WO2020102953A1

WO2020102953A1 - Cell analysis method, cell analysis device and computer-readable storage medium

Info

Publication number: WO2020102953A1
Application number: PCT/CN2018/116275
Authority: WO
Inventors: 叶波; 余珊; 陈巧妮; 邢圆
Original assignee: 深圳迈瑞生物医疗电子股份有限公司
Priority date: 2018-11-19
Filing date: 2018-11-19
Publication date: 2020-05-28
Also published as: CN117782948A; CN113039551A; CN113039551B

Abstract

A cell analysis method and device, and a computer-readable storage medium applied to the cell analysis device. The method comprises: acquiring digital images of cells in a blood sample (101); classifying the acquired digital images of the cells so as to respectively form classification information of each of the cells (102); and outputting the digital images of the cells and the classification information of the cells (103).

Description

Cell analysis method, cell analysis equipment and computer readable storage medium

Technical field

The present invention relates to medical device technology, and in particular, to a method for analyzing cells, a cell analysis device, and a computer-readable storage medium.

Background technique

The existing automatic blood cell digital image analysis system has the functions of cell identification and pre-classification. Generally, the classification accuracy of white blood cells in normal samples is high. However, the processed blood smears are often abnormal samples, and the cells contained are mostly abnormal cells such as naive or primitive cells. Abnormal leukocyte morphology is complex and changeable, and its morphology is susceptible to treatments such as medication and radiotherapy, and it is no longer typical. This may lead to poor accuracy of the pre-classification results of the automatic blood cell digital image analysis system, requiring a large number of inspection technicians Manual intervention and adjustment to ensure the accuracy of the output results.

In the process of manual re-examination of the cell classification, when the classification of the corresponding cell is found to be wrong, the inspection technician first needs to select the cell, then determine the target type to which the cell should belong, and select the view where the target type should belong Area), and finally drag the cell to the view (or area) where the target type should belong or change the type of the cell through control commands, such as the right mouse button. The operation process is more complicated, such as dragging the cell When dragging from the current view of the target type to the correct view, since one page of the display screen can only hold a limited number of views, when the correct view that the cell should belong to is not in the current page, you need to advance Select the correct view to this page, and then drag. Especially when the number of samples to be re-examined is large, the efficiency of traditional cell re-examination methods is low, which will greatly prolong the sample turnaround time, and it is easy to cause errors in the re-examination results due to fatigue of the inspection technician.

Summary of the invention

Embodiments of the present invention provide a method for analyzing cells, a cell analysis device, and a computer-readable storage medium, which can classify acquired digital images of the cells to form classification information for each of the cells, and output the The digital image of the cell and the classification information of the cell can assist the user to manually re-examine the pre-classification of the cell, which greatly improves the efficiency of the re-examination. In addition, the cell classification information can also assist users in directly classifying artificial cells.

The technical solutions of the embodiments of the present invention are implemented as follows.

An embodiment of the present invention provides a method for analyzing cells, which is applied to a cell analysis device. The method includes:

Obtain digital images of cells in blood samples;

Classify the acquired digital images of the cells to form classification information of each of the cells, wherein the classification information of each of the cells includes cell type information and corresponding probability information;

The digital image of the cell and the classification information of the cell are output.

An embodiment of the present invention also provides a cell analysis device. The cell analysis device includes:

A control device configured to adjust the relative position of the digital imaging device and the blood sample;

Digital imaging device, including lens group and digital camera;

An image acquisition device configured to acquire a digital image of cells in a blood sample;

An image processing device configured to classify the acquired digital images of the cells to form classification information of each of the cells, wherein the classification information of each of the cells includes cell type information and corresponding probability information;

The display output device is configured to output a digital image of the cell and classification information of the cell.

