Method and system for obtaining sufficient cytological images in cytopathology

Abstract

Exemplary methods and systems for identifying suspect cytological specimens containing target cells are disclosed. One exemplary method includes acquiring a first image associated with a first region of the cytological specimen through a first objective lens, acquiring a second set of one or more images associated with a first subregion of the first region through a second objective lens, identifying a first number of target cells, and determining whether the first number of target cells is greater than a threshold number. If the first number of target cells is greater than the threshold number, the exemplary method prompts a warning. If not, the exemplary method further includes acquiring a third set of one or more images associated with a second subregion through a second objective lens, and identifying a second number of target cells associated with the second subregion.

Description

[[Technical Field]]

[0001] (Cross-Reference to Related Application) This application claims priority based on U.S. Provisional Application No. 63 / 403,660 filed on September 2, 2022, which is incorporated into this application by reference in its entirety.

[0002] Embodiments of the present invention generally relate to methods and systems for acquiring cytological images in cytopathology. [[Background Art]]

[0003] Unless otherwise specifically stated herein, the approaches described in this section are not prior art to the claims of the present application, and inclusion in this section does not constitute an admission that they are prior art.

[0004] Cytopathology is a subfield of pathology that studies and diagnoses diseases at the cellular level, and generally involves acquiring cytological images at the cellular level. Acquiring cytological images of cytological specimens includes digital imaging of the glass slide on which the cytological specimen is distributed. Image digitalization typically includes scanning a glass slide with a cytological specimen to generate a digital slide of the image of the glass slide. The scanning may be performed by a whole-slide image scanner. The digital slide of the image can be displayed on a display (e.g., a computer monitor, etc.) instead of a microscope. [[Summary of the Invention]] [[Problem to be Solved by the Invention]]

[0005] Digitizing images from glass slides efficiently and accurately can be challenging. One reason is that cytological specimens on glass slides may contain single cells and cell populations distributed in three-dimensional space. This three-dimensional distribution of cells and cell populations within the cytological specimen makes focusing difficult. One conventional method to address this focusing difficulty is to acquire a focused image of the entire glass slide, but this method is extremely time-consuming. Another conventional method to address the focusing difficulty is to manually mark a symmetrical area of ​​the glass slide and acquire a focused image of the marked area. However, manual approaches can lead to marking the wrong area, resulting in the acquired focused image not containing many of the target cells in the cytological specimen. [Brief explanation of the drawing]

[0006] [Figure 1] An illustrative diagram showing how cytological specimens are distributed in a three-dimensional space on a glass slide. [Figure 2] This diagram illustrates an example configuration of a multi-objective lens module, showing how to acquire images related to layers of a cytological specimen through the first objective lens. [Figure 3A] A diagram illustrating a portion of a cytological specimen. [Figure 3B] Figure showing images related to a portion of a cytological specimen obtained through the first objective lens. [Figure 4A] This diagram illustrates an example configuration of a multi-objective lens module, showing how to acquire an image related to the second subpart of a cytological specimen through the second objective lens. [Figure 4B] Figure showing an image related to the second sub-part obtained through the second objective lens. [Figure 4C] This diagram illustrates an example configuration of a multi-objective lens module, showing how to acquire images related to multiple second subparts of a cytological specimen through the second objective lens. [Figure 4D] Figure showing an image related to the second sub-part obtained through the second objective lens. [Figure 4E]This diagram illustrates an example configuration of a multi-objective lens module, showing how to acquire an image related to the second subpart of a cytological specimen through the second objective lens. [Figure 4F] Figure showing an image related to the second sub-part obtained by the second objective lens 220. [Figure 5] A diagram illustrating a system for acquiring images related to cytological specimens. [Figure 6] A flowchart illustrating the process of obtaining images related to target cells distributed in a cytological specimen. [Figure 7] A flowchart illustrating an exemplary process for obtaining images related to target cells distributed in a cytological specimen, arranged according to some embodiments of this disclosure. [Modes for carrying out the invention]

[0007] The following detailed description refers to the accompanying drawings, which constitute part of this specification. In the drawings, unless otherwise specified in the context, similar symbols generally indicate similar components. The exemplary embodiments described in the detailed description, drawings, and claims are not intended to be limiting. Other embodiments may be used and other modifications may be made without departing from the spirit or scope of the subject matter presented herein. It will be readily apparent that the aspects of this disclosure generally described herein and shown in the figures can be arranged, substituted, combined, and designed in a variety of different configurations, all of which fall within the scope expressly assumed herein. Hereinafter, a cytological specimen is considered suspicious if the target cells distributed within the space of the cytological specimen contain cells at risk of disease.

[0008] Figure 1 illustrates a cytological specimen 110 distributed in three-dimensional space on a glass slide 120, arranged according to several embodiments of the present disclosure. The cytological specimen 110 may contain multiple cells and impurities. For example, the cytological specimen 110 may contain dust or blemishes 131, target cells 141 (such as 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 blemishes 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., at different depths within the volume of the cytological specimen 110).

[0009] Figure 2 shows an example configuration of a multi-objective lens module 200, arranged according to some embodiments of the present disclosure, in which an image related to a layer of a cytology specimen 210 is acquired through a first objective lens 220. In relation to Figure 1, in Figure 2, the cytology specimen 210 may correspond to a cytology specimen 110. In Figure 2, the multi-objective lens module 200 includes a first objective lens 220 configured to acquire an image related to the cytology specimen 210. More specifically, the first objective lens 220 is configured to acquire an image related to one or more layers of the cytology specimen 210. In some embodiments, each layer may correspond to a depth of field associated with the field of view of the first objective lens 220.

