Intra-image focus correction of a line-scan imager

Abstract

Apparatus and associated methods relate to performing intra-image focus correction of a line-scan imager. A focus-correcting line-scan imager includes a linear imaging sensor and a linear focusing sensor, both of which receive light focused by a common objective lens. The linear imaging sensor is coplanar with a focal plane of the objective lens, but the linear focusing sensor is canted with respect to the focal plane such that light-sensitive pixels at a first end of the linear focusing sensor are located farther than the focal plane from the objective lens and light-sensitive pixels at a second end opposite the first end of the linear focusing sensor are located nearer than the focal plane to the objective lens. Location where the objective lens provides best focus within the linear focusing sensor is determined based on a comparison of corresponding images obtained by the linear imaging sensor and the linear focusing sensor.

Description

[0001] P2024-0510-WG-P15766W001

[0002] INTRA-IMAGE FOCUS CORRECTION OF A LINE-SCAN IMAGER

[0003] CROSS-REFERENCE TO RELATED APPLICATION^ )

[0004] This application claims the benefit of U.S. Provisional Application No. 63 / 681,726, filed August 9, 2024, and entitled “INTRA-IMAGE FOCUS CORRECTION OF A LINE-SCAN IMAGER,” the disclosure of which is hereby incorporated by reference in its entirety.

[0005] BACKGROUND

[0006] Microscopic analysis of cells and tissues often requires the preparation of very thin, high-quality sections (slices) of a fresh tissue. Such sectioning typically results in a thin slice of the fresh tissue characterized by tissue dimensions (e.g., width and length dimensions) that are large in directions orthogonal to the fresh tissue’s thickness dimension. Such tissue sections can be stained to better highlight of and / or increase contrast between various structures or portions within the section. Most fresh tissues are soft and flexible, making them difficult to section into sections of uniform thickness throughout the tissue section’ s two-dimensional expanse of width and length. As a result of this soft and flexible nature of these fresh tissues, sections thereof can have thicknesses that vary over these other dimensions of these sections (z.e., over the length and width dimensions). These tissue sections are typically mounted onto glass slides, which present a planar surface onto which the tissue sections are mounted. Such mounting results in the tissue projecting above the planar surface of the glass slide in a non-uniform fashion due to these thickness variations. The glass slides are often viewed under a microscope by a technician and / or imaged by a slide scanner over a two-dimensional expanse of the tissue section.

[0007] SUMMARY

[0008] Some embodiments are related to a system for imaging a tissue section with intra-image focus correction. The system includes a line-scan imager, and a line-scan focuser. The line -scan imager includes a linear imaging sensor having a first row of light- sensitive pixels. The line-scan imager also includes an objective lens having an optical axis oriented such that a focal plane of the objective lens is coplanar with the first row of lightsensitive pixels. The line-scan focuser includes a linear focusing sensor having a second row of light-sensitive pixels that is canted with respect to the focal plane of the objective lens such that light-sensitive pixels at a first end of the linear focusing sensor are located farther than the focal plane from the objective lens and light-sensitive pixels at a second end opposite the first end of the linear focusing sensor are located nearer than the focal plane to the objective lens. The line-scan focuser includes a focusing processor configured to determine a location of best focus within the linear focusing sensor based on a comparison of images obtained by the linear imaging sensor and the linear focusing sensor. The line-scan focuser also includes a focusing mechanism configured to position the objective lens relative to the linear imaging sensor based on the location of best focus as determined by the focusing processor.

[0009] Some embodiments relate to a method for imaging a tissue section using intra-image focus correction. In the method, an objective lens is positioned relative to a linear imaging sensor via a focusing mechanism. A first image of a linear portion of the tissue section is captured via the linear imaging sensor coplanar with a focal plane of the objective lens. A second image of the linear portion of the tissue section is captured via a linear focusing sensor that is canted with respect to the focal plane. A location of best focus within the linear focusing sensor is determined. The tissue section is repositioned relative to the linear imaging sensor and the linear focusing sensor in a direction transverse to each of the optical axis and the linear imaging sensor. The objective lens is repositioned relative to the linear imaging sensor, based on the location of best focus, thereby refocusing the tissue section onto the focal plane. The above steps, except the first position of the objective lens, are repeated to obtain a two-dimensional image of the tissue section.

[0010] BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The material described herein is illustrated by way of example and not by way of limitation in the accompanying figures. For simplicity and clarity of illustration, elements illustrated in the figures are not necessarily drawn to scale. For example, the dimensions of some elements may be exaggerated relative to other elements for clarity. Further, where considered appropriate, reference labels have been repeated among the figures to indicate corresponding or analogous elements. In the figures:

[0012] FIG. 1 is a perspective view of a lab technician inserting and / or removing glass slide with tissue sections into a slide scanner that has intra-image focus-correction capability.

[0013] FIG. 2 is a perspective view of an example configuration of optical components of a slide scanner that has intra-image focus-correction capability.

[0014] FIGS. 3 A and 3B are plan and front elevation views of optical components depicted in FIG. 2.

[0015] FIGS. 4A and 4B are images of a tissue section captured by a linear imaging sensor and a linear focusing sensor of an auto-focusing line-scan imager, respectively. FIGS. 5 is a graph of the contrast image for each of the images captured by the linear imaging sensor and the linear focusing sensor.

[0016] FIG. 6 is a graph of a ratio of contrast images taken by the linear imaging sensor and the linear focusing sensor as fit to a peak-locating curve.

[0017] FIG. 7 is a flow chart of an example of a method for focusing a line-scan imager while imaging in real time.

[0018] FIGS. 8A and 8B are illustrations of alternate embodiments of configuring a linear focusing sensor for determining a metric of focus of a line-scan imager.