Memory, configured to store executable instructions;

The processor, configured to execute the executable instructions stored in the memory, executes:

Obtain digital images of cells in blood samples;

An embodiment of the present invention also provides a computer-readable storage medium that stores executable instructions and is configured to cause a processor to execute the executable instructions to implement:

Obtain digital images of cells in blood samples;

In the embodiment of the present invention, by acquiring a digital image of cells in a blood sample; classifying the acquired digital images of the cells to form classification information of each of the cells; outputting the digital image of the cells and the Cell classification information; wherein the cell classification information of each cell includes cell type information and probability information, the cell type information includes at least two predefined cell types to be selected, and the probability information includes corresponding cells to be selected The type is the probability value of the target type of the cell. The user can accurately know the probable cell type of the cell and the probability that the corresponding cell type to be selected is the target type of the cell through the classification information of the cell, and the user finds that the cell type identification error At this time, the target type of the cell can be adjusted easily and quickly according to the classification information of the cell, especially the probability information.

BRIEF DESCRIPTION

FIG. 1 is a schematic flowchart of an optional method for analyzing a cell provided by an embodiment of the present invention;

2 is a schematic structural diagram of an optional cell analysis device according to an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of an optional cell analysis device according to an embodiment of the present invention;

4A to 4C are schematic diagrams of an optional display interface of a cell analysis device provided by an embodiment of the present invention;

5A to 5D are schematic diagrams of an optional display interface of a cell analysis device provided by an embodiment of the present invention;

6 is a schematic diagram of a cell analysis device provided by an embodiment of the present invention;

7 is a schematic diagram of a cell analysis system provided by an embodiment of the present invention;

8 is a schematic structural diagram of an optional cell analysis device provided by an embodiment of the present invention.

detailed description

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail below with reference to the accompanying drawings. The present invention should not be construed as being limited to the provided embodiments. On the contrary, the contents described in the embodiments of the present invention make the present invention comprehensive and complete, and convey the idea of the embodiments of the present invention to those skilled in the art, so the ordinary Other embodiments obtained by the technical personnel without making creative work fall within the protection scope of the present invention.

It should be noted that, in the embodiments of the present disclosure, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a method or server including a series of elements includes not only the explicitly recorded Elements, but also include other elements that are not explicitly listed, or include elements inherent to the implementation method or server. If there are no more restrictions, the element defined by the sentence "include a ..." does not exclude that there are other related elements in the method or server including the element (such as the steps in the method or the unit in the server) For example, the unit may be part of a circuit, part of a processor, part of a program or software, etc.).

For example, the method for visually analyzing cells provided by the embodiments of the present disclosure includes a series of steps, but the method for visually analyzing cells provided by the embodiments of the present disclosure is not limited to the described steps. Similarly, the terminal provided by the embodiments of the present disclosure includes A series of units are provided, but the terminal provided by the embodiments of the present disclosure is not limited to include the units explicitly described, and may also include units required to obtain relevant information or perform processing based on the information. It should be noted that in the following description, it refers to "one embodiment", which describes a subset of all possible embodiments, but it can be understood that "one embodiment" may be the same subset or different sub-sets of all possible embodiments. Set, and can be combined with each other without conflict.

Before describing the present invention in further detail, the terms and terms involved in the embodiments of the present invention will be described. The terms and terms involved in the embodiments of the present invention are applicable to the following explanations.

1) Classification information of cells, including cell type information and probability information.

2) Cell type information, which is used to characterize the at least two predefined cell types to be selected, including but not limited to: neutrophil nucleus granulocytes, neutrophil neutrophils, eosinophils, Basophils, lymphocytes, blasts, naive granulocytes, atypical lymphocytes, naive lymphocytes, naive mononuclear cells, nucleated red blood cells, megakaryocytes.

3) Probability information, which is used to characterize the probability that the corresponding cell type to be selected is the target type of the cell. The probability information may be presented directly in the form of probability values, or indirectly in the form of probability magnitudes or in sorted form Present, for example, divided into a first possibility, a second possibility, a third possibility, etc. according to the probability value.

4) The target type of the cell, the classification of the cell output after the image processing device processes the digital image of the cell in the blood sample, that is, the type of cell to be selected to which the digital image of the cell belongs.

5) A display output device configured to output digital images and classification information of the corresponding cells to the display interface. For example, the display output device may be a display output interface (ie, an electrical interface) for outputting digital / analog image signals, and can output image signals to an external display. The display output device may further include a display output interface and a display device, wherein the display device is connected to the display output interface for receiving the signal output by the display output interface and correspondingly displaying the digital image and classification of the cells information.

