberrations) in the image will be.

[0005] Typically, commercially available objectives will be designed for a specific thickness of such a sample holder or cover. When a sample holder with a bottom plate of a different thickness is used, deviations from the specified thickness can cause significant degradation of image quality due to spherical aberration introduced by the sample holder. A correction collar may be provided that allows compensation for spherical aberration. In common laboratory practice, a technician uses a microscope to adjust the spherical aberration correction setting by (1) manually rotating the collar while observing a live image on a computer screen and / or (2) using a scale located on the microscope objective to set the collar to a known sample holder thickness.

[0006] Therefore, in some objective imaging systems, a portion of the lens system that constitutes the objective can be moved relative to the optical axis. Such objectives are known in the art as objectives with correction collars. When using a correction collar, a clear image can be obtained despite variations or deviations in the thickness and / or curvature of the sample holder. However, correcting aberrations using a correction collar is not simple, and usually only a skilled technician can find the position where the image is clearest. Summary of the Invention [Means for solving the problem]

[0007] To address, in whole or in part, some of the problems generally described herein and / or other problems that may be observed by one of ordinary skill in the art, the present disclosure provides methods, processes, systems, apparatus, instruments, and / or devices as illustrated by way of example in the implementations set forth below.

[0008] According to one embodiment, an optical imaging system is provided, comprising: a sample stage configured to hold a sample to be imaged on the top surface of a sample coverslip; an objective lens positioned directly below the sample stage and configured to image the sample on the top surface of the sample coverslip; an optical detector configured to capture a sample image of at least the sample on the sample coverslip; and a processor programmed to: derive reference image data from a reference image captured from a calibration coverslip; derive test image data from a test image captured from light reflected from the top surface of the sample coverslip at the focal plane of the objective lens; process the reference image data and the test image data; generate a calculated point spread function associated with the objective lens and other optical components in use (via deconvolution of the test image data); and deconvolve the sample image using the calculated point spread function, thereby reducing artifacts from the sample image.

[0009] According to another embodiment, an optical imaging system includes an objective lens, a sample stage configured to position a top surface of a) a sample coverslip for holding a sample or b) a calibration coverslip at a focal plane of the objective lens, and an optical detector configured to capture at least a) a sample image of the sample on the sample coverslip, b) a reference image from light reflected back through the calibration coverslip from the focal plane of the objective lens, and c) a test image from light reflected back through the sample coverslip from the focal plane of the objective lens, and to derive reference image data from the reference image and test image data from the test image. An optical imaging system is provided, comprising: a processor programmed to: extract, deconvolve inspection image data using reference image data as an initial point spread function for the objective lens and cover slip; generate a deconvolved test image through at least one deconvolution of the inspection image data that is associated with calculated point spread functions for the objective lens and other optical components in use, including the sample cover slip; and deconvolve a sample image using the point spread function calculated from the deconvolution of the inspection image data, thereby reducing artifacts from the sample image.

[0010] According to another embodiment, a computerized method for imaging a sample is provided, comprising the steps of capturing a sample image of the sample using an objective lens positioned directly below a sample coverslip that holds the sample; acquiring reference image data obtained from a calibration coverslip; capturing test image data obtained from light reflected from the top surface of the sample coverslip at the focal plane of the objective lens; processing the reference image data and the test image data to generate a calculated point spread function associated with the objective lens and other optical components in use (via deconvolution of the test image data); and deconvolving the sample image with the calculated point spread function, thereby reducing artifacts from the sample image.

[0011] Other devices, apparatus, systems, methods, features, and advantages of the present technology will be, or become, apparent to one with skill in the art upon examination of the following figures and detailed description, and it is intended that all such additional systems, methods, features, and advantages be included within this description, be within the scope of the present technology, and be protected by the accompanying claims. The present specification also provides, for example, the following: (Item 1) 1. An optical imaging system, comprising: a sample stage configured to hold a sample to be imaged on the top surface of the sample coverslip; an objective lens positioned beneath the sample stage and configured to image the sample on the top surface of the sample coverslip; an optical detector configured to capture a sample image of at least the sample on the sample coverslip; A processor, the processor comprising: deriving reference image data from a reference image captured from a calibration coverslip; deriving test image data from a test image captured from light reflected from a top surface of the sample coverslip at a focal plane of the objective lens; processing the reference image data and the test image data to generate, via deconvolution of the test image data, a calculated point spread function associated with the objective lens and other optical components in use; deconvolving the sample image with the calculated point spread function, thereby reducing artifacts from the sample image; and a processor programmed to An optical imaging system comprising: (Item 2) Item 10. The system of item 1, wherein the processor is programmed to perform blind deconvolution of the inspection image data to determine the calculated point spread function. (Item 3) The processor: 3. The system of claim 1, further comprising: a processor configured to: determine an optimized point spread function for improving image quality of the sample image by repeatedly deconvolving the test image and repeatedly comparing a) the deconvolved image of the test image with b) the reference image until an incremental improvement in the deconvolved image is achieved with respect to the calculated point spread function. (Item 4) Item 4. The system of item 3, wherein the processor is programmed to determine the optimized point spread function based on at least one criterion of spot size, spot shape, spot intensity, and spot location obtained with only incremental refinement after successive deconvolutions of the inspection image. (Item 5) Item 4. The system of item 3, wherein the processor is programmed to determine the optimized point spread function by repeated deconvolution of the inspection image and repeated comparison of a) the deconvolved image of the inspection image with b) a calibrated reference image until improvement in successive deconvolved images is progressively improved. (Item 6) 6. The system of any of items 1-5, further comprising a light source, the light source providing a projection point of light to be focused onto the focal plane for the test image or for the reference image. (Item 7) Item 7. The system of item 6, further comprising a controller configured to control at least one of a) vertical displacement of the objective lens relative to the coverslip, b) lateral displacement of the objective lens relative to the coverslip, c) exposure duration of the optical detector, d) intensity of the light source illuminating the sample, e) insertion of an optical filter into the optical path, f) positioning of an off-axis aperture in the optical path, and g) autofocus adjustment. (Item 8) 8. The system of claim 7, wherein the controller is configured to control at least one of a position of the objective lens and a position of the sample stage to focus light from the light source onto the focal plane of the objective lens. (Item 9) Under control of the controller, the optical detector captures a plurality of sample images from the sample coverslip at different lateral positions; Under control of the controller, the optical detector captures a plurality of calibrated reference images from the calibration coverslip at different lateral positions; Under control of the controller, the optical detector captures a plurality of test images from light reflected back through the sample coverslip from the focal plane of the objective lens at the different lateral positions; the processor generating, for each lateral position, a separate deconvolved inspection image associated with a separate calculated point spread function; the processor deconvolving the plurality of sample images with the individual calculated point spread functions for each lateral position, thereby reducing artifacts from the plurality of sample images. Item 9. The system according to item 8. (Item 10) Under control of the controller, the optical detector captures a plurality of sample images from different sample coverslips; Under control of the controller, the optical detector captures a plurality of calibrated reference images from a plurality of calibration coverslips, each associated with a different sample coverslip; Under control of the controller, the optical detector captures a plurality of test images from light reflected back from and through the different sample cover slips; the processor generates, for each sample coverslip, a separate deconvolved test image associated with a separate calculated point spread function; the processor deconvolves the plurality of sample images with the individual calculated point spread function for each coverslip, thereby reducing artifacts from the plurality of sample images. Item 9. The system according to item 8. (Item 11) 11. The system of any of items 1-10, further comprising a correction collar to compensate for optical aberrations caused by the sample coverslip. (Item 12) 12. The system of any of items 1-11, wherein the processor is configured to store in memory individual optimized point spread functions for a plurality of coverslips having individual standard thicknesses. (Item 13) 13. The system of any of items 1-12, wherein the processor stores in memory individual optimized point spread functions associated with different types of cover slips having different optical thicknesses. (Item 14) The processor further comprises: The position z displaced from the sample coverslip with respect to the sample image 1 Retrieving the captured image from said position z 1 and determining the spherical aberration associated with Each of them is at the position z 1 deriving a set of selectable point spread functions having different spherical aberrations associated with the point spread functions; selecting a starting point spread function associated with the identified spherical aberration from the set of selectable point spread functions; Using the origin spread function, the position z 1 deconvolving the sample image in 14. The system of any of items 1-13, configured to: (Item 15) The processor further comprises: The sample is displaced from the cover slip to a first position z 1 retrieving the first sample image taken in First position z 1 determining a first calculated point spread function for reducing artifacts from the first sample image; The first position z 1 The sample was removed farther from the coverslip than the second position z 2 retrieving a second image taken at Using the first calculated point spread function as the starting point spread function in the deconvolution, 2 determining a second calculated point spread function for reducing artifacts from the second sample image taken in 15. The system of any of items 1-14, configured to: (Item 16) 16. The system of any of items 1-15, further comprising an off-axis aperture in an optical path to the optical detector. (Item 17) 17. The system of any of items 1-16, further comprising a beam splitter positioned directly below the objective lens for directing light from a light source through the objective lens onto the sample coverslip or the calibration coverslip. (Item 18) 1. An optical imaging system, comprising: An objective lens, a sample stage configured to position a top surface of a) a sample coverslip for holding a sample or b) a calibration coverslip in a focal plane of the objective; an optical detector configured to capture at least a) a sample image of the sample on the sample coverslip, b) a reference image from light reflected back from the focal plane of the objective through the calibration coverslip, and c) a test image from light reflected back from the focal plane of the objective through the sample coverslip; A processor, the processor comprising: extracting reference image data from the reference image; extracting inspection image data from the inspection image; deconvolving the test image data using the reference image data as an initial point spread function for the objective lens and the coverslip; generating a calculated point spread function for the objective lens and other optical components in use, including the sample coverslip, through at least one deconvolution of the inspection image data; deconvolving the sample image using the calculated point spread function from the deconvolution of the inspection image data, thereby reducing artifacts from the sample image; a processor programmed to An optical imaging system comprising: (Item 19) Item 19. The system of item 18, wherein the processor utilizes the reference image data as the initial point spread function for generating an optimized point spread function associated with the sample coverslip. (Item 20) 1. A computerized method for imaging a sample, comprising: capturing a sample image of the sample using an objective lens positioned beneath a sample coverslip holding the sample; acquiring reference image data acquired from a calibration coverslip; capturing inspection image data obtained from light reflected from a top surface of the sample coverslip at a focal plane of the objective lens; processing the reference image data and the test image data to generate, via deconvolution of the test image data, a calculated point spread function associated with the objective lens and other optical components in use; deconvolving the sample image with the calculated point spread function, thereby reducing artifacts from the sample image; and A computerized method comprising: [Brief explanation of the drawings]

[0012] The present technology can be better understood by reference to the following figures, in which components are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the technology, in which like reference numbers designate corresponding parts throughout the different views.

[0013] [Figure 1] FIG. 1 shows a conventional configuration of a background microscope using a correction collar.

[0014] [Figure 2]FIG. 2 is a schematic diagram of one embodiment of the optical system of the present technology that uses a standard reference plate to derive an optimized point spread function associated with the objective lens and other optical components in use, including a coverslip.

[0015] [Figure 3] Figure 3 is an intensity profile from a single bead taken at optimized compensation color settings.