[0010] In some embodiments, the first objective lens 220 has a first field of view, a first depth of field, and a first magnification. The first field of view may correspond to any of regions A, B, C, D, E, F, G, H, I, and J of the cytological specimen 210.

[0011] In some embodiments, the first depth of field is defined as the distance between the closest, sharp, and in-focus object to the first objective lens 220 and the furthest, sharp, and in-focus object in the first field of view. Examples of these distances may be distances 1, 2, and 3 shown in Figure 2. The first depth of field may correspond to the layers of the cytology specimen 210. Therefore, distance 1 may correspond to the first layer of the cytology specimen 210, distance 2 to the second layer, and distance 3 to the third layer.

[0012] The first layer may be the layer furthest from the glass slide (e.g., glass slide 120), and the third layer may be the layer adjacent to the glass slide (e.g., glass slide 120). Although Figure 2 shows three layers, it should be noted that the cytological specimen 210 may contain more or fewer layers depending on the first depth of field of the first objective lens 220. In these embodiments, regions and distances may define parts of the cytological specimen 210. For example, J3 corresponds to the lower right front part of the cytological specimen 210. Thus, in one embodiment shown in Figure 2, the cytological specimen 210 may be defined by parts A1, A2, A3, B1, B2, B3, C1, C2, C3, D1, D2, D3, E1, E2, E3, F1, F2, F3, G1, G2, G3, H1, H2, H3, I1, I2, I3, J1, J2, and J3.

[0013] In some embodiments, the cytological specimen 210 may include, but is not limited to, dust 211, blemishes 212, type 1 cells 213 (e.g., epithelial cells at risk of malignancy), type 2 cells 214 (e.g., red blood cells), type 3 cells 215 (e.g., normal epithelial cells), and type 4 cells 216 (e.g., other cells). The dust 211, blemishes 212, type 1 cells 213, type 2 cells 214, type 3 cells 215, and type 4 cells 216 may be distributed within the cytological specimen 210 based on their specific gravity. The blemishes 212 may be manually marked by the user on the upper surface of the cytological specimen 210. The dust 211 and cells 213, 214, 215, and 216 may have significant differences in density and be distributed in different layers of the cytological specimen 210. For example, dust 211 may have a lower specific gravity than cells 213, 214, 215, and 216 and be distributed in the first layer of the cytological specimen 210. Cells 213, 214, 215, and 216 may have a higher specific gravity than dust 211 and be distributed in the second layer of the cytological specimen 210.

[0014] In some embodiments, the cells 213, 214, 215, and 216 in the cytological specimen 210 typically have a size ranging from about 5 micrometers to about 20 micrometers. It should be noted that the selection of the first objective lens 220 may be based on the first depth of field of the first objective lens 220. Since the first depth of field of the first objective lens 220 may be much larger than the size of the cells 213, 214, 215, and 216 in the cytological specimen 210, images of cells 213, 214, 215, and 216 may be acquired at the same first depth of field (e.g., distance 1, distance 2, or distance 3).

[0015] In some embodiments, before acquiring an image of all layers of the cytology specimen 210 via the first objective lens 220, the first objective lens 220 is configured to acquire images of each layer (e.g., the first, second, and third layers) of several regions randomly selected from all of regions A, B, C, D, E, F, G, H, I, and J. For example, the first objective lens 220 may be configured to acquire images of portions H1, H2, and H3 within a randomly selected region H, where region H corresponds to the first field of view of the first objective lens 220, and distances 1, 2, or 3 correspond to the first depth of field of the objective lens 200. Several regions other than region H, for example, regions F, I, and J, may also be randomly selected. Thus, the first objective lens 220 may also acquire images of portions F1, F2, F3, I1, I2, I3, J1, J2, and J3 of the cytology specimen 210.

[0016] In some embodiments, the images of portions F1, F2, F3, H1, H2, H3, I1, I2, I3, J1, J2, and J3 are processed by a processor comprising an artificial intelligence engine, and it is identified whether the images of portions F1, F2, F3, H1, H2, H3, I1, I2, I3, J1, J2, and J3 include images of cells 213, 214, 215, and 216. The artificial intelligence engine may include a machine learning function. The artificial intelligence engine may be trained based on known sample images of cells corresponding to cells 213, 214, 215, and 216 having various contrasts, brightness, shapes, and other image characteristics. In some embodiments, for illustrative purposes only, the artificial intelligence engine may identify that images of cells 213, 214, 215, and 216 are not included in the images of portions F1, F3, H1, H3, I1, I3, J1, and J3, but are mainly included in the images of portions F2, H2, I2, and J2. Accordingly, the artificial intelligence engine may conclude that cells 213, 214, 215, and 216 are distributed in the second layer of the cytological specimen 210. After reaching such a conclusion, the first objective lens 220 is configured to acquire an entire image of the second layer of the cytological specimen 210 (that is, the images of portions A2, B2, C2, D2, E2, F2, H2, I2, and J2).

[0017] With reference to FIG. 2, FIG. 3A is a diagram illustrating a portion 310 of a cytological specimen 210, and FIG. 3B shows an image 330 associated with the portion 310 acquired through the first objective lens 220 arranged in accordance with some embodiments of the present disclosure. In some embodiments, the portion 310 may correspond to portion H2 of the cytological specimen 210, and cells 321, 322, 323, 324, 325, and 326 may be distributed within the portion 310. In some embodiments, the cell 324 may be an epithelial cell at malignant risk, and cells 321, 322, 323, 325, and 326 may include normal epithelial cells, red blood cells, and other cells.