When the display output device is implemented as a display output interface, different display devices can be connected to the display output interface configuration according to the usage environment, for example, at least two displays can be connected to achieve simultaneous detection by multiple persons, and a projector can also be connected for Teaching demonstration.

6) In response, it is used to indicate the condition or state on which the performed operation depends, and when the dependent condition or state is satisfied, the performed operation or operations may be real-time or may have a set delay; Unless otherwise specified, there are no restrictions on the order of execution of multiple operations.

7) Smear, that is, a substrate smeared with a specimen, for example, a glass slide after smearing blood evenly and staining blood cells therein.

FIG. 1 is a schematic flowchart of an optional method for analyzing cells provided by an embodiment of the present invention. The method can be applied to a cell analysis device. The cell analysis device includes: a control device, a digital imaging device, an image acquisition device, and an image. Processing device and display output device. The steps shown will be explained with reference to FIG. 1.

Step 101: Acquire a digital image of cells in a blood sample.

In one embodiment of the method of the present invention, the digital imaging device includes a lens group and a digital camera. Due to the large number of smears of blood samples to be tested, an automatic placement device for automatically placing the smear to the imaging position of the lens group is also provided, which can increase the processing speed of the system and reduce the workload of medical personnel .

In one embodiment of the method of the present invention, the automatic placement device includes a mechanical conveyor. The mechanical transfer part may be implemented as a robot arm, for example, for clamping the smear to a position facing the lens of the lens group. The mechanical conveying section may also be implemented as a transmission belt for conveying the smear to a position facing the lens of the lens group, for example.

For example, when a smear box containing multiple blood sample smears is placed in the receiving section, the automatic transmission device transports the smear box from the receiving section to the area where the lens group is located, in order from the compartment of the smear box Remove the smear, or remove the smear from the designated compartment of the smear box, and place it in the imaging position of the lens group. After the shooting is completed, send the smear back to the smear box; when a smear box is completed, the smear The film box is returned to the accommodating part, and then the next smear box is photographed, so that the smear can be photographed in batches with high efficiency.

In an embodiment of the method of the present invention, in the cell analysis device to which the cell image processing method provided in the embodiment of the present invention is applied, the digital imaging device, such as the lens group, may use a microscope objective lens. The cell analysis device may further include a loading part for placing one or more smears. As an example, the shape has a square shape, a round row, etc., and a light hole is provided at the position where the smear is placed to ensure the brightness of the captured image; A fixing portion (such as a jig) may also be provided for clamping the smear to keep the position stable.

Specifically, a partial structure of the cell analysis device will be described with reference to FIG. 3. FIG. 3 is an optional structural diagram of the cell analysis device provided by the embodiment of the present invention. The cell analysis device includes an accommodating portion 301 for accommodating one or more smears at a time; smear 302, stage 303, first objective 304, second objective 305, third objective 306, eyepiece 307, digital imaging Device 308, wherein the digital imaging device 308 includes: a lens group and a digital camera.

Step 102: Classify the acquired digital images of the cells to form classification information for each of the cells.

Step 103: Output a digital image of the cell and classification information of the cell.

Wherein, the classification information of each of the cells includes cell type information and probability information, the cell type information includes at least two predefined cell types to be selected, and the probability information includes the corresponding cell type to be selected is the cell The probability value of the target type. The at least two predefined cell types to be selected can be set by the user according to actual conditions.

Thus, with the technical solution shown in this embodiment, since the probability information included in the cell classification information includes the probability value that the corresponding cell type to be selected is the target type of the cell, the user can be accurate and intuitive To know the cell type to be selected that the cell may belong to and the probability value corresponding to the cell type to be selected, so as to provide an effective reference for the manual re-examination process of the cell. Especially when the cell analysis equipment has pre-classified the cells, the cell classification information, especially the probability information, can assist the user to manually re-examine the pre-classification of the cells, which effectively improves the re-examination efficiency and reduces the user's work burden.