[0016] [Figure 4] FIG. 4 is an intensity profile from the same bead as in FIG. 3 taken with non-optimized compensation color settings.

[0017] [Figure 5] FIG. 5 is the intensity profile of an image captured from the same bead as in FIG. 3 taken at non-optimized correction color settings and then processed according to the present technique.

[0018] [Figure 6A] FIG. 6A is a depiction of a captured image taken from a 4.0 μm bead under non-optimal imaging conditions.

[0019] [Figure 6B] FIG. 6B is a processed image depiction of the captured image of the beads of FIG. 6A, processed according to the present technique.

[0020] [Figure 6C] FIG. 6C is a depiction of a captured image taken from the same 4.0 μm bead of FIG. 6A, but now taken under optimal imaging conditions.

[0021] [Figure 6D] FIG. 6D is a graphical comparison of the intensity profiles of FIGS. 6A, 6B, and 6C.

[0022] [Figure 7] FIG. 7 is a flow chart illustrating the computerized method of the present technique.

[0023] [Figure 8] FIG. 8 is a flow chart illustrating another computerized method of the present technique. DETAILED DESCRIPTION OF THE INVENTION

[0024] Detailed Description All numerical values ​​herein are assumed to be modified by the term "about" or "approximately," whether explicitly stated or not, and the terms "about" and "approximately" generally refer to a range of numbers that one of ordinary skill in the art would consider equivalent to the recited value (i.e., having the same function or result). In some cases, the terms "about" and "approximately" may include numbers rounded to the nearest significant figure. The recitation of numerical ranges by endpoints includes all numbers within that range (e.g., 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.80, 4, and 5).

[0025] As used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the content clearly dictates otherwise. As used in this specification and the appended claims, the term "or" is generally employed in its sense, including "and / or," unless the content clearly dictates otherwise. In describing the depicted embodiments of the disclosed technology, as illustrated in the accompanying figures, specific terminology is employed for clarity and ease of description. However, it is to be understood that the disclosure of this patent specification is not intended to be limited to the specific terminology so selected, and that each specific element includes all technical equivalents that operate in a similar manner. Furthermore, it is to be understood that various elements and / or features of different illustrative embodiments may be combined with and / or substituted for each other where possible within the scope of this disclosure and the appended claims.

[0026] Various embodiments of the disclosed technology are described hereinafter with reference to the figures. Note that the figures are not drawn to scale, and that elements of similar structure or function are represented by like reference numerals throughout the figures. Note also that the figures are intended only to facilitate the description of the embodiments. They are not intended as a comprehensive description of the technology or as a limitation on the scope of the disclosed technology, which is defined solely by the appended claims and their equivalents. In addition, illustrated embodiments of the disclosed technology need not have all of the aspects or advantages shown. For example, an aspect or advantage described in connection with a particular embodiment of the disclosed technology is not necessarily limited to that embodiment and can be practiced in any other embodiment, even if not so illustrated.

[0027] As used in the following description, the terms "imaging," "image capture," "capturing an image," or "detecting an image" refer to any process for collecting optical data from an image capture device. The image may be captured as a digital image for storage, i.e., for measuring optical properties such as intensity, color, or other types of data.

[0028] As used in the following description, the term "coverslip" shall refer to any sample-holding structure configured to support a container into which a sample may be deposited. In particular, the term "coverslip" may include trays or similar structures containing such sample-holding structures, known in the art by terms including "microwell," "consumable," "microtiter," and "microplate." Alternatively, "plate" shall be understood to refer to a structure capable of holding a single sample well or multiple sample wells.

[0029] As used herein, the term "sample" generally refers to a material known or suspected to contain an analyte. Sample types can be cellular or tissue, including, but not limited to, expanded cell cultures for organoid and similar experiments. Samples from cell printing can also be used. In other examples, samples can be used directly from the source or following pre-processing to modify the sample's properties. Samples can be derived from any biological source, such as physiological fluids, including blood, interstitial fluid, saliva, ocular lens fluid, cerebrospinal fluid, sweat, urine, breast milk, ascites fluid, mucus, synovial fluid, peritoneal fluid, vaginal fluid, amniotic fluid, or the like. Samples can be pre-processed prior to use, such as preparing plasma from blood, diluting viscous fluids, and the like. Pre-processing methods can involve filtration, precipitation, dilution, distillation, concentration, inactivation of interfering components, chromatography, separation steps, and the addition of reagents. In addition to physiological fluids, other liquid samples such as water, foodstuffs, and the like can be used for conducting environmental or food-producing assays. In addition, solid materials known or suspected to contain analytes can also be used as samples. In some cases, it may be beneficial to modify the solid sample to release the analyte and form a liquid medium.

[0030] As used herein, the term "light" generally refers to electromagnetic radiation, which is quantizable as photons. As pertaining to this disclosure, light can propagate in wavelengths ranging from ultraviolet (UV) to infrared (IR). In this disclosure, the term "light" is not intended to be limited to electromagnetic radiation in the visible range. In this disclosure, the terms "light," "photons," and "radiation" are used interchangeably.

[0031] As used herein, "optimized" means "improved over the original state" and does not necessarily mean that the improvement has been maximized (unless otherwise specified).

[0032] As used herein, "incremental improvement" means that the difference between the starting value and the improvement is within 30% of the starting value, within 20% of the starting value, within 10% of the starting value, or preferably within 5% of the starting value, and more preferably within 1% of the starting value.

[0033] Solutions to some of the problems identified above have been attempted. For example, U.S. Patent No. 7,825,360 (the entire content of which is incorporated by reference) addressed some of the above problems with an optical device including a focusing mechanism for changing the distance between an objective lens and a sample, an optical thickness detection unit for detecting the optical thickness of a cover glass, an operation unit for calculating an amount of aberration correction based on the optical thickness of the cover glass detected by the optical thickness detection unit, a driver unit for driving a correction collar based on the amount of aberration correction calculated by the operation unit, and an imaging sensor for forming an image of the sample passing through the objective lens.

[0034] In FIG. 1 (reproduced from the '360 patent), objective lens 5 is positioned below sample S mounted on stage 16 for observing the sample. An objective lens holder is present to hold objective lens 5. Objective lens 5 is raised and lowered by drive unit 17. Excitation light source 12 includes illumination optical system 13 for directing excitation light from light source 12 to objective lens 5 and sample S via fluorescence cube 14. Fluorescence cube 14 includes a dichroic mirror, an excitation filter, and an internal fluorescence filter. Light for observing sample S passing through objective lens 5 is collected and imaged by lens barrel 6 onto imaging device 15, such as a CCD or equivalent. Alternatively, sample S can be illuminated from above using a transillumination light source and transillumination system, not shown in FIG. 1, for collecting the light from the light source onto sample S.

[0035] A beam emitted from a light source 2 to be used for measuring a sample holding member, focusing, and the like is reflected via a collimating lens 3 onto a half mirror 8 and onto a dichroic mirror 9 disposed between the objective lens 5 and a phosphor cube 14, illuminating a sample S via the objective lens 5. As shown in FIG. 1 , a shielding plate 4 is located between the collimating lens 3 and the half mirror 8, approximately conjugate with the pupil of the objective lens 5. The shielding plate 4 shields half of the beam, using the optical axis of the collimating lens 3 as a boundary to limit the beam of the light source 2 to half.

[0036] Illumination light from the light source 2 returned from the sample S passes through an objective lens 5, a dichroic mirror 9, and a half mirror 8 and is collected by a lens barrel 6 to form an image on a two-segment sensor 7. Because half of the light beam is blocked by the shielding plate 4, the light returned from the sample S passes through an optical path symmetrical to that of the illumination light, using the optical axis as the center, and is projected onto the two-segment sensor 7. In this case, the division direction of the two-segment sensor 7 and the division direction of the shielding plate 4 are arranged relative to each other.

[0037] The objective lens 5 includes a correction collar 5A and a correction collar driver unit 25 for driving the correction collar 5A. The objective lens identification unit, the correction collar driver unit 25, the driver unit 17, the imaging device 15, and the two-segment sensor 7 are electrically connected to an operating device 27. The objective lens 5 can be (1) a dry objective lens with a large numerical aperture (the amount of correction is 0.11 mm to 0.23 mm), in which various aberrations occur significantly due to the non-uniform thickness of the cover glass; (2) an objective lens for thick glass, the thickness of which varies widely within a very wide range of up to 2 mm, and which is expected to be used with glass Petri dishes or equivalents; or (3) an objective lens for plastic, the thickness of which varies widely within a very wide range of up to 2 mm, and which is expected to be used with plastic containers.

[0038] As described in one example in the '360 patent, the mechanical thickness of the reference cover glass was set to 0.17 mm. If the actual mechanical thickness of the cover glass of sample S was 0.13 mm, this was converted to an optical thickness of 0.11166 mm by the refractive index of the cover glass (ne = 1.5255). The reference thickness of 0.17 mm was converted to 0.08539 mm, respectively. In this case, the optical difference is 0.02627 (the mechanical thickness is 0.044 mm). In the subflow for detecting the optical thickness, the actual optical thickness of sample S (mechanical thickness 0.13 mm) was calculated, and the amount of drive of the correction collar was calculated based on the difference in optical thickness relative to the reference cover glass (S1-7) (mechanical difference 0.04 mm).

[0039] As explained in the '360 patent, the correction collar was then matched to the actual thickness of the cover glass by driving the correction collar to correct as much as possible the aberration caused by the thickness difference of 0.02627 mm (mechanical difference of 0.04 mm). At this moment, the space between the sample 18 and the objective lens 5 deviates from the state in which the sample 18 is focused. The calculation of the amount of correction collar drive varies depending on the type of objective lens, and the amount of correction collar drive is calculated by a correction collar drive calculation data table or function that is characteristic of the objective lens. The correction collar drive calculation data table or function that is characteristic of the objective lens is stored in the operating device 27, and the thickness difference of the correction collar (mechanical difference of 0.04 mm) and the amount of correction collar drive are calculated by the characteristic correction collar drive calculation data table or function.

[0040] As explained in the '360 patent, the aberration of the objective lens 5 was changed by moving the correction collar, and the amount of movement was calculated by an aberration correction calculation data table or function that would be characteristic of the objective lens being used. This aberration correction calculation data table or function was stored in the operating device 27, and the amount of correction of the focus position was calculated by the thickness difference of the correction collar (mechanical difference of 0.04 mm) and the aberration correction calculation data table or function.

[0041] As another proposed solution, U.S. Patent No. 10,317,658 (the entire contents of which are incorporated by reference) describes a microscope including one or more objective lenses having an optical system, each configured to at least collect observation light from a sample, a correction collar provided on each of the one or more objective lenses and configured to move the optical system in the direction of the optical axis of the optical system by rotating it about each of the one or more objective lenses to correct aberrations, a switching unit for switching the position of the one or more objective lenses, and a focusing unit. Similar to the '360 patent, the operation of the correction collar in the '658 patent also corrects aberrations (or spherical aberrations) according to the thickness of the slide or culture container in which the sample S is placed or stored.

[0042] In yet another proposed solution, U.S. Pat. No. 9,383,567 (the entire contents of which are incorporated by reference) describes an objective lens unit with a rotatable correction collar. In the '567 patent, the objective lens 41 has a non-uniform shape along the circumferential direction of the side of the objective lens. The correction collar moves the lens group (optical system) in the direction of the optical axis in response to rotation. This action of the correction collar in the '567 patent allows aberrations (spherical aberrations) to be corrected according to the thickness of a glass slide on which the sample S is mounted, or according to the thickness of a growing container in which the sample S is stored.

