Method and system to obtain images of target cells in cytopathology

Abstract

Example methods and systems for capturing images of target cells in cytological specimens have been disclosed. One example method includes: acquiring a first set of Z-stacked images and a second set of Z-stacked images; generating a first focal map; capturing target cell images based on the first focal map; and generating a second focal map. The distance between the first set of target cells associated with the first focal map and the second set of target cells associated with the second focal map along the z-axis of the cytological specimen is different from the distance between the third set of target cells associated with the first focal map and the fourth set of target cells associated with the second focal map along the z-axis of the cytological specimen.

Description

Technical Field

[0001] [Reference to Related Applications] This application claims priority to U.S. Provisional Patent Application No. 63 / 652,051, filed May 26, 2024, the entire contents of which are incorporated herein by reference.

[0002] Several specific embodiments of the present invention are generally related to methods and systems for obtaining cytological images in cytopathology. Prior Technology

[0003] Unless otherwise stated herein, the methods described in this section are not prior art to the claims in this application and cannot be acknowledged as prior art simply because they are included in this section.

[0004] Cytopathology is a branch of pathology that specializes in the study and diagnosis of diseases at the cellular level, typically involving obtaining cytological images at the cellular level. Obtaining cytological images of a cytological specimen requires digitizing the image of the specimen spread on a glass slide. This digitization process usually involves scanning the slide using a whole-slide imaging scanner to generate a whole-slide image (WSI). This whole-slide image can then be viewed on a display device (such as a computer screen) instead of a microscope.

[0005] Accurately and efficiently capturing target cells on glass slides using image digitization technology is challenging. Traditional digital imaging methods typically establish one or more focal maps based on the focal positions of multiple regions on the slide and attempt to capture target cell images based on these focal maps. However, traditional methods often set different focal maps to maintain arbitrary spacing between them, without considering the three-dimensional distribution characteristics and different depth distribution of target cells in the cytological specimen suspension. Therefore, relying on traditional focal maps to capture target cells distributed on glass slides is not only inefficient but also lacks accuracy, resulting in poor image quality of the captured target cells. Summary of the Invention

[0006] none Simple Explanation of the Diagram

[0007] Figure 1 is an example diagram showing that the cytological specimens are distributed in three-dimensional space and arranged on a glass slide according to several embodiments disclosed herein; Figure 2 illustrates how an example scanner acquires images associated with a cytological specimen, arranged according to several embodiments disclosed herein; Figure 3 is an exploded view of an example cytological specimen, including the components configured in accordance with several embodiments disclosed herein; Figure 4A is a conventional exploded view of an example cytological specimen; Figure 4B shows another conventional exploded view of an example cytological specimen; Figure 5A is an exploded view of an example cytological specimen, the configuration of which conforms to several embodiments of this disclosure; Figure 5B illustrates another breakdown diagram of a conventional cytological specimen, the configuration of which conforms to some embodiments disclosed herein; Figure 6 illustrates an exemplary scanner configured to generate one or more focus maps on a cytological specimen and acquire images of target cells in the cytological specimen based on the focus maps, the configuration of which conforms to several embodiments of this disclosure; Figure 7 is a flowchart illustrating an exemplary process for obtaining images of target cells in a cytological specimen, the configuration of which conforms to several embodiments disclosed herein; Figure 8 shows an exemplary apparatus configured according to several embodiments of the present disclosure for performing various embodiments of the present disclosure. Implementation

[0008] In the following detailed description, reference will be made to the accompanying drawings, which form a part thereof. In the drawings, like symbols generally identify like components unless otherwise stated herein. The illustrative specific embodiments described in the embodiments, the brief description of the drawings, and the claims are not limiting. Other specific embodiments and other changes may be utilized without departing from the spirit or scope of the subject matter presented herein. It will be readily understood that the forms of this disclosure, as generally described in this specification and illustrated in the drawings, can be configured, replaced, combined, and designed in a variety of different configurations, all of which are explicitly considered herein. In the illustration, a cytological specimen is considered suspicious when the target cells distributed in the cytological specimen space include cells at risk of disease.

[0009] In this disclosure, a cytological specimen comprises multiple two-dimensional regions in the Cartesian xy-plane, each of which is referred to as a "region". The "z-stacked image" of a region refers to a series of images of that region captured at different focal positions along the z-axis (i.e., depth of field) of the cytological specimen. The "focal image" refers to the three-dimensional representation formed by the optimal focal positions of multiple regions in the cytological specimen.

[0010] Figure 1 is an example diagram illustrating the distribution of the cytological specimen 110 in three-dimensional space on a glass slide 120 according to a partial embodiment disclosed herein. The cytological specimen 110 may contain multiple cells and impurities. For example, the cytological specimen 110 may contain dust or markers 131, target cells 141 (e.g., epithelial cells at risk of malignancy), and non-target cells 151, 153, 155, 157, and 159 (e.g., red blood cells, normal epithelial cells, and other cells). The dust or markers 131, target cells 141, and non-target cells 151, 153, 155, 157, and 159 are distributed in three-dimensional space on the glass slide 120 (i.e., located at different depths within the volume of the cytological specimen 110).