In an embodiment of the method of the present invention, the outputting the digital image of the cell and the classification information of the cell may include: selecting from the candidate cell types of the cell type information according to the probability information The target type of the cell; based on the selected target type, a digital image of the cell and corresponding classification information are output. For example, when the cell analysis device automatically selects eosinophils as the target type of a certain cell, a digital image showing the cell is output in the display area belonging to the eosinophil type of the display output device, and In the case of a digital image, all the cell types to be selected and their corresponding probability values of the cell are displayed in the corresponding area of the display output device, as shown in FIG. 4A, for example, to assist the user in re-examination.

In an embodiment of the method of the present invention, the selecting the target type of the cell from the candidate cell types of the cell type information according to the probability information may include: selecting the maximum probability from the probability information The cell type to be selected corresponding to the value is selected as the target type of the cell. Since the cell type information includes at least two predefined candidate types, the cell analysis device directly and automatically selects the candidate cell type corresponding to the maximum probability value in the probability information through the solution shown in this embodiment. The target type of the cell, so that the user does not need to pre-sort artificial cells.

In an embodiment of the present invention, the output of the digital image of the cell and the classification information of the cell includes: directly in the form of probability values or indirectly in the form of probability grading or ranking related to the probability values The probability information is output. As shown in FIGS. 4A to 4C, FIGS. 4A to 4C are schematic diagrams of an optional display interface of a cell analysis device provided by an embodiment of the present invention. When the digital image of a cell is selected, the classification information of the selected cell can be displayed on the display interface of the cell analysis device, wherein the classification information of the cell includes cell type information and probability information, The cell type information is used to characterize at least two predefined cell types to be selected for the cell, and the probability information is used to characterize the probability that the cell type to be selected is the target type of the cell accordingly. In FIG. 4A, the probability information of the selected cell is directly shown in the form of a probability value; in FIG. 4B, the probability information of the selected cell is shown indirectly in the possibility classification, which is classified according to the probability value; In FIG. 4C, two cell images in the display interface of FIG. 4A or 4B are exemplarily enlarged, wherein the probability information of all cells is displayed indirectly in a sorted manner according to the probability value.

In an embodiment of the method of the present invention, the method may further include: adjusting the target type of the selected cell in response to human-computer interaction and according to the probability information. During the process of automatic pre-classification of the cells based on the digital image of the cells in the blood sample by the cell analysis device, the output cell classification information may be wrong, therefore, through human-computer interactive operation (that is, human-computer interactive instructions , Including but not limited to: control instructions issued by external control devices, user's voice instructions), the user can instruct the cell analysis device to automatically adjust the target type of the misclassified cell according to the probability information, or the user according to the probability The information manually adjusts the target type of misclassified cells to avoid misdetection again. This embodiment is particularly advantageous when the probability information included in the classification information of the cells is sorted in ascending or descending order.

In an embodiment of the method of the present invention, before adjusting the target type of the selected cell, the method further includes: in response to human-computer interaction and according to the probability information, screening the probability value corresponding to the current target type is less than the corresponding Cells of the first threshold and output a digital image of the selected cells, and / or filter cells with a difference between the probability value corresponding to the current target type and the adjacent probability value in the corresponding probability information less than the second threshold and output Digital image of selected cells. Preferably, the first threshold is set in relation to the cell type to be selected. Specifically, each type of target to be selected has a corresponding first threshold. For example, when a cell type to be selected is more complex, the corresponding first threshold value is larger, or when it is difficult to distinguish between two cell types to be selected, the first corresponding to the two cell types to be selected The threshold can also be set relatively large. In other words, the setting of the first threshold determines the reliability of the cell type to be selected as the target type of the cell. The larger the first threshold, the lower the credibility, that is, the greater the possibility of classification errors. In addition, when the probability value corresponding to the current target type of the cell is not much different from the probability values of other candidate cell types of the cell, the possibility of the cell being misclassified is relatively large, and the setting of the second threshold can Reduce such classification errors. Therefore, through the technical solution shown in this embodiment, cells that may be misclassified can be quickly and accurately screened out and provided to the user for re-examination, so as to improve the re-examination efficiency.