[0043] In yet another proposed solution, U.S. Pat. No. 8,053,711 (the entire contents of which are incorporated by reference) describes a spherical aberration adjustment system including multiple objective lenses, at least one of which has a spherical aberration collar. The objective lenses were mounted on an objective lens holder, which positioned one of the multiple objective lenses in an imaging position. A drive mechanism in the '711 patent was coupled to one objective lens by a mechanical link, which was configured to transmit motion from the drive mechanism to the spherical aberration collar. A control system in the '711 patent was configured to operate the drive mechanism to move the spherical aberration collar of one objective lens in the imaging position to a specific spherical aberration adjustment setting. In the '711 patent, the drive mechanism moved the spherical aberration collars of the multiple objective lenses based on the thickness of the sample holder.

[0044] As described in the '711 patent, the spherical aberration correction system operated in one of two modes: protocol-driven and manual. In the protocol-driven mode, the spherical aberration collars of the individual objective lenses were automatically adjusted by the spherical aberration correction system based on a protocol in a spherical aberration software program related to the thickness of the sample holder or the resulting thickness of the sample holder and the depth of the sample medium recorded in the protocol. For example, if the sample holder had a thickness of 0.2 mm, then at least one of the spherical aberration collars of the objective lenses was adjusted to a spherical aberration setting of 0.2 mm. In the manual mode of operation, the spherical aberration correction collars of the objective lenses were adjusted to a user-defined setting selected via a graphical user interface (GUI). For example, a user would use a spherical aberration graphical user interface (GUI) stored on a computer to adjust at least one of the spherical aberration collars to a spherical aberration adjustment setting, such as 0.2 mm, based on the thickness of the sample holder 121, which was 0.2 mm.

[0045] Thus, while certain types of compensation mechanisms have been proposed in the art for correcting aberrations due to variations in substrate thickness variations, these compensation mechanisms often require complex optics, user interaction, and translation of optical components to provide the compensation. Accordingly, the technology described herein generally relates to improved methods, apparatus, and systems for compensating for errors in optical microscopy imaging.

[0046] In one embodiment, the present technology is directed to an optical system that is an inverted microscope configuration, such as the system shown in FIG. 2 . However, the general teachings of the present technology are equally applicable to non-inverted microscope configurations (as described below). With inverted microscope configurations, image quality is known to degrade due to mismatches in coverslip thickness with the intended design of the microscope objective. More specifically, due to the thickness of the coverslip (or variations in the thickness of the coverslip at different lateral positions), the acquired image can be distorted by differences in refractive index between the coverslip material and the sample (or medium holding the sample) to be imaged. As described in the background section, image quality distortions due to these refractive index differences between the coverslip material and the sample are often more pronounced for higher numerical aperture (NA) systems, which are often used to collect more light and provide brighter images and are often used for higher spatial resolution in both the lateral and axial directions. As discussed above, microscope objective correction collars are traditionally used to alter and potentially correct spherical aberration in images due to the presence of a coverslip in the optical path. However, this process is tedious, often involves user interaction, and in some cases does not satisfactorily correct the sample image. In practice, thickness variations and mismatches for coverslips compared to the intended design of the microscope objective often limit the available sample consumable and objective combinations, which can result in unsatisfactory losses in resolution and signal.

[0047] The present technique addresses these issues through the following illustrative embodiment, which allows images taken from non-optimal conditions to be processed with restoration of sufficient optical quality for image analysis, thereby avoiding the need for user interaction and corrective color adjustments (and the associated time loss) for every image being captured. Indeed, in one embodiment of the present technique (described in more detail below, where a projected point light source is used to obtain an acceptable point spread function for deconvolution of the sample image), the inventive procedure is particularly applicable to high numerical aperture systems, where numerical aperture values ​​range from 0.65 to 0.95 or higher, and allows sufficient optical quality for image analysis to be obtained without the need for user interaction and corrective color adjustments.

[0048] FIG. 2 is a diagram depicting various embodiments of an optical system 200 of the present technology. The system 200 includes an objective lens 202 for a microscope system, an optical detector 204, a light source 206, a first imaging lens 208, a beam splitter 210, a second imaging lens 240, and an aperture 230 (e.g., an aperture offset from the optical axis Z). The objective lens 202 is configured to perform imaging and / or optical measurements on a sample that may be deposited in a sample well 214. Other components of the microscope system include an optional sample holding structure, such as a sample stage 216, which supports a sample coverslip 212 with reference surfaces 212 a, 212 b. The sample stage 216 (as indicated by the double arrow below the stage 216 on FIG. 2 ) can move either the sample coverslip 212 or a calibration coverslip 250 into position for imaging.

[0049] The system 200 of FIG. 2 can be implemented as a module or subsystem of a microscope system. Other components used by the microscope system 200 to image a sample, such as an excitation light source, filters, a beam splitter, and a sample image capture device, are represented in FIG. 2 as a sample imaging component 221. The sample imaging component 221 may include, for example, lenses, filters, or other optical devices that form an optical path including the objective lens 202 and the sample coverslip 212 when the microscope system is used to image a sample. The microscope system 200 may also use different light sources or different sample image capture devices based on the type of imaging or measurement being performed. The optical device may be inserted below the objective lens of FIG. 2 at 221a, above the beam splitter 210, or above the off-center aperture 230 at 221b. The sample well 214 in the embodiment shown in FIG. 2A is formed on the sample coverslip 212, providing a bottom reference surface 212a and an upper reference surface 212b for the autofocus procedure.

[0050] A self-calibration of the best focus position can be performed for a given objective lens 202. The process of focusing that objective lens 202 for subsequent imaging or optical measurements may then be performed with minimal further imaging. The calibration of the best focus position may be stored as a reference calibration gradient, which may be stored or included in the system data storage system 223 along with data characterizing the objective lens 202.

[0051] A first imaging lens 208 collimates light from a light source 206 along optical path 201 and passes the collimated light to a beam splitter 210. The beam splitter 210 reflects a portion of the light along optical path 203 toward the objective lens 202 and toward a sample coverslip 212 on optical path 205. The sample coverslip 212 reflects the light back to the objective lens 202 and toward the beam splitter 210 on optical path 207. The beam splitter 210 passes a portion of the light along optical path 209 toward an off-center aperture 230. The light passing through the off-center opening on aperture 230 is less than the total light beam impinging on aperture 230. The remainder of aperture 230 blocks the portion of the light beam not passing through the aperture. The light passing through aperture 230 is directed to a second imaging lens 240 and an optical detector 204. The off-center aperture 230 operates by sampling a portion of the wavefront from the objective lens 202. The sampled portion of light is focused and directed toward the detector by the second imaging lens 240, but is constrained by the off-center aperture as an asymmetric marginal ray. This allows the light and the location of best focus to be visualized without changing any of the component settings. It should be understood that different off-center apertures with different sizes and / or positions can be used to allow for sensitivity adjustment or different sized pupil diameters of the objective lens 202. It should also be understood that an off-center aperture can also be placed between the first imaging lens 208 and the beam splitter 210. In this configuration, the detection side does not require the off-center aperture 230. Here, the same principle is at work: there is an on-axis beam, asymmetrically sampled at the edge of the pupil of the objective 202 by the off-center aperture, creating an off-axis beam that strikes a smaller imaging lens 240, which responds to defocusing of the microscope objective by translating and blurring the image of the projected source as the objective is moved away from best focus. In one embodiment, the off-center aperture is switchable (mechanically or optically), which can improve accuracy.In one embodiment, the aperture can be moved to allow beam profiling with more light due to improved coupling efficiency through the projected source imaging system.

[0052] 2 only show light along the optical paths that form the light beam that impinges on optical detector 204. The portion of light that is not shown is the portion of light that is blocked by the blocking portion of aperture 230.

[0053] The objective lens 202 is configured to move along optical paths 203 and 205 on the z-axis (shown in FIG. 2 ), which is perpendicular to the xy-plane along which the sample coverslip 212 extends. The following description, as a way of providing clarity, refers to the position of the objective lens as being on the z-axis and the position of the reference image, or lateral position, as being on the xy-plane. It should be understood that the use of "x-axis," "z-axis," or "y-axis" to provide spatial designations is not intended to be limiting. Any suitable coordinate system may be used. Furthermore, it should be noted that example implementations may involve the objective lens 202 traveling in a non-vertical direction.

[0054] The sample coverslip 212 may include a sample well 214 that may be positioned for imaging a sample that may be deposited therein in accordance with the normal function and operation of the microscope system, as shown in FIG. 2. In the exemplary system shown in FIG. 2, the sample coverslip 212 has a first surface 212a and a second surface 212b, which may be considered the bottom surface of the sample well. The first surface 212a and / or the second surface 212b may be at least partially reflective, thereby providing a reflective surface for use during an autofocus procedure or for use for inspection imaging with the sample's surface 212b in the focal plane of the objective lens 202 (described in detail below). A reflective surface may also be provided on the coverslip, or on the surface of a slide, or other planar material positioned in the optical path proximate to the bottom surface of the sample well 214.

[0055] The objective lens 202 may be moved along the z-axis using a linear actuation motor controlled by the controller 220. The objective lens 202 is represented diagrammatically in FIG. 2 as including a linear actuation motor that moves the objective lens 202. The objective lens 202 includes selected optics configured to focus light from the light source 206 onto a sample S on a sample coverslip 212 held by a sample stage 216, thereby enabling the microscope system 200 to capture an image of the sample S. During an autofocus procedure, the objective lens 202 is controlled to focus on the surface 212 a or 212 b. In some implementations, the motor that moves the objective lens 202 may be a stepper motor or a servo motor with a linear actuator.

[0056] Light along the optical path passing through the decentered aperture 230 travels through a second imaging lens 240 to the optical detector 204, where the projected source is imaged on the detector plane. When defocused, the light beam on the optical detector 204 expands in size, shifts its position on the optical detector 204, and has lower intensity and / or contrast. When the objective lens 202 is focused, the image is captured at maximum intensity, minimum size, and its highest contrast. The process of focusing the objective lens 202 involves moving the objective lens 202 to find the best focus position on the z-axis. Each beam spot in each image captured at each z-position of the objective lens 202 appears at a position in the image plane that is offset from the spot position on the previous image.

[0057] In an exemplary implementation, the optical detector 204 in the auto-focusing system 200 may be a linear array detector, a 2D sensor array such as a charge-coupled device, a position-sensitive diode, an image capture device, or any suitable device that can be controlled by the controller 220 to capture images of the reference image as the objective lens 202 is controlled to move to a series of z-positions. The light source 206 in the auto-focusing system 200 may be any suitable light-emitting device, such as a laser, a light-emitting diode (LED) or LED array, a white light, a fluorescent light source, an infrared light, or the like.

[0058] The controller 220 may be implemented using any computer-programmable system having a hardware interface connected to at least the optical detector 204 and a motor configured to move the objective lens 202. In some implementations, the controller 220 may also be a component of a microscope system 221 in which the objective lens is auto-focused. The auto-focusing procedure may be a function stored as software in a data storage medium 223 to which the controller 220 has access.