[0018] In some embodiments, since the first objective lens 220 is positioned above the portion 310, image 330 may be a top view of the portion 310. Image 330 in Figure 3B has an image region H' that corresponds to region H related to the portion 310 in Figure 3A. The correspondence between region H and image region H' may include one or more elements related to enlargement, reduction, rotation, and twisting.

[0019] In some embodiments, the artificial intelligence engine is used to process image 330 based on contrast, brightness, shape, and other image characteristics associated with cells 321 to 326, so as to identify images 331, 333, and 335 in image 330 that include images of cells 321 to 326. For example, it is identified that images associated with cells 321 and 322 are included in image 331, images associated with cells 323, 324, and 326 are identified to be included in image 333, and images associated with cells 321, 322, and 325 are identified to be included in image 335. Each of images 331, 333, and 335 has a spatial relationship within the image region H' of image 330. Based on the spatial relationship between images 331, 333, 335 and the image region H', and the correspondence between the image region H' and the region H, referring to FIG. 3A, sub-regions 341, 343, and 345 within the region H can be identified. In some embodiments, sub-regions associated with cells distributed in any of portions A2, B2, C2, D2, E2, F2, H2, I2, and J2 may be identified in a similar manner. In some embodiments, sub-portions 351, 353, and 355 of portion 310 are defined by distance 2 and sub-regions 341, 343, and 345, respectively. In some embodiments, as described above, cell 324 may be an epithelial cell at risk of malignancy (i.e., a target cell), and cells 321, 322, 323, 325, and 326 may include normal epithelial cells, red blood cells, and other cells (i.e., non-target cells). Sub-portions associated with target cells (i.e., sub-portions 351 and 353) may be examined in more detail than sub-portions associated with non-target cells (i.e., sub-portion 355). Sub-portions 351, 353, and 355 are further examined with a second objective lens to obtain more detailed images associated with target cell 324. For example, the second objective lens is configured to acquire images associated with sub-portions 351 and 353 rather than sub-portion 355. For example, the second objective lens may be configured to acquire images associated with sub-portions 351 and 353, but not acquire images associated with sub-portion 355.

[0020] Depending on the embodiment, as described above, the sub-parts of sections A2, B2, C2, D2, E2, F2, G2, I2, and J2 can be further examined with a second objective lens to obtain more detailed images related to cells 321 to 326.

[0021] Figure 4A shows how an exemplary multiple objective lens module, arranged according to some embodiments of the present disclosure, acquires an image relating to a second sub-part of a cytological specimen through a second objective lens. In Figure 4A, the second objective lens 410 of the multiple objective lens module is configured to acquire one or more images relating to a sub-part 420. In relation to Figure 3A, the sub-part 420 may correspond to a sub-part 353 defined by distance 2 and sub-region 343.

[0022] In some embodiments, the second objective lens 410 has a second field of view, a second depth of field, and a second magnification. In some embodiments, relating to Figure 3A, the second field of view may correspond to a sub-region 343. In some other embodiments, relating to Figure 3A, the second field of view may correspond to only a portion of the sub-region 343. For illustrative purposes only, Figure 4A shows the second field of view corresponding to the sub-region 343.

[0023] In some embodiments, the second magnification is higher than the first magnification. For illustrative purposes only, the first and second magnifications are not limited to these, but can be 4x and 20x, respectively.

[0024] In some embodiments, the second depth of field is smaller than the first depth of field. For illustrative purposes only, the first and second depths of field can be, but are not limited to, about 50 micrometers and about 1 micrometer, respectively. In some embodiments, the second depth of field refers to the distance between the closest, sharp, and in-focus object to the second objective lens 410 and the furthest, sharp, and in-focus object to the second objective lens 410 in the second field of view. Examples of distances may be, for example, distance 21, distance 22, distance 23, or distance 24 shown in Figure 4A. Distance 2 shown in Figure 4A corresponds to the same distance 2 shown in Figures 2 and 3A.

[0025] In some embodiments, the second depth of field may correspond to a sublayer of the layers of the cytology specimen 210 (for example, the second layer defined by distance 2 shown in Figure 2). Thus, distance 21 may correspond to the first sublayer of the second layer of the cytology specimen 210, distance 22 to the second sublayer of the second layer of the cytology specimen 210, distance 23 to the third sublayer of the second layer of the cytology specimen 210, and distance 24 to the fourth sublayer of the second layer of the cytology specimen 210. Although four sublayers are described for illustrative purposes only, the number of sublayers may be greater or less depending on the second depth of field of the second objective lens 410.

[0026] In some embodiments, the second objective lens 410 is configured to acquire an image related to the sub-part 420. More specifically, the second objective lens 410 is configured to acquire images of the second sub-parts 421, 422, 423, and / or 424 of the sub-part 420. In some embodiments, the second objective lens 410 is configured to focus on the second sub-parts 421, 422, 423, and / or 424 at a second depth of field.

[0027] In some embodiments, the focus may be based on the determination of an artificial intelligence engine. The artificial intelligence engine may have machine learning capabilities. For example, the artificial intelligence engine may be trained on sample images of target cells (e.g., epithelial cells with a malignant risk) having various contrasts, brightness, shapes, and other image characteristics. Thus, the artificial intelligence engine can identify images of target cells and decide to focus on those identified images. For example, of the images of second sub-parts 421, 422, 423, and 424 acquired through the second objective lens 410, the artificial intelligence engine can identify that the images of second sub-parts 422 and 423 contain images of target cells. Based on the determination by the artificial intelligence engine, the second objective lens 410 is driven to focus on the second sub-parts 422 and 423 and save the images of second sub-parts 422 and 423.

