A method and system for analyzing formed components of a sample

Abstract

The present invention provides a method and system for analyzing formed elements in samples. Multiple component images are obtained by vertically capturing multiple layers of the sample. For each target formed element captured in the sample, a first sub-image with a resolution meeting the required resolution and a second sub-image with a different resolution are obtained from these component images. A second target image is then obtained based on the first and second sub-images of each target formed element. Each target formed element in the second target image is derived from the first and second sub-images, resulting in higher image quality than a single component image, thereby improving the detection efficiency of formed elements.

Description

Technical Field

[0001] This invention relates to the field of medical devices, and more specifically to a method and system for analyzing formed components in a sample. Background Technology

[0002] In current microscopic morphological examinations of urine's formed elements, the high-transmittance components of urine cause only slight changes in light intensity, resulting in low contrast under a bright-field microscope (e.g., ...). Figure 1 As shown in the bright-field image on the left, these elements are not easily detected, leading to missed detections. To address this issue, current methods primarily utilize optical phase contrast (PH) / dark-field contrast (DF) / differential interference contrast (DIC) to enhance the optical contrast of components, thereby strengthening the image and aiding in the detection of formed elements. Figure 1 The phase contrast image on the right is shown.

[0003] However, the above solution has the following drawbacks:

[0004] 1) Additional optical components and optical path switching devices are required, increasing cost and mechanical complexity.

[0005] 2) Due to the changes involved in the optical path, it is not possible to acquire both bright field and optically enhanced images simultaneously; acquiring both requires additional time.

[0006] 3) Optical image enhancement, such as phase contrast and dark field imaging, reduces the light transmission of the microscope objective, resulting in an exposure time that is many times longer to acquire an enhanced image with sufficient contrast. This limits the imaging speed of urine formed element detection instruments, thereby reducing the test statistics and detection speed.

[0007] Therefore, the detection efficiency of existing formed element analysis equipment needs to be improved. Summary of the Invention

[0008] This invention provides a method and system for analyzing formed elements in samples, aiming to improve the detection efficiency of formed elements.

[0009] One embodiment provides a formed element analysis system for a sample, comprising:

[0010] Sample-carrying components are used to carry samples;

[0011] A sample feeding component is used to pick up a sample and provide the picked-up sample to the sample carrying component;

[0012] An imaging component is used to capture images of the sample from the sample-bearing component to obtain images of the formed elements in the sample;

[0013] Processor, used for:

[0014] The imaging component is controlled to perform multi-layer imaging of the sample on the sample carrier in the vertical direction to obtain multiple component images of the sample corresponding to each layer.

[0015] Acquiring a first target sub-image and / or a first target image includes: for the formed elements of the target being photographed in the sample, acquiring a first sub-image from multiple component images where the clarity of the formed elements of the target meets the clarity requirements, wherein the first sub-image is a sub-image in the component images that contains part or all of the formed elements of the target; obtaining a first target sub-image containing the complete formed elements of the target based on the acquired first sub-image of the formed elements of the target, and / or obtaining a first target image containing multiple formed elements of the target based on the acquired first sub-image of multiple formed elements of the target;

[0016] Acquiring a second target sub-image and / or a second target image includes: for the target formed element captured in the sample, acquiring a second sub-image from multiple component images where the target formed element has a different clarity from that in a first sub-image, the second sub-image being a sub-image in the component images containing part or all of the target formed element; obtaining a second target sub-image containing the complete target formed element based on the acquired first and second sub-images of the target formed element; and / or, based on the acquired first and second sub-images of multiple target formed elements, or based on the first target image and multiple second sub-images of target formed elements, obtaining a second target image containing multiple target formed elements.

[0017] One embodiment provides a formed element analysis system for a sample, comprising:

[0018] Sample-carrying components are used to carry samples;

[0019] A sample feeding component is used to pick up a sample and provide the picked-up sample to the sample carrying component;

[0020] An imaging component is used to capture images of the sample from the sample-bearing component to obtain images of the formed elements in the sample;

[0021] Processor, used for:

[0022] Determine whether the sample belongs to a preset category; if the sample belongs to a preset category, acquire a first target sub-image and / or a first target image, and acquire a second target sub-image and / or acquire a second target image; otherwise, acquire a first target sub-image and / or a first target image; wherein, the first target sub-image and / or the first target image is obtained based on a first sub-image whose clarity of the target formed elements meets the clarity requirements, and the second target sub-image and / or the second target image is obtained based on a first sub-image whose clarity of the target formed elements meets the clarity requirements and a second sub-image whose clarity is different from that of the first sub-image.

[0023] One embodiment provides a method for analyzing the formed elements of a sample, comprising:

[0024] The sample carrying the sample is controlled to be photographed in multiple layers in the vertical direction to obtain multiple component images of the sample in each layer;

[0025] Acquiring a first target sub-image and / or a first target image includes: for the formed elements of the target being photographed in the sample, acquiring a first sub-image from multiple component images where the clarity of the formed elements of the target meets the clarity requirements, wherein the first sub-image is a sub-image in the component images that contains part or all of the formed elements of the target; obtaining a first target sub-image containing the complete formed elements of the target based on the acquired first sub-image of the formed elements of the target, and / or obtaining a first target image containing multiple formed elements of the target based on the acquired first sub-image of the formed elements of the target;

[0026] Acquiring a second target sub-image and / or a second target image includes: for the target formed element captured in the sample, acquiring a second sub-image from multiple component images where the target formed element has a different clarity from that in a first sub-image, the second sub-image being a sub-image in the component images containing part or all of the target formed element; obtaining a second target sub-image containing the complete target formed element based on the acquired first and second sub-images of the target formed element; and / or, based on the acquired first and second sub-images of multiple target formed elements, or based on the first target image and multiple second sub-images of the target formed element, obtaining a second target image containing multiple target formed elements.

[0027] One embodiment provides a formed element analysis system for a sample, comprising:

[0028] Sample-carrying components are used to carry samples;

[0029] A sample feeding component is used to pick up a sample and provide the picked-up sample to the sample carrying component;

[0030] An imaging component is used to capture images of the sample from the sample-bearing component to obtain images of the formed elements in the sample;

[0031] Processor, used for:

[0032] The imaging component is controlled to perform multi-layer imaging of the sample on the sample carrier in the vertical direction to obtain multiple component images of the sample corresponding to each layer.

[0033] A first image containing the target formed element is obtained from at least one of a plurality of component images, wherein the sharpness of the target formed element in the first image meets a sharpness requirement; and,

[0034] A second image containing the target formed element is obtained from at least two of the multiple component images, or a second image containing the target formed element is obtained from the first image and at least one of the multiple component images; wherein the contrast of the target formed element in the second image is greater than its contrast in the first image.

[0035] According to the above embodiments, a sample formed element analysis method and system involves capturing multiple component images of a sample in a vertical direction. For each target formed element captured in the sample, a first sub-image with a resolution meeting the required resolution and a second sub-image with a different resolution are obtained from these component images. A second target image is then obtained based on the first and second sub-images of each target formed element. Each target formed element in the second target image is derived from the first and second sub-images, resulting in a higher image quality than a single component image, thereby improving the detection efficiency of formed elements. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the formation of distinct field images and phase contrast images in the prior art;

[0037] Figure 2 A structural block diagram of an embodiment of the sample formation analysis system provided by the present invention;

[0038] Figure 3 A flowchart of an embodiment of the sample formation analysis method provided by the present invention;

[0039] Figure 4 A schematic diagram of multi-layer imaging for the imaging component;

[0040] Figure 5 for Figure 3 A flowchart of an embodiment of step 2;

[0041] Figure 6A schematic diagram for obtaining the first target sub-image and / or the first target image;

[0042] Figure 7 for Figure 3 Flowchart of step 3 of the embodiment;

[0043] Figure 8 A schematic diagram illustrating the principle of resolution calculation;

[0044] Figure 9 A schematic diagram illustrating the capture of a positive-focus image when the formed elements are within the depth of field of the focal plane;

[0045] Figure 10 A schematic diagram of a telephoto image taken when the formed elements are outside the depth of field of the focal plane (away from the objective lens direction);

[0046] Figure 11 A schematic diagram of a near-focus image taken when the formed elements are outside the depth of field of the focal plane (near the objective lens);

[0047] Figure 12 A schematic diagram illustrating the process of obtaining the second target sub-image from the first and second sub-images;

[0048] Figure 13 A schematic diagram for determining the second sub-image based on the focal plane of the first sub-image;

[0049] Figure 14 A flowchart of another embodiment of the sample formation analysis method provided by the present invention;

[0050] Figure 15 The flowchart is a further embodiment of the sample formed element analysis method provided by the present invention. Detailed Implementation

[0051] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of this application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to this application are not shown or described in the specification. This is to avoid obscuring the core parts of this application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0052] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0053] The serial numbers assigned to components in this document, such as "first" and "second," are used only to distinguish the described objects and have no sequential or technical meaning. The terms "connection" and "linkage" used in this application, unless otherwise specified, include both direct and indirect connections (linkages).

[0054] This invention improves the detection efficiency of formed elements by acquiring multiple component images and then using multi-image fusion to improve image quality. The following are some detailed examples.