[0019] DETAILED DESCRIPTION

[0020] Because of the varying thickness tsection(x, y) of tissue sections, intra-image focusing challenges can arise. Best focus for a first location (xi, yi) of the tissue section might be different than best focus for a second location (x2, ys) of the tissue section. Thus, fixed-focus imaging of a tissue section can result in some portions of the image being in focus, or nearly so, while other portions of the image are somewhat or significantly out of focus. It would be advantageous to capture images of tissue sections that are in focus, or nearly so, throughout the two-dimensional expanse of these images. Such in-focus imaging of tissue sections can be achieved by providing intra-image focus-correction capability for a line-scan imager, as the tissue section is being scanned thereby.

[0021] Apparatus and associated methods relate to performing intra-image focus correction for a line-scan imager. A focus-correcting line-scan imager includes a linear imaging sensor and a linear focusing sensor, both of which receive light focused thereon by an objective lens that is common to both the linear imaging sensor and the linear focusing sensor. The linear imaging sensor is coplanar with a focal plane of the objective lens, but the linear focusing sensor is canted with respect to the focal plane such that lightsensitive pixels at one end of the linear focusing sensor are located farther than the focal plane from the objective lens and light-sensitive pixels at an opposite of the linear focusing sensor are located nearer than the focal plane to the objective lens. A location where the objective lens provides best focus within the linear focusing sensor is determined based on a comparison of corresponding images obtained by the linear imaging sensor and the linear focusing sensor.

[0022] FIG. 1 is a perspective view of a lab technician inserting and / or removing glass slide with tissue sections into a slide scanner that has intra-image focus-correction capability. In FIG. 1, technician 10 is using slide scanner 12 to capture a two-dimensional image of tissue sections mounted on glass slides. Such two-dimensional images are typically obtained line-by-line by a line-scan imager as each glass slide is positioned and sequentially repositioned to cause sequential line portions of the tissue section mounted thereon to be focused upon the line-scan imager.

[0023] Each line of the two-dimensional image is captured by the line-scan imager with the glass slide positioned accordingly. Linear portions of the tissue section are captured by a linear imaging sensor. A two-dimension image of each tissue section is created by capturing sequential images of adjacent linear portions of the tissue section by a linear imaging sensor. To do so, a scanning mechanism is configured to position and sequentially reposition the glass slide containing the tissue section relative to the line-scan imager. Slide scanner 12 includes an intra-image focus corrector to ensure that the resulting two-dimensional image is in focus or nearly so for each of the images of the linear portions of tissue section 14. The intra-image focus corrector corrects focus for each linear portion (or section) of adjacent linear portions (or sections) of tissue section 14 that is being captured. To accomplish such in-focus imagery, slide scanner 12 includes a linear focusing sensor that is located adjacent to the linear imaging sensor so as to share a common objective lens therewith. Each of the linear imaging sensor and the linear focusing sensor captures images of a linear portion of the tissue section. The configuration of these components (i.e., the linear imaging sensor, the linear focusing sensor, and the objective lens) permits determination of a best-focus position of the objective lens with respect to the linear imaging sensor, as will be described below.

[0024] FIG. 2 is a perspective view of an example configuration of optical components of a slide scanner that has intra-image focus-correction capability. FIGS. 3A and 3B are plan and front elevation views of optical components depicted in FIG. 2. In FIG. 2, glass slide 12 presents tissue section 14 for imaging by line-scan imager 16. Line-scan imager 16 includes linear imaging sensor 18, objective lens 20, slide scanning mechanism 22. Typically, a slide holder is used by slide scanning mechanism 22 to secure glass slide 12 to slide scanning mechanism 22. Linear imaging sensor 18 has a row of light-sensitive imaging pixels ipo, ip i ... ipN, which is aligned along longitudinal axis 24 of linear imaging sensor 18. Longitudinal axis 24 of linear imaging sensor 18 defines an x-direction. Objective lens 20 is located between linear imaging sensor 18 and the slide holder. Objective lens 20 has optical axis 26 defining a z-direction. Herein, planes normal to optical axis 26 are termed focal planes of objective lens 20. Objective lens 20 is oriented such that focal plane 28 of objective lens 20 is coplanar with the row of light-sensitive imaging pixels ipo, ipi ... ipN of linear imaging sensor 18. When properly focused, focal plane 28 coincides with image plane 30 of line scan imager 18. Image plane 30 is the plane where an image of tissue section 14 is in best focus. Therefore, such coincidence of image plane 30 with the row of light-sensitive imaging pixels ipo, ipi ... ipN ensures that images of linear portions of tissue section 14 are properly focused by objective lens 20 upon the row of light-sensitive imaging pixels ipo, ipi ... ipN of linear imaging sensor 18. As depicted in FIG. 2, image plane 30 is not coincident with focal plane 28. Therefore, images captured by linear imaging sensor 18 are sub-optimally focused. Focusing correction will improve the quality of images captured by linear imaging sensor 18, as will be described below.

[0025] Continuing with the description of how line-scan imager 16 operates, after linear imaging sensor 18 captures a first image of a first linear portion of tissue section 14, scanning mechanism 22 moves the slide holder, thereby repositioning tissue section 14. Tissue section 14 is typically repositioned so that a second image of a next linear portion of tissue section 14, which is typically adjacent to the first linear portion, is positioned for capture by linear imaging sensor 18. In some embodiments, between these two image captures by linear imaging sensor 18, tissue sample 14 will be repositioned so that the first linear position of tissue section 14 can be captured by another linear sensor used for focusing purposes as will be described in detail below. Such repositioning of tissue section 14 is thus done in a width-wise manner (i.e., in a y-direction perpendicular to both the x- direction and z-direction), so as to cause each of the light-sensitive imaging pixels ipo, ipi . .. ipN of linear imaging sensor 18 to receive light from a location within the next linear portion of tissue section 14 that is adjacent the corresponding location within first linear portion of tissue section 14 that was previously imaged by linear imaging sensor 18. This sequential alternation of repositioning of tissue section 14 and capturing of images of linear portions of tissue section 14 can be repeated until a two-dimensional image with the desired dimensions of tissue section 14 is obtained.