[0059] As illustrated in FIG. 2, an objective lens 202 positioned below a sample coverslip 212 images the sample S through the transparent coverslip. The sample S is positioned above / above the sample coverslip 212. The image focal plane 202a for the objective lens 202 is shown in FIG. 2 coincident with the upper reference surface 212b on which the sample S resides. The sample coverslip 212 can be a consumable substrate (e.g., a clear plastic bottom on a microtiter plate). The sample well 214 shown in FIG. 2 can be a single well as shown, or can represent multiple wells, each having a bottom that holds a sample to be imaged. The optical system also utilizes a calibration coverslip 250, which can be moved to replace the sample coverslip 212.

[0060] 2 , the light source 206 can be, for example, a laser or other light source such as a light-emitting diode or white light or fluorescent lamp or tungsten-halogen lamp, used to project a narrow beam of light onto a sample coverslip 212 (to generate a test image), and when in position, can be used to project a narrow beam of light onto a calibration coverslip 250 (to generate a reference image). A first imaging lens 208 directs the laser beam onto a beam splitter 210, which directs the laser beam through the objective lens 202 onto the sample coverslip 212, where the laser light passes through surface 212a and then encounters surface 212b on the interior (well) side of the sample coverslip 212. The laser light reaching surface 212b is reflected back through the sample coverslip 212 and imaged onto an image plane 204a of the optical detector 204 through the objective lens 202 and the second imaging lens 240. Although light source 206 is shown as a single fixture in FIG. 2, it may comprise multiple light-emitting devices such as those described above, each controlled as needed by controller 220 for the particular imaging required.

[0061] In one embodiment, when capturing the test and reference images, a laser or LED (or another narrow light beam source) may be used for exposure. In one embodiment, when capturing the sample image, white light, a fluorescent light source, infrared light, or the like may be used for exposure. In one embodiment, narrow-spectrum wavelength light, such as a laser or LED, may be used to excite the fluorescent emission of the sample or to capture light backscattered from the sample.

[0062] In one embodiment of the present technology, the aperture 230 selects a portion of the image light that passes through the second imaging lens 240 to the image plane 204a of the optical detector 204. In one embodiment of the present technology, the aperture 230 is positioned offset from the optical axis so that the selected light is light passing through the peripheral (or edge) region of the objective lens 202, where spherical aberration is most severe. In one embodiment of the present technology, autofocus is used in positioning either the objective lens 202, or the first imaging lens 208, or the second imaging lens 240, or the sample coverslip 212, or the calibration coverslip 250 to generate the sharpest image for a reference image to be taken with the calibration coverslip 250 instead of or for a test image to be taken with the sample coverslip 212 in place. The acquired image can be stored in the memory device 223 of the controller 220 for subsequent processing, for example, in the processor 220a of the controller 220 or any other processor in communication with the controller 220. 2 as modules of the controller 220, either the processor 220a or the storage device 223 may be located remotely from the controller 220 and communicate with the controller 220 as needed to exchange data, processing results, and instructions therebetween. A correction color 252 (as shown in FIG. 2) may be used to help improve image quality.

[0063] When a sample coverslip 212 is present in the optical path, in one embodiment, a focused laser beam (or other focused light) is reflected from the top surface of the sample substrate (e.g., from surface 212b of the sample coverslip 212) and imaged onto image plane 204a to obtain an inspection image associated with the particular coverslip in use. In this embodiment, the image focal plane 202a for the objective lens 202 coincides with the top reference surface 212b of the sample coverslip 212. The autofocused reflected light from the surface of the sample substrate can be considered a single-point reflected light.

[0064] Similarly, when a calibration coverslip 250 is present in the optical path (instead of the sample coverslip 212), in one embodiment, a focused laser beam (or other focused light) is reflected from the top surface of the calibration coverslip 250 (e.g., from the top of a glass slide of uniform thickness that does not necessarily hold a sample) and imaged onto the image plane 204a of the optical detector 204 to acquire a reference image. In this embodiment, the image focal plane 202a for the objective lens 202 coincides with the top surface of the calibration coverslip 250.

[0065] 2, both the test image taken with the sample coverslip 212 and the reference image taken with the calibration coverslip 250 can be acquired and stored in the memory device 223 of the controller 220, for subsequent processing, e.g., in the processor 220a. In one embodiment of the present technology, the test image taken with light reflected back from the sample coverslip 212 and the image taken with light reflected back through the calibration coverslip 250 need not be created using only a single point of reflected light (as detailed above). Multiple image points can be directed at the sample coverslip 212 and / or the calibration coverslip 250. In other words, an array of image points can be used, which will in effect map the optical wavefront of light propagating from the objective lens 202 toward the imaging plane of the optical detector 204.

[0066] Furthermore, in one embodiment, the light reflected from the top surface of the sample coverslip 212 and / or calibration coverslip 250 may be a line image (or other well-defined shape). Such a line image would (in effect) allow assessment of the PSF across the lateral direction of the sample coverslip 212 and / or calibration coverslip 250.

[0067] In one embodiment of the present technology, as described above, the aperture 230 is set off-axis to better assess the effect of spherical aberration on the focus for the sample coverslip 212 and / or the calibration coverslip 250; the effect of spherical aberration is typically greater the closer the light passes to the edge of the pupil in the lens system. Nevertheless, in one embodiment of the present technology, the aperture 230 can be set on-axis or at other positions within the optical paths 201, 203, 205, 207, and 209. In one embodiment of the present technology, varying the off-axis position of the aperture 230 allows for interrogation of the optical wavefront of light propagating from the objective lens 202 toward the image plane of the optical detector 204. The position, size, and shape of the aperture can be specific to the particularities of the imaging system and the PSF for deconvolution. The simplest aperture can be a single eccentric circle large enough to transmit the peripheral ray bundle from the objective lens at sufficient light levels. However, a single asymmetric aperture may reduce light levels and introduce bias into the measured inspection or sample image, which can be compensated for when needed by adding an aperture on the opposite side of the light beam, a series of apertures, an annular ring, or the like.

[0068] In theory, in the absence of any errors such as spherical aberration and / or defocusing, the image of light reflected back from the top surface of the calibration coverslip 250 (when in place) should resemble an ideal point spread function. Any deviation from the ideal point spread function would represent an error in the optical system that would blur the optical image.

[0069] In one embodiment, the processor 220a of the controller 220 generates a calculated point spread function (PSF) required to enhance the image of the inspection image taken from light reflected from the surface 212b of the sample coverslip 212. c To determine the initial point spread function (PSF), i The technique is configured to perform deconvolution (e.g., blind deconvolution) using a reference image taken from the calibration coverslip 250 as the PSF (P s ). In general, blind deconvolution is a deconvolution technique that allows for the recovery of an “original image” from a single “blurred” image or set thereof, without necessarily assuming any prior knowledge of the image or PSF. While typical linear and nonlinear deconvolution techniques may utilize a known PSF to recover the original (unblurred) image, for blind deconvolution, the PSF is estimated from the image, allowing the deconvolution to be performed. Here, the technique utilizes an initial (or estimated) point spread function (PSF) of the optical system, including the objective lens 202, the sample coverslip 212, and other optical components in the optical path from the sample S to the image plane 204a of the optical detector 204, as described above. i ) is used as the calculated PSF of the optical system. In one embodiment, the deconvolution technique of the present technology can be performed iteratively, whereby each iteration of the algorithm used for deconvolution c Alternatively, the deconvolution techniques used in the present technique may be performed non-iteratively, in which case one application of the algorithm used for deconvolution is to refine the calculated PSF of the optical system. c successfully provides a good estimate of

[0070] One deconvolution process is illustrated by equation (1) below. [ka] where f is the original undistorted image, g is the distorted noisy image, and h is the PSF of the system. [ka] is the convolution operator and n is any corrupting noise.

[0071] The bidirectional Lucy-Richardson restoration algorithm for recovering the PSF is shown below in equation (2). [ka]

[0072] During the ceremony, [ka] is the estimate of f (original image) after k iterations, * is the correlation operator, [ka] is called the Richardson-Lucy (RL) function. [ka] is referred to as the deblurred image. These and other suitable image processing techniques for deconvolution in the present technology are described by David S.C. Biggs and Mark Andrews in "Acceleration of iterative image restoration algorithms" in APPLIED OPTICS, vol. 36, no. 8, March 10, 1997, pp. 1766-1755, the entire contents of which are incorporated herein by reference, and by Sage et al. in "DeconvolutionLab2: An open-source software for deconvolution microscopy" in Methods, vol. 115, February 15, 2017, pp. 28-41, the entire contents of which are incorporated herein by reference. Commercial image deconvolution software that can be used for this technique is available from companies such as Media Cybernetics (1700 Rockville Pike, Suite 240 Rockville, Maryland USA 20852), R2018a (version 9.4.813654) from Matlab® (1 Apple Hill Drive, Natick, MA 01760-2098), and others.

[0073] Here, in one embodiment, the processor 220a of the controller 220 calculates a reference image, h (the initial PSF of the optical system), taken using light projected and focused from the light source 206 and reflected from the top surface of the calibration coverslip 250 in equations (1) and (2). i ) and an inspection image taken using light projected and focused from the light source 206 and reflected from the top surface of the sample coverslip 212 is used as g (a distorted, noisy image). Iterations using equation (2) provide an estimate of the original image, if not blurred by optical imperfections. [ka] During this iteration of the deconvolution process, the quality of the test image may be improved, for example, until the shape, size, intensity, and / or location of the spots is optimal (and / or similar to the reference image) or until the deconvolution is complete. [ka] Once the acceptability criteria for the convolution are met, the parameters and calculation settings for deconvolution, for example according to that of equation (2), are stored in storage device 223, and processor 220a now processes all pixels of the sample image taken from sample S on sample coverslip 212 and calculates the calculated PSF arrived at through deconvolution. c can be used to generate an optimal image.

[0074] Thus, with the configuration illustrated in FIG. 2 , an optical imaging system includes (in this embodiment) an objective 202, a sample stage 216 configured to position the top surface of a) a sample coverslip 212 for holding a sample S or b) a calibration coverslip 250 in the focal plane 202 a of the objective 202, an optical detector 204 configured to capture at least a) a sample image of the sample on the sample coverslip 212, b) a reference image from light reflected back from the focal plane 202 a of the objective 202 through the calibration coverslip 250, and c) a test image from light reflected back from the focal plane 202 a of the objective 202 through the sample coverslip 212, and An optical imaging system is provided, comprising a processor 220a programmed to: extract image data, extract inspection image data from the inspection image, deconvolve the inspection image data using the reference image data as an initial point spread function for the objective lens and cover slip, generate a deconvolved test image through at least one deconvolution of the inspection image data that is associated with calculated point spread functions for the objective lens and other optical components in use, including the sample cover slip, and deconvolve a sample image using the point spread function calculated from the deconvolution of the inspection image data, thereby reducing (or eliminating) artifacts from the sample image.

[0075] In one embodiment of the present technology, the PSF calculated by blind deconvolution c The determination of PSF may be performed during the process of capturing sample image data from each well. c The determination may also be made after the image (sample image data) has been stored.