[0055] like Figure 2 As shown, the formed component analysis system for samples provided by the present invention includes: a processor 10, an imaging component 20, a sample carrying component 30, and a sample loading component 40.

[0056] The sample carrier component 30 is used to carry the sample.

[0057] The sample feeding component 40 is used to pick up the sample and provide the picked-up sample to the sample carrying component 30.

[0058] Imaging component 20 is used to photograph the sample on sample carrier 30 to obtain component images of the formed elements in the sample. Imaging component 20 may include a microscope and an image acquisition device. The image acquisition device is used to acquire component images of the sample after magnification by the microscope. The component images vary depending on the sample and may be one or more of the following: cell images, crystal images, sediment images, and parasite images.

[0059] The processor 10 controls other components of the formed element analysis system to achieve the various functions of the system. For example, the processor 10 controls the formed element analysis system to perform formed element analysis on a sample, and the specific process can be as follows: Figure 3 As shown, it includes the following steps:

[0060] Step 1: The processor 10 controls the imaging component 20 to perform multi-layer imaging of the sample on the sample-bearing component in the vertical direction to obtain multiple component images of the sample corresponding to each layer. For example... Figure 4As shown, sample a has a certain height. The focusing distance L of imaging component 20 is usually fixed. Since the relative height (vertical height) between imaging component 20 and sample-bearing component 30 is determined, the position of the focal plane is also determined. Imaging component 20 performs multiple images of sample a in the vertical direction, which is equivalent to imaging component 20 capturing sample a at multiple different vertical heights. Each image corresponds to one focal plane (as shown by the dotted line in the figure). Thus, due to the different focal plane heights, the sharpness of the same formed element in the various component images obtained will differ.

[0061] The sample carrier component 30 can have various structures, such as a counting cell or a flow chamber. The specific structures of the two are different. Although the principle of multi-layer imaging is the same, the specific methods are slightly different, which will be explained below.

[0062] In one embodiment, the sample carrying component 30 includes a counting cell. The counting cell is used to contain a sample provided by the sample dispensing component 40. The sample may be a sample that has undergone sedimentation, centrifugation, or sheath flow enrichment.

[0063] Imaging component 20 has a corresponding focusing distance L for capturing samples from the counting cell at the focusing distance L to obtain a composition image.

[0064] The formed element analysis system also includes a drive device 60 for driving the imaging component 20 and the counting cell to move relative to each other at least in the vertical direction. Specifically, it can drive the imaging component 20 to move in the vertical direction, or drive the counting cell to move in the vertical direction, or drive both of them to move in the vertical direction.

[0065] The processor 10 controls the imaging assembly 20 to perform multi-layer imaging of the sample on the sample carrier 30 to obtain multiple component images of the sample at each layer. Specifically, this may include: controlling the driving device 60 to drive the imaging assembly 20 and the counting cell to move relative to each other in the vertical direction, so that the imaging assembly 20 is at different vertical heights relative to the counting cell at a focusing distance L (e.g., ...). Figure 4 Images are taken on multiple layers of the sample (as shown), thus obtaining multi-layer images of the sample (e.g., ...). Figure 4 The dashed lines in the image correspond to multiple component images.

[0066] In another embodiment, the sample carrier 30 includes a flow chamber. The flow chamber is used for the sample provided by the sample delivery component 40 to be enveloped by a flat sheath fluid flow to form a planar laminar flow.

[0067] The imaging assembly 20 includes multiple imaging components capable of capturing images simultaneously.

[0068] The processor 10 controls the imaging assembly to perform multi-layer imaging of the sample on the sample carrier 30 to obtain multiple component images of the sample corresponding to each layer. Specifically, this may include: multiple imaging components located above the flow chamber, with different imaging parts focused at different heights of the laminar flow passing through the flow chamber (i.e., the focal plane heights of each imaging part are different, such as...). Figure 4 (As shown by multiple dashed lines), multiple imaging components are controlled to simultaneously capture images of the laminar flow passing through the flow chamber, thereby performing multi-layer imaging of the sample to obtain multiple component images of the sample corresponding to each layer.

[0069] Step 2: Processor 10 acquires the first target sub-image and / or the first target image. The specific process can be as follows: Figure 5 As shown, steps 21 and 22 are included.

[0070] Step 21: For the formed elements of the target sample being photographed ( Figure 6 a1, a2, and a3 in the image are examples. The processor 10 processes multiple component images ( Figure 6 In the example (A1, A2, and A3 in the figure), the first sub-image that meets the sharpness requirement for the formed element of the target is obtained. This first sub-image is a sub-image containing part or all (in the example in the figure, all) of the formed element of the target in the component image. A component image typically contains multiple formed elements, and the target formed element can be one or more of them, as shown by a1-a3 in the figure. Multiple component images (such as A1-A3) are equivalent to taking multiple photos of these target formed elements at different heights. Due to differences in the size, thickness, etc., of the target formed elements, some target formed elements (such as a2) are clearer in component image A1, while others (such as a3) are clearer in component image A2. Therefore, for each target formed element, a sub-image with relatively high sharpness is selected from the various component images as the first sub-image, or the first sub-image is obtained through further calculation.

[0071] The processor 10 acquires a first sub-image of the target formed elements whose sharpness meets the sharpness requirements from multiple component images. There are several ways to achieve this, and two examples are given below.

[0072] The first method involves the processor 10 detecting target formed elements in each component image, such as detecting target formed elements a1-a3 in each component image A1-A3, and calculating the sharpness of the detected target formed elements in the sub-image containing the target formed elements. Specifically, there are multiple ways to detect target formed elements from component images, and one or more of these methods can be used for target formed element detection. Examples of these methods are given below.

[0073] The first detection method involves the processor 10 detecting target formed elements from a component image based on an image segmentation method. Image segmentation, the process of dividing an image into multiple regions or objects, is a crucial step in detecting target formed elements. Specifically, image segmentation can employ one or more of the following segmentation methods:

[0074] 1. Thresholding Segmentation: Thresholding segmentation methods based on image grayscale values, such as histogram bimodal method, iterative thresholding method, Otsu thresholding method, etc. Alternatively, improved two-dimensional maximum entropy thresholding segmentation methods can be used, combining point grayscale and region grayscale features of the image for segmentation.

[0075] 2. Edge detection: Differential operators are used to detect edges in an image, such as the Laplace operator and the Canny operator. Edge detection helps to identify formed element boundaries in an image.

[0076] 3. Morphological segmentation: This method uses mathematical morphology operations such as opening and closing, dilation, and erosion to segment images. Morphological segmentation helps remove noise and fill holes in images.

[0077] 4. Machine Learning Methods: Image segmentation is performed using machine learning methods such as Support Vector Machines and Random Forests. This method typically requires a large amount of training data to train the model.

[0078] 5. Based on feature extraction methods, image features are extracted from the component images, and classification and recognition are performed based on the extracted image features to obtain the target formed elements in the component images. Specifically, one or more of the following methods can be used for feature extraction:

[0079] (1) Shape features: Extract the shape features of objects in the image, such as area, perimeter, and roundness.

[0080] (2) Texture features: Extract the texture features of objects in the image, such as gray-level co-occurrence matrix, wavelet transform, etc.

[0081] (3) Color features: In color images, color features of objects can be extracted, such as RGB values, HSV values, etc.

[0082] The second detection method involves processor 10 detecting target formed elements from a component image using a machine learning model. The machine learning model can be pre-trained using component images labeled with target formed elements. Processor 10 then inputs the component image into the machine learning model to obtain the target formed elements labeled by the model within the component image. In some examples, the machine learning model can be a neural network. For instance, a backpropagation (BP) neural network: a commonly used neural network model for image classification and recognition. By training a BP neural network, it learns to map features of the input image to corresponding category labels. Processor 10 inputs the image features into the trained BP neural network to obtain the categories of formed elements output by the BP neural network, thus identifying the target formed elements in the component image.

[0083] After the processor 10 detects the target formed elements, the positions of each target formed element in the component image are known. The component image is typically a two-dimensional image; if a coordinate system is established using the component image, the pixel coordinates of each target formed element are also known. The processor 10 can determine the same target formed element in different component images based on its position in the component image. For example, it can determine whether a target formed element is the same by its pixel coordinates in the two-dimensional component image. For the same target formed element: the processor 10 calculates the sharpness of the target formed element in the sub-image containing it. Based on the sharpness of the target formed element, it obtains one or more sub-images from the multiple component images containing the target formed element whose sharpness meets the sharpness requirements, thus obtaining the first sub-image of the target formed element. Understandably, the first sub-image of the target formed element obtained in this way is a sub-image in the component image that contains the complete target formed element; that is, the sub-image contains the complete target formed element. In some examples, meeting the sharpness requirement includes having a sharpness not lower than a first sharpness. In some examples, meeting the sharpness requirement includes having a sharpness of the target formed element in its first sub-image that is higher than the sharpness of the target formed element in other sub-images.

[0084] Therefore, in some examples, if there is only one sub-image of the target formed element whose clarity meets the clarity requirement, then that sub-image can be used as the first sub-image of the target formed element.