[0011] Figure 2 illustrates how an exemplary scanner 220 acquires images associated with a cytological specimen 210, configured in accordance with embodiments disclosed herein. Referring to Figure 1, the cytological specimen 210 shown in Figure 2 may correspond to a cytological specimen 110. In Figure 2, the scanner 220 is configured to acquire associated images of one or more layers of the cytological specimen 210. In some embodiments, each layer may correspond to a specific depth of field of an objective lens included with the scanner 220.

[0012] In some embodiments, an objective lens included in the scanner 220 has a first field of view and a first depth of field. The first field of view may correspond to any of the regions A, B, C, D, E, F, G, H, I, and J of the cytological specimen 210.

[0013] In some embodiments, within the first field of view, the distance between the nearest and farthest clearly focused object of the scanner 220 is referred to as the first depth of field. Some example distances can be seen with reference to d1, d2, d3, and d4 shown in Figure 2. Therefore, d1 corresponds to the first layer of the cytological specimen 210; d2 corresponds to the second layer of the cytological specimen 210; d3 corresponds to the third layer of the cytological specimen 210; and d4 corresponds to the fourth layer of the cytological specimen 210.

[0014] Referring to Figure 1, the first layer may be the layer furthest from a glass slide (e.g., glass slide 120), and the fourth layer may be the layer immediately adjacent to that glass slide (e.g., glass slide 120). It should be noted that although Figure 2 shows a four-layer structure, the cytological specimen 210 may contain more or fewer layers depending on the first depth of field of the objective lens included in the scanner 220. In these embodiments, a region and a distance can jointly define a "part" of the cytological specimen 210. For example, J3 is the portion of the cytological specimen 210 corresponding to region J and distance d3. Therefore, in the embodiment shown in Figure 2, the cytological specimen 210 can be defined by 40 parts, namely A1, A2, A3, A4, B1, B2, B3, B4, C1, C2, C3, C4, D1, D2, D3, D4, E1, E2, E3, E4, F1, F2, F3, F4, G1, G2, G3, G4, H1, H2, H3, H4, I1, I2, I3, I4, J1, J2, J3, and J4.

[0015] In some embodiments, FIG3 illustrates an exploded view of a cytological specimen 300, which includes various components arranged in accordance with embodiments disclosed herein. In some embodiments, cytological specimen 300 may correspond to cytological specimen 210. In some embodiments, for ease of illustration, components A1, B1, C1, D1, E1, F1, G1, H1, I1, and J1 correspond to the first layer of cytological specimen 300 and are located within the same depth of field as shown in FIG3. In some embodiments, for illustrative purposes, components A1, A2, A3, and A4 correspond to different layers of cytological specimen 300 associated with the same region A along a Z-axis, as shown in FIG3.

[0016] Referring to Figure 2, in some embodiments, the cytological specimen 210 may include, but is not limited to, dust 211, markers 212, first-type cells 213 (e.g., epithelial cells at risk of malignancy), second-type cells 214 (e.g., red blood cells), third-type cells 215 (e.g., normal epithelial cells), and fourth-type cells 216 (e.g., other cells). Dust 211, first-type cells 213, second-type cells 214, third-type cells 215, and fourth-type cells 216 may be distributed throughout the cytological specimen 210. Markers 212 may be manually marked on the top surface of the cytological specimen 210 by a user. Dust 211 and cells 213, 214, 215, and 216 may be distributed in different layers of the cytological specimen 210. In some embodiments, cells 213 are target cells, while cells 214, 215, and 216 are non-target cells.

[0017] In some embodiments, before acquiring images related to the cytology specimen 210 using the scanner 220, the scanner 220 is configured to generate one or more focus maps in multiple regions of the cytology specimen 210. Each focus map is used to guide the scanner 220 to dynamically adjust its focus as it moves to different regions of the cytology specimen 210, ensuring that each region of the cytology specimen 210 can be imaged according to the focus information specified by the focus map.

[0018] To generate a focal image, scanner 220 is configured to randomly select regions on the cytological specimen 210 and determine an optimal focal position for each selected region based on some imaging characteristics of the scanned region. Alternatively, scanner 220 can be configured to select regions on the cytological specimen 210 based on information obtained from a previous scan (e.g., a low-magnification scan) and determine an optimal focal position for each selected region based on some imaging characteristics of the scanned region. For example, scanner 220 can be set to randomly select regions B, D, F, H, I, and J from all regions and acquire relevant images of these selected regions. For instance, scanner 220 is set to acquire images of portions H1, H2, H3, and / or H4 of randomly selected region H. Here, region H corresponds to a first field of view of the objective lens included in scanner 220, and d1, d2, d3, or d4 correspond to a first depth of field range of the objective lens. Similarly, scanner 220 can also acquire images of some of B1, B2, B3, B4, D1, D2, D3, D4, F1, F2, F3, F4, I1, I2, I3, I4, J1, J2, J3 and J4 in cytological specimen 210.

[0019] Traditionally, scanner 220 is configured to determine the optimal focus position for a region based on the contrast, edge, and color values ​​of images of related parts. Continuing with the example of region H, after acquiring images of parts H1, H2, H3, and H4, scanner 220 compares the images of parts H1, H2, H3, and H4 and determines which image has the highest contrast, the sharpest edges, or the brightest color values. Referring to Figure 2, assuming that the image of part H2 has the highest contrast among the images of related parts in region H, scanner 220 sets this depth of field of part H2 as the optimal focus position for region H. Similarly, for ease of explanation, scanner 220 also sets the depth of field of parts B2, D2, F2, I2, and J2 as the optimal focus positions for regions B, D, F, I, and J, respectively. Figure 4A shows a conventional exploded view of cytological specimen 400. In Figure 4A, the depth of field of locations B2, D2, H2, F2, I2, and J2 correspond to the optimal focal positions of the first focal image within regions B, D, H, F, I, and J, respectively.