[0076] In one embodiment of the present technology, test images are collected across a predetermined number of wells or plates in use, and settings for individual blind deconvolutions per well are stored for later processing of sample images from each well. cDetermination of the initial PSF is performed by comparing the reference images from each well. i Use as the calculated PSF of each well for subsequent deconvolution of the sample image data c A deconvolution, such as a blind deconvolution, may be performed to arrive at

[0077] Thus, in one embodiment of the present technology, under the control of controller 220, optical detector 204 captures multiple sample images (at different lateral positions) from a sample coverslip. Under the control of controller 220, optical detector 204 captures multiple calibrated reference images (at different lateral positions) from a calibration coverslip. Under the control of controller 220, optical detector 204 captures multiple test images (at different lateral positions) from light reflected back through the sample coverslip from the focal plane of the objective. Processor 220a generates a separate deconvolved test image associated with a separate calculated point spread function for each lateral position, and then deconvolves the multiple sample images using the separate calculated point spread function for each lateral position, thereby reducing or removing artifacts from the multiple sample images.

[0078] In one embodiment of the present technology, the processor 220a calculates the PSF c is the PSF where the variance between wells is calculated for a given number of wells. c Although it may be possible to determine that a given well is within the range of 100 nm, it is not guaranteed to perform a blind deconvolution of the test image for each and every well. Instead, it is possible to determine a single estimated point spread function (PSF) for the multi-well sample holder. e ) is the calculated PSF for processing the sample image data for each well. c(1) and (2). Thus, in one embodiment of the present technology, it may not be necessary to strictly use the iterative analysis described above with reference to equations (1) and (2) for each well being imaged. Furthermore, because the deconvolution techniques of the present technology can be applied to each pixel in the sample image, the processed image can be presented to depict intensity in the XZ plane.

[0079] In one embodiment of the present technology, under control of controller 220, optical detector 204 captures multiple sample images (from different sample coverslips). Under control of controller 220, optical detector 204 captures multiple calibrated reference images (from multiple calibration coverslips). Under control of controller 220, optical detector 204 captures multiple test images (from different sample coverslips) from light reflected back through the different sample coverslips. Processor 220a generates a separate deconvolved test image associated with a separate calculated point spread function for each sample coverslip, and then deconvolves the multiple sample images using the separate calculated point spread function for each coverslip, thereby reducing or removing artifacts from the multiple sample images.

[0080] Also, while the deconvolution discussed above is preferred for the present technique, other image processing techniques can also be used in the present technique. In one embodiment, the processor 220a generates a stored point spread function (PSF) associated with a particular set of sample coverslips. s ), here configured to perform deconvolution using the PSF for that particular set of sample coverslips (and the optics in use). s has already been determined. In another embodiment, the individual PSFs for each sample well set (and optics in use) s However, the processor 220a calculates the stored PSF for each well (and optical system in use). sIt may already be determined that it is possible to perform deconvolution using these stored PSFs. s Using {overscore (R)} may then enable linear and non-linear deconvolution techniques to recover the original image from the sample image data without the need for blind deconvolution iterations.

[0081] Regardless of the deconvolution technique utilized, in one embodiment, the processor 220a is configured to calculate a quality metric indicative of the improvement in the deconvolved image achieved after deconvolution. The quality metric can be a measure of full width half maximum (FWHM) improvement in resolution of pixels in the restored image. The quality metric can be a measure of signal-to-noise ratio improvement of pixels in the restored image. The quality metric can be a measure of signal intensity improvement of pixels in the restored image. In one embodiment, the processor is programmed to determine an optimized point spread function for improving the image quality of the sample image by repeated deconvolution of the test image and repeated comparison of a) the deconvolved image of the test image with b) a calibrated reference image until there is a progressive improvement in the deconvolved image with respect to the calculated point spread function.

[0082] In one embodiment of the present technology, one or more metrics may be used to generate a single merit value to be associated with the deconvolved image, each of which may have a set of criteria for acceptable feature size, shape, location, intensity gradient, peak intensity value, or cross-correlation.

[0083] In one embodiment of the present technology, light projected and focused from light source 206 and reflected from the top surface of calibration coverslip 250 may be imaged as a single spot image representing a reference image, such that the reflected light from calibration coverslip 250 is aligned with an initial point spread function PSF i It is used to obtain the PSF. iis applied to the reflected light focused onto the top surface 212b of the sample surface coverslip 212 for the test image with deconvolution of the test image, which can then be compared to the reference image in multiple ways, such as those described above, with each metric being assigned a weight to penalize the divergence metric in one embodiment.

[0084] In other words, the merit function quality score is a weighted sum of various metrics, such as spot size, spot morphology, spot intensity, spot location, spot gradient, and cross-correlation, for comparing the deconvolved test image with a reference image taken from the calibration coverslip 250; the better the quality score, the better the optimized, calculated PSF that should then be used to deconvolve the sample image. c Successive deconvolutions result in a final computed PSF that is ideally within a preset percentage difference from the calibrated reference image or within a preset percentage difference change in the merit function. c can be used to form

[0085] In one embodiment, processor 220a is programmed to determine the optimized point spread function based on at least one criterion of acquired spot size, spot shape, spot intensity, and spot location that is only incrementally improved (within 30%, 20%, 10%, or preferably within 5%, or more preferably within 1% of a change in one or more of the criteria described above) after successive deconvolutions of the inspection image. In another embodiment, processor 220a is programmed to determine the optimized point spread function based on a criterion of a merit function quality score after subsequent deconvolutions that changes the quality score by less than 30%, 20%, 10%, or preferably less than 5%, or more preferably less than 1%.

[0086] In one embodiment, processor 220a is programmed to determine an optimized point spread function for improving the image quality of the sample image by repeated blind deconvolution of the test image and repeated comparison of a) the deconvolved image of the test image with b) a calibrated reference image until the improvement in successive deconvolved images is incremental (within 30%, or within 20%, or within 10%, or preferably within 5%, and more preferably within 1% of the change in one or more of the criteria described above).

[0087] In one embodiment of the present technology, once an acceptable PSF (hereinafter PSF0) is determined using reference and test images taken of light reflected from the respective top surfaces of the calibration coverslip 250 and sample coverslip 212, the objective lens 202 can be adjusted to focus at a region above the top surface of the coverslip, e.g., a distance z1 above the coverslip. The sample image taken at z1 can then be processed, e.g., using the Lucy-Richardson method (or other deconvolution technique) described above to deconvolve the sample image at z1 (perhaps through iterative blind convolution) until an acceptable quality metric is obtained. At that point, the processor 220a can store the new PSF (hereinafter PSF1) associated with z1. The objective lens 202 can then be adjusted to focus at a distance z2, which is further above the coverslip than z1. Again, the sample image taken at z2 can then be processed, for example, using the Lucy-Richardson method (or other deconvolution technique) described above to deconvolve the image at z2 (perhaps through iterative blind convolution) until an acceptable quality metric is obtained, for example, based on the criteria set forth above, and processor 220a can then store the obtained PSF (hereinafter PSF2) associated with z2. In this way, imaging through the depths of the well holding the sample to be measured can be performed with restoration of image quality at each depth.

[0088] In another embodiment, the processor 220a generates selectable point spread functions PSF associated with different levels of spherical aberration. s A set of PSFs is available for image processing. The processor 220a then selects a point spread function PSF from the set. sand use the selected one to deconvolve the image. In one embodiment, the quality score described above is used to select the selectable point spread function PSF that produced the better deconvolved image. s and thus can be used to select one PSF to use for deconvolving the sample image at a particular level or height within sample well 214. In one embodiment, processor 220a selects from a set of selectable point spread functions based on the level of spherical aberration expected from known perturbations. For example, point spread function PSF s may have point spread functions that differ from each other according to increments of spherical aberration (e.g., by 0.2 waves of spherical aberration). In one embodiment, the quality scores described above select a selectable point spread function PSF that produces a better deconvolved image and is therefore considered an optimized PSF. s can be used to determine

[0089] In one embodiment, a model (or a database in the processor) can select a PSF for each lateral image taken across the sample well 214 based on a stored point spread function, for example, associated with known or expected lateral variations in the thickness of the sample coverslip 212 being used. s Associate with.

[0090] Generally, most sample coverslips will exhibit increasing spherical aberration with depth, but there may be some sample coverslips where the sample and sample medium are not at significantly different refractive indices and therefore the spherical aberration may be less pronounced. Nevertheless, in one embodiment, because the amount of spherical aberration may increase with defocusing within the sample, a model can be used to predict the number of waves of spherical aberration for a given defocusing from an ideal focus position similar to the case where no defocusing and no spherical aberration are present. In one embodiment, above the sample bottom, the model (or a database in processor 220b) would associate different selectable point spread functions with, for example, known depths within sample well 214 or known heights z1 above top side 212b of sample coverslip 212 and the expected spherical aberration at these locations. A processor, such as processor 220a, would select one of the selectable point spread functions based on the measured or expected spherical aberration.

[0091] In one embodiment, selectable point spread functions available to processor 220a for deconvolution are stored in a database during an initial manufacturing process. During this process, the point spread functions are calculated with reference to the system components provided by the manufacturer, including the optical lenses, the depth across each sample well 214, and / or known or predicted lateral variations in the thickness of the sample coverslip 212, which may create spherical aberration or defocusing in the system. Selectable point spread functions are available to enable rapid deconvolution of images when utilizing known system components.

[0092] In one embodiment of the present technology, if an object is at height z2, the imaging process may first ensure that the spherical aberration model is correct at that height z2 before correcting for image blur or aberrations that are blurring the image of the object via deconvolution. In other words, defocusing may be determined at height z2, but in one embodiment, the processor will use multiple z-planes around height z2 to select a particular height from the multiple planes that has high (or otherwise acceptable) contrast (e.g., an acceptable signal-to-noise ratio greater than 2, or an acceptable signal-to-background ratio greater than 2, or a signal gradient that consistently rises above the noise) and apply the PSF to its image. Thus, in one embodiment, the processor will evaluate the appropriate level for convolution with the expected spherical aberration at different heights above the sample coverslip, and then further improve image quality by digitally or mechanically refining focus, such as by use of correction collar 252, across those depths where acceptable contrast is found. This embodiment may be particularly useful when the reference image has determined a PSF associated with optical system errors due to, for example, spherical aberration or defocusing or other imaging artifacts, and the sample S to be imaged is not simply on the top surface 212b of the sample coverslip, but instead is at a depth of, for example, 100 microns above the sample coverslip 212. In that case, the objective lens 202 or substrate stage 216 is moved to image the sample S at a position 100 microns between the coverslip and the sample S, so there will be no elements to reflect light back to the optical detector until the sample begins to come into focus.At that point, the processor will select PSFs in storage 223 (or otherwise accessible to processor 220a) associated with heights of, for example, 96 microns, 98 microns, 100 microns, 102 microns, and 104 microns, perform deconvolution of sample images taken at these locations, and determine one of the selectable PSFs that produces a deconvolved image with the best contrast, quality score, or merit function. In one embodiment of the present technology, objects at different depths are imaged using the optimized PSFs identified for each of the different depths, so that a model for predicting spherical aberration with depth can be built and refined.

[0093] Example In the examples below, the Lucy-Richardson method and blind deconvolution were used to deconvolve images in the examples below, using the procedure for deconvolution described above. In the examples given below, a precision coverslip from Thorlabs (Newton, New Jersey, US) was glued onto a microscope slide using Norland 61 optical adhesive (Cranbury, New Jersey, US) and served as a calibration coverslip. The Thorlabs precision coverslip had a thickness tolerance of 5 μm and was positioned on the side of the microscope with the precision coverslip facing downward toward the light source 206.