[0085] In some examples, from multiple component images containing sub-images of the target formed element, the sub-image with the highest sharpness is selected as the first sub-image of the target formed element. That is, meeting the sharpness requirement means that the sharpness of the target formed element in its first sub-image is higher than the sharpness of the target formed element in other sub-images. In some examples, the sharpness of the first sub-image of the target formed element selected in this way is not lower than that of the first sharpness processor 10, which selects the sub-image with the highest sharpness from multiple component images containing the target formed element as the first sub-image of the target formed element. The highest sharpness of the first sub-image of the target formed element generally means that the target formed element was substantially within the depth of field of the focal plane when the component image to which the first sub-image belongs was captured.

[0086] In some examples, if there are multiple sub-images whose sharpness meets the required sharpness for the target formed element, these multiple sub-images are processed to obtain a first sub-image of the target formed element. Processing multiple sub-images to obtain a single first sub-image is equivalent to image enhancement. For example, a weighted calculation is performed on the multiple sub-images to obtain the first sub-image of the target formed element; since each sub-image used for the weighted calculation is a component image containing the complete target formed element, the first sub-image of the target formed element obtained after the weighted calculation is also a component image containing the complete target formed element. Alternatively, these multiple sub-images can be input into a pre-trained machine learning model or deep learning model (such as a neural network model) to obtain the first sub-image of the target formed element output by the machine learning model or deep learning model.

[0087] It should be noted that, in this paper, a sub-image containing part or all of the formed elements of the target in a component image can refer to the image corresponding to a sub-region of the component image, or it can refer to the image obtained by weighted calculation of the images corresponding to the same sub-region in multiple component images.

[0088] The first method for obtaining a first sub-image of a target formed element requires detecting the target formed element from the component images. The sub-image (e.g., an image of the bounding rectangle region of the target formed element, or an image of the target formed element itself) contains the target formed element. The second method, however, involves dividing the component images into regions. Specifically, in the second method for obtaining a first sub-image of a target formed element, each component image is divided into multiple regions according to the same rules, with each region corresponding to a sub-image. The processor 10 calculates the sharpness of the target formed element in each region, or each sub-image. For each divided region, the processor 10 selects from the sub-images of the same region in multiple component images the sub-images where the sharpness of the target formed element meets the sharpness requirements, thus obtaining the first sub-image of that target formed element. Understandably, the first sub-image of the target formed element obtained in this way may be a sub-image containing a portion of the target formed element in the component image (i.e., the sub-image contains a portion of the target formed element, or contains a portion of the target formed element but not the entire target formed element), or it may be a sub-image containing the complete target formed element in the component image (i.e., the sub-image contains the complete target formed element). In some examples, meeting the sharpness requirement includes having a sharpness not lower than a first sharpness. In some examples, meeting the sharpness requirement includes having a sharpness of the target formed element in its first sub-image that is higher than the sharpness of other sub-images of the target formed element in the same region.

[0089] Therefore, in some examples, if the sharpness of the formed elements of the target in the same region meets the sharpness requirements, then this sub-image can be used as the first sub-image of the formed elements of the target.

[0090] In some examples, the sub-image with the highest sharpness of the target formed element is selected from sub-images within the same region as the first sub-image of that target formed element. That is, the sharpness of the target formed element in its first sub-image is higher than the sharpness of other sub-images of the target formed element within the same region. In some examples, the sharpness of the first sub-image of the target formed element selected in this way is not lower than the first sharpness.

[0091] In some examples, if multiple sub-images of the target formed element within the same region meet the sharpness requirements, these multiple sub-images are processed to obtain the first sub-image of the target formed element. Processing multiple sub-images to obtain a single first sub-image is equivalent to image enhancement. For example, weighted calculations can be performed on the multiple sub-images to obtain the first sub-image of the target formed element. Alternatively, these multiple sub-images can be input into a pre-trained machine learning model or deep learning model (such as a neural network model) to obtain the first sub-image of the target formed element output by the machine learning model or deep learning model. In the second method for obtaining the first sub-image of the target formed element, since the component image is divided into regions, when generating the first target sub-image or the first target image, the first sub-images of each region need to be synthesized together. Whether through weighted processing or obtaining the first sub-image using an AI model, the transition between the first sub-images of different regions within the first target sub-image or the first target image can be made smoother.

[0092] In the process of acquiring the first sub-image of the target formed elements, regardless of whether the first method or the second method is used, the processor 10 needs to calculate the sharpness of the target formed elements in the sub-image. There are many ways to calculate the sharpness, and several examples are given below.

[0093] In one approach, processor 10 can calculate feature values ​​of a sub-image containing the target formed element. These feature values ​​may include one or more of the following: grayscale variance, gradient value, information entropy, and spectral entropy. Processor 10 can determine the sharpness of the target formed element in the sub-image based on these feature values. The feature values ​​reflect the intensity of the target feature in the sub-image; that is, they are indicators of sharpness. The target feature is the non-background feature in the sub-image. Here, the background refers to the liquid without formed elements, i.e., the image background or background image. The target feature can be distinguished from the background by transmittance or refractive index; for example, an image in which the transmittance or refractive index differs from the background is considered a target feature. Figure 8 As shown, because light converges more when an image is sharp compared to when it's blurry, the intensity of target features is higher. Therefore, image processing algorithms can be used to calculate the intensity changes of target features to evaluate sharpness. Commonly used image processing algorithms include image gradient operators, Sobel operators, and Laplacian operators. Eigenvalues ​​can reflect the intensity changes of target features, from... Figure 8It is known that a clear image has a large intensity variation, so the larger the feature value, the clearer the image. Therefore, the processor 10 can take the sub-image in which the feature value of the sub-image containing the target formed component is not lower than a preset first threshold (corresponding to the first clarity) as the first sub-image, or take the sub-image with the largest feature value among the feature values ​​of the sub-image containing the target formed component as the first sub-image.

[0094] In another approach, the processor 10 can also calculate the number of target features in the sub-image containing the target formed element. Generally, the clearer the image, the more target features are obtained. Therefore, the clarity of the target formed element in the sub-image can be determined based on the number of target features. For example, the processor 10 can use the sub-image containing the target formed element where the number of target features is not less than a preset second threshold (corresponding to the first clarity) as the first sub-image, or use the sub-image containing the target formed element where the number of target features is the largest among the feature values ​​as the first sub-image.

[0095] Step 22: The processor 10 obtains the first target sub-image and / or the first target image based on the first sub-image of the formed elements of the target.

[0096] The following explains how step 22 obtains the first target sub-image.

[0097] In some embodiments, the processor 10 obtains a first target sub-image containing the complete target formed element based on the acquired first sub-image of the target formed element. In some examples, the first sub-image of the target formed element is a sub-image of the component image containing the complete target formed element, and the processor 10 uses the first sub-image of the target formed element as the first target sub-image of the target formed element. In some examples, the processor 10 obtains a first target image containing multiple target formed elements based on the acquired first sub-images of multiple target formed elements, detects the target formed element based on the first target image to obtain a sub-image of the target formed element in the first target image, and uses it as the first target sub-image of the target formed element.

[0098] In some examples, the sharpness of the first sub-image of the target formed elements is no less than that of the first sharpness.

[0099] Considering that in some embodiments, the clarity of the first sub-image of the target formed elements is not lower than the first clarity, the clarity of the first target sub-image obtained from the first sub-image of the target formed elements is also not lower than the first clarity.

[0100] In some examples, the sharpness of the first target sub-image of the target formed elements is not lower than that of the first sharpness.

[0101] The following explains how step 22 obtains the first target image.

[0102] In some embodiments, a first target image containing multiple target formed elements is obtained from the first sub-images of the acquired target formed elements. The target formed element images in each of the first sub-images are relatively clear, such as a1 in A3, a2 in A1, and a3 in A2. Therefore, the first target image B obtained by combining these relatively clear target formed element images is clearer overall than the component images (A1, A2, and A3). The position and size (proportion) of each target formed element in the first target image B are the same as the position and size of each target formed element in the component images, which is equivalent to improving the clarity of the component images through this step. The target formed elements corresponding to the first sub-image are usually located within the depth of field of the focal plane of the imaging component 20 during shooting. The first sub-images of each target formed element may be located within the depth of field of different focal planes (different vertical heights), therefore, this step can be called depth-of-field fusion.

[0103] In some examples, the sharpness of the first sub-image of the target formed elements is no less than that of the first sharpness.

[0104] Considering that in some embodiments, the clarity of the first sub-image of each target formed component is not lower than the first clarity, the clarity of the first target image obtained from the first sub-image of each target formed component is not lower than the first clarity.

[0105] In some examples, the sharpness of the first target image is no less than that of the first sharpness.

[0106] The processor 10 can also output a first target image and / or a first target sub-image of at least one target formed element. For example, the formed element analysis system may also include a human-computer interaction device 50. The human-computer interaction device 50 is used to output visual information and receive user input; for example, it may include a display and an input device for receiving user input. The input device may be a mouse, keyboard, various buttons, a touchscreen, etc. The processor 10 can output the first target image B to the display, allowing the display to show the first target image B, so that medical personnel can see a high-resolution panoramic image containing multiple target formed elements. Of course, the processor 10 can also output one or more first target sub-images of target formed elements, for example, outputting one or more first target sub-images of target formed elements to the display, allowing the display to show multiple first target sub-images of target formed elements, so that medical personnel can see a high-resolution image of the target formed elements.