[0020] Traditionally, for areas not scanned by scanner 220 (i.e., non-scanned areas), scanner 220 estimates the optimal focal position of these non-scanned areas (e.g., areas A, C, E, and G) based on the optical focal position of scanned areas (e.g., areas B, D, F, H, I, and J). This estimation can employ several technically feasible methods, such as linear or bilinear interpolation, triangulation, or more advanced computational methods, to generate a smooth, continuous master focus map covering the entire cytological specimen. Figure 4A also shows a conventional exploded view of the cytological specimen 400, where the depth of field of sites A2, C2, E2, and G2 is estimated as the optimal focal position of the first focus map for areas A, C, E, and G, respectively. Since the first focus map is associated with the highest contrast, sharpest edges, or brightest color values, it can also be called the "master focus map." After generating the master focus map, scanner 220 focuses on all areas of the cytological specimen based on this master focus map to obtain images of each area of ​​the cytological specimen.

[0021] Figure 4B illustrates another conventional exploded view of the cytology specimen 400. After generating the primary focus image, the scanner 220 can be configured to generate one or more additional secondary focus images. These additional secondary focus images are generated by offsetting the optimal focus position of the primary focus image by a fixed distance, allowing the scanner 220 to capture images from different layers of the cytology specimen 400. In other words, the conventional approach is to generate such secondary focus images in parallel with the primary focus image, meaning that the distance between any corresponding focus position in the primary and secondary focus images remains constant throughout the entire cytology specimen. This method aims to extract more comprehensive image information from each layer of the cytology specimen. This distance can be set by the user based on experience. For example, as shown in Figure 4B, the depth of field of sites A4, B4, C4, D4, E4, F4, G4, H4, I4, and J4 can be determined as the optimal focus position of the secondary focus image for all regions A, B, C, D, E, F, G, H, I, and J in the cytology specimen 400, respectively. After generating the secondary focus image, the scanner 220 will focus on all areas of the cytology specimen 400 based on the secondary focus image, thereby acquiring additional images of each area of ​​the cytology specimen 400.

[0022] However, traditional methods have drawbacks. First, the primary focus image is generated based on the highest contrast, sharpest edges, and / or brightest color values. But the highest contrast, sharpest edges, and / or brightest color values ​​do not necessarily correspond to the presence of target cells. For example, referring to Figures 2 and 4A, target cells are mainly distributed in the third layer of the cytology specimen 210. However, the primary focus image is generated in the second layer of the cytology specimen 210, resulting in a reduced number of images of target cells distributed in the cytology specimen 210 obtained based on the primary focus image. Second, the secondary focus image is generated by shifting the optimal focus position of the primary focus image by a fixed distance set by the user. However, this fixed distance set by the user also fails to take into account image characteristics and the presence of target cells. For example, referring to Figures 2 and 4B, the secondary focus image is generated for the fourth layer of the cytology specimen 210. However, target cells are not distributed in the fourth layer of the cytology specimen 210.

[0023] Figure 5A shows an exploded view of a cytology specimen 500, configured in accordance with some embodiments disclosed herein. In some embodiments, the cytology specimen 500 may correspond to a cytology specimen 300. Referring to Figure 2, in some embodiments, the scanner 220 is configured to generate a master focus image on the cytology specimen 500. More specifically, the scanner 220 is configured to randomly select multiple regions on the cytology specimen 210 and determine the optimal focus position based on the imaging characteristics of each region. More specifically, the scanner 220 is configured to randomly select several regions from all regions A, B, C, D, E, F, G, H, I, and J, and acquire their Z-stack images. For example, the scanner 220 may be configured to randomly select regions B, D, F, H, I, and J, and acquire Z-stack images associated with these selected regions. More specifically, for illustrative purposes, the scanner 220 is configured to acquire a first set of Z-stack images associated with region B (i.e., images associated with sites B1, B2, B3, and / or B4). Similarly, scanner 220 can also acquire images related to regions D, F, H, I and J (i.e., images related to parts D1, D2, D3, D4, F1, F2, F3, F4, H1, H2, H3, H4, I1, I2, I3, I4, J1, J2, J3 and / or J4).

[0024] In some embodiments, scanner 220 processes the Z-stacked images using an artificial intelligence engine to identify whether the Z-stacked images contain images of target cells 213. This artificial intelligence engine may have machine learning capabilities. It can be trained on known cell sample images corresponding to target cells 213 with different contrast, brightness, shape, and other image features. In some embodiments, scanner 220 is configured to compare images contained in the same Z-stacked image and identify the image containing the highest number of target cells in that same Z-stacked image. For example, in region F, scanner 220 is configured to compare images contained in a third Z-stacked image (such as images of regions F1, F2, F3, and F4) and identify the image of region F3, which has the highest number of target cells in the third Z-stacked image (as shown in FIG. 2). Similarly, referring to FIG. 5A, scanner 220 is configured to identify the images containing the highest number of target cells in regions B2, D2, H3, I3, and J3 in the first, second, fourth, fifth, and sixth Z-stacked images, respectively. Therefore, the scanner 220 is configured to set the optimal focus positions of regions B, D, F, H, I, and J as the depth of field of parts B2, D2, F3, H3, I3, and J3, respectively.