[0026] The repositioning of tissue section 14 can be performed in one or two image directions. For example, scanning mechanism 22 can cause tissue section 14 to move in a width- wise image direction i.e., a direction in which images of line portions of tissue section 14 are translated in a y-direction), from which the sequence of images captured can be combined to form a two-dimension image of a rectangular portion of tissue section 14. Such a two-dimensional image is composed of a sequence of images of adjacent linear portions of tissue section 14, which were sequentially captured by linear image sensor 18. In some embodiments, this two-dimensional image can be a first rectangular image of a larger image that can be created. For example, after obtaining this first rectangular image, scanning mechanism 22 can be configured to reposition tissue section 14 in a lengthwise image direction the x-direction), which is a direction transverse the width- wide image direction used to produce the first rectangular image, thereby preparing for image capture of a linear portion of tissue section 14 that is adjacent to an end one of the images of the first rectangular image. A second rectangular image can then be obtained by altematingly capturing an image of the aligned linear section of tissue section 14 and repositioning tissue section 14 in the width- wise image direction, as was performed for fist rectangular image. The first and second rectangular images can then be concatenated to form this larger image. In this same manner, third, fourth, fifth, etc. rectangular images can be created to form ever larger images.

[0027] Even if focal plane 28 is coincident with image plane 30 during capture of a first image of a first linear portion of tissue section 14, due to thickness variations, repositioning of tissue section 14 can result in a change in location of image plane 30, thereby destroying the coincidence of image plane 30 with focal plane 28. If not corrected, after many such repositioning events of tissue section 14, a large displacement of image plane 30 from focal plane 28 can result, thereby causing images captured by linear imaging sensor 18 to be sub-optimally focused. Such a change in location of image plane 30 can be caused by differences in thickness of the linear portions of tissue section 14 at the various positions to which tissue section 14 is moved for image capture. Typically, thickness variation of tissue section 14 is gradual as adjacent linear portions of tissue section 14 have little thickness differences. Thus, if a first image of a first linear portion of tissue section 14 is at or near best focus, then so too would typically be a next image of a next linear portion adjacent to the first linear portion of tissue section 14. But, as tissue section 14 is repositioned many times, differences in image quality can arise due to sub-optimal focus, if focusing correction is not performed.

[0028] Such focusing correction of line-scan imager 16 is performed by line-scan focuser 32. Line-scan focuser 32 includes linear focusing sensor 34, focusing processor 36, and focusing mechanism 38. Linear focusing sensor 34 is canted (z.e., non-coplanar) with respect to focal plane 28, thereby rendering linear focusing sensor 34 not parallel with linear imaging sensor 18. A projection in the direction of the optical axis of linear focusing sensor 34 onto focal plane 28, however, is parallel with linear imaging sensor 18. Linear focusing sensor 34 is located such that the projection in the direction of optical axis 26 of linear focusing sensor 34 onto focal plane 28 is proximate linear imaging sensor. Linear focusing sensor 34 is offset in a y-direction from linear imaging sensor 18. Such an offset can be measured as a number of linear portions of tissue section 14 that are imaged therebetween by objective lens 20. As tissue section 14 is aligned such that linear imaging sensor 18 can capture an image of a first linear portion of tissue section 14, linear focusing sensor 34 is aligned to capture an image of a second linear portion parallel to the first linear portion of tissue section 14. An image of this second linear portion of tissue section 14 will either: i) be captured by linear imaging sensor 18 after a few repositioning events of tissue section 14; or ii) has already been captured by linear imaging sensor 18 a few repositioning events ago. Thus, each image captured by linear imaging sensor 18 can have an corresponding image (i.e., an image of the same linear portion of tissue section 14) captured by linear focusing sensor 34.

[0029] Linear focusing sensor 34 has a row of light-sensitive imaging pixels fpo, fpi ... fpN. Longitudinal axis 40 of linear focusing sensor 34 is out of plane with focal plane 28 by cant angle 0. Because linear focusing sensor 34 is canted with respect to focal plane 28, light-sensitive pixels (e.g., fpo and fpi) at a first end of linear focusing sensor 34 are located farther than focal plane 28 from objective lens 20. Also, light-sensitive pixels (e.g., fpN-i and fpN) at a second end opposite the first end of linear focusing sensor 34 are located nearer than the focal plane 28 to objective lens 20. In this way, linear focusing sensor 34 traverses focal plane 28, thereby intercepting focal plane 28 at intercept location 42 between the first and second ends of linear focusing sensor 34. Typically, such an intersection is at a central location of linear focusing sensor 34.

[0030] Because linear focusing sensor 34 is canted with respect to focal plane 28, linear focusing sensor 34 cannot be coplanar with image plane 30 of objective lens 20. Image plane 30 is necessarily perpendicular to optical axis 26, and therefore parallel with the focal planes of objective lens 20. As such, longitudinal axis 40 of linear focusing sensor 34 is out of plane with image plane 30 by cant angle 9. Also, linear focusing sensor 34 intercepts image plane 30 at intercept location 44. Images captured by linear focusing sensor 34 will enjoy optimal focus only in light-sensitive imaging pixels fpx, fpythat are located proximate intercept location 44 where linear focusing sensor 34 intercepts image plane 30. As one goes away from intercept location 44 along the row of light-sensitive imaging pixels e.g., fpo and fpi) of linear focusing sensor 34, captured image data becomes increasingly out of focus. Thus, the location (i.e., intercept location 44) of best focus of the captured image data by the row of light-sensitive imaging pixels (e.g., fpo and fpi) is indicative of the de-focus condition of linear imaging sensor 18. For example, if intercept location 44 coincides with intercept location 42, then linear imaging sensor 18 is optimally focused. If, however, intercept location 44 is not coincident with intercept location 42, then linear imaging sensor 18 is not optimally focused. Where out of focus, imagery suffers a loss of contrast. This loss of contrast (and conversely non-loss of contrast) is used to determine the location within linear focusing sensor 34 where focus is optimal.