[0028] In another embodiment, the focus may be determined based on the image characteristics of the target cells (e.g., epithelial cells at risk of malignancy). Such image characteristics may include, but are not limited to, color information related to the target cells. For example, the nuclei of epithelial cells are stained blue by hematoxylin, while red blood cells remain red because they do not have nuclei to be stained. Therefore, focusing on the blue areas rather than the red areas in the image increases the probability of obtaining an image of the target cells.

[0029] In some embodiments, the image of the target cell may include a first predetermined range in the RGB (red, green, blue) domain or the HSV (hue, saturation, lightness) domain. For example, the first predetermined range may include a range of values ​​for each of R, G, and B. Alternatively, the first predetermined range may include a range of hue values. On the other hand, the image of a non-target cell (such as a normal epithelial cell, red blood cell, or other cell) may include a second predetermined range in the RGB domain or the HSV domain that is different from the first predetermined range. For example, the second predetermined range may include a different range of values ​​for R, G, and B, or the second predetermined range may include a different range of hue values. For example, among the images of second sub-parts 421, 422, 423, and 424 acquired through the second objective lens 410, the images of second sub-parts 422 and 423 may include the first predetermined range. This allows the second objective lens 410 to be configured to focus on the second sub-parts 422 and 423 and save images of the second sub-parts 422 and 423, but not to focus on the second sub-parts 421 and 424.

[0030] Figure 4B shows an image 430 related to a second sub-part 422 acquired through a second objective lens 220 arranged according to some embodiments of the present disclosure. Image 430 includes images of a target cell 441 and non-target cells 443 and 445. The image of the target cell 441 includes a first predetermined range in the RGB domain or HSV domain, and the images of the non-target cells 443 and 445 include a second predetermined range in the RGB domain or HSV domain. The first predetermined range in the RGB domain or HSV domain related to the image of the target cell 441 causes the second objective lens 410 to focus on the second sub-part 422.

[0031] In some embodiments, the artificial intelligence engine is configured to determine whether the number of target cells in image 430 is greater than a threshold number of target cells. In some embodiments, in relation to Figure 2, in response to determining a first number greater than the threshold number, the artificial intelligence engine is configured to prompt a warning indicating that the cytological specimen 210 is suspicious. For example, the warning may alert a physician that the patient associated with the cytological specimen 210 may have a disease.

[0032] In some embodiments, in response to the determination that the first number is less than or equal to a threshold number, the second objective lens 410 is further configured to focus on another second sub-part adjacent to the second sub-part 422 (e.g., the second sub-part 423) and acquire an image related to the second sub-part 423.

[0033] Figure 4C illustrates how an exemplary multi-objective lens module acquires images related to multiple second subparts (e.g., second subparts 422 and 423) of a cytological specimen through a second objective lens, and Figure 4D shows an image 450 related to a second subpart 423 acquired through a second objective lens 220, arranged according to some embodiments of the present disclosure. Image 450 includes an image of a target cell 451. The image of the target cell 451 may include a first predetermined range in the RGB domain or the HSV domain.

[0034] In some embodiments, the artificial intelligence engine is configured to determine whether the number of target cells in images 430 and 450 is greater than a threshold number of target cells. In some embodiments, in relation to Figure 2, in response to determining a first number greater than the threshold number, the artificial intelligence engine is configured to prompt a warning indicating that the cytological specimen 210 is suspicious. For example, the warning may alert a physician that the patient associated with the cytological specimen 210 may be ill.

[0035] In some embodiments, in relation to Figure 3A, in response to the determination that the number of first target cells in images 430 and 450 is less than or equal to a threshold number, the second objective lens 410 is further configured to acquire one or more images relating to another sub-part containing target cells (e.g., a sub-part 351 defined by distance 2 and sub-region 341). This other sub-part is a different sub-part from sub-part 420.

[0036] Figure 4E illustrates how an exemplary multi-objective lens module, arranged according to some embodiments of the present disclosure, acquires an image related to a second subpart of a cytological specimen (e.g., second subpart 462) through a second objective lens. In Figure 4E, the second objective lens 410 is configured to acquire one or more images related to subpart 460. In relation to Figure 3A, subpart 460 may correspond to subpart 351 defined by distance 2 and subregion 341. Figure 4F shows an image 470 related to the second subpart 462 acquired through a second objective lens 220 arranged according to some embodiments of the present disclosure. Image 470 includes an image of target cells 471. The image of target cells 471 may include a first predetermined range in the RGB domain or the HSV domain.

[0037] In some embodiments, relating to Figures 4B and 4D, the artificial intelligence engine is configured to determine whether the number of target cells in images 430, 450, and 470 is greater than a threshold number of target cells. In some embodiments, relating to Figure 2, in response to determining a first number greater than the threshold number, the artificial intelligence engine is configured to prompt a warning indicating that the cytological specimen 210 is suspicious. For example, the warning may alert a physician that the patient associated with the cytological specimen 210 may be ill.

[0038] In some embodiments, as relating to Figures 4B and 4D, in response to the determination that the number of first target cells in images 430, 450, and 470 is less than or equal to a threshold number, the second objective lens 410 is further configured to acquire one or more images relating to another sub-part containing the target cells. This other sub-part is a sub-part different from sub-parts 420 and 460.

[0039] The above process may be repeated until the number of first target cells exceeds a threshold number, or, if the number of first target cells does not continue to exceed a threshold number, it may be repeated until the second objective lens 410 acquires an image from all sub-parts of the cytological specimen that contain the target cells.