[0094] Figure 3 depicts the intensity distribution of an optical image taken from an individual 0.5 μm bead at best focus and with optimal correction color settings. For later comparison purposes, the intensity distribution in Figure 3 represents the "ideal" image intensity distribution for a 0.5 μm bead.

[0095] Figure 4 depicts the intensity distribution of an optical image taken from an individual 0.5 μm bead at best focus but with non-optimal correction color settings. As expected, the image resolution and signal intensity shown in Figure 4 are degraded compared to the ideal image intensity distribution depicted in Figure 3.

[0096] Figure 5 depicts the intensity distribution of a mathematically processed image using the deconvolution techniques of the present technology to process the non-optimal optical image of Figure 4. As is apparent, both the image resolution and signal strength for the mathematically processed image are improved using the mathematical deconvolution of the present technology.

[0097] Figure 6A depicts another optical image taken from a 4.0 μm bead at best focus but under non-optimal imaging conditions with non-optimal compensation color settings. Figure 6B depicts the image data of Figure 6A processed using the deconvolution process of the present technique. Figure 6C depicts an optical image taken from the same 4.0 μm bead of Figure 6A at best focus, but now under optimal imaging conditions. The more closely the processed image data of Figure 6B matches the optical image data of Figure 6C, the better the deconvolution of the present technique recovered from the non-optimal image what the actual image of the object should be.

[0098] FIG. 6D is a graphical comparison of the intensity profiles of FIGS. 6A, 6B, and 6C. This comparison shows that the deconvolution process of the present technique allows images taken under non-optimal conditions to "recover" many of the attributes found in images taken under optimal imaging conditions, such as peak intensity and FWHM, thereby allowing useful image data to be obtained without the user having to spend time and effort making actual optical adjustments, such as with correction collars or other means. Viewed from a different perspective, FIGS. 6A, 6B, 6C, and 6D illustrate the effectiveness of the present technique for recovering useful image data from images that would otherwise be blurred due to optical system errors resulting from, for example, spherical aberration or defocusing or other imaging artifacts. Computer Control

[0099] It should be understood that the controller 220 diagrammatically illustrated in FIG. 2 is merely representative of various computing devices, including one or more types of user devices, such as, for example, user input devices (e.g., keypads, touch screens, mice, and the like), user output devices (e.g., display screens, printers, visual indicators or alerts, audible indicators or alerts, and the like), and computing devices. The controller 220 (or processor 220a) may have a software-controlled graphical user interface (GUI) for display by the output devices and one or more devices for loading media-readable material (e.g., logical instructions embodied in software, data, and the like). The controller 220 (or processor 220a) includes an operating system (e.g., Microsoft Windows® software) for controlling and managing its various functions, and thus comprises a processor.

[0100] 7 is a flowchart detailing a computerized method of the present technique for imaging a sample, which may be implemented in or with the processor 220a in the controller 220. Although a numerical set of numbers follows to label the steps, the present technique is not necessarily limited to the steps occurring in the given numerical order.

[0101] In step 701, a reference image of the spot reflected from the top surface of a calibration coverslip (or other standard reference slide) is recorded (for subsequent image processing in processor 220). In a preferred option, aperture 230 samples (selects) only a portion of the wavefront of the spot that is offset from the optical axis and reflected and passes through a peripheral region of objective lens 202 where spherical aberration varies by the fourth power.

[0102] In step 703, a biological image of the sample on the sample coverslip is recorded (for subsequent image processing in processor 220a).

[0103] In step 705, a test image of the spot reflected from the top surface of the sample coverslip is recorded (for subsequent image processing in processor 220a) (steps 703 and 705 may occur in reverse order).

[0104] In step 707, the test image is deconvolved (by processor 220a) to obtain a calculated point spread function of the optical imaging system including the sample coverslip. In a preferred option, a deconvolution imaging process (described above, including but not limited to blind deconvolution and the Lucy-Richardson algorithm) is used in conjunction with the test image, which serves as a first approximation of the point spread function of the optical system, to recover the point spread function. In a preferred option, using a test image taken with an offset aperture will result in the recovered PSF having reduced spherical aberration.

[0105] In step 709, the calculated point spread function is used to deconvolve the biological image (e.g., by processor 220a). If there is little variation in sample coverslip thickness, the calculated PSF does not need to be recalculated very often when imaging from well to well. In another embodiment of the present technology, based on the amount of defocusing observed in the test image, a test image library for defocusing can be used to determine an appropriate test image to be used in the steps of a) setting the focus position for the objective lens and b) obtaining an appropriate calculated PSF for the optical component in use.

[0106] As described above, in one embodiment, the technique is directed to an inverted microscope configuration, such as the system shown in FIG. 2 , but the technique can also be applied to a non-inverted microscope configuration. In that case, a laser beam or other light reflected from a calibration coverslip would be reflected from a focal plane on the side of the calibration coverslip (or other reference plate) facing the objective lens 202 (i.e., the objective lens 202), and the imaging optics would be above a sample coverslip 212 supporting the sample to be imaged. The same steps would then follow, with test and reference images being taken from the laser beam or other light reflected from a focal plane on the side of the calibration coverslip or calibration coverslip facing the objective lens 202. The reference image would be used as an initial PSF for deconvolution of the test image, from which a calculated PSF for the optical system, including the objective lens 202 and the imaging optics, would be derived. A sample image taken from the sample coverslip supporting the sample would then be deconvolved using the optimal PSF.

[0107] In one embodiment of the present technology in a non-inverted configuration, the reflections used for the reference and test images will be taken from an image focal plane located on the inner surface of the coverslip, closer to the sample. The sample position of interest can proceed through a thick sample or (as described below with an immersion objective) into the depth of the well holding the sample.

[0108] In one embodiment of the present technology in a non-inverted configuration, when the objective lens is immersed in the sample medium, mismatches between the refractive index and the objective lens design relative to the sample will result in spherical aberration. In this case, a process similar to that described above for inverted configuration imaging at different depths within the sample well can be applied. That is, a sample image taken at z1 (closest to the bottom of the coverslip) can then be processed using, for example, the Lucy-Richardson method (or other deconvolution technique) described above to deconvolve the sample image at z1 (perhaps through iterative blind convolution) until an acceptable quality metric is obtained. At that point, processor 220a can store a new PSF (hereafter PSF1) associated with z1. Objective lens 202 can then be adjusted to focus at distance z2, which is further below the coverslip than z1. Again, the sample image taken at z2 may then be processed, for example, using the Lucy-Richardson method (or other deconvolution technique) described above to deconvolve the image at z2 (perhaps through iterative blind convolution) until an acceptable quality metric is obtained, for example, based on the criteria set forth above, and processor 220a may then store the obtained PSF (hereinafter PSF2) associated with z2. In this way, in a non-inverted configuration, imaging at greater depths of the well holding the sample to be measured can be performed with image quality restoration at each depth.

[0109] Similar to that described above, in an embodiment with a non-inverted configuration, because the amount of spherical aberration may increase with depth into the sample, or, viewed differently, with height above the top surface 212b of the sample coverslip 212, a model can be used to predict the number of waves of spherical aberration for a given depth within the sample well 214 or height z1 above the top surface 212b of the sample coverslip 212. In one embodiment, the model (or a database accessible by processor 220a) may associate different selectable point spread functions with expected spherical aberration. A processor such as processor 220a would select at least one of the selectable point spread functions (associated with different displacements about height z1), perform deconvolution of the sample image at the different displacements, and select a particular PSF based, for example, on the observed contrast obtained following sample image deconvolution (as described above).

[0110] Whether in an inverted or non-inverted microscope configuration, FIG. 8 is a flowchart detailing another computerized method of the present technique for imaging a sample, which may be implemented within controller 220 or in conjunction with processor 220a.

[0111] In step 801, a sample image (e.g., of a sample on a top surface of a sample holder) is captured. In step 803, reference image data is captured, obtained from a focused laser beam or other light reflected from a surface (e.g., top surface) of a standard reference plate, such as a calibration coverslip. In step 805, test image data is captured, obtained from a focused laser beam or other light reflected from a surface (e.g., top surface) of a sample holder, such as a sample coverslip. In step 807, the reference image data and test image data are processed to generate an optimized point spread function associated with the optical component in use. In step 809, the sample image is deconvolved with the optimized point spread function, thereby reducing or removing artifacts from the sample image.

[0112] It should be understood that one or more of the processes, subprocesses, and process steps described herein may be implemented by hardware, firmware, software, or a combination of two or more of the foregoing, for example, on one or more electronically or digitally controlled devices. The software may reside in software memory (not shown) within a suitable electronic processing component or system, such as, for example, controller 220 and / or processor 220a, depicted diagrammatically in FIG. 2. The software memory may include an ordered list of executable instructions for implementing logical functions (i.e., logic that may be implemented in digital form, such as digital circuitry or source code, or in analog form, such as analog sources, such as analog electrical, sound, or video signals). The instructions may be executed within a processing module, including, for example, one or more microprocessors, general-purpose processors, processors, digital signal processors (DSPs), or combinations of application-specific integrated circuits (ASICs). Furthermore, the schematic diagrams illustrate a logical division of functions with physical (hardware and / or software) implementations, without being limited by the architecture or physical layout of the functions. The system embodiments described herein may be implemented in a variety of configurations and may operate as hardware / software components in a single hardware / software unit or in separate hardware / software units.

[0113] The executable instructions may be implemented as a computer program product having instructions stored therein that, when executed by processor 220a and / or controller 220, direct the electronic system to perform commands, such as reading data from storage device 223, from an image file containing measured or processed data, etc. In one embodiment, the executable instructions enable controller 220 or processor 220a to store measured or processed image data, such as, for example, reference image data from a reference image and / or inspection image data from an inspection image (e.g., an initial inspection image), in memory 223 (or internal memory of processor 220a). Further, the executable instructions may enable the controller 220 or processor 220a to: a) deconvolve the inspection image data using the reference image data as an initial point spread function for the objective lens and coverslip; b) generate calculated point spread functions for the objective lens and other optical components in use, including the sample coverslip, through at least one deconvolution of the inspection image data; and c) deconvolve the sample image using the point spread function calculated from the deconvolution of the inspection image data, thereby reducing or removing artifacts from the sample image.

[0114] In one embodiment, the executable instructions enable the controller 220 to control at least one of a) vertical displacement of the objective lens relative to the coverslip, b) lateral displacement of the objective lens relative to the coverslip, c) exposure duration of the optical detector, d) intensity of the light source illuminating the sample, e) insertion of an optical filter into the optical path, f) positioning of an off-axis aperture in the optical path, g) autofocus adjustment, and h) other optical detector or camera settings.