[0031] Focusing processor 36 is configured to determine the location of best focus within linear focusing sensor 34. Focusing processor 36 determines such a location by determining the location where the contrast is best in the imagery captured by linear focusing sensor, as contrast suffers as a result of poor focus. But image contrast alone can be insufficient to determine such a location, because the nature of tissue section 14 being imaged can have high and low contrast regions. Such high and low contrast regions are then replicated in the imagery captured by linear focusing sensor 34. Imagery captured by linear imaging sensor 18, however, is uniform across the row of light-sensitive imaging pixels ipo, ipi ... ipN regardless of focusing condition. Thus, each of the row of light- sensitive imaging pixels ipo, ip i ... ipN of linear imaging sensor 18, is equally in or out of focus with the focus condition of the other light-sensitive imaging pixels ipo, ipi ... ipN of linear imaging sensor 18. As such, contrast of images captured by linear imaging sensor 18 can be used to normalize the contrast of images captured by linear focusing sensor 34.

[0032] Thus, focusing processor 36 is configured to determine a location of peak normalized contrast of linear focusing sensor 34. Focusing processor 36 does this by making a first contrast image from one or more adjacent images captured by linear focusing sensor 34. Focusing processor 36 also makes a second contrast image of the corresponding one or more adjacent images captured by linear imaging sensor 18. Because linear imaging sensor 18 and linear focusing sensor 34 are offset from one another, these contrast images will be created after correspond images are captured by both linear imaging sensor 18 and linear focusing sensor 34. In one embodiment, the one of these two contrast images that has not had its true corresponding contrast image created as of yet can be stored in memory for use when the true corresponding contrast image is created (i.e., after the tissue has been repositioned such that its image moves past the offset to the other of the linear imaging sensor 18 and linear focusing sensor 34). In another embodiment, scanning mechanism 22 can position tissue sample 14 so that a first image of a first linear portion is captured by linear imaging sensor 18 then repositioned so that a second image of the first linear portion is captured by linear focusing sensor 34. Then scanning mechanism repositions tissue sample 14 to repeat this image capturing for an adjacent linear portion. Both of these methods can be described as synchronizing acquisition of images. Contrast images formed using such synchronized images truly correspond with one another.

[0033] Regardless of the embodiment, focusing processor 36 then takes a ratio of the first contrast image and the second contrast image (using either true corresponding or approximately corresponding contrast images). Such a ratio results in a normalized-contrast image. A peak contrast location along the row of the normalized-contrast image data identifies the location (e.g., between fpxand fpy) along the row of light-sensitive imaging pixels fpo, fpi ... fpN near intercept location 44 where linear focusing sensor 34 intercepts image plane 30. A distance along linear focusing sensor 34 between such a peak contrast location and intercept location 42 is indicative of the focusing condition of linear imaging sensor 18. If, for example, the peak contrast location coincides with intercept location 42, then linear imaging sensor is coplanar with image plane 30, and therefore linear imaging sensor is optimally focused. If, however, the peak contrast location is not coincident with intercept location 42, then the distance AF therebetween is indicative of the out of focus condition of linear imaging sensor 18.

[0034] There are various ways to make focusing corrections to line-scan imager 16. For example, linear imaging sensor 18 can be translated in the z-direction to the location of image plane 30 via an image sensor translation mechanism. In another embodiment, tissue section 14 can be translated in the z-direction via a slide holder translation mechanism. In the embodiment depicted, objective lens 20 is translated in the z-direction via focusing mechanism 38. Focusing mechanism 38 is configured to position the objective lens relative to the linear imaging sensor 18 (and relative to the tissue section 14) based on the location of best focus as determined by focusing processor 36. In such an embodiment, for example, focusing processor 36 can determine the distance AD between image plane 30 and focal plane 28. Using basic trigonometry:

[0035] AD = AFsin(0) (1)

[0036] Knowing the distance^? between objective lens 20 and focal plane 28, focusing processor 36 can then determine the distance u between objective lens 20 and image plane 30 as: u = fp — D (2)

[0037] Knowing the distance u as well as the focal length / of objective lens 20, focusing processor 36 can then determine, using the thin lens equation, for example, the distance v between objective lens 20 and the surface of tissue section 14 as: Knowing all the above, focusing processor 36 can determine a change in location AL of objective lens 20 along the optical axis for best focus line-scan imager 28, using again, for example, the thin lens equation:

[0038] — fp+ —L + — v- L = - f (4)

[0039] Equation (4) reduces to a quadratic in AL, which focusing processor 36 can readily solve.

[0040] The above method of determining a metric of focus of images captured by linear imaging sensor 18 is a simple example for an embodiment in which focus correction is performed by translating objective lens 20 along optical axis 26. Different equations might govern such focus correction for different optical systems and methods. Moreover, for different methods of focus correction, the calculations performed are those appropriate to the method employed for focus correction.

[0041] Linear imaging sensor 18 and linear focusing sensor 34 share the optical system designed for imaging of tissue section 14. Such a shared optical system enjoys various advantages. For example, the cost of such a line-scan imager that has intra-image focus -correction capability can be small in comparison with imagers employing separate optical systems for imaging and focus correction. Such a combined system can be easier to calibrate, as both linear imaging sensor 18 and linear focusing sensor 34 can be easily and precisely located regarding optical paths to each sensor. The size of such a combined system as line-scan imager 16 can be small with respect to imagers employing separate optical systems for imaging and focus correction. Such advantages as mentioned immediately above can be translated into cost savings, maintenance requirements, reliability, etc.