[0107] In some examples where the sample carrying component 30 includes a flow chamber, since the flow chamber is used for the sample provided by the sample feeding component 40 to be wrapped by a flat sheath fluid flow to form a planar laminar flow, multiple formed elements in the sample may pass through the flow chamber in batches. In such examples, it is more meaningful and in line with user needs for the processor 10 to acquire and output a first target sub-image of one or more target formed elements.

[0108] Step 3: The processor 10 acquires the second target sub-image and / or the second target image, for example, acquiring the second target sub-image and / or the second target image based on the second sub-image of the formed components of the target. The specific process can be as follows: Figure 7 As shown, steps 31 and 32 are included.

[0109] Step 31: For the sample, the formed elements of the target being photographed ( Figure 6 a1, a2, and a3 in the image are examples. The processor 10 processes multiple component images ( Figure 6 In the example of A1, A2, and A3, the target formed element is obtained in a second sub-image (a1 in A1, a2 in A2, and a3 in A3 are examples) where the formed element is different from the first sub-image in terms of sharpness. The second sub-image is a sub-image in which the formed element of the target is partially or completely contained in the component image.

[0110] Understandably, if the first sub-image of a target formed element is a sub-image containing a portion of the target formed element, generally, the second sub-image of the target formed element is also a sub-image containing a portion of the target formed element; if the first sub-image of a target formed element is a sub-image containing the complete target formed element, generally, the second sub-image of the target formed element is also a sub-image containing the complete target formed element.

[0111] The following explains how step 32 obtains the second target sub-image.

[0112] In some embodiments, the processor 10 obtains a second target sub-image containing the complete target formed element based on the acquired first and second sub-images of the target formed element. In some examples, both the first and second sub-images of the target formed element are sub-images containing the complete target formed element in the component image, and the processor 10 calculates the difference between the first and second sub-images of the target formed element to obtain the second target sub-image of the target formed element. In some examples, the processor 10 obtains a second target image containing multiple target formed elements based on the acquired first and second sub-images of multiple target formed elements, or based on a first target image and multiple second sub-images of target formed elements; the processor 10 detects the target formed element based on the second target image to obtain a sub-image of the target formed element in the second target image and uses it as the second target sub-image of the target formed element.

[0113] In some examples, the contrast of the second target sub-image is not lower than that of the first contrast base.

[0114] The following explains how step 32 obtains the second target image.

[0115] Processor 10 can obtain a second target image containing multiple target formed elements based on the acquired first and second sub-images of multiple target formed elements. For example, for each target formed element, there are first and second sub-images with different resolutions. Image enhancement can be performed on the target formed element based on the first and second sub-images, so that a second target image can be obtained (e.g., synthesized) after image enhancement of each target formed element, which is equivalent to image enhancement of the component image. Processor 10 can also obtain a second target image containing multiple target formed elements based on the first target image and the second sub-images of multiple target formed elements. For example, if each target formed element in the first target image has high resolution, the first sub-image in the first target image can be enhanced using the second sub-images of these target formed elements to obtain the corresponding second target image. These two methods are essentially the same, and the resulting second target images are also similar. Obviously, the image quality of the second target image is better than that of the first target image, thus increasing the accuracy of formed element detection and improving detection efficiency.

[0116] In some examples, the contrast of the second target image is not lower than that of the first image.

[0117] like Figure 9-11 As shown, since formed elements may be translucent, they act like lenses, adjusting the distribution of light passing through them. Therefore, in a defocused state, due to the readjustment of light distribution, the image of the formed elements exhibits higher contrast in brightness and darkness. Taking the highly transparent component in urine, the shadow of red blood cells, as an example... Figure 9 As shown, when the erythrocytes are located within the depth of field of the object-side focal plane of the imaging component 20, the resulting positive-focus image is clear, but the contrast is insufficient. Figure 10 As shown, the erythrocytes are located outside the depth of field of the object-side focal plane of the imaging component 20 and far from the imaging component. In the telephoto image captured here, the cells are not clear but the contrast is relatively good. Figure 11 As shown, the shadowed red blood cells are located outside the depth of field of the object-side focal plane of the imaging component 20 but close to the imaging component. In the near-focus image captured here, the cells are not clear but the contrast is relatively good. It can be seen that in the defocus state (both far focus and near focus are defocused), although the edge sharpness of the shadowed red blood cells is reduced, the contrast of the image is improved compared to the in-focus state due to the existence of aberrations.

[0118] like Figure 12 As shown, this embodiment utilizes this characteristic to enhance the contrast of the formed elements of the target. Specifically, in this embodiment, the sharpness of the second sub-image of the formed elements of the target is lower than that of its first sub-image, but the contrast of the second sub-image is higher than that of its first sub-image. The resulting second target image not only retains the sharpness of the first sub-images of each formed element of the target but also retains the contrast of the second sub-image, thus achieving optical enhancement. Figure 1 , Figure 9-12 It is evident that this scheme is well-suited for bright-field images, which are component images obtained by capturing samples under illumination. Bright-field images contain relatively transparent formed components with low contrast. This embodiment uses bright-field component images as an example for illustration. Using the detection method provided by this invention, no additional optical devices or optical path switching devices are required, nor are phase difference and dark-field imaging necessary. Multiple component images acquired through depth-of-field fusion are processed accordingly (step 3), resulting in a fast and efficient process that significantly improves imaging or detection efficiency while enhancing image quality. Thus, the processor 10 can identify the types of formed components of each target from the first target image and / or the second target image.

[0119] Similarly, in step 31, there are various methods for obtaining the second sub-image of the target formed elements. Several examples are given below.

[0120] The first method, for the same target formed element: The processor 10, based on the calculated sharpness of the target formed element in a sub-image containing the target formed element, obtains a sub-image (e.g., a single sub-image) from multiple component images containing the target formed element whose sharpness differs from that of the first sub-image, to obtain a second sub-image of the target formed element. For example, from multiple component images containing the target formed element, a sub-image whose sharpness differs from that of the first sub-image can be selected as the second sub-image of the target formed element. Optionally, the sharpness of the target formed element in its second sub-image is lower than a second sharpness, and / or, the second sharpness is less than or equal to the first sharpness, that is, the sharpness of the second sub-image is lower than the sharpness of the first sub-image. The relatively low sharpness of the second sub-image of the target formed element indicates that the target formed element was likely not located within the depth of field of the focal plane when the component image to which the second sub-image belongs was captured. Figure 9-11As can be seen from the principle, the contrast of the second sub-image is often higher than that of the first sub-image. Specifically, in the aforementioned steps, if the sharpness calculated by the processor 10 is the feature value of the sub-image containing the target formed element, then the processor 10 can use the sub-image containing the target formed element whose feature value is lower than a preset third threshold (corresponding to the second sharpness) as the second sub-image, where the third threshold is less than or equal to the first threshold. In the aforementioned steps, if the sharpness calculated by the processor 10 is the number of target features in the sub-image containing the target formed element, then the processor 10 can use the sub-image containing the target formed element whose number of target features is lower than a preset fourth threshold (corresponding to the second sharpness) as the second sub-image, where the fourth threshold is less than or equal to the second threshold.

[0121] The second method involves the processor 10 selecting, from multiple component images of the same region, a sub-image of the target formed element whose sharpness differs from that of the first sub-image, to obtain a second sub-image of the target formed element. For example, selecting a sub-image of the same region where the target formed element has a different sharpness from its first sub-image is used as the second sub-image of the target formed element. Optionally, the sharpness of the target formed element in its second sub-image is lower than a second sharpness, and / or the second sharpness is less than or equal to the first sharpness. Specifically, if the sharpness calculated by the processor 10 in the aforementioned steps is a feature value of the sub-image containing the target formed element, then the processor 10 can use a sub-image whose feature value is lower than a preset third threshold (corresponding to the second sharpness) as the second sub-image, where the third threshold is less than or equal to the first threshold. If the sharpness calculated by the processor 10 in the aforementioned steps is the number of target features in the sub-image where the target formed elements are located, then the processor 10 can use the sub-image where the number of target features in the sub-image where the target formed elements are located is lower than a preset fourth threshold (corresponding to the second sharpness) as the second sub-image, where the fourth threshold is less than or equal to the second threshold.