In an embodiment of the method of the present invention, the method further includes recording the adjustment process of the target type of each cell, including recording the type of the candidate cell to be adjusted and the number of times the type of the selected cell is adjusted. For example, the current target type of cell 1 and cell 2 is eosinophils, and the current target type of cell 3 is basophils. If the user finds that the classification of these three cells is wrong and adjusts accordingly, the cell The analysis device records that the number of times the cell whose current target type is eosinophils is adjusted is 2, and the number of times the cell whose current target type is eosinophils is adjusted is 1. Further, according to the recorded adjustment process, when the number of times of adjustment of cells whose target type is the same candidate cell type exceeds the adjustment threshold, the first threshold for the same candidate cell type is increased. For example, when the user finds that cells classified as eosinophils always have classification errors, the user can raise the first threshold for eosinophils, that is, eosinophils are selected as the target type of cells Credibility declines. This can reduce classification errors.

In addition, the method may further include labeling the adjusted cells, for example, cells that have been adjusted different times may be labeled differently, so as to display the cell adjustment process for the user.

In an embodiment of the method of the present invention, the outputting the digital image of the cell and the classification information of the cell may include: sorting the probability information included in the classification information of the cell; based on the The sorting result of the probability information outputs the classification information of the cells in a preset output mode. Further, during the display process, when the mouse arrow is directed to the corresponding cell image or the corresponding display area by controlling an external control device and man-machine interaction, the corresponding cell image or all cells in the display area Probability information will be triggered to display. Wherein, the outputting the classification information of the cell by a preset output method may include: outputting the classification information of the cell in a preset display area of the display output device, and further, it may also be used as an external operation device (such as a mouse) Or a stylus) is placed on a cell, a virtual interface is displayed in response to the operation of the external operation device, and the classification information of the cell is displayed on the virtual interface. Referring again to FIGS. 4A to 4C, the classification information of the cells in the selected state is shown in a specific area of the display interface in FIG. 4A, and the classification of the cells in the selected state is shown in the form of a virtual pop-up window in FIG. 4B. The information is displayed in the same display area of the cell image in FIG. 4C, and the classification information of the cell can be further displayed in the cell image (not shown in the figure). Since different users have different usage habits, with the above technical solutions provided by the embodiments of the present invention, users can flexibly set the output display area of the cell classification information according to their own usage habits.

In an embodiment of the method of the present invention, the outputting the classification information of the cells by a preset output method may further include: outputting the classification of the cells in a predetermined output language or expression in a fixed display area information. Since the application environment of the cell analysis device is different, the language type displayed in the fixed display area can be set, and at the same time, preset image indication information can be used to replace the type information of the cell expressed in text information. In order to avoid the misdetection of the cell caused by the difference in the language used between different users.

In an embodiment of the method of the present invention, the sorting the probability information included in the classification information of the cells may include: sorting the probability information included in the classification information of the cells in ascending or descending order. Through the technical solution shown in this embodiment, sorting the probability information in ascending or descending order can make it easier for the user to visually observe the cell types to be selected with the highest probability value of the selected cell in order to better assist the user in identifying the wrong Change the target type.

In an embodiment of the method of the present invention, the adjusting the target type of the selected cell may include: the cell analysis device automatically changing the target type of the selected cell to another target type in the probability information of the selected cell, The probability information corresponding to the additional target type is only smaller than the probability information corresponding to the target type before change. Therefore, the cell analysis device automatically changes the target type of the selected cell to the probability information of the selected cell in response to the human-computer interaction operation. The probability information corresponding to the characterization of the selected cell is only less than the probability information corresponding to the target type before the change. The target type can effectively improve the operation efficiency of changing the target type of the selected cell, and realize the automatic adjustment of the cell analysis device. Further, when the user finds that the classification of the selected cell is wrong, the target type of the selected cell can be manually changed to the correct cell type to be selected through human-computer interaction, because the cell analysis device can The probability information included in the classification information is sorted in ascending or descending order. At this time, the probability information in the classification information of the cell can play a role in assisting the user to make a quick judgment, so as to achieve accurate classification of the cells and avoid misdetection.