[0025] In other embodiments, the AI ​​engine can replace the random selection of cytological specimen 500 regions to directly identify the specific location of target cell 213 in a region (e.g., region F). For example, the AI ​​engine can infer one or more bounding boxes of target cell 213. For each bounding box that reaches a confidence score threshold, the AI ​​engine can acquire Z-stacked images (e.g., F1, F2, F3, F4) within that bounding box. For illustrative purposes, scanner 220 is configured to compare the images of regions F1, F2, F3, and F4, determine that the image of region F3 has the highest contrast, sharpest edge, brightest color value, or sharpest edge among the images of F1, F2, F3, and F4, and determine the depth of field of region F3 as the optical focal point of region B.

[0026] In some embodiments, for regions not scanned by scanner 220 (e.g., regions A, C, E, and G), scanner 220 can estimate the optimal focal position of these unscanned regions based on the optimal focal position of scanned regions (e.g., regions B, D, F, H, I, and J). This estimation can be based on several technically feasible methods, such as linear or bilinear interpolation, triangulation, or more advanced computational methods, to generate a smooth, continuous master focus map covering the entire cytological specimen. Figure 5A also shows an exploded view of the cytological specimen 500, where the depth of field of sites A2, C2, E2, and G3 is estimated as the optimal focal position of the master focus map in regions A, C, E, and G, respectively. Accordingly, in some embodiments, the master focus map includes the optimal focal position corresponding to the depth of field of sites A2, B2, C2, D2, E2, F3, G3, H3, I3, and J3. Compared to conventional methods, the master focus map in some embodiments of this disclosure is generated based on images of the highest number of target cells in each region of the cytological specimen 500. Therefore, the main focus image in some embodiments of this disclosure corresponds to the state of existence of the target cell.

[0027] Figure 5B shows another exploded view of the cytology specimen 500, configured in accordance with some embodiments disclosed herein. Referring to Figure 2, in some embodiments, the scanner 220 is configured to generate a primary focus image on the cytology specimen 500. More specifically, after generating the primary focus image, the scanner 220 is further configured to compare images contained in the same Z-stack image and identify images containing target cells within that Z-stack image. For example, in region F, the scanner 220 is configured to compare images contained in a third Z-stack image and identify the image associated with region F2 in that third Z-stack image containing target cells, as shown in Figure 2. Similarly, referring to Figure 5A, the scanner 220 is configured to identify images located in the first, second, fourth, fifth, and sixth Z-stack images, respectively, containing regions B1, D1, H1, I1, and J2 of the target cells.

[0028] In some embodiments, for unscanned areas not scanned by scanner 220, scanner 220 is configured to estimate the optimal focal position of unscanned areas (e.g., areas A, C, E, and G) based on the optical focal position of scanned areas (e.g., areas B, D, F, H, I, and J). This estimation can be based on several technically feasible methods, such as linear or bilinear interpolation, triangulation, or more advanced computational methods, to generate a smooth and continuous secondary focal map on the cytological specimen. Figure 5B also presents an exploded view of cytological specimen 500, where the depths of field of sites A1, C1, E1, and G2 are estimated as the optimal focal positions of the secondary focal map in areas A, C, E, and G, respectively. Accordingly, in some embodiments, the secondary focal map includes optimal focal positions corresponding to the depths of field of sites A1, B1, C1, D1, E1, F2, G2, H1, I1, and J2.

[0029] In some embodiments, the optimal focus positions associated with the secondary focus image are all located above or below the optimal focus positions corresponding to the primary focus image. In other words, the secondary focus image and the primary focus image do not overlap. Referring to Figures 5A and 5B, the optical focus positions associated with the secondary focus image (i.e., parts A1, B1, C1, D1, E1, F2, G2, H1, I1, and J2) are all located above the depth-of-field range corresponding to the optical focus positions of the primary focus image (i.e., parts A2, B2, C2, D2, E2, F3, G3, H3, I3, and J3). Therefore, when generating the secondary focus image, the scanner 220 is configured to identify all areas located above or below the relevant area of ​​the primary focus image.

[0030] Compared to traditional methods, according to some embodiments of this disclosure, the secondary focus map is generated based on images of target cells in each region of the cytological specimen 500. Therefore, the secondary focus map in some embodiments of this disclosure also corresponds to the presence of target cells. Furthermore, unlike traditional methods where any corresponding focus position between the primary and secondary focus maps maintains a fixed distance, in this embodiment, the distance between any corresponding focus position between the two focus maps varies on the cytological specimen 500 depending on the distribution of target cells. For example, referring to Figures 5A and 5B, in regions A, B, C, D, E, F, G, and J, the focus position of the secondary focus map is one layer above the focus position of the primary focus map, while in regions H and I, it is two layers above the primary focus map. In other words, at the corresponding focus positions of the primary and secondary focus maps, there is a difference in the distance between the target cells related to the primary focus map and the target cells related to the secondary focus map.