[0122] The third method involves the processor 10 acquiring a sub-image of the formed element of the target as a second sub-image of the formed element, such as... Figure 13As shown, the focal plane of the sub-image is located in the distant direction (away from the microscope objective) of the first sub-image of the target formed element, and the distance between the focal plane and the first sub-image of the target formed element is a preset distance L1. The vertical height between the imaging component 20 and the sample-carrying component 30 when capturing the component image is known; that is, the focal plane height of the component image is known, which means the focal plane heights of the first and second sub-images are known, or their relative positions are known. Thus, the processor 10 can select a sub-image located below the focal plane of the first sub-image (away from the objective) and at a distance of the preset distance L1 from the focal plane of the first sub-image of the target formed element as the second sub-image. Depth of field is the distance within the range before and after focusing, where a clear image is presented. Objects within the depth of field range of the focal plane are all clear in photographs. To ensure a contrast improvement in the second sub-image, the preset distance L1 does not exceed 10 times the depth of field; this embodiment uses a depth of field not exceeding 9 times as an example. When the imaging component 20 performs multi-layer imaging of the sample in the vertical direction (step 1), if the spacing between two adjacent layers is too large, that is, if the focal plane distance between two consecutive component images is too large, some formed elements of the target may not be within the depth of field of the focal plane. Therefore, in this embodiment, when the imaging component 20 performs multi-layer imaging of the sample in the vertical direction, the focal plane distance between two consecutive component images does not exceed the depth of field of the imaging component 20. The imaging component 20 can use continuous scanning imaging to accelerate the acquisition of multi-layer component images. The spacing between the continuously scanned component images is less than the depth of field, that is, the difference in distance between the imaging component 20 and the sample between two consecutive images does not exceed the depth of field of the imaging component 20. This ensures that each formed element is always within the depth of field of the focal plane in one of its all formed element images, thereby ensuring the clarity of the first sub-image. The range of continuous scanning imaging covers the component enrichment surface at the current position.

[0123] Regardless of which of the three methods is used, the selected second sub-image has a higher contrast than the first sub-image. Then, the second target image obtained based on the first and second sub-images of each target formed component has enhanced contrast. It can be seen that the present invention not only improves the clarity of the component images through depth-of-field fusion, but more importantly, it further improves the contrast of the first target image after the processing in step 3 above. Moreover, as can be seen from step 3, multiple component images used in depth-of-field fusion are sufficient, without the need to additionally capture images such as optical phase difference (PH) / dark field (DF) / differential interference difference (DIC). The entire image processing process is not complicated, which greatly improves image quality and detection efficiency.

[0124] In step 32, the processor 10 obtains a second target image containing multiple target formed components, which can be achieved through difference calculation. This will be explained in detail below.

[0125] In one embodiment, each target has constituent elements, such as Figure 12 As shown, the processor 10 can perform difference calculations on its first and second sub-images to obtain a difference sub-image (second target sub-image) of the formed elements of the target. When performing difference calculations on the first and second sub-images, the first sub-image is clearer than the second, so the clarity of the first sub-image is preserved in the difference sub-image. Similarly, the second sub-image has higher contrast than the first sub-image, so the contrast of the second sub-image is also preserved in the difference sub-image. Figure 12 It can also be seen that the cells in the difference sub-images not only have clear outlines but also have large differences in brightness (high contrast), making them easier to classify. Of course, in other embodiments, the processor 10 can also input the first and second sub-images of the target formed elements into a pre-trained model (such as a neural network model), which outputs a difference sub-image (the second target sub-image) that retains the clarity of the first sub-image and the contrast of the second sub-image. Furthermore, the processor 10 obtains a second target image containing multiple target formed elements based on the difference sub-images of multiple target formed elements, for example, by stitching together the difference sub-images of each target formed element.

[0126] In another embodiment, the processor 10 obtains a third target image containing multiple target formed components from the acquired second sub-images of multiple target formed components. The specific method can be the same as obtaining a first target image from the first sub-images of multiple target formed components; for example, both methods involve stitching together the first / second sub-images of each target formed component to obtain the first / second target image. Then, the processor 10 performs a difference calculation between the first target image and the third target image to obtain the second target image containing multiple target formed components. Similarly, when performing a difference calculation between the first and third target images, the first sub-image in the first target image is clearer than the second sub-image in the third target image, so the clarity of the first sub-image is retained in the second target image. Likewise, the second sub-image in the third target image has higher contrast than the first sub-image in the first target image, so the contrast of the second sub-image is also retained in the second target image. Of course, in other embodiments, the processor 10 can also input the first and third target images into a pre-trained model (such as a neural network model), which outputs a second target image that retains the clarity of the first target image and the contrast of the third target image. After obtaining the second target image, if all the formed components of the target are identified, the second target sub-image of each formed component can be obtained. This has been explained in the previous content and will not be repeated here.

[0127] It is evident that regardless of the method used, the resulting second target image retains the advantages of the first / second sub-images, thereby improving image contrast. In some embodiments, the contrast of the second target image is not lower than that of the first target image, but is higher than that of the first target image.

[0128] The processor 10 can output a second target image and / or at least one second target sub-image of the formed elements of the target, for example, by outputting the second target image to a display for the display to show the second target image. The second target image seen by the user has higher contrast, which can help determine whether the analysis system has correctly classified the formed elements of the target, and can also be used for manual classification based on the second target image. The processor 10 can also output one or more second target sub-images of the formed elements of the target, for example, by outputting one or more second target sub-images of the formed elements of the target to a display for the display to show multiple second target sub-images of the formed elements of the target.

[0129] In some examples where the sample carrying component 30 includes a flow chamber, since the flow chamber is used for the sample provided by the sample feeding component 40 to be wrapped by a flat sheath fluid flow to form a planar laminar flow, multiple formed elements in the sample may pass through the flow chamber in batches. In such examples, it is more meaningful and in line with user needs for the processor 10 to acquire and output a second target sub-image of multiple target formed elements.

[0130] As can be seen from the above, the analysis system and method provided by this invention require no additional hardware. By acquiring multiple component images through multi-layer imaging, simultaneous acquisition of bright-field and enhanced images is achieved, avoiding the speed limitations of traditional optical image enhancement methods. Compared to directly using two component images of different resolutions for fusion enhancement, this invention uses sub-images of formed elements as units for fusion enhancement, i.e., block enhancement, solving the problems of urine images not settling, not being on the same focal plane, and enhancing very large components. Compared to depth-of-field fusion, contrast enhancement of formed elements is performed using a second sub-image with higher contrast from multiple component images.

[0131] The formed element analysis system can be a urine formed element analyzer. To save computational resources (and reduce costs), in some embodiments, the above analysis method can be performed only on specific types of samples, while conventional analysis methods can be used for other types of samples. For example... Figure 14 As shown, the analysis method provided in this embodiment includes the following steps:

[0132] Step 1': The processor 10 determines whether the sample belongs to the preset category. If the sample belongs to the preset category, step 2' is executed; if the sample does not belong to the preset category, step 3' is executed.

[0133] The processor 10 determines whether a sample belongs to a preset category. For example, it can obtain sample information, which can be obtained from the hospital's LIS system, based on user operations on the analysis system, or from other medical devices. Sample information may include at least one of the following: the sample's test items, measurement methods, dry chemical measurement results, formed element measurement results, and physical measurement results. Based on the sample information, the processor determines whether the sample belongs to a preset category.

[0134] Step 2': When the sample belongs to a preset category, the processor 10 acquires a first target sub-image and / or a first target image, and acquires a second target sub-image and / or acquires a second target image. The first target image is obtained based on a first sub-image whose clarity meets the clarity requirements of the target formed elements. The second target image is obtained based on a second sub-image whose clarity differs from that of the first sub-image. The specific methods by which the processor 10 acquires the first target sub-image and / or the first target image, and acquires the second target sub-image and / or acquires the second target image, are described in the foregoing embodiments. In other words, step 2' includes steps 1, 2, and 3 described above, and the detailed process has been explained in the foregoing and will not be repeated here.

[0135] Step 3': When the sample is not of a preset category, the processor 10 acquires the first target sub-image and / or the first target image, but does not acquire the second target sub-image and / or the second target image. The specific method by which the processor 10 acquires the first target sub-image and / or the first target image is described in the foregoing embodiments. That is, this step 3' includes steps 1 and 2 as described above, but does not include step 3. The detailed process has been described in the foregoing and will not be repeated here.

[0136] It is evident that for samples of the preset categories, sampling... Figure 3 The analysis method shown improves the contrast of the depth-of-field fused image. For samples that are not in the preset category, no contrast is required, so depth-of-field fusion can be performed, which reduces computational cost and achieves low cost.

[0137] In one embodiment, if the sample's testing items include erythrocyte phase analysis, the processor 10 can determine that the sample belongs to a preset category. Specifically, the processor 10 can obtain the sample's testing items from the hospital's LIS system; if it includes erythrocyte phase analysis, then it executes the test. Figure 3 The analysis method shown.

[0138] In one embodiment, if the measurement method for the sample includes red blood cell phase detection, phase contrast measurement, or phase difference measurement, the processor 10 can determine that the sample belongs to a preset category. Users typically operate the analysis system to set the measurement method or measurement mode, and the processor 10 can determine whether the sample belongs to the preset category based on the measurement method set by the user.

[0139] In one embodiment, if the dry chemical measurement results of a sample include a positive result for a preset dry chemical measurement item, the processor 10 can determine that the sample belongs to a preset category. A positive result for a preset dry chemical measurement item may include at least one of occult blood, protein, and nitrite. The processor 10 can obtain the dry chemical measurement results of the sample from the dry chemical analyzer to determine whether it is a sample of the preset category.

[0140] In one embodiment, if the formed element measurement results of a sample include a positive result for a preset formed element measurement item, the processor 10 can determine that the sample belongs to a preset category. A positive result for a preset formed element measurement item may include at least one of red blood cells, casts, bacteria, crystals, and bacteria. If the analysis system has previously performed these formed element measurements on the sample, it can determine whether the sample belongs to the preset category based on the results of these measurements.