[0042] FIGS. 4A and 4B are images of a tissue section captured by a linear imaging sensor and a linear focusing sensor of an auto-focusing line-scan imager, respectively. In FIG. 4A, first rectangular image 46 is a concatenation of a sequence of images captured by linear imaging sensor 18 (depicted in FIG. 2). First rectangular image 46 depicts a rectangular region of a tissue section 14. Most of the contrast in first rectangular image 46 is due to the actual contrast of light reflected from different regions of tissue section 14. These different regions of tissue section 14 have different colors and different measures of reflectivity. First rectangular image 46 was captured by linear imaging sensor 18 when linescan imager 16 was not optimally focused, as will be explained below. Linear imaging sensor 18 can be any kind of a linear optical sensor, such as, for example, a monochromatic linear optical sensor or a color linear optical sensor. In FIG. 4B, second rectangular image 48 is a concatenation of a sequence of images captured by linear focusing sensor 34 (depicted in FIGS. 2-3B). Second rectangular image 48 depicts the same rectangular region of tissue section 14 as was depicted in first rectangular image 46 shown in FIG. 4A. Most of the contrast in second rectangular image 48 is again due to the actual contrast of light reflected from different regions of tissue section 14. Second rectangular image 48 was captured by linear focusing sensor 34, which transverses image plane 30, and thus has some of the light-sensitive imaging pixels fpo, fpi ... fpN that are at or near optimal or best focus location. Others of the light-sensitive imaging pixels fpo, fpi . . . fpN of linear focusing sensor 34 will not be at a best focus location. Thus, image data produced by these sub-optimally located lightsensitive imaging pixels fpo, fpi ... fpN will produce images with lower contrast than would in-focus pixels producing image data from the same region of tissue section 14. Linear focusing sensor 34 also can be any kind of a linear optical sensor, such as, for example, a monochromatic linear optical sensor or a color linear optical sensor.

[0043] Linear imaging sensor 18 and linear focusing sensor 34 need not be the same type of sensor. For example, linear imaging sensor 18 can be a color linear optical sensor, while linear focusing sensor 34 can be a monochromatic linear optical sensor. While these two sensors (i.e., linear imaging sensor 18 and linear focusing sensor 34) may be different from one another, contrast of their imagery should be comparable such that proper normalization of the contrast images created from images captured by linear focusing sensor 34 can be obtained. Thus, if linear imaging sensor 18 is a color linear optical sensor, while linear focusing sensor 34 is a monochromatic optical sensor, then normalization of images captured by linear focusing sensor 34 should use a monochromatic combination of the color channels of linear imaging sensor 18.

[0044] FIGS. 5 is a graph of the contrast image for each of the images captured by the linear imaging sensor and the linear focusing sensor. In FIG. 5, graph 50 includes horizontal axis 52, vertical axis 54, first contrast / pixel-index relation 56 and second contrast / pixel-index 58. Horizontal axis 52 is indicative of index n of light-sensitive pixels for both the row of light-sensitive imaging pixels ipo, i i ... ipN of linear imaging sensor 18 and the row of light-sensitive imaging pixels fpo, fpi ... fpN of linear focusing sensor 34. Vertical axis 54 is indicative of a metric of contrast of contrast images formed from images captured by each of linear imaging sensor 18 and linear focusing sensor 34. First contrast / pixel-index relation 56 is data from a row of a contrast image formed from first rectangular image 46 (depicted in FIG. 4A) captured by linear imaging sensor 18. Second contrast / pixel-index relation 58 is data from a row of a contrast image formed from second rectangular image 48 (depicted in FIG. 4B) captured by linear focusing sensor 34. Variations in contrast are evident in both first and second contrast / pixel-index relations 56 and 58. Note, however, that first contrast / pixel-index relation 56 is greater than second contrast / pixel-index relation 58 at both ends of the row of their respective light-sensitive imaging pixels. Specifically, first contrast / pixel-index relation 56 is less than second contrast / pixel-index relation 58 in the central region of the row of their respective lightsensitive imaging pixels. Such relative magnitudes of contrast indicate that second rectangular image 48 captured by linear focusing sensor 34 is highest somewhere in that central region of the row of light-sensitive imaging pixels fpo, fp i . . . fpN of linear focusing sensor 34.

[0045] FIG. 6 is a graph of a ratio of contrast images taken by the linear imaging sensor and the linear focusing sensor as fit to a peak-locating curve. In FIG. 6, graph 60 includes horizonal axis 62, vertical axis 64, normalized-contrast / pixel-index relation 66 and peak-locating curve 68. Horizontal axis 62 is indicative of index n of light sensitive pixel for both the row of light-sensitive imaging pixels ipo, ipi ... ipN of linear imaging sensor 18 and the row of light-sensitive imaging pixels fpo, fpi ... fpN of linear focusing sensor 34. Vertical axis 64 is indicative of a metric of normalized contrast of a normalized contrast image formed from a ratio of contrast images formed from images captured by each of linear imaging sensor 18 and linear focusing sensor 34. Normalized-contrast / pixel-index relation 66 is a ratio of second contrast / pixel-index relation 58 and first contrast / pixel-index 56, which are depicted in FIG. 5. Normalized-contrast / pixel-index relation 66 has some high frequency variation superimposed upon a curve that exhibits a broad peak with diminishing tails at each end. Peak-locating curve 68 is a model function, such as, for example, a Gaussian function. Such a Gaussian function has parameters that can be adjusted to best fit normalized-contrast / pixel-index relation 66. Such parameters control the magnitude of the peak, the location of the peak as well as the breadth of the peak. From the fit peak-finding function, the parameter indicative of the location of the peak is indicative of the location within linear focusing sensor 34 where optimal focus is found.