[0040] Figure 5 shows an exemplary system 500 for acquiring images related to cytological specimens, arranged according to some embodiments of the present disclosure. The system 500 includes, but is not limited to, a computing unit 510, a camera 520, a multiple objective lens module 530, a stage 550, and a light source 560. In some embodiments, the stage 550 is configured to transport and move glass slides. Cytological specimens 540 are distributed on glass slides.

[0041] In some embodiments, the arithmetic unit 510 includes a processor, a memory subsystem, and a communication subsystem. The artificial intelligence engine described above can be implemented as a set of executable instructions executed by the processor and stored in the memory subsystem.

[0042] In some embodiments, the computing unit 510 is configured to generate control signals and transmit them to the camera 520, the multiple objective lens module 530, the stage 550, and the light source 560 via a communication subsystem. For example, the computing unit 510 is configured to control the light source 560 to generate light in a specific wavelength range relevant to the image characteristics of target cells distributed within the cytological specimen 540. For example, the specific wavelength range may correspond to color information relevant to the target cells, as described above. An example of a specific wavelength range may be about 530 nm to about 630 nm. Alternatively, another example of a specific wavelength range is about 450 nm to about 560 nm, preferably about 450 nm to about 530 nm.

[0043] In some embodiments, the computing unit 510 is configured to control the movement of the stage 550. As the stage 550 carries the glass slide and the cytological specimen 540, the movement of the stage 550 allows the area of ​​the cytological specimen 540 (for example, areas A, B, C, D, E, F, G, H, I, or J shown in Figure 2, or sub-areas 341, 343, or 345 in Figure 3A) and the field of view of the objective lens 533 or 535 to be aligned with the optical path of the light generated by the light source 560 and shown in Figure 5. Thus, the light generated by the light source 560 can pass through the area and the objective lens 533 or 535 and reach the camera 520.

[0044] In some embodiments, the computing unit 510 is configured to control the lens switching module 531 of the multiple objective lens module 530 to switch between different objective lenses (e.g., objective lenses 533 and 535). In other embodiments, the computing unit 510 is configured to control the camera 520, the multiple objective lens module 530, and / or the stage 550 to have objective lens 533 or objective lens 535 focus on a portion, sub-part, and / or second sub-part of the cytological specimen 540. In relation to Figure 2, objective lens 533 may correspond to the first objective lens 220, and in relation to Figure 4A, objective lens 555 may correspond to the second objective lens 410.

[0045] In some embodiments, the computing unit 510 is configured to control the camera 520 to acquire images of the in-focus portion of the cytological specimen 540. The computing unit 510 is also configured to control the camera 520 to transmit the acquired images to the computing unit 510 for further processing. Such processing includes, but is not limited to, identifying images related to target cells or non-target cells in the cytological specimen 540, and comparing one or more images acquired through the second objective lens 535 with images acquired through the first objective lens 533 to determine whether the ratio of the number of target cells in the images acquired through the second objective lens 535 to the number of target cells in the images acquired through the first objective lens 533 is within a predetermined range.

[0046] Figure 6 is a flowchart illustrating an exemplary process 600 for acquiring images related to target cells distributed within a cytological specimen, arranged according to several embodiments of the present disclosure. Process 600 may include one or more operations, functions, or actions performed by hardware, software, and / or firmware, as shown in blocks 610, 620, 630, 640, 650, and / or 660. Each block is not limited to the embodiments described. The outlined steps and operations are provided only as examples, and some steps and operations may be optionally consolidated into fewer steps and operations or extended into additional steps and operations without impairing the essence of the disclosed embodiments. Although the blocks are shown in order, these blocks may be executed in parallel or in an order different from that described herein. In some embodiments, process 600 can be applied to various scanning techniques for scanning a cytological specimen using corresponding different full-slide imaging scanners. Such scanning techniques include, but are not limited to, area scanning techniques or line scanning techniques.

[0047] Process 600 may begin with block 610 “Acquire a first image relating to a first region of the cytological specimen.” In some embodiments, relating to Figures 2 and 5, the processor 510 is configured to control the movement of the stage 550 so that a first region of the cytological specimen 210 (e.g., region H in Figure 2) is aligned with a first field of view of the first objective lens 220 / 533. The processor 510 is configured to control the light source 560 to generate light in a specific wavelength range corresponding to the color information relating to the target cells described above in the cytological specimen 210. An example of a specific wavelength range may be about 530 nm to about 630 nm. Alternatively, another example of a specific wavelength range is about 450 nm to about 560 nm, preferably about 450 nm to about 530 nm. The light passes through the first region and the first objective lens 220 / 533. The processor 510 is configured to control the camera 520 to acquire an image relating to the first region through the first objective lens 220 / 533.

[0048] In some embodiments, the first objective lens 220 / 533 has a first depth of field in the first field of view. The first depth of field may be less than the thickness of the cytological specimen 210. Therefore, in order to obtain a clear and in-focus image relating to the entire first region, the camera 520 is configured to sequentially acquire images of different layers within the first region of the cytological specimen 210 according to the first depth of field. For example, relating to Figure 2, the camera 520 is configured to sequentially acquire images of portions H1, H2, and H3. In some embodiments, the processor 510 is configured to identify target cells from the images of portions H1, H2, and H3 based on contrast, brightness, shape, and other image characteristics relating to target cells in the cytological specimen 210.