[0031] Figure 6 illustrates an exemplary scanner 600 configured according to several embodiments of the present disclosure, used to generate one or more focus maps on a cytological specimen and acquire images of target cells in the cytological specimen based on the focus maps. Referring to Figure 2, in some embodiments, scanner 600 corresponds to scanner 220. Scanner 600 includes, but is not limited to, a computing device 610, a camera 620, an objective lens module 630, a stage 650, and a light source 660. In some embodiments, stage 650 is configured to carry and move a glass slide. Cytological specimens 640 are distributed on the glass slide.

[0032] In some embodiments, the computing device 610 includes a processor, a memory subsystem, and a communication subsystem. The aforementioned artificial intelligence engine can be implemented as an executable instruction set stored in the memory subsystem and executed by the processor.

[0033] In some embodiments, the computing device 610 is configured to generate control signals and transmit these control signals to the camera 620, objective lens module 630, stage 650, and light source 660 via the communication subsystem. For example, the computing device 610 may be configured to control the light source 660 to generate light within a specific wavelength range, which is associated with the image characteristics of target cells in the cytological specimen 640. For example, this specific wavelength range may correspond to the aforementioned color information associated with the target cells. An example of the specific wavelength range may be from about 530 nanometers to about 630 nanometers. Alternatively, the specific wavelength range may be from about 450 nanometers to about 560 nanometers, and more preferably from about 450 nanometers to about 530 nanometers.

[0034] In some embodiments, the computing device 610 is configured to control the movement of the stage 650. When the stage 650 carries the glass slide and the cytology specimen 640, its movement can align a region of the cytology specimen 640 (e.g., regions A, B, C, D, E, F, G, H, I, or J shown in FIG. 2) with a field of view of the objective lens 633, so that the light emitted by the light source 660 (as shown in FIG. 6) forms an optical path. Therefore, the light emitted by the light source 660 can pass through the region and the objective lens 633 and reach the camera 620.

[0035] In some embodiments, the computing device 610 is configured to control the lens switching module 631 of the objective lens module 630 to switch between different objectives (e.g., objectives 633 and 635). In other embodiments, the computing device 610 is configured to control the movement of the camera 620, the objective lens module 630, and / or the stage 650 so that objective lens 633 or objective lens 635 is focused on a portion of the cytological specimen 640. Referring to FIG2, objective lens 633 may correspond to the first objective lens 220.

[0036] In some embodiments, the computing device 610 is configured to control the camera 620 to acquire an image of a focused portion of one of the cytological specimens 640. The computing device 610 is also configured to control the camera 620 to transmit the acquired image to the computing device 610 for further processing. Such processing includes, but is not limited to, identifying images in the cytological specimen 640 that are associated with target cells or non-target cells, and comparing Z-stacked images of the same region of the cytological specimen 640 to determine the number of target cells in the Z-stacked images. In some embodiments, the computing device 610 is configured to generate one or more focus maps on the cytological specimen 640 based on the number of target cells captured in the Z-stacked images.

[0037] Figure 7 is a flowchart illustrating an exemplary process 700 configured according to several embodiments of this disclosure for obtaining images related to target cells distributed in a cytological specimen. Process 700 may include one or more operations, functions, or actions, as shown in blocks 710, 720, 730, 740, 750, 760, and / or 770, which may be performed by hardware, software, and / or firmware. The configuration of each block should not be considered a limitation on the embodiments described. The steps and operations outlined are provided as examples only; some steps and operations may be optional, combined into fewer steps and operations, or extended into additional steps and operations without affecting the nature of the disclosed embodiments. Although the blocks are arranged sequentially in the figures, these blocks may also be executed in parallel and / or in a different order than described herein. In some embodiments, process 700 may be applied to a variety of scanning methods to scan cytological specimens, in conjunction with corresponding different whole-slide imaging scanners. Such scanning methods may include, but are not limited to, area scanning or line scanning methods.

[0038] Process 700 begins at step 710: "Acquiring a first set of Z-stacked images including the first image and the second image". In some embodiments, referring to Figures 2 and 5A, scanner 220 acquires a first Z-stacked image associated with a first region during a focusing process targeting a first region of cytology specimen 500. In some embodiments, during this focusing process, scanner 220 is configured to focus on target cells distributed within the first region. In some embodiments, scanner 220 is configured to randomly select and scan the first region (e.g., region B) from all regions of cytology specimen 500. In other embodiments, scanner 220 is configured to preferentially select and scan the first region of cytology specimen 500 (e.g., a low-magnification scan of scanner 220) based on image information obtained from a previous scan of cytology specimen 500 (e.g., showing some regions occupied and others empty). Exemplary image information may include, but is not limited to, an intensity value of an image obtained from the previous scan (e.g., showing image pixels as non-white or low whiteness), and the expected size, structure, and appearance of the cells. After selecting region B, scanner 220 is further configured to acquire a first Z-stacked image associated with region B. For ease of explanation, scanner 220 is configured to acquire a first image associated with portion B2 and a second image associated with portion B1. In some embodiments, the first image is associated with a first group of target cells distributed within region B, and the second image is associated with a second group of target cells distributed within region B. In some embodiments, the number of the first group of target cells corresponds to the highest number of target cells distributed in the corresponding part of region B.

[0039] In some embodiments, scanner 220 is configured to determine whether the number of images included in the first Z-stacked image reaches a threshold. For example, suppose the first Z-stacked image has a threshold of four images. Scanner 220 is further configured to acquire a third image associated with portion B3 and a fourth image associated with portion B4 to reach the threshold of four images.