[0046] FIG. 7 is a flow chart of an example of a method for focusing a line-scan imager while imaging in real time. In FIG. 7, method 70 begins at step 72 where objective lens 20 is initially positioned along optical axis 26 (which are depicted in the earlier figures). Method 70 then advances to steps 74 and 78, where linear imaging sensor 18 and linear focusing sensor 34 capture first and second images, respectively. From step 74, method 70 then advances to step 76, and from step 78, method 70 advances to step 80, where contrast curves or images are generated for each of the first and second images captured in steps 74 and 78, respectively. Contrast images can be formed in a variety of manners. For example, a gradient kernel can be convolved with the first and second images or image sections (i.e., group of adjacent images). One-dimensional gradient kernels can be any size, from 1x2 to 1x16 or even larger. Two-dimensional gradient kernels can be any size as well, from 2x2 to 16x16 or even larger. After such gradient image data is generated, that gradient image data can be squared (and the square root of the result can also be taken, if desired), so as to generate contrast data. Various other methods of generating contrast images can be performed as are known in the art. After generating such contrast images, method 70 then advances to step 82 from both steps 76 and 80, where a ratio of the contrast images generated in steps 76 and 80 is formed. Then, at step 84, a peak-finding curve is fit to the ratio formed at step 82. Then at step 86, objective lens 18 is moved to a calculated position along optical axis 26, based on the location of the peak of the peak-finding curve fit to the ratio at step 84.

[0047] FIGS. 8A and 8B are illustrations of alternate embodiments of configuring a linear focusing sensor for determining a metric of focus of a line-scan imager. In FIG. 8A, linear focusing sensor 34 has prism 88 interposed above the row of light-sensitive imaging pixels fpo, fpi ... fpN- Prism 88 causes the optical path length to vary across the row of light-sensitive imaging pixels fpo, fpi ... fpN. Such linear varying of the optical path length across the row of light-sensitive imaging pixels fpo, fpi ... fpN. can be used as an alternative to canting linear focusing sensor 34 as depicted in FIGS. 2-3A. In FIG. 8B individual glass blocks pmo, pmi ... pmN are situated immediately above each of lightsensitive imaging pixels fpo, fpi ... fpN. These individual glass blocks pmo, pmi . . . pmN are sized so that the optical path length linearly increases across the row of light-sensitive imaging pixels fpo, fpi ... fpN of linear imaging sensor 40.

[0048] These embodiments, depicted in FIGS. 8A and 8B, can be relatively easy to manufacture and calibrate, and can permit linear focusing sensor 34 to be positioned parallel with linear imaging sensor 18. Such embodiments can permit linear focusing sensor 34 to be positioned parallel with linear imaging sensor as well. For example, both linear imaging sensor 18 and linear focusing sensor 34 can be mounted to a common circuit board. In some embodiments, a spacer or standoff can be interposed between the linear focusing sensor 34 and such a circuit board, thereby raising linear focusing sensor 34 above an elevation of linear imaging sensor 18. Such relative elevations of linear imaging sensor 18 and linear focusing sensor 34 can be configured to substantially equalize the optical paths of light-sensitive imaging pixels fp(N-n / 2 and fp(N+n / 2 at or near the center of imaging focusing sensor 40 and of light-sensitive imaging pixels ipo, ipi ... ipN of linear imaging sensor 18.

[0049] Discussion of Possible Embodiments

[0050] The following are non-exclusive descriptions of possible embodiments of the present invention.

[0051] Some embodiments are related to a system for imaging a tissue section with intra-image focus correction. The system includes a line-scan imager, and a line-scan focuser. The line -scan imager includes a linear imaging sensor having a first row of lightsensitive pixels. The line-scan imager also includes an objective lens having an optical axis oriented such that a focal plane of the objective lens is coplanar with the first row of light- sensitive pixels. The line-scan focuser includes a linear focusing sensor having a second row of light-sensitive pixels that is canted with respect to the focal plane of the objective lens such that light-sensitive pixels at a first end of the linear focusing sensor are located farther than the focal plane from the objective lens and light-sensitive pixels at a second end opposite the first end of the linear focusing sensor are located nearer than the focal plane to the objective lens. The line-scan focuser includes a focusing processor configured to determine a location of best focus within the linear focusing sensor based on a comparison of images obtained by the linear imaging sensor and the linear focusing sensor. The line-scan focuser also includes a focusing mechanism configured to position the objective lens relative to the linear imaging sensor based on the location of best focus as determined by the focusing processor.

[0052] The system of the preceding paragraph can optionally include, additionally and / or alternatively, any one or more of the following features, configurations, and / or additional components:

[0053] A further embodiment of the foregoing system can further include a scanning mechanism configured to position the tissue section being scanned relative to the line-scan imager and further configured to move the tissue section being scanned in a scanning direction that is transverse to both the optical axis and to the first row of light- sensitive pixels of the linear imaging sensor.

[0054] A further embodiment of any of the foregoing systems, wherein the objective lens can be configured to concurrently focus a first linear portion of the tissue section onto the linear imaging sensor and a second linear portion of the tissue section onto the linear focusing sensor.

[0055] A further embodiment of any of the foregoing systems, wherein the first and second linear portions of the tissue section that are concurrently imaged by the linear imaging sensor and the linear focusing sensor, respectively, can be oriented perpendicular to the scanning direction.

[0056] A further embodiment of any of the foregoing systems, wherein the linear imaging sensor and the linear focusing sensor can be oriented such that the first and second linear portions of the tissue section, which are concurrently focused thereon, are parallel to one another.