[0049] Block 610 can be repeatedly run on different regions of the cytological specimen (e.g., regions F, I, and J in Figure 2). For example, the processor 510 is configured to further acquire images of portions F1, F2, F3, I1, I2, I3, J1, J2, and J3 of the cytological specimen 210 and to identify target cells from the images of F1, F2, F3, I1, I2, I3, J1, J2, and J3 based on contrast, brightness, shape, and other image characteristics related to target cells within the cytological specimen 210.

[0050] In some embodiments, in response to the identification that target cells are largely dispersed in a specific layer (e.g., a second layer) within a region (e.g., regions F, H, I, and J) of the cytological specimen 210, the processor 510 is configured to acquire a first image (e.g., an image of portion H2) associated with that region (e.g., region H) of the cytological specimen. Finally, for example, the processor 510 is configured to acquire images of A2, B2, C2, D2, E2, F2, G2, H2, I2, and J2 associated with the respective regions A, B, C, D, E, F, G, H, I, and J.

[0051] Block 610 may be followed by Block 620, “Acquire a second set of images related to a first subregion of the cytological specimen.” In some embodiments, the processor 510 is configured to sequentially identify images related to the target cell from images A2, B2, C2, D2, E2, F2, G2, H2, I2, or J2. For example, relating to Figures 3A and 3B, the processor 510 is configured to identify image 333 related to the target cell. As previously mentioned, based on the spatial relationship between image 333 and image region H', and the correspondence between image region H' and region H, a subregion 343 within region H can be identified. Similarly, the processor 510 may be configured to identify image 331 related to the target cell and identify a subregion 341 within region H.

[0052] In some embodiments, as described above in relation to Figure 3A, the second objective lens is configured to acquire a more detailed image of the target cells 324 in the first sub-region 343. In some embodiments, as described above in relation to Figures 4A and 5, the processor 510 is configured to control the movement of the stage 550 so that the first sub-region of the cytological specimen (e.g., sub-region 343 in Figure 4A) is aligned with the second field of view of the second objective lens 410 / 535. The processor 510 is configured to control the light source 560 to generate light in a specific wavelength range corresponding to the color information of the target cells in the cytological specimen as described above. An example of a specific wavelength range may be about 530 nm to about 630 nm. Alternatively, another example of a specific wavelength range is about 450 nm to about 560 nm, preferably about 450 nm to about 530 nm. The light passes through the first sub-region and the second objective lens 410 / 535. The processor 510 is configured to control the camera 520 to acquire a second set of images related to the first sub-region through the second objective lens 410 / 535. The second set of images may include one or more images of the second sub-parts 421, 422, 423, and 424 of the sub-part 420.

[0053] In some embodiments, relating to Figures 4A, 4B, 4C, 4D, and 5, the processor 510 is configured to control and focus the second objective lens 410 / 535 to acquire a second set of images of the second sub-parts 422 and 423 based on the image characteristics of the target cell, such as color information associated with the target cell. For example, the second set of images (e.g., images 430 and 450) contains color information associated with the target cell, while images associated with the first sub-region 343 other than the second set of images (e.g., images of the second sub-parts 421 and 424) do not contain color information associated with the target cell.

[0054] Block 620 may be followed by Block 630, “Identifying the First Target Cell Count.” In some embodiments, the first target cell count in the image associated with the first subregion 343 is identified. More specifically, the target cell counts in images 430 and 450 are identified as the first target cell count in Block 630.

[0055] Block 630 may be followed by Block 640, "Determine the number of first target cells greater than the threshold number." In response to the determination in Block 630 that the number of first target cells identified is greater than the threshold number of target cells, Block 640 is followed by Block 650, "Prompt a warning." In some embodiments, as shown in Figure 2, Block 650 displays a warning indicating that the cytological specimen 210 is suspicious. On the other hand, if the determination in Block 630 is that the number of first target cells identified is less than or equal to the target cell threshold value, Block 640 is followed by Block 660, "Acquire a third image set related to the second subregion of the cytological specimen."

[0056] In block 660, in some embodiments, as described above in relation to Figures 3A, 4E, and 4F, the second objective lens is configured to acquire a more detailed image of the target cells 324 in the second sub-region 341. In some embodiments, as described above in relation to Figures 4A and 5, the processor 510 is configured to control the movement of the stage 550 so that the second sub-region of the cytological specimen (e.g., sub-region 341 in Figure 4E) is aligned with the second field of view of the second objective lens 410 / 535. The processor 510 is configured to control the light source 560 to generate light in a specific wavelength range corresponding to the color information of the target cells in the cytological specimen as described above. An example of a specific wavelength range may be about 530 nm to about 630 nm. Alternatively, another example of a specific wavelength range is about 450 nm to about 560 nm, preferably about 450 nm to about 530 nm. The light passes through the second sub-region and the second objective lens 410 / 535. The processor 510 is configured to control and focus the second objective lens 410 / 535 to acquire a third set of images of the second sub-region 462 based on the image characteristics of the target cells, such as color information associated with the target cells. Furthermore, in some embodiments, the number of second target cells in the image associated with the second sub-region 341 is identified. More specifically, the number of target cells in image 470 is identified as the number of second target cells in block 660.

[0057] In some embodiments, the processor 510 is configured to determine whether the sum of the number of first target cells associated with the first subregion 343 and the number of second target cells associated with the second subregion 341 is greater than the threshold number of target cells specified in block 640. In response to the determination that the sum is greater than the threshold number, the processor 510 is configured to issue a warning indicating that the cytological specimen 210 is suspicious, as shown in Figure 2.