[0141] In one embodiment, if the physical measurement results of a sample include a positive result for a preset physical measurement item, the processor 10 can determine that the sample belongs to a preset category. A positive preset physical measurement item may include: urine color matching a preset specific color, such as red or orange; or it may include: transparency meeting preset requirements, such as turbidity or slight turbidity. The processor 10 can obtain the physical measurement results of the sample from the urine analyzer to determine whether it is a sample of the preset category.

[0142] If the above one or more judgment methods fail to determine whether the sample belongs to the preset category, then the sample is determined not to belong to the preset category. In other words, if the above judgment conditions for the preset category are not met, the sample is determined not to belong to the preset category, and only depth fusion needs to be performed.

[0143] like Figure 15 As shown, the formed element analysis system provided by the present invention can also employ formed element analysis... Figure 15 The process shown is in Figure 15 In the illustrated embodiment, the analysis process includes the following steps:

[0144] Step 1: The processor 10 controls the imaging component 20 to perform multi-layer imaging of the sample on the sample carrier 30 in the vertical direction to obtain multiple component images of the sample at each layer. The specific process is the same as above. Figure 3 Step 1 of the illustrated embodiment will not be described in detail here.

[0145] Step 2: The processor 10 obtains a first image containing the target formed element from at least one of the multiple component images. The sharpness of the target formed element in the first image meets the sharpness requirements.

[0146] Step 3: The processor 10 obtains a second image containing the target formed element based on at least two of the multiple component images, or obtains the second image containing the target formed element based on the first image and at least one of the multiple component images. The contrast of the target formed element in the second image is greater than its contrast in the first image. The processor 10 can output the first image and / or the second image, for example, to a display, so that the user can see the target formed element with high clarity and high contrast on the screen. Alternatively, the analysis system can also use the high-clarity first image and the high-contrast second image to perform formed element analysis, such as determining the type of the target formed element, thus improving the accuracy and efficiency of the analysis.

[0147] In some embodiments, the at least two component images used to obtain the second image include the component image used to obtain the first image. For example, one of the component images used to obtain the second image is the component image used to obtain the first image. That is, if the processor 10 obtains the first image based on at least one component image and obtains the second image based on at least two component images, then the "at least one component image" and the "at least two component images" contain the same component image, such as component image A1.

[0148] Of course, in other embodiments, the at least two component images used to obtain the second image are not the same component images used to obtain the first image. That is, the component images used to obtain the first image are completely different from the component images used to obtain the second image.

[0149] In one embodiment, the first image may include a first target sub-image and / or a first target image. Step 2” specifically includes: for the target formed element of the sample being photographed, the processor 10 obtains a first sub-image of the target formed element whose clarity meets the clarity requirements from multiple component images. The first sub-image is a sub-image in the component images that contains part or all of the target formed element; based on the obtained first sub-image of the target formed element, a first target sub-image containing the complete target formed element is obtained, and / or, based on the obtained first sub-images of multiple target formed elements, a first target image containing multiple target formed elements is obtained. That is, the specific implementation process of step 2” can include step 2 of the aforementioned embodiment, and will not be repeated here.

[0150] In some embodiments, the first image may also be the component image with the highest clarity selected directly from multiple component images, and this component image is used as the first image.

[0151] In one embodiment, the second image may include a second target sub-image and / or a second target image. Step 3" may specifically include: for the formed elements of the target sample being captured, the processor 10 acquires at least two third sub-images of different sharpness from multiple component images of the formed elements of the target, wherein the third sub-image is a sub-image in the component images that contains part or all of the formed elements of the target, and obtains a second target sub-image containing the complete formed elements of the target based on the at least two third sub-images of the acquired formed elements of the target; and / or, obtains a second target image containing multiple formed elements based on the at least two third sub-images of the acquired multiple formed elements of the target. In the aforementioned embodiments, the first sub-image is a sub-image with certain sharpness requirements, and the second sub-image is a sub-image with lower sharpness than the first sub-image. However, in this embodiment, the third sub-image does not have these requirements; for example, at least two sub-images of different sharpness can be selected as the third sub-image.

[0152] The processor 10 obtains a second target sub-image containing the complete target formed element based on at least two third sub-images of the acquired target formed element. The specific principle and process are the same as in the aforementioned embodiment where "a second target sub-image containing the complete target formed element is obtained based on the first and second sub-images of the acquired target formed element" is used; only the first and second sub-images need to be replaced with at least two third sub-images. For example, if at least two third sub-images include the aforementioned first and second sub-images, meaning one of the at least two third sub-images of the target formed element is the first sub-image of the target formed element, then step 3 is the same as step 3 in the aforementioned embodiment. If at least two third sub-images are all first sub-images, or all second sub-images, then step 3 is slightly different from step 3 in the aforementioned embodiment, but the principle is the same; simply replace all the first / second sub-images in step 3 of the aforementioned embodiment with first sub-images or all with second sub-images. This will not be elaborated further here.

[0153] Similarly, the processor 10 obtains a second target image containing multiple target formed elements based on at least two third sub-images of the acquired multiple target formed elements. The specific principle and process are the same as in the aforementioned embodiment, where "a second target image containing multiple target formed elements is obtained based on the first and second sub-images of the acquired multiple target formed elements, or based on the first target image and the second sub-images of the multiple target formed elements." Only the first and second sub-images need to be replaced with at least two third sub-images. For example, if at least two third sub-images include the aforementioned first and second sub-images, meaning one of the at least two third sub-images of the target formed elements is the first sub-image of the target formed element, then step 3 is the same as step 3 in the aforementioned embodiment. If at least two third sub-images are all first sub-images, or all second sub-images, then step 3 is slightly different from step 3 in the aforementioned embodiment, but the principle is the same; simply replace all the first / second sub-images in step 3 of the aforementioned embodiment with first sub-images or all with second sub-images. This will not be elaborated upon here.

[0154] In some embodiments, the processor 10 may also obtain a second image containing the target formed element based on a first image and at least one of a plurality of component images. In such an example, if the first image is the component image with the highest sharpness selected directly from the plurality of component images, then the second image containing the target formed element is obtained based on at least one of the first image and other component images (other component images refer to component images that are not the first image), for example, by subtracting the first image from the other component images to obtain the second image. Furthermore, if the first image is obtained based on at least two component images, then the second image containing the target formed element can be obtained based on the first image and at least one of the plurality of component images, for example, by subtracting the first image from any one of the plurality of component images to obtain the second image. Alternatively, obtaining the second image containing the target formed element based on the first image and at least one of the plurality of component images may involve first obtaining the aforementioned third sub-image based on at least one of the plurality of component images, and then obtaining the second image based on the first image and the third sub-image.

[0155] This document describes various exemplary embodiments with reference to them. However, those skilled in the art will recognize that changes and modifications can be made to the exemplary embodiments without departing from the scope of this document. For example, various operational steps and components for performing operational steps can be implemented in different ways depending on the specific application or considering any number of cost functions associated with the operation of the system (e.g., one or more steps can be deleted, modified, or combined with other steps).

[0156] Furthermore, as those skilled in the art will understand, the principles herein can be reflected in a computer program product on a computer-readable storage medium pre-loaded with computer-readable program code. Any tangible, non-transitory computer-readable storage medium may be used, including magnetic storage devices (hard disks, floppy disks, etc.), optical storage devices (CD-ROMs, DVDs, Blu-ray discs, etc.), flash memory, and / or the like. These computer program instructions may be loaded onto a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to form a machine, such that instructions, which execute on the computer or other programmable data processing apparatus, can generate means for performing a specified function. These computer program instructions may also be stored in a computer-readable storage medium that can instruct the computer or other programmable data processing apparatus to operate in a particular manner, such that instructions stored in the computer-readable storage medium can form an article of manufacture, including means for implementing the specified function. The computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to perform a series of operational steps on the computer or other programmable apparatus to produce a computer-implemented process, such that instructions, which execute on the computer or other programmable apparatus, can provide steps for implementing the specified function.

[0157] While the principles herein have been illustrated in various embodiments, numerous modifications to the structure, arrangement, proportions, elements, materials, and components, particularly suited to specific environmental and operational requirements, may be used without departing from the principles and scope of this disclosure. These modifications and other alterations or alterations will be included within the scope of this document.

[0158] The foregoing specific descriptions have been described with reference to various embodiments. However, those skilled in the art will recognize that various modifications and changes can be made without departing from the scope of this disclosure. Therefore, considerations for this disclosure are to be illustrative rather than restrictive, and all such modifications are to be included within its scope. Similarly, advantages, other advantages, and solutions to problems with respect to various embodiments have been described above. However, benefits, advantages, solutions to problems, and any elements that produce these, or make them more explicit, should not be construed as critical, essential, or necessary. The term “comprising” and any other variations thereof as used herein are non-exclusive inclusion, meaning that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed or not part of the process, method, system, article, or apparatus. Furthermore, the term “coupled” and any other variations thereof as used herein refer to physical connections, electrical connections, magnetic connections, optical connections, communication connections, functional connections, and / or any other connections.