[0115] In one embodiment, the executable instructions enable the controller 220 to control at least one of the position of the objective lens 202 or the distance of the sample coverslip 212 from the objective lens 202 to focus light from the light source onto the focal plane of the objective lens. In one embodiment, under the control of the controller 220, the optical detector 204 can capture multiple sample images from the sample coverslip at different lateral positions, capture multiple calibrated reference images from a calibration coverslip at different lateral positions, and capture multiple test images from light reflected back through the sample coverslip from the focal plane of the objective lens at different lateral positions, and the processor 220a can generate a separate deconvolved test image associated with a separate calculated point spread function for each lateral position, and deconvolve the multiple sample images using the separate calculated point spread function for each lateral position, thereby reducing and removing artifacts from the multiple sample images. In another embodiment, under the control of controller 220, optical detector 204 can capture multiple sample images from different sample coverslips, can capture multiple calibrated reference images from multiple calibration coverslips, each associated with a different sample coverslip, and can capture multiple test images from different sample coverslips from light reflected back through the different sample coverslips; processor 220a can generate a separate deconvolved test image associated with a separate calculated point spread function for each sample coverslip, and can deconvolve the multiple sample images using the separate calculated point spread function for each coverslip, thereby reducing or eliminating artifacts from the multiple sample images.

[0116] The computer program product may be embodied in any non-transitory computer-readable storage medium for use by or in connection with an instruction execution system, apparatus, or device, such as an electronic computer-based system, processor-containing system, or other system that may selectively fetch and execute instructions from the instruction execution system, apparatus, or device. In the context of this disclosure, a computer-readable storage medium is any non-transitory means that may store a program for use by or in connection with the instruction execution system, apparatus, or device. The non-transitory computer-readable storage medium may optionally be, for example, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device. A non-exhaustive list of more specific examples of non-transitory computer-readable media includes an electrical connection having one or more wires (electronic), a portable computer diskette (magnetic), a random access memory (electronic), a read-only memory (electronic), an erasable programmable read-only memory (electronic), such as flash memory, a compact disc memory (optical), such as a CD-ROM, CD-R, CD-RW, and a digital versatile disc memory, i.e., a DVD (optical). A non-transitory computer-readable storage medium may even be paper or another suitable medium on which the program is printed, so that the program can be captured electronically, for example, via optical scanning of the paper or other medium, and then compiled, interpreted, or otherwise processed in a suitable manner if necessary, and then stored in a computer or machine memory, such as storage device 223 or the internal memory of processor 220a.

[0117] It should also be understood that the term "in signal communication" as used herein means that two or more systems, devices, components, modules, or sub-modules can communicate with each other via signals traveling over some type of signal path. The signals may be communication, power, data, or energy signals that may communicate information, power, or energy from a first system, device, component, module, or sub-module to a second system, device, component, module, or sub-module along the signal path between the first and second systems, devices, components, modules, or sub-modules. The signal path may include physical, electrical, magnetic, electromagnetic, electrochemical, optical, wired, or wireless connections. The signal path may also include additional systems, devices, components, modules, or sub-modules between the first and second systems, devices, components, modules, or sub-modules.

[0118] More generally, terms such as "communicate" and "in communication with" (e.g., a first component "communicates" or "is in communication with" a second component) are used herein to indicate a structural, functional, mechanical, electrical, signal, optical, magnetic, electromagnetic, ionic, or fluid relationship between two or more components or elements. Thus, the fact that one component is described as communicating with a second component is not intended to exclude the possibility that additional components may be present between and / or operatively associated with or engaged with the first and second components.

[0119] It should also be understood that "receiving and transmitting data" as used herein means that two or more systems, devices, components, modules, or sub-modules can communicate with each other via signals traveling over some type of signal path. The signals may be communication, power, data, or energy signals that may communicate information, power, or energy from a first system, device, component, module, or sub-module to a second system, device, component, module, or sub-module along the signal path between the first and second systems, devices, components, modules, or sub-modules. The signal path may include physical, electrical, magnetic, electromagnetic, electrochemical, optical, wired, or wireless connections. The signal path may also include additional systems, devices, components, modules, or sub-modules between the first and second systems, devices, components, modules, or sub-modules.

[0120] Exemplary Description of the Technology The following numbered description of the present technology describes some of the inventive aspects of the present technology.

[0121] Description 1. An optical imaging system comprising: a sample stage configured to hold a sample to be imaged on a top surface of a sample coverslip; an objective lens positioned directly below the sample stage and configured to image the sample on the top surface of the sample coverslip; an optical detector configured to capture a sample image of at least the sample on the sample coverslip; and a processor programmed to: derive reference image data from a reference image captured from a calibration coverslip; derive test image data from a test image captured from light reflected from the top surface of the sample coverslip at the focal plane of the objective lens; process the reference image data and the test image data; generate a calculated point spread function associated with the objective lens and other optical components in use (via deconvolution of the test image data); and deconvolve the sample image using the calculated point spread function, thereby reducing or removing artifacts from the sample image.

[0122] Statement 2. The system of statement 1, wherein the processor is programmed to perform blind deconvolution of the examination image data to determine a calculated point spread function.

[0123] Statement 3. The system of statements 1 or 2, wherein the processor is programmed to determine an optimized point spread function for improving image quality of the sample image by repeated deconvolution of the test image and repeated comparison of a) the deconvolved image of the test image with b) a calibrated reference image until there is incremental improvement in the deconvolved image with respect to the calculated point spread function.

[0124] Statement 4. The system of any of the preceding statements, wherein the processor is programmed to determine an optimized point spread function based on at least one criterion of spot size, spot shape, spot intensity, and spot location, obtained with only incremental refinement after successive deconvolutions of the inspection image.

[0125] Statement 5. The system of any of the preceding statements, wherein the processor is programmed to determine an optimized point spread function for improving image quality of the sample image by repeated blind deconvolution of the test image and repeated comparison of a) the deconvolved image of the test image with b) a calibrated reference image until the improvement in successive deconvolved images is incremental.

[0126] Statement 6. The system of any of the preceding statements, further comprising a light source that provides light to be focused onto the focal plane.

[0127] Statement 7. The system of any of the above statements, further comprising a controller configured to control at least one of: a) vertical displacement of the objective lens relative to the coverslip; b) lateral displacement of the objective lens relative to the coverslip; c) exposure duration of the optical detector; d) intensity of the light source illuminating the sample; e) insertion of an optical filter (e.g., a spectral filter and / or a spatial filter) into the optical path; f) positioning of an off-axis aperture in the optical path; g) autofocus; and h) other optical detector or camera settings.

[0128] Statement 8. The system of any of the preceding statements, wherein the controller controls at least one of the position of the objective lens and the position of the sample stage to focus light from the light source onto a focal plane of the objective lens.

[0129] Statement 9. The system of claim 8, wherein, under control of the controller, the optical detector captures a plurality of sample images from a sample coverslip at different lateral positions; under control of the controller, the optical detector captures a plurality of calibrated reference images from a calibration coverslip at different lateral positions; under control of the controller, the optical detector captures a plurality of test images from light reflected back from the focal plane of the objective through the sample coverslip at different lateral positions; the processor generates a separate deconvolved test image associated with a separate calculated point spread function for each lateral position; and the processor deconvolves the plurality of sample images using the separate calculated point spread function for each lateral position, thereby reducing or removing artifacts from the plurality of sample images.

[0130] Statement 10. The system of statement 8, wherein under control of the controller, the optical detector captures a plurality of sample images from different sample cover slips; under control of the controller, the optical detector captures a plurality of calibrated reference images from a plurality of calibration cover slips, each associated with a different sample cover slip; under control of the controller, the optical detector captures a plurality of test images from the different sample cover slips from light reflected back through the different sample cover slips; the processor generates a separate deconvolved test image associated with a separate calculated point spread function for each sample cover slip; and the processor deconvolves the plurality of sample images using the separate calculated point spread function for each cover slip, thereby reducing or removing artifacts from the plurality of sample images.

[0131] Statement 11. The system of any of the preceding statements, further comprising a correction collar that compensates for optical aberrations caused by the sample coverslip.

[0132] Statement 12. The system of any of the preceding statements, wherein the processor stores in memory individual optimized point spread functions for a plurality of coverslips having individual standard thicknesses.

[0133] Statement 13. The system of any of the preceding statements, wherein the processor stores in memory separate optimized point spread functions associated with different types of coverslips having different optical thicknesses.

[0134] Statement 14. The system of any of the above statements, wherein the processor is further configured to: retrieve, for the sample image, an image taken from a position z1 displaced from the sample coverslip; identify a spherical aberration associated with the position z1; retrieve a set of selectable point spread functions, each having a different spherical aberration associated with a different height displaced from the position z1; select from the set of selectable point spread functions a starting point spread function associated with the identified spherical aberration; and deconvolve the sample image at the position z1 using the starting point spread function.

[0135] Statement 15. The system of any of the above statements, wherein the processor is further configured to: take a first sample image taken at a first position z1 displaced from the sample coverslip; determine a first calculated point spread function for reducing artifacts from the first sample image at the first position z1; take a second image taken at a second position z2 that is removed farther from the sample coverslip than the first position z1; use the first calculated point spread function as a starting point spread function in the deconvolution; and determine a second calculated point spread function for reducing artifacts from the second sample image taken at the second position z2.

[0136] Statement 16. The system of any of the preceding statements, further comprising an off-axis aperture in the optical path to the optical detector.

[0137] Statement 17. The system of any of the preceding statements, further comprising a beam splitter positioned beneath the objective lens for directing light from the light source through the objective lens onto the sample coverslip or calibration coverslip.

[0138] Statement 18. An optical imaging system, comprising: an objective; a sample stage configured to position a top surface of a) a sample coverslip for holding a sample or b) a calibration coverslip at a focal plane of the objective; and an optical detector configured to capture at least a) a sample image of the sample on the sample coverslip, b) a reference image from light reflected back through the calibration coverslip from the focal plane of the objective; and c) a test image (e.g., an initial test image) from light reflected back through the sample coverslip from the focal plane of the objective; and deriving reference image data from the reference image and a test image from the test image. an optical imaging system comprising: a processor programmed to retrieve image data; deconvolve inspection image data using reference image data as an initial point spread function for the objective lens and cover slip; generate a deconvolved test image through at least one deconvolution of the inspection image data that is associated with calculated point spread functions for the objective lens and other optical components in use, including the sample cover slip; and deconvolve a sample image using the point spread function calculated from the deconvolution of the inspection image data, thereby reducing or removing artifacts from the sample image.

[0139] Statement 19. The system of statement 18, wherein the processor utilizes the reference image data as an initial point spread function for generating an optimized point spread function associated with the sample coverslip.

[0140] Statement 20. A computerized method for imaging a sample, the computerized method including the steps of: capturing a sample image of the sample using an objective lens positioned directly beneath a sample coverslip that holds the sample; acquiring reference image data obtained from a calibration coverslip; capturing test image data obtained from light reflected from the top surface of the sample coverslip at the focal plane of the objective lens; processing the reference image data and the test image data to generate a calculated point spread function associated with the objective lens and other optical components in use (via deconvolution of the test image data); and deconvolving the sample image with the calculated point spread function, thereby reducing or removing artifacts from the sample image. The method can be implemented using any of the components of an optical imaging system described in statements 1-19.

[0141] Description 21. A computer-readable medium storing instructions that, when executed by a computer, cause the computer to perform the method steps of capturing a sample image of a sample using an objective lens positioned beneath a sample coverslip holding the sample, acquiring reference image data obtained from a calibration coverslip, capturing test image data obtained from light reflected from the top surface of the sample coverslip at the focal plane of the objective lens, processing the reference image data and the test image data to generate a calculated point spread function associated with the objective lens and other optical components in use (via deconvolution of the test image data), and deconvolving the sample image with the calculated point spread function, thereby reducing or removing artifacts from the sample image. The computer-readable medium can be stored, for example, within controller 220 and / or processor 220a or other similar computing device in communication with controller 220 and / or processor 220a.