[0058] In some embodiments, in response to the determination that the total is less than or equal to a threshold number, the processor 510 is configured to control the second objective lens 410 / 535 to focus and acquire an image related to another subregion based on the image characteristics of the target cells, e.g., color information associated with the target cells. In relation to Figure 3A, the other subregion may be a subregion of the first region H. Alternatively, in relation to Figure 2, the other subregion may be a subregion of another region (e.g., regions A, B, C, D, E, F, G, H, I, or J). The total number of target cells acquired from various subregions through the second objective lens is continuously compared to the threshold number of target cells specified in block 640. If the total exceeds the threshold number, in relation to Figure 5, the processor 510 is configured to issue a warning and process 600 terminates. If the total does not exceed the threshold number, in relation to Figure 5, the processor 510 controls the second objective lens 535 and process 600 is repeatedly executed until images of all subregions on the cytological specimen related to the target cells are acquired.

[0059] Figure 7 is a flowchart illustrating an exemplary process 700 for acquiring images related to target cells distributed within a cytological specimen, arranged according to several embodiments of the present disclosure. Process 700 may include one or more operations, functions, or actions performed by hardware, software, and / or firmware, as shown in blocks 710, 720, 730, and / or 740. Each block is not limited to the embodiments described. The outlined steps and operations are provided only as examples, and some steps and operations may be optionally consolidated into fewer steps and operations or extended into additional steps and operations without impairing the essence of the disclosed embodiments. Although the blocks are shown in order, these blocks may be executed in parallel or in an order different from that described herein. In some embodiments, process 700 can be applied to various scanning techniques for scanning a cytological specimen using corresponding different full-slide imaging scanners. Such scanning techniques include, but are not limited to, area scanning techniques or line scanning techniques.

[0060] Process 700 may begin with block 710 “Acquire one or more first images relating to a first region of the cytological specimen.” In some embodiments, relating to Figures 2 and 5, the processor 510 is configured to control the movement of the stage 550 so that a first region of the cytological specimen 210 (e.g., region H in Figure 2) is aligned with a first field of view of the first objective lens 220 / 533. The processor 510 is configured to control the light source 560 to generate light in a specific wavelength range corresponding to the color information relating to the target cells described above in the cytological specimen 210. An example of a specific wavelength range may be about 530 nm to about 630 nm. Alternatively, another example of a specific wavelength range is about 450 nm to about 560 nm, preferably about 450 nm to about 530 nm. The light passes through the first region and the first objective lens 220 / 533. The processor 510 is configured to control the camera 520 to acquire an image related to the first region through the first objective lens 220 / 533.

[0061] In some embodiments, the first objective lens 220 / 533 has a first depth of field in the first field of view. The first depth of field may be less than the thickness of the cytological specimen 210. Therefore, in order to obtain a clear and in-focus image relating to the entire first region, the camera 520 is configured to sequentially acquire images of different layers within the first region of the cytological specimen 210 according to the first depth of field. For example, relating to Figure 2, the camera 520 is configured to sequentially acquire images of portions H1, H2, and H3. In some embodiments, the processor 510 is configured to identify target cells from the images of portions H1, H2, and H3 based on contrast, brightness, shape, and other image characteristics relating to target cells in the cytological specimen 210.

[0062] Block 710 may be followed by Block 720, “Acquire further images relating to other areas of the cytological specimen.” In some embodiments, in Block 720, the operations performed in Block 710 may be repeated for different areas of the cytological specimen (e.g., areas F, I, and J in Figure 2). For example, the processor 510 is configured to acquire further images of portions F1, F2, F3, I1, I2, I3, J1, J2, and J3 of the cytological specimen 210 and to identify target cells from the images of F1, F2, F3, I1, I2, I3, J1, J2, and J3 based on contrast, brightness, shape, and other image characteristics relating to target cells within the cytological specimen 210.

[0063] Block 720 may be followed by Block 730, "Identify the number of first target cells and determine if the number of first target cells is greater than the threshold number." In some embodiments, the number of first target cells is identified from the first image acquired in Block 710 and further images acquired in Block 720. In some embodiments, the number of first target cells may be greater than the threshold number of target cells. If the target cells are epithelial cells at risk of malignancy and the threshold number is the threshold number used by physicians to diagnose cancer, then process 700 can conclude that the patient providing the cytological specimen may be a cancer patient. In relation to Figure 5, it should be noted that it is not necessary to use the second objective lens 535 to reach this conclusion. In Block 730, in relation to Figures 2 and 5, the processor 510 is configured to prompt a warning indicating that the cytological specimen 210 is suspicious.

[0064] Block 730 may be followed by Block 740, "Acquire a second set of images related to subregions of the cytological specimen." If the number of first target cells identified in Block 730 is greater than a threshold number, and therefore the patient is very likely to be a cancer patient, then, as relating to Figure 5, the images acquired through the second objective lens 535 can be used to examine the size and / or morphology of some of the most important target cells in the cytological specimen. Thus, in some embodiments, as relating to Figure 5, the processor 510 controls the movement of the stage 550 so that the second objective lens 535 acquires one image from each subregion of the cytological specimen, reducing the scanning time of the cytological specimen and acquiring images of the most important target cells in the cytological specimen.

[0065] The above examples can be implemented by hardware (including hardware logic circuits), software, firmware, or a combination thereof. The above examples can be implemented by any suitable arithmetic unit, computer system, wearable, etc. An arithmetic unit includes a processor, memory units, and physical NIC(s) that can communicate with each other via a communication bus or the like. An arithmetic unit may include a non-temporary computer-readable medium that stores instructions or program code that causes the processor to execute the processes described herein with reference to Figures 6 and 7 in response to execution by the processor. An arithmetic unit can communicate with a wearable and / or one or more sensors.