Claims

1. A system for analyzing the formed elements of a sample, characterized in that, include: Sample-carrying components are used to carry samples; A sample feeding component is used to pick up a sample and provide the picked-up sample to the sample carrying component; An imaging component is used to capture images of the sample from the sample-bearing component to obtain images of the formed elements in the sample; Processor, used for: The imaging component is controlled to perform multi-layer imaging of the sample on the sample carrier in the vertical direction to obtain multiple component images of the sample corresponding to each layer. Acquiring a first target sub-image and / or a first target image includes: for the formed elements of the target being photographed in the sample, acquiring a first sub-image from multiple component images where the clarity of the formed elements of the target meets the clarity requirements, wherein the first sub-image is a sub-image in the component images that contains part or all of the formed elements of the target; obtaining a first target sub-image containing the complete formed elements of the target based on the acquired first sub-image of the formed elements of the target, and / or obtaining a first target image containing multiple formed elements of the target based on the acquired first sub-image of the formed elements of the target; Acquiring a second target sub-image and / or a second target image includes: for the target formed element captured in the sample, acquiring a second sub-image from multiple component images where the target formed element has a different clarity from that in a first sub-image, the second sub-image being a sub-image in the component images containing part or all of the target formed element; obtaining a second target sub-image containing the complete target formed element based on the acquired first and second sub-images of the target formed element; and / or, based on the acquired first and second sub-images of multiple target formed elements, or based on the first target image and multiple second sub-images of the target formed element, obtaining a second target image containing multiple target formed elements.

2. The formed element analysis system as described in claim 1, characterized in that, The processor, for the formed elements of the target image captured in the sample, acquires a first sub-image from multiple component images whose sharpness meets the sharpness requirements, including: For each of the component images: detect the target formed elements in the component image and calculate the sharpness of the detected target formed elements in the sub-image containing the target formed elements; For the same target formed element: Based on the calculated sharpness of the target formed element in the sub-image containing the target formed element, sub-images whose sharpness meets the sharpness requirement are obtained from multiple component images containing the target formed element, to obtain a first sub-image of the target formed element; optionally, meeting the sharpness requirement includes the sharpness of the target formed element in its first sub-image being not lower than a first sharpness, or meeting the sharpness requirement includes the sharpness of the target formed element in its first sub-image being higher than the sharpness of the target formed element in other sub-images, or, if there are multiple sub-images whose sharpness meets the sharpness requirement, then these multiple sub-images are processed, preferably by weighted calculation, to obtain the first sub-image of the target formed element.

3. The formed element analysis system as described in claim 2, characterized in that, The processor detects target formed elements in the component image in at least one of the following ways: Detecting formed elements of a target from a component image based on image segmentation methods; Target formed elements are detected from composition images using machine learning models.

4. The formed element analysis system as described in claim 2, characterized in that, The processor, for the formed elements of the target image captured in the sample, obtains a second sub-image from multiple component images, wherein the formed elements of the target have a different sharpness than the first sub-image, including: For the same target formed element: based on the calculated sharpness of the target formed element in the sub-image containing the target formed element, obtain a sub-image from the sub-images containing the target formed element in multiple component images where the sharpness of the target formed element differs from that in the first sub-image, to obtain a second sub-image of the target formed element; optionally, the sharpness of the target formed element in its second sub-image is lower than the second sharpness, or the second sharpness is less than or equal to the first sharpness.

5. The formed element analysis system as described in claim 1, characterized in that, The processor, for the formed elements of the target image captured in the sample, acquires a first sub-image from multiple component images whose sharpness meets the sharpness requirements, including: Each component image is divided into multiple regions according to the same rules, and each region corresponds to a sub-image; For each region, from the sub-images of the same region in multiple component images, select the sub-images of the same region in which the clarity of the target formed element meets the clarity requirement, to obtain the first sub-image of the target formed element; optionally, meeting the clarity requirement includes the clarity of the target formed element in its first sub-image being not lower than a first clarity, or meeting the clarity requirement includes the clarity of the target formed element in its first sub-image being higher than the clarity of other sub-images of the target formed element in the same region, or if there are multiple sub-images of the target formed element in the same region in which the clarity meets the clarity requirement, then process the multiple sub-images, preferably by weighted calculation, to obtain the first sub-image of the target formed element.

6. The formed element analysis system as described in claim 5, characterized in that, The processor, for the formed elements of the target image captured in the sample, obtains a second sub-image from multiple component images, wherein the formed elements of the target have a different sharpness than the first sub-image, including: For each region, from the sub-images of the same region in the multiple component images, a sub-image of the target formed element that has a different sharpness from that in the first sub-image is selected to obtain a second sub-image of the target formed element; optionally, the sharpness of the target formed element in its second sub-image is lower than the second sharpness; optionally, the second sharpness is less than or equal to the first sharpness.

7. The formed element analysis system as described in claim 1, 4, or 6, characterized in that, The processor, for the formed elements of the target image captured in the sample, obtains a second sub-image from multiple component images, wherein the formed elements of the target have a different sharpness than the first sub-image, including: A sub-image of the target formed element is acquired as the second sub-image of the target formed element, wherein the focal plane of the sub-image is located in the distant direction of the first sub-image of the target formed element and the distance between the focal plane of the first sub-image of the target formed element and the focal plane of the first sub-image of the target formed element is a preset distance; optionally, the preset distance does not exceed 10 times the depth of field.

8. The formed element analysis system as described in claim 1, characterized in that, The processor obtains a first target sub-image containing the complete target formed elements based on the acquired first sub-image of the target formed elements, including at least one of the following: The first sub-image of the target formed element is a sub-image in the component image that contains the complete target formed element, and the first sub-image of the target formed element is taken as the first target sub-image of the target formed element; A first target image containing multiple target formed elements is obtained from the first sub-images of the acquired target formed elements. Target formed elements are detected from the first target image to obtain a sub-image of the target formed element in the first target image, which is then used as the first target sub-image of the target formed element.

9. The formed element analysis system as described in claim 1, characterized in that, The processor obtains a second target sub-image containing the complete formed elements of the target based on the acquired first and second sub-images of the target formed elements, including at least one of the following: The first and second sub-images of the formed element of the target are both sub-images containing the complete formed element of the target in the component image. The second target sub-image of the formed element is obtained by calculating the difference between the first and second sub-images of the formed element of the target. Based on the first and second sub-images of multiple target formed elements, or based on the first target image and the second sub-images of multiple target formed elements, a second target image containing multiple target formed elements is obtained; target formed elements are detected based on the second target image to obtain a sub-image of the target formed element in the second target image and used as the second target sub-image of the target formed element.

10. The formed element analysis system as described in claim 1 or 9, characterized in that, The processor obtains a second target image containing multiple target formed elements based on the acquired first and second sub-images of multiple target formed elements, or based on the first target image and the second sub-images of multiple target formed elements, including: For each formed element of a target, the difference between its first and second sub-images is calculated to obtain a difference sub-image of that formed element. Based on the difference sub-images of multiple formed elements of targets, a second target image containing multiple formed elements of targets is obtained; or, A third target image containing multiple target formed elements is obtained from a second sub-image of multiple target formed elements. The difference between the first target image and the third target image is calculated to obtain a second target image containing multiple target formed elements.

11. The formed element analysis system according to any one of claims 1 to 10, characterized in that, The sharpness of the first sub-image of the target formed element is not lower than the first sharpness; and / or, the sharpness of the first sub-image of the target formed element is higher than the sharpness of the target formed element in other sub-images; And / or, the sharpness of the first target sub-image with the target forming elements is not lower than the first sharpness; And / or, the clarity of the first target image is not lower than the first clarity; And / or, the contrast of the second target sub-image with the target formed elements is not lower than the first contrast; and / or, the contrast of the second target image is not lower than the first contrast.

12. The formed element analysis system according to any one of claims 1 to 11, characterized in that, The processor is also used to control the output: The first target image and / or at least one target sub-image with target constituents; and, The second target image and / or at least one target sub-image with forming elements.

13. The formed element analysis system according to any one of claims 1 to 12, characterized in that, The processor is further configured to: determine whether the sample belongs to a preset category; and when the sample belongs to a preset category, execute the acquisition of the first target sub-image and / or the first target image, and execute the acquisition of the second target sub-image and / or the second target image; Conversely, the acquisition of the first target sub-image and / or the first target image is performed, but the acquisition of the second target sub-image and / or the second target image is not performed; Optionally, the processor determines whether the sample belongs to a preset category, including: Obtain sample information, which includes at least one of the following: sample testing items, measurement methods, dry chemical measurement results, formed element measurement results, and physical measurement results; Determine whether the sample belongs to a preset category based on the sample information; Optionally, if the sample's test items include red blood cell phase testing, the measurement methods include red blood cell phase detection, phase contrast measurement or phase difference measurement, the dry chemical measurement results include a positive result for a preset dry chemical measurement item, the formed element measurement results include a positive result for a preset formed element measurement item, or the physical measurement results include a positive result for a preset physical measurement item, then the sample belongs to the preset category; otherwise, the sample does not belong to the preset category.