[0057] A further embodiment of any of the foregoing systems, wherein the linear imaging sensor and the linear focusing sensor can be oriented such that the first and second images of linear portions of the tissue section can be sequentially captured by the linear imaging sensor and the linear focusing sensor as the tissue section is positioned at first and second positions relative to the line-scan imager, respectively.

[0058] A further embodiment of any of the foregoing systems, wherein each of the light-sensitive pixels of the linear imaging sensor can correspond to a corresponding one of the light-sensitive pixels of the linear focusing sensor. Each pair of corresponding lightsensitive pixels is located so as to be able to sequentially image the same region of the tissue section as the tissue section is positioned at first and second positions relative to the line-scan imager, respectively.

[0059] A further embodiment of any of the foregoing systems, wherein the focusing mechanism can be configured to move the objective lens relative to the linear imaging sensor in directions parallel to the optical axis thereby changing focus of the line-scan imager.

[0060] A further embodiment of any of the foregoing systems, wherein the light- sensitive pixels of the linear focusing sensor can be located at linearly varying distances from the objective lens with the nearest of the light-sensitive pixels in a nearest focal plane to the objective lens and a farthest of the light-sensitive pixels in a farthest focal plane from the objective lens.

[0061] A further embodiment of any of the foregoing systems, wherein the objective lens can be configured to focus the tissue section on a first side of the objective lens onto an image plane on a second side opposite the first side of the objective lens. A further embodiment of any of the foregoing systems, wherein the location of best focus can correspond to a location of a focused one of the second row of lightsensitive pixels of the linear focusing imager.

[0062] A further embodiment of any of the foregoing systems, wherein the focused one of the second row of light-sensitive pixels can be located where the image plane intersects the row of light-sensitive pixels of the linear focusing sensor.

[0063] A further embodiment of any of the foregoing systems, wherein the focusing processor can be further configured to: i) determine a first contrast image of a first image captured by the linear imaging sensor, the first contrast image being a function of pixel location of the linear imaging sensor; ii) determine a second contrast image of a second image captured by the linear focusing sensor, the second contrast image being a function of pixel location of the linear focusing sensor; and / or iii) determine a ratio of the second contrast image to the first contrast image as a function of pixel locations of each of the linear focusing sensor and the linear imaging sensor.

[0064] A further embodiment of any of the foregoing systems, wherein the focusing processor can be further configured to: i) determine the first contrast image as a function of pixel location of the linear imaging sensor based, at least in part, on the first image; ii) determine the second contrast image as a function of pixel location of the linear focusing sensor based, at least in part, on the second image; and / or iii) determine a ratio function of the second contrast image to the first contrast image as a function of pixel locations of each of the linear focusing sensor and the linear imaging sensor.

[0065] A further embodiment of any of the foregoing systems, wherein the focusing processor can be further configured to fit the ratio function to a peak-locating function.

[0066] A further embodiment of any of the foregoing systems, wherein the peaklocating function, as fit to the ratio function, can identify a location of best focus, such that a distance defined between the objective lens and a pixel at the location of best focus is a best focal distance.

[0067] A further embodiment of any of the foregoing systems, wherein the focusing mechanism can be further configured to position the objective lens relative to the linear imaging sensor at the best focal distance.

[0068] Some embodiments relate to a method for imaging a tissue section using intra-image focus correction. In the method, an objective lens is positioned relative to a linear imaging sensor via a focusing mechanism. A first image of a linear portion of the tissue section is captured via the linear imaging sensor coplanar with a focal plane of the objective lens. A second image of the linear portion of the tissue section is captured via a linear focusing sensor that is canted with respect to the focal plane. A location of best focus within the linear focusing sensor is determined. The tissue section is repositioned relative to the linear imaging sensor and the linear focusing sensor in a direction transverse to each of the optical axis and the linear imaging sensor. The objective lens is repositioned relative to the linear imaging sensor, based on the location of best focus, thereby refocusing the tissue section onto the focal plane. The above steps, except the first position of the objective lens, are repeated to obtain a two-dimensional image of the tissue section.

[0069] The method of the preceding paragraph can optionally include, additionally and / or alternatively, any one or more of the following features, configurations, and / or additional components:

[0070] A further embodiment of the foregoing method can further include: i) determining a first contrast image of a first image captured by the linear imaging sensor, the first contrast image being a function of pixel location of the linear imaging sensor; ii) determining a second contrast image of a second image captured by the linear focusing sensor, the second contrast image being a function of pixel location of the linear focusing sensor; and / or iii) determining a ratio of the second contrast image to the first contrast image as a function of pixel locations of each of the linear focusing sensor and the linear imaging sensor.

[0071] A further embodiment of any of the foregoing methods can further include: i) determining the first contrast image as a function of pixel location of the linear imaging sensor based, at least in part, on the first image; ii) determining the second contrast image as a function of pixel location of the linear focusing sensor based, at least in part, on the second image; and / or iii) determining a ratio function of the second contrast image to the first contrast image as a function of pixel locations of each of the linear focusing sensor and the linear imaging sensor.

[0072] While the invention has been described with reference to an exemplary embodiment(s), it will be understood by those skilled in the art that various changes may be made, and equivalents may be substituted for elements thereof without departing from the scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention is not limited to the particular embodiment(s) disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.

Claims

CLAIMS:

1. A system for imaging a tissue section with intra-image focus correction, the system comprising: a line-scan imager including: a linear imaging sensor having a first row of light-sensitive pixels; and an objective lens having an optical axis oriented such that a focal plane of the objective lens is coplanar with the first row of light-sensitive pixels; and a line-scan focuser including: a linear focusing sensor having a second row of light-sensitive pixels that is canted with respect to the focal plane of the objective lens such that light-sensitive pixels at a first end of the linear focusing sensor are located farther than the focal plane from the objective lens and light-sensitive pixels at a second end opposite the first end of the linear focusing sensor are located nearer than the focal plane to the objective lens; a focusing processor configured to determine a location of best focus within the linear focusing sensor based on a comparison of images obtained by the linear imaging sensor and the linear focusing sensor; and a focusing mechanism configured to position the objective lens relative to the linear imaging sensor based on the location of best focus as determined by the focusing processor.