As shown in FIG. 4A, there are ten types of cells to be selected in this embodiment. The probability that the target type of the selected cell is the type of the selected cell is 0.0421, the probability of the type of the selected cell type 2 is 0.1122. The probability of selecting cell type 3 is 0.0561, the probability of selecting cell type 4 is 0.0252, the probability of selecting cell type 5 is 0.0351, the probability of selecting cell type 6 is 0.0070, the probability of selecting cell type 7 It is 0.0168, the probability of being the candidate cell type 8 is 0.7013, the probability of being the candidate cell type 9 is 0.0028; the probability of being the candidate cell type 10 is 0.0014. In FIG. 4A, the classification information of the cells in the selected state is output in a descending order of the probability information of the 10 candidate cell types. According to the solution shown in the embodiment of the present invention, the cells in the selected state are classified as the cell type 8 with the highest probability value. When the user finds that the classification of the selected cell is wrong, the user can manually The target type of the selected cell is changed to the correct cell type to be selected. Since the cell type to be selected corresponds to the probability information, the user can intuitively observe that the probability that the cell is the cell type to be selected 2 is only less than that of the cell type to be selected 8. The probability value, so you can directly change the target type of the selected cell from the candidate cell type 8 to the candidate cell type 2. Similarly, in FIG. 4B and FIG. 4C, the classification information of the cells in the selected state is also output in descending order.

In an embodiment of the method of the present invention, the method may further include: for different cells whose target type is the same candidate cell type, the probability information of the different cells corresponding to the same candidate cell type Perform ascending or descending sorting; based on the ascending or descending sorting results of the probability information of the different cells corresponding to the same cell type to be selected, output a digital image of the different cells. In the process of classifying the digital pictures of each cell in advance by the cell analysis device, the probability that the target type of different cells is the same cell type to be selected is usually different. For example, the target type of the first cell is classified as The probability of eosinophils is 0.95, the probability that the target type of the second cell is classified as eosinophil is 0.72, the probability that the target type of the third cell is classified as eosinophil is 0.80, and the probability of the fourth cell The probability that the target type is classified as eosinophils is 0.89, and the probability that the fourth cell is classified as eosinophils is 0.6. Through the cell analysis device, the probability of classifying the target types of these four cells as eosinophils can be sorted in ascending or descending order, and the digital images of these four cells can be output in ascending or descending order of the probability, for example, according to The first cell (probability 0.95)-the fourth cell (probability 0.89)-the third cell (probability 0.80)-the second cell (probability 0.72) in the order of eosinophils A digital image of each cell belonging to eosinophils is output in the display area of. Thus, for example, when the user performs a re-examination in the display area of the eosinophil type, the user can quickly and intuitively find cell images that may not belong to eosinophils, usually cell images with a low probability.

Further, when the user finds the misclassified cells according to the ascending or descending sorting results of the probability information corresponding to the same cell type to be selected for the different cells, the target type of the misclassified cells can be adjusted in a corresponding manner. In particular, the user can set: when the probability value of different cells whose target type is the same to-be-selected cell type corresponding to the same to-be-selected cell type is less than a predetermined threshold, uniformly adjust the target types of the different cells . In other words, the user can set a credible threshold according to the actual situation. The credible threshold indicates that when the probability that the target type of the cell is a certain cell type is lower than the credible threshold, the result is considered unreliable and needs to be Adjust the target type of the cell. For example, when the user performs a retest in the display area of a cell type to be selected, the user observes that when the probability that some cells belong to the cell type to be selected is below a certain threshold, the classification of these cells is wrong , Then the user can set the threshold as a trusted threshold and modify the target types of these cells in batches through instructions. Through the technical solution shown in this embodiment, the setting of the predetermined threshold, that is, the reliable threshold can facilitate the user to perform batch adjustment processing and improve the efficiency of cell re-examination.