14. The formed component analysis system according to any one of claims 1 to 13, characterized in that, The sample carrying component includes a counting cell for holding a sample provided by the sample loading component; the imaging component has a focusing distance for capturing a sample in the counting cell at the focusing distance to obtain a component image; the formed component analysis system further includes a driving device for driving the imaging component and the counting cell to move relative to each other at least in the vertical direction. The processor controls the imaging component to perform multi-layer imaging of the sample on the sample carrier component to obtain multiple component images corresponding to the sample in each multi-layer. This includes controlling the driving device to drive the imaging component and the counting cell to move relative to each other in the vertical direction, so that the imaging component can perform imaging at different vertical heights relative to the counting cell at the focusing distance, thereby performing multi-layer imaging of the sample to obtain multiple component images corresponding to the sample in each multi-layer. or, The sample carrying component includes a flow chamber, which is used for the sample provided by the sample delivery component to be enveloped by a flat sheath fluid flow to form a planar laminar flow. The imaging component includes multiple imaging components capable of simultaneous imaging. The processor controls the imaging component to perform multi-layer imaging of the sample in the sample carrying component to obtain multiple component images of the sample corresponding to each layer. This includes: multiple imaging components are located above the flow chamber and different imaging parts are focused at different heights of the laminar flow passing through the flow chamber; controlling multiple imaging components to simultaneously image the laminar flow passing through the flow chamber, thereby performing multi-layer imaging of the sample to obtain multiple component images of the sample corresponding to each layer.

15. A system for analyzing the formed elements of a sample, characterized in that, include: Sample-carrying components are used to carry samples; A sample feeding component is used to pick up a sample and provide the picked-up sample to the sample carrying component; An imaging component is used to capture images of the sample from the sample-bearing component to obtain images of the formed elements in the sample; Processor, used for: Determine whether the sample belongs to a preset category; When the sample is of a preset category, a first target sub-image and / or a first target image are acquired, as well as a second target sub-image and / or a second target image are acquired; otherwise, a first target sub-image and / or a first target image are acquired; wherein, the first target sub-image and / or the first target image are obtained based on a first sub-image whose clarity of the target formed elements meets the clarity requirements, and the second target sub-image and / or the second target image are obtained based on a first sub-image whose clarity of the target formed elements meets the clarity requirements and a second sub-image whose clarity differs from that of the first sub-image.

16. The formed element analysis system as described in claim 15, characterized in that, The processor determines whether the sample belongs to a preset category, including: Obtain sample information, which includes at least one of the following: sample testing items, measurement methods, dry chemical measurement results, formed element measurement results, and physical measurement results; Determine whether the sample belongs to a preset category based on the sample information.

17. The formed element analysis system as described in claim 16, characterized in that, If the sample's test items include red blood cell phase testing, the measurement methods include red blood cell phase detection, phase contrast measurement, or phase difference measurement, and the dry chemical measurement results contain a positive result for a preset dry chemical measurement item, the formed element measurement results contain a positive result for a preset formed element measurement item, or the physical measurement results contain a positive result for a preset physical measurement item, then the sample belongs to the preset category; otherwise, the sample does not belong to the preset category.

18. The formed component analysis system according to any one of claims 15 to 17, characterized in that, The acquisition of the first target sub-image and / or the first target image includes: for the target formed element captured in the sample, acquiring a first sub-image from multiple component images where the clarity of the target formed element meets the clarity requirements, wherein the first sub-image is a sub-image in the component images that contains part or all of the target formed element; obtaining a first target sub-image containing the complete target formed element based on the acquired first sub-image of the target formed element; and / or obtaining a first target image containing multiple target formed elements based on the acquired first sub-images of multiple target formed elements. And / or, The second target sub-image and / or obtaining the second target image includes: for the target formed element of the sample being photographed, obtaining a second sub-image from multiple component images where the target formed element has a different clarity from that of the first sub-image, the second sub-image being a sub-image in the component image containing part or all of the target formed element; obtaining a second target sub-image containing the complete target formed element based on the obtained first and second sub-images of the target formed element; and / or, based on the obtained first and second sub-images of multiple target formed elements, or based on the first target image and multiple second sub-images of the target formed element, obtaining a second target image containing multiple target formed elements.

19. A method for analyzing the formed elements of a sample, characterized in that, include: The sample carrying the sample is controlled to be photographed in multiple layers in the vertical direction to obtain multiple component images of the sample in each layer; Acquiring a first target sub-image and / or a first target image includes: for the formed elements of the target being photographed in the sample, acquiring a first sub-image from multiple component images where the clarity of the formed elements of the target meets the clarity requirements, wherein the first sub-image is a sub-image in the component images that contains part or all of the formed elements of the target; obtaining a first target sub-image containing the complete formed elements of the target based on the acquired first sub-image of the formed elements of the target, and / or obtaining a first target image containing multiple formed elements of the target based on the acquired first sub-image of the formed elements of the target; Acquiring a second target sub-image and / or a second target image includes: for the target formed element captured in the sample, acquiring a second sub-image from multiple component images where the target formed element has a different clarity from that in a first sub-image, the second sub-image being a sub-image in the component images containing part or all of the target formed element; obtaining a second target sub-image containing the complete target formed element based on the acquired first and second sub-images of the target formed element; and / or, based on the acquired first and second sub-images of multiple target formed elements, or based on the first target image and multiple second sub-images of the target formed element, obtaining a second target image containing multiple target formed elements.

20. The formed element analysis method as described in claim 19, characterized in that, The step of obtaining a second target image containing multiple target formed elements based on the first and second sub-images of the acquired multiple target formed elements, or based on the first target image and the second sub-images of the multiple target formed elements, includes: For each formed element of a target, the difference between its first and second sub-images is calculated to obtain a difference sub-image of that formed element. Based on the difference sub-images of multiple formed elements of targets, a second target image containing multiple formed elements of targets is obtained; or, A third target image containing multiple target formed elements is obtained from a second sub-image of multiple target formed elements. The difference between the first target image and the third target image is calculated to obtain a second target image containing multiple target formed elements.

21. The formed element analysis method as described in claim 19, characterized in that, Also includes: Determine whether the sample belongs to a preset category; When the sample is a preset category, the process of obtaining the first target sub-image and / or the first target image is performed, and the process of obtaining the second target sub-image and / or the second target image is performed. Conversely, the acquisition of the first target sub-image and / or the first target image is performed, but the acquisition of the second target sub-image and / or the second target image is not performed.

22. The formed element analysis method according to any one of claims 19 to 21, characterized in that, The sharpness of the first sub-image of the target formed element is not lower than the first sharpness; and / or, the sharpness of the first sub-image of the target formed element is higher than the sharpness of the target formed element in other sub-images; And / or, the sharpness of the first target sub-image with the target forming elements is not lower than the first sharpness; And / or, the clarity of the first target image is not lower than the first clarity; And / or, the contrast of the second target sub-image with the target formed elements is not lower than the first contrast; and / or, the contrast of the second target image is not lower than the first contrast.

23. A system for analyzing the formed elements of a sample, characterized in that, include: Sample-carrying components are used to carry samples; A sample feeding component is used to pick up a sample and provide the picked-up sample to the sample carrying component; An imaging component is used to capture images of the sample from the sample-bearing component to obtain images of the formed elements in the sample; Processor, used for: The imaging component is controlled to perform multi-layer imaging of the sample on the sample carrier in the vertical direction to obtain multiple component images of the sample corresponding to each layer. A first image containing the target formed element is obtained from at least one of a plurality of component images, wherein the sharpness of the target formed element in the first image meets a sharpness requirement; and, A second image containing the target formed element is obtained from at least two of the multiple component images, or a second image containing the target formed element is obtained from the first image and at least one of the multiple component images; wherein the contrast of the target formed element in the second image is greater than its contrast in the first image.

24. The formed element analysis system as described in claim 23, characterized in that, Of the at least two component images used to obtain the second image, one is the component image used to obtain the first image; or, The at least two component images used to obtain the second image are not component images used to obtain the first image.

25. The formed element analysis system as described in claim 23, characterized in that, The processor obtains a first image containing the target formed elements based on at least one of the plurality of component images, including: For the target formed element of the sample being photographed, a first sub-image of the target formed element whose clarity meets the clarity requirements is obtained from multiple component images. The first sub-image is a sub-image in the component images that contains part or all of the target formed element. A first target sub-image containing the complete target formed element is obtained based on the obtained first sub-image of the target formed element, and / or a first target image containing multiple target formed elements is obtained based on the obtained first sub-images of multiple target formed elements. The first image includes the first target sub-image and / or the first target image.

26. The formed element analysis system as described in claim 23, characterized in that, The processor obtains a second image containing the target formed elements based on at least two of the plurality of component images, including: For the formed elements of the target sample being photographed, at least two third sub-images of different sharpness are obtained from multiple component images of the target formed elements. The third sub-images are sub-images in the component images that contain part or all of the target formed elements. Optionally, one of the at least two third sub-images of the target formed elements is a first sub-image of the target formed elements. A second target sub-image containing the complete target formed elements is obtained based on the at least two third sub-images of the target formed elements. And / or, a second target image containing multiple target formed elements is obtained based on the at least two third sub-images of multiple target formed elements. The second image includes the second target sub-image and / or the second target image.

Images