[0040] Step 710 can be followed by step 720, "acquiring a second Z-stacked image including the third and fourth images." In some embodiments, referring to Figures 2 and 5A, the scanner 220 is configured to acquire a second Z-stacked image associated with a second region during focusing on a second region of the cytology specimen 500. In some embodiments, during focusing, the scanner 220 is configured to focus on target cells distributed in the corresponding area of ​​the second region. In some embodiments, the scanner 220 is configured to randomly select the second region of the cytology specimen 500 (e.g., region I). After selecting region I, the scanner 220 is further configured to acquire a second Z-stacked image associated with region I. For ease of explanation, the scanner 220 is configured to acquire a third image associated with a portion I3 and a second image associated with a portion I1. In some embodiments, the third image is associated with a third group of target cells distributed in region I, and the fourth image is associated with a fourth group of target cells distributed in region I. In some embodiments, the number of the third group of target cells corresponds to the highest number of target cells distributed in region I.

[0041] In some embodiments, scanner 220 is configured to determine whether the number of images included in the second Z-stacked image reaches a threshold. For example, assuming the second Z-stacked image needs to contain four images as a threshold, scanner 220 will further acquire a fifth image associated with portion I2 and a sixth image associated with portion I4.

[0042] Block 720 can be followed by block 730, "Generate First Focus Image". Referring to Figures 2 and 5A, scanner 220 is configured to generate a first focus image on cytological specimen 500. In some embodiments, this first focus image corresponds to the main focus image described in Figure 5A. In some embodiments, the first focus image includes portions B2 and I3.

[0043] In some embodiments, scanner 220 is configured to acquire Z-stacked images of the cytological specimen 500 associated with one or more additional regions (e.g., region D). In some embodiments, in response to scanner 220 determining that the number of regions with acquired Z-stacked images is less than a threshold number of regions, scanner 220 is configured to acquire Z-stacked images associated with other regions (e.g., regions F, H, and J). In some embodiments, based on the Z-stacked images, the first focus map further includes regions D2, F3, H3, and J3 as shown in FIG. 5A.

[0044] Step 730 can be followed by step 740, "capturing the image based on the first focal map". Referring to Figures 2 and 5A, the scanner 220 is configured to capture the target cell image based on the first focal map including regions B2, D2, F3, H3, I3, and J3. In some embodiments, the scanner 220 is also configured to capture the target cell image by estimating the first focal map including regions A2, C2, E2, and G3.

[0045] Step 740 can be followed by step 750, "generating a second focus image". Referring to Figures 2 and 5A, the scanner 220 is configured to generate a second focus image on the cytological specimen 500. In some embodiments, this second focus image corresponds to the secondary focus image discussed in the description of Figure 5B. In some embodiments, the second focus image includes portions B1 and I1.

[0046] In some embodiments, based on the distribution of target cells, the distance between target cells at any corresponding focal position in the first focal map and the second focal map is adjusted according to the region of the cytological specimen 500. For example, a first distance between a group of target cells distributed in part B2 of the first focal map and a group of target cells distributed in part B1 of the second focal map is different from a second distance between a group of target cells distributed in part I3 of the first focal map and a group of target cells distributed in part I1 of the second focal map.

[0047] Block 750 can be followed by block 760, where "the number of target cells captured is below a threshold." In some embodiments, referring to Figures 2 and 5A, the scanner 220 is configured to determine that the number of target cells captured in block 740 is below a threshold number of target cells. In some embodiments, when the number of target cells captured in block 740 is determined to be greater than or equal to the threshold number, process 700 ends, and there is no need to continue capturing target cell images, thereby saving scanning time.

[0048] In other embodiments, in response to the determination that the number of target cells captured in block 740 is lower than the threshold number, block 760 may then proceed to block 770, which involves "capturing the image based on the second focal map". Referring to Figures 2 and 5B, scanner 220 is configured to capture target cell images based on the second focal map including regions B1, D1, F2, H1, I1, and J2. In some embodiments, scanner 220 is also configured to capture target cell images based on the second focal map including regions A1, C1, E1, and G2 through estimation.

[0049] The above examples can be implemented through hardware (including hardware logic circuits), software, or firmware, or a combination thereof. Figure 8 illustrates an example device 800 configured to perform the embodiments of this disclosure. For example, device 800 can be implemented as a server for performing the process 700 shown in Figure 7. Device 800 can also be implemented as a relay network device configured to perform the process 700 in Figure 7. Device 800 can be any computing device or network device suitable for practicing one or more embodiments of this disclosure. It should be noted that the device 800 described herein is merely an example, and any other technically feasible configuration is within the scope of this disclosure.

[0050] As shown in the figure, device 800 includes, but is not limited to, an interconnect (bus) 830 connecting at least one processor 840, a computer-readable medium 810, an input / output (I / O) device interface 850, and a network interface 860. Processor 840 can be any suitable processor, implemented as a central processing unit (CPU), a graphics processing unit (GPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), any other type of processing unit, or a combination of different processing units, such as a CPU configured to work in conjunction with a GPU or a digital signal processor (DSP). Generally, processor 840 can be any technically feasible hardware unit capable of processing data and / or executing executable instructions (including process 870). In some embodiments, process 870 may correspond to the aforementioned process 700.

[0051] Network interface 860 is configured to connect device 800 to one or more networks, enabling device 800 to communicate with other devices on such networks.