[0066] The technologies described above can be implemented using special-purpose hardwired circuitry, software and / or firmware combined with programmable circuitry, or a combination thereof. Special-purpose hardwired circuitry may take the form of, for example, application-specific integrated circuits (ASICs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), etc. The term "processor" is broadly interpreted to include processing units, ASICs, logic units, programmable gate arrays, etc.

[0067] Some aspects of the embodiments disclosed herein can be equivalently implemented on an integrated circuit as one or more computer programs running on one or more computers (e.g., one or more programs running on one or more computing systems), one or more programs running on one or more processors (e.g., one or more programs running on one or more microprocessors), firmware, or substantially any combination thereof, in whole or in part. In view of this disclosure, it is also possible to design the circuit and / or write the software and / or firmware code.

[0068] The software and / or other instructions for implementing the technologies described herein are stored on non-temporary computer-readable media and can be executed by one or more general-purpose or dedicated programmable microprocessors. As used herein, “computer-readable media” includes any mechanism that provides (i.e., stores and / or transmits) information in a form accessible to machines (such as computers, network devices, personal digital assistants (PDAs), mobile devices, manufacturing tools, and any device with one or more sets of processors). Computer-readable media include recordable / non-recordable media (such as read-only memory (ROM), random-access memory (RAM), magnetic disks or optical storage media, flash memory devices, etc.).

[0069] From the foregoing, it will be understood that various embodiments of the present disclosure are described herein for illustrative purposes and that various modifications are possible without departing from the scope and spirit of the present disclosure. Accordingly, the various embodiments disclosed herein are not intended to be limiting.

Claims

1. A method for identifying suspected cytological specimens containing target cells, A first image relating to a first region of the cytological specimen is acquired through a first objective lens having a first field of view, a first magnification, and a first depth of field in the first field of view. Acquiring a second set of one or more images related to a first sub-region of the first region through a second objective lens having a second field of view, a second magnification, and a second depth of field in the second field of view, From the second set of one or more images, the number of first target cells distributed within the space defined by the first subregion and the first depth of field of the cytological specimen is identified. To determine whether the number of the first target cells is greater than the threshold number of target cells, In response to the determination that the number of the first target cells is greater than the threshold number of target cells, a warning is issued indicating that the cytological specimen is suspicious. In response to the determination that the number of first target cells is less than or equal to the threshold number of target cells, a third set of one or more images related to the second sub-region of the first region is acquired through the second objective lens, and the number of second target cells distributed in the second space defined by the second sub-region and the first depth of field is identified. To determine whether the sum of the number of the first target cells and the number of the second target cells is greater than the threshold number, including, method.

2. The first region corresponds to the first field of view. The method according to claim 1.

3. The second field of view corresponds to one or more parts of the first subregion. The method according to claim 1.

4. The second magnification is greater than the first magnification, and the second depth of field is smaller than the first depth of field. The method according to claim 1.

5. Before obtaining the first image mentioned above, To acquire images related to other regions of the cytological specimen through the first objective lens, Based on the images related to the other regions, the first layer of the cytological specimen containing the target cells, corresponding to the first depth of field, Acquiring an image of the first layer through the first objective lens, wherein one of the images of the first layer is the first image; This also includes, The method according to claim 1.

6. The image characteristics related to the target cell include contrast, brightness, shape, and color information related to the target cell. The method according to claim 1.

7. The aforementioned color information corresponds to light having a first wavelength range of approximately 530 nm to approximately 630 nm, a second wavelength range of approximately 450 nm to approximately 560 nm, or a third wavelength range of approximately 450 nm to approximately 530 nm. The method according to claim 6.

8. The aforementioned color information includes the range of R, G, and B values ​​in the RGB (red, green, blue) domain, or the range of hue values ​​in the HSV (hue, saturation, lightness) domain. The method according to claim 6.

9. In response to the determination that the sum is greater than the threshold number, the warning is issued. In response to the determination that the sum is less than or equal to the threshold number, a fourth set of one or more images relating to the third subregion of the first region is acquired through the second objective lens, or a fifth set of one or more images relating to the fourth subregion of the second region of the cytological specimen is acquired. The method according to claim 1.

10. Obtaining a second set of one or more images is possible. This includes driving the second objective lens to focus on a first space of the cytological specimen defined by the first sub-region and the second depth of field, and acquiring an image of the first space which is one of the second sets of one or more images, The first space is associated with the second layer of the cytological specimen. The method according to claim 1.

11. The process further includes, before obtaining the second set of one or more images, obtaining further images related to other areas of the cytological specimen through the first objective lens, Identifying the number of the first target cells is done before acquiring the second set of one or more images, and after acquiring the further images. The method according to claim 1.

12. The number of first target cells distributed within the space is identified from the first image and the further images. The space is defined by the first region, the first depth of field, the other region, and the first depth of field. The method according to claim 11.

13. Obtaining the second set of one or more images further includes obtaining a single image relating to the first subregion. The method according to claim 12.

14. The aforementioned warning is issued before the second set of the one or more images is obtained. The method according to claim 11.

15. A non-temporary computer-readable medium for identifying suspicious cytological specimens, In response to execution by one or more processors, instructions are stored in one or more of the processors that cause them to perform the operation according to any one of claims 1 to 14. A non-temporary computer-readable medium.

16. A system for identifying suspicious cytological specimens, One or more processors, A non-temporary computer-readable medium connected to one or more of the processors, which stores instructions that cause one or more of the processors to perform the operation according to any one of claims 1 to 14 in response to execution by one or more of the processors, Equipped with, system.

Images