2. The system of claim 1 , further comprising: a scanning mechanism configured to position the tissue section being scanned relative to the line-scan imager and further configured to move the tissue section being scanned in a scanning direction that is transverse to both the optical axis and to the first row of light- sensitive pixels of the linear imaging sensor.

3. The system of claim 2, wherein the objective lens is configured to concurrently focus a first linear portion of the tissue section onto the linear imaging sensor and a second linear portion of the tissue section onto the linear focusing sensor.

4. The system of claim 3, wherein the first and second linear portions of the tissue section that are concurrently imaged by the linear imaging sensor and the linear focusing sensor, respectively, are oriented perpendicular to the scanning direction.

5. The system of claim 3, wherein the linear imaging sensor and the linear focusing sensor are oriented such that the first and second linear portions of the tissue section, which are concurrently focused thereon, are parallel to one another.

6. The system of claim 5, wherein the linear imaging sensor and the linear focusing sensor are oriented such that the first and second images of linear portions of the tissue section can be sequentially captured by the linear imaging sensor and the linear focusing sensor as the tissue section is positioned at first and second positions relative to the linescan imager, respectively.

7. The system of claim 6, wherein each of the light-sensitive pixels of the linear imaging sensor corresponds to a corresponding one of the light-sensitive pixels of the linear focusing sensor, each pair of corresponding light-sensitive pixels located so as to be able to sequentially image the same region of the tissue section as the tissue section is positioned at first and second positions relative to the line-scan imager, respectively.

8. The system of claim 1 , wherein the focusing mechanism is configured to move the objective lens relative to the linear imaging sensor in directions parallel to the optical axis thereby changing focus of the line-scan imager.

9. The system of claim 1, wherein the light-sensitive pixels of the linear focusing sensor are located at linearly varying distances from the objective lens with the nearest of the light-sensitive pixels in a nearest focal plane to the objective lens and a farthest of the light-sensitive pixels in a farthest focal plane from the objective lens.

10. The system of claim 9, wherein the objective lens is configured to focus the tissue section on a first side of the objective lens onto an image plane on a second side opposite the first side of the objective lens.

11. The system of claim 10, wherein the location of best focus corresponds to a location of a focused one of the second row of light-sensitive pixels of the linear focusing imager.

12. The system of claim 11 , wherein the focused one of the second row of lightsensitive pixels is located where the image plane intersects the row of light-sensitive pixels of the linear focusing sensor.

13. The system of claim 1 , wherein the focusing processor is further configured to: determine a first contrast image of a first image captured by the linear imaging sensor, the first contrast image being a function of pixel location of the linear imaging sensor; determine a second contrast image of a second image captured by the linear focusing sensor, the second contrast image being a function of pixel location of the linear focusing sensor; and determine a ratio of the second contrast image to the first contrast image as a function of pixel locations of each of the linear focusing sensor and the linear imaging sensor.

14. The system of claim 13, wherein the focusing processor is further configured to: determine the first contrast image as a function of pixel location of the linear imaging sensor based, at least in part, on the first image; determine the second contrast image as a function of pixel location of the linear focusing sensor based, at least in part, on the second image; and determine a ratio function of the second contrast image to the first contrast image as a function of pixel locations of each of the linear focusing sensor and the linear imaging sensor.

15. The system of claim 14, wherein the focusing processor is further configured to: fit the ratio function to a peak-locating function.

16. The system of claim 15, wherein the peak-locating function, as fit to the ratio function, identifies a location of best focus, such that a distance defined between the objective lens and a pixel at the location of best focus is a best focal distance.

17. The system of claim 16, wherein the focusing mechanism is further configured to position the objective lens relative to the linear imaging sensor at the best focal distance.

18. A method for imaging a tissue section using intra-image focus correction, the method comprising the steps of: i) positioning, via a focusing mechanism, an objective lens relative to a linear imaging sensor; ii) capturing, via the linear imaging sensor coplanar with a focal plane of the objective lens, a first image of a linear portion of the tissue section;iii) capturing, via a linear focusing sensor that is canted with respect to the focal plane, a second image of the linear portion of the tissue section; iv) determining a location of best focus within the linear focusing sensor; v) repositioning the tissue section relative to the linear imaging sensor and the linear focusing sensor in a direction transverse to each of the optical axis and the linear imaging sensor; vi) repositioning the objective lens relative to the linear imaging sensor, based on the location of best focus, thereby refocusing the tissue section onto the focal plane; and vii) repeating steps ii) - vi) to obtain a two-dimensional image of the tissue section.

19. The method of claim 18 further comprising: determining a first contrast image of a first image captured by the linear imaging sensor, the first contrast image being a function of pixel location of the linear imaging sensor; determining a second contrast image of a second image captured by the linear focusing sensor, the second contrast image being a function of pixel location of the linear focusing sensor; and determining a ratio of the second contrast image to the first contrast image as a function of pixel locations of each of the linear focusing sensor and the linear imaging sensor.

20. The method of claim 19 further comprising: determining the first contrast image as a function of pixel location of the linear imaging sensor based, at least in part, on the first image; determining the second contrast image as a function of pixel location of the linear focusing sensor based, at least in part, on the second image; and determining a ratio function of the second contrast image to the first contrast image as a function of pixel locations of each of the linear focusing sensor and the linear imaging sensor.

Images