[0052] Processor 840, I / O device interface 850, and network interface 860 are configured to read data from and write data to computer-readable medium 810. Computer-readable medium 810 may store instructions or program code that, when executed by the processor, cause the processor to perform the process described with reference to FIG7. In some embodiments, computer-readable medium 810 includes cache 820 for storing data related to process 870.

[0053] Some or all of the contents of the embodiments of this disclosure can be implemented equivalently in the form of integrated circuits, or as a single or multiple programs, firmware, or any combination of the above forms running on one or more processors. Based on the contents of this disclosure, it is feasible to design related circuits and / or write software and firmware code.

[0054] Software and / or other instructions used to implement the techniques described herein may be stored in a non-transitory computer-readable storage medium (e.g., 810) and may be executed by one or more dedicated programmable microprocessors (e.g., processor 840). As used in this specification, "computer-readable storage medium" encompasses any mechanism capable of providing (i.e., storing and / or transmitting) information in a machine-readable form (e.g., a computer, a network device, a personal digital assistant (PDA), a mobile device, a manufacturing tool, any device with one or more processors, etc.). Computer-readable storage media may include recordable / non-recordable media (e.g., read-only memory (ROM), random access memory (RAM), magnetic disk or optical storage media, flash memory devices, solid-state storage devices, etc.).

[0055] As can be seen from the foregoing, the various embodiments described herein are for illustrative purposes only, and various modifications can be made without departing from the scope and spirit of this disclosure. Therefore, the various embodiments disclosed herein should not be considered limiting.

[0056] 110: Cytological specimen 120: Glass slide 131: Mark 141: Target cells 151, 153, 155, 157, 159: Non-target cells 210: Cytological specimen 211: Dust 212: Mark 213: Type I cells 214: Second type of cell 215: Third type of cell 216: Fourth type of cell 220: Scanner 300: Cytological Specimen 400: Cytological Specimen 500: Cytological Specimen 600: Scanner 610: Computing device 620: Camera 630: Objective lens module 631: Lens Switching Module 633: Objective lens 635: Objective lens 640: Cytological Specimen 650: Platform 660: Light source 700: Process 710, 720, 730, 740, 750, 760, 770: Square 800: Device 810: Computer-readable media 820: Cache 830: Interconnection Line (Bus) 840: Processor 850: Input / Output (I / O) Device Interface 860: Network Interface 870: Process A, B, C, D, E, F, G, H, I, J: Area A1, A2, A3, A4, B1, B2, B3, B4, C1, C2, C3, C4, D1, D2, D3, D4, E1, E2, E3, E4, F 1. F2, F3, F4, G1, G2, G3, G4, H1, H2, H3, H4, I1, I2, I3, I4, J1, J2, J3, J4: part d1, d2, d3, d4: Layers

Claims

1. A method for acquiring images of target cells in a cytological specimen, comprising: A first Z-stacked image associated with a first region of the cytological specimen is acquired through a lens. The first Z-stacked image includes a first image taken from a surface of the cytological specimen at a first depth of field and a second image taken from the surface of the cytological specimen at a second depth of field. The first image is associated with a first group of target cells distributed within the first region, and the second image is associated with a second group of target cells distributed within the first region. The number of the first group of target cells corresponds to the highest number of target cells distributed within the first region. A second Z-stacked image associated with a second region of the cytological specimen is acquired through the lens. The second Z-stacked image includes a third image taken from the surface of the cytological specimen at a third depth of field and a fourth image taken from the surface of the cytological specimen at a fourth depth of field. The third image is associated with cells distributed within the second region. A third group of target cells within the domain is associated with the fourth image with a fourth group of target cells distributed within the second region, wherein a number of the third group of target cells corresponds to the highest number of target cells distributed within the second region; a first focal map is generated based on the first region, the second region, the first depth of field, and the third depth of field; an image of the target cells is captured through the lens based on the first focal map; a second focal map is generated based on the first region, the second region, the second depth of field, and the fourth depth of field, wherein a first distance between the first group of target cells and the second group of target cells in a z-axis direction of the cytological specimen is different from a second distance between the third group of target cells and the fourth group of target cells in the z-axis direction of the cytological specimen; and an image of the target cells is captured through the lens based on the second focal map when the number of target cells obtained according to the first focal map is lower than a first threshold.

2. The method as described in claim 1, wherein the second focus image is positioned above the first focus image when the second depth of field is less than the first depth of field and the fourth depth of field is less than the third depth of field.

3. The method as described in claim 1, wherein the second focus image is located below the first focus image when the second depth of field is greater than the first depth of field and the fourth depth of field is greater than the third depth of field.

4. The method as described in claim 1, further comprising: Determine whether the number of images included in the first Z-stacked image reaches a second threshold; In response to the determination that the number of images does not reach the second threshold, the distance between a stage used to carry the cytological specimen and the lens is adjusted; and additional images related to the first region are obtained through the lens.

5. The method as described in claim 1, wherein the first Z-stacked image is acquired during a first focusing process performed by the lens on the first region, and the second Z-stacked image is acquired during a second focusing process performed by the lens on the second region.

6. The method as claimed in claim 5, wherein the lens is configured to focus on the target cell in the first region during the first focusing process and on the target cell in the second region during the second focusing process.

7. The method as described in claim 1, further comprising acquiring Z-stacked images related to other regions of the cytological specimen through the lens.