[0142] Statement 22. A method for imaging a sample, comprising: capturing a sample image; capturing reference image data obtained from light reflected from a surface of a standard reference plate; capturing test image data obtained from light reflected from a surface of a sample holder; processing the reference image data and the test image data to generate an optimized (or calculated) point spread function associated with the optical components in use; and deconvolving the sample image with the optimized point spread function, thereby reducing or removing artifacts from the sample image. The method can be implemented using any of the components of an optical imaging system described in Statements 1-19.

[0143] Description 23. A computer-readable medium storing instructions that, when executed by a computer, cause the computer to perform the method steps of capturing a sample image, capturing reference image data obtained from light reflected from a surface of a standard reference plate, capturing test image data obtained from light reflected from a surface of a sample holder, processing the reference image data and the test image data to generate an optimized (or calculated) point spread function associated with the optical components in use, and deconvolving the sample image with the optimized point spread function, thereby reducing or removing artifacts from the sample image. The computer-readable medium can be stored, for example, within the controller 220 and / or processor 220a or other similar computing device in communication with the controller 220 and / or processor 220a.

[0144] Statement 24. An optical imaging system comprising: means for capturing a sample image of a sample; means for acquiring reference image data obtained from a calibration coverslip; means for capturing test image data obtained from light reflected from the top surface of the sample coverslip; means for processing the reference image data and the test image data to generate a calculated point spread function associated with the objective lens and other optical components in use (via deconvolution of the test image data); and means for deconvolving the sample image with the calculated point spread function, thereby reducing or removing artifacts from the sample image.

[0145] Statement 25. An optical imaging system comprising: means for capturing a sample image; means for capturing reference image data obtained from light reflected from a surface of a standard reference plate; means for capturing test image data obtained from light reflected from a surface of a sample holder; means for processing the reference image data and the test image data to generate an optimized (or calculated) point spread function associated with the optical component in use; and means for deconvolving the sample image with the optimized point spread function, thereby reducing or removing artifacts from the sample image.

[0146] Statement 26. An optical imaging system comprising: an objective lens; a sample stage configured to position the top surface of a) a sample coverslip for holding a sample or b) a calibration coverslip at the focal plane of the objective lens; means for capturing at least a) a sample image of the sample on the sample coverslip, b) a reference image from light reflected back through the calibration coverslip from the focal plane of the objective lens, and c) a test image (e.g., an initial test image) from light reflected back through the sample coverslip from the focal plane of the objective lens; and means for deconvolving the sample image using a calculated point spread function obtained from deconvolution of at least one of the test images using the reference image as an initial point spread function for the deconvolution.

[0147] Description 27. An optical imaging system comprising: means for capturing a sample image of a sample; and means for deconvolving the sample image, thereby reducing or removing artifacts from the sample image, wherein the means for deconvolving retrieves, for the sample image, an image taken from a position z1 displaced from a sample coverslip; ascertains a spherical aberration associated with the position z1; retrieves a set of selectable point spread functions, each having a different spherical aberration associated with a different height displaced from z1; selects from the set of selectable point spread functions an origin spread function associated with the identified spherical aberration; and deconvolves the sample image at position z1 using the origin spread function.

[0148] Description 28. An optical imaging system comprising: means for capturing a sample image of a sample; and means for deconvolving the sample image to thereby reduce or remove artifacts from the sample image, wherein the means for deconvolving takes a first sample image taken at a first position z1 displaced from a sample coverslip, determines a first calculated point spread function to reduce artifacts from the first sample image at z1, and takes a second image taken at a second position z2 displaced further from the sample coverslip than z1, and uses the first calculated point spread function as a starting point spread function in the deconvolution to determine a second calculated point spread function to reduce artifacts from the second sample image at z2.

[0149] It should be understood that various aspects or details of the technology may be changed without departing from the scope of the technology. Furthermore, the foregoing description is for purposes of illustration only, and not for purposes of limitation, and the technology is defined by the claims.

Claims

1. 1. An optical imaging system, comprising: a sample stage configured to support a sample coverslip, the sample coverslip configured to hold a sample to be imaged on an upper surface of the sample coverslip, the sample stage being capable of moving either the sample coverslip or a calibration coverslip into position for imaging; an objective lens positioned beneath the sample stage and configured to image the sample on the top surface of the sample coverslip; an optical detector configured to capture a sample image of at least the sample on the sample coverslip; A processor, the processor comprising: deriving reference image data from a reference image captured from said calibration coverslip comprising a reference plate; deriving test image data from a test image captured from light reflected from a top surface of the sample coverslip at a focal plane of the objective lens; processing the reference image data and the test image data to generate, via deconvolution of the test image data, a calculated point spread function associated with the objective lens and other optical components in use; deconvolving the sample image with the calculated point spread function, thereby reducing artifacts from the sample image; and a processor programmed to An optical imaging system comprising:

2. The system of claim 1 , wherein the processor is programmed to perform a blind deconvolution of the inspection image data to determine the calculated point spread function.

3. The processor:

3. The system of claim 1, further comprising: a processor configured to: determine an optimized point spread function for improving image quality of the sample image by repeatedly deconvolving the test image and repeatedly comparing a) the deconvolved image with b) the reference image until there is a progressive improvement in the deconvolved image of the test image with respect to the calculated point spread function.

4. 4. The system of claim 3, wherein the processor is programmed to determine the optimized point spread function based on at least one criterion of spot size, spot shape, spot intensity, and spot location obtained with only incremental refinement after successive deconvolutions of the inspection image.

5. 4. The system of claim 3, wherein the processor is programmed to determine the optimized point spread function by repeated deconvolution of the inspection image and repeated comparison of a) the deconvolved image of the inspection image with b) a calibrated reference image until improvement in successive deconvolved images is progressively improved.

6. The system of any of claims 1-5, further comprising a light source, the light source providing a projection point of light to be focused onto the focal plane for the test image or for the reference image.

7. 7. The system of claim 6, further comprising a controller configured to control at least one of: a) vertical displacement of the objective lens relative to the coverslip; b) lateral displacement of the objective lens relative to the coverslip; c) exposure duration of the optical detector; d) intensity of the light source illuminating the sample; e) insertion of an optical filter into the optical path; f) positioning of an off-axis aperture in the optical path; and g) autofocus adjustment.

8. 8. The system of claim 7, wherein the controller is configured to control at least one of a position of the objective lens and a position of the sample stage to focus the light from the light source to the focal plane of the objective lens.

9. Under control of the controller, the optical detector captures a plurality of sample images from the sample coverslip at different lateral positions; Under control of the controller, the optical detector captures a plurality of calibrated reference images from the calibration coverslip at the different lateral positions; Under control of the controller, the optical detector captures a plurality of test images from light reflected back through the sample coverslip from the focal plane of the objective lens at the different lateral positions; the processor generating, for each lateral position, a separate deconvolved inspection image associated with a separate calculated point spread function; the processor deconvolving the plurality of sample images with the individual calculated point spread functions for each lateral position, thereby reducing artifacts from the plurality of sample images. The system of claim 8.

10. Under control of the controller, the optical detector captures a plurality of sample images from different sample coverslips; Under control of the controller, the optical detector captures a plurality of calibrated reference images from a plurality of calibration coverslips, each associated with a different sample coverslip; Under control of the controller, the optical detector captures a plurality of test images from light reflected back from and through the different sample cover slips; the processor generates, for each sample coverslip, a separate deconvolved test image associated with a separate calculated point spread function; the processor deconvolves the plurality of sample images with the individual calculated point spread function for each coverslip, thereby reducing artifacts from the plurality of sample images. The system of claim 8.

11. The system of any of claims 1-10, further comprising a correction collar for compensating for optical aberrations caused by the sample coverslip.

12. The system of any of claims 1-11, wherein the processor is configured to store in memory individual optimized point spread functions for a plurality of coverslips having individual standard thicknesses.

13. The system of any of claims 1-12, wherein the processor stores in memory individual optimized point spread functions associated with different types of cover slips having different optical thicknesses.

14. The processor further comprises: The position z displaced from the sample coverslip with respect to the sample image １ Retrieving the captured image from said position z １ and determining the spherical aberration associated with Each of them is at the position z １ deriving a set of selectable point spread functions having different spherical aberrations associated with the point spread functions; selecting a starting point spread function associated with the identified spherical aberration from the set of selectable point spread functions; Using the origin spread function, the position z １ deconvolving the sample image in A system according to any one of claims 1 to 13, configured to:

15. The processor further comprises: a first position z displaced from the sample coverslip; １ retrieving a first sample image taken at First position z １ determining a first calculated point spread function for reducing artifacts from the first sample image in The first position z １ a second position z removed further from the sample coverslip than ２ retrieving the second sample image taken at Using the first calculated point spread function as the starting point spread function in the deconvolution, ２ determining a second calculated point spread function for reducing artifacts from the second sample image taken in A system according to any one of claims 1 to 14, configured to:

16. The system of any preceding claim, further comprising an off-axis aperture in the optical path to the optical detector.

17. 17. The system of claim 1, further comprising a beam splitter positioned directly below the objective lens for directing light from a light source through the objective lens onto the sample coverslip or the calibration coverslip.

18. 1. An optical imaging system, comprising: An objective lens, a sample stage configured to position the top surface of a) a sample coverslip for holding a sample or b) a calibration coverslip with a reference plate at the focal plane of the objective lens, the sample stage being capable of moving either the sample coverslip or the calibration coverslip into position for imaging; an optical detector configured to capture at least a) a sample image of the sample on the sample coverslip, b) a reference image from light reflected back from the focal plane of the objective through the calibration coverslip, and c) a test image from light reflected back from the focal plane of the objective through the sample coverslip; A processor, the processor comprising: extracting reference image data from the reference image; extracting inspection image data from the inspection image; deconvolving the test image data using the reference image data as an initial point spread function for the objective lens and the sample coverslip; generating a calculated point spread function for the objective lens and other optical components in use, including the sample coverslip, through at least one deconvolution of the inspection image data; deconvolving the sample image using the calculated point spread function from the deconvolution of the inspection image data, thereby reducing artifacts from the sample image; a processor programmed to An optical imaging system comprising:

19. 20. The system of claim 18, wherein the processor utilizes the reference image data as the initial point spread function for generating an optimized point spread function associated with the sample coverslip.

20. 1. A computerized method for imaging a sample, comprising: capturing a sample image of the sample using an objective lens positioned directly beneath a sample coverslip holding the sample, the sample coverslip being positioned by a sample stage configured to move either the sample coverslip or a calibration coverslip; acquiring reference image data acquired from the calibration coverslip, the calibration coverslip comprising a reference plate; capturing inspection image data obtained from light reflected from a top surface of the sample coverslip at a focal plane of the objective lens; processing the reference image data and the test image data to generate, via deconvolution of the test image data, a calculated point spread function associated with the objective lens and other optical components in use; deconvolving the sample image with the calculated point spread function, thereby reducing artifacts from the sample image; and A computerized method comprising:

Images