8. The method as described in claim 7, further comprising: Determine whether the number of regions including the first region, the second region, and the other region reaches a third threshold. And in response to the situation where the number of samples in the area does not reach the third threshold, the lens is used to obtain a Z-stack image of another area of ​​the cytological specimen slide.

9. The method as described in claim 1, further comprising processing the first region preferentially over all regions based on image information obtained from a previous scan of one of the cytological specimens to obtain the first Z-stacked image.

10. A system for acquiring images of target cells distributed in a cytological specimen, comprising: A lens; One processor; And a non-transitory computer-readable medium having instructions stored thereon, which, when executed by a processor, cause the processor to: acquire, through the lens, a first Z-stacked image associated with a first region of the cytological specimen, wherein the first Z-stacked image includes a first image acquired from a surface of the cytological specimen at a first depth of field and a second image acquired from the surface of the cytological specimen at a second depth of field, and the first image is associated with a first group of target cells distributed in the first region and the second image is associated with a second group of target cells distributed in the first region, wherein a number of the first group of target cells corresponds to the highest number of target cells distributed in the first region; acquire, through the lens, a second Z-stacked image associated with a second region of the cytological specimen, wherein the second Z-stacked image includes a third image acquired from the surface of the cytological specimen at a third depth of field and a second image acquired from the surface of the cytological specimen at a fourth depth of field. A fourth image, and the third image associated with a third group of target cells distributed in the second region, and the fourth image associated with a fourth group of target cells distributed in the second region, wherein a number of the third group of target cells corresponds to the highest number of target cells distributed in the second region; a first focal map is generated based on the first region, the second region, the first depth of field, and the third depth of field; an image of the target cells is captured through the lens according to the first focal map; a second focal map is generated based on the first region, the second region, the second depth of field, and the fourth depth of field, wherein a first distance between the first group of target cells and the second group of target cells in a z-axis direction of the cytological specimen is different from a second distance between the third group of target cells and the fourth group of target cells in the z-axis direction of the cytological specimen; and in response to when a number of target cells obtained according to the first focal map is lower than a first threshold, an image of the target cells is captured through the lens according to the second focal map.

11. The system as described in claim 10, in response to a second focus image being located above the first focus image when the second depth of field is less than the first depth of field and the fourth depth of field is less than the third depth of field.

12. The system as described in claim 10, wherein the second focus map is located below the first focus map when the second depth of field is greater than the first depth of field and the fourth depth of field is greater than the third depth of field.

13. The system of claim 10, wherein the non-transitory computer-readable medium stores additional instructions that, when executed by the processor, cause the processor to: determine whether a number of images contained in the first Z-stacked image reaches a second threshold; in response to determining that the number of images does not reach the second threshold, adjust a distance between a stage for carrying a cytological specimen and the lens; and acquire additional images related to the first region through the lens.

14. The system as claimed in claim 10, wherein the first Z-stacked image is acquired during a first focusing process performed by the lens on the first region, and the second Z-stacked image is acquired during a second focusing process performed by the lens on the second region.

15. The system as claimed in claim 14, wherein the lens is configured to focus on the target cell in the first region during the first focusing process and on the target cell in the second region during the second focusing process.

16. The system of claim 10, wherein the non-transitory computer-readable medium stores additional instructions that, when executed by the processor, cause the processor to acquire Z-stacked images related to other regions of the cytological specimen through the lens.

17. The system of claim 16, wherein the non-transitory computer-readable medium stores additional instructions that, when executed by the processor, cause the processor to: determine whether the number of regions including the first region, the second region, and the other regions reaches a third threshold; and, in response to determining that the number of regions does not reach the third threshold, acquire a Z-stacked image of the other region of the cytological specimen slide through the lens.

18. A non-transitory computer-readable storage medium comprising a set of instructions, which, when executed by a processor of a computing device, cause the processor to perform a method for acquiring images of target cells distributed in a cytological specimen, the method comprising: A first Z-stacked image associated with a first region of the cytological specimen is acquired through a lens. The first Z-stacked image includes a first image taken from a surface of the cytological specimen at a first depth of field and a second image taken from the surface of the cytological specimen at a second depth of field. The first image is associated with a first group of target cells distributed within the first region, and the second image is associated with a second group of target cells distributed within the first region. The number of the first group of target cells corresponds to the highest number of target cells distributed within the first region. A second Z-stacked image associated with a second region of the cytological specimen is acquired through the lens. The second Z-stacked image includes a third image taken from the surface of the cytological specimen at a third depth of field and a fourth image taken from the surface of the cytological specimen at a fourth depth of field. The third image is associated with the target cells distributed within the second region. A third group of target cells within a region is associated with the fourth image with a fourth group of target cells distributed within the second region, wherein a number of the third group of target cells corresponds to the highest number of target cells distributed within the second region; a first focal map is generated based on the first region, the second region, the first depth of field, and the third depth of field; an image of the target cells is captured through the lens based on the first focal map; a second focal map is generated based on the first region, the second region, the second depth of field, and the fourth depth of field, wherein a first distance between the first group of target cells and the second group of target cells in a z-axis direction of the cytological specimen is different from a second distance between the third group of target cells and the fourth group of target cells in the z-axis direction of the cytological specimen; and in response to a number of target cells obtained according to the first focal map being lower than a first threshold, an image of the target cells is captured through the lens based on the second focal map.

Images