This is described with reference to FIG. 5A, which is a schematic diagram of an optional display interface of the cell analysis device provided by an embodiment of the present invention. In FIG. 5A, the probability information corresponding to neutrophils of different cells belonging to neutrophils is sorted in descending order, and digital images of different cells belonging to neutrophils are output in this descending order. Specifically, in this embodiment, the number of cells belonging to neutrophils is 10, wherein the probability that cell 1 is judged to be neutrophil is 0.75; the probability that cell 2 is judged to be neutrophil is 0.55; The probability of cell 3 being broken into neutrophils is 0.99; the probability of cell 4 being judged to be neutrophils is 0.95; the probability of cell 5 being judged to be neutrophils is 0.7; cell 6 is judged to be neutrophils The probability of cells is 0.65; the probability of cells 7 being judged to be neutrophils is 0.9; the probability of cells 8 being judged to be neutrophils is 0.6; the probability of cells 9 being judged to be neutrophils is 0.85; cell 10 The probability of being judged to be neutrophils was 0.8. Through the cell analysis device, the descending sorting is performed according to the probability that the 10 cells belong to neutrophils, and according to the sorting result, a digital image of the 10 cells and corresponding probability information are output. Of course, it is also possible to sort the probability information of the cells 11-cell 20 in the display area of eosinophils and the cells 21-cell 30 in the display area of basophils. In this embodiment, only the probability of the current target type is displayed in the corresponding cell image.

Further, when the user finds here that the cells 2, 6 and 8 currently belong to neutrophils and the cells 12, 18, 19, 20 currently belong to eosinophils and the cells 22 currently belong to basophils , 28, 30 When the classification error occurs, the user can select these misclassified cell images at the same time, by double-clicking an area, as shown in Figure 5B, or through the drop-down menu, as shown in Figure 5C for these misclassified cells The image is uniformly adjusted and the adjusted cells are marked. The adjusted display interface is shown in FIG. 5D, in which the display of the probability information of the adjusted cells also changes accordingly. Specifically, cells 6 and 8 that originally belonged to neutrophils were adjusted to eosinophils, cell 2 that originally belonged to neutrophils was adjusted to basophils; cells that originally belonged to eosinophils 18 19 and 19 are adjusted to neutrophils, cells 12 and 20 that originally belonged to eosinophils are adjusted to basophils; cells 22 and 28 that originally belonged to eosinophils are adjusted to eosinophils; The cells 30 originally belonging to basophils are adjusted to neutrophils. Of course, the user can also directly instruct the cell sub-device through human-computer interaction that the probability value of belonging to neutrophils is less than the first predetermined threshold (here, for example, 0.7) and the probability value of belonging to eosinophils is less than the second The target type of cells with a predetermined threshold (here, for example, 0.8) and a probability value that belongs to basophils is less than the third predetermined threshold (here, for example, 0.75) are automatically adjusted uniformly, which can further effectively improve the changes The operation efficiency of the cell type realizes the automatic adjustment of the cell analysis equipment. In this way, the efficiency of rechecking digital images of cells in blood samples can be greatly improved.

In an embodiment of the method of the present invention, the classifying the acquired digital image of the cell to form classification information of the cell includes: through a neural network algorithm, especially a deep neural network algorithm, to The acquired digital images of the cells are classified to form classification information of the cells. Through the technical solution shown in this embodiment, the cell analysis device can automatically classify the digital images of the cells in the blood sample, so as to increase the speed of the digital image processing of the cells in the blood sample by the cell analysis device .

In an embodiment of the method of the present invention, the method further comprises: receiving a count control instruction, in response to the count control instruction, counting the number of cells in a selected area of the digital image of the blood sample, and recording the selected The number of target type cells in a given area. Through the technical solution shown in this embodiment, the number of cells in the blood sample can also be counted, so that the user can count the proportion of mutated cells in the blood sample.

The cell analysis device of the present invention will be described with reference to FIG. 2. The advantages described above for the method of the present invention are also applicable to the cell analysis device of the present invention.

2 is a schematic diagram of an optional structure of a cell analysis device 200 provided by an embodiment of the present invention. The cell analysis device 200 includes: a control device (not shown in the figure) configured to adjust the relative position of the digital imaging device and the blood sample Location; digital imaging device (not shown), including lens group and digital camera. The image acquisition device 201 is configured to acquire digital images of cells in the blood sample; the image processing device 202 is configured to classify the acquired digital images of the cells to form classification information of each of the cells; display output device 203, configured to output a digital image of the cell and classification information of the cell. Wherein, the classification information of each of the cells includes cell type information and probability information, the cell type information includes at least two predefined cell types to be selected, and the probability information includes the corresponding cell type to be selected is the cell The probability value of the target type. The at least two predefined cell types to be selected can be set by the user according to actual conditions.

In an embodiment of the cell analysis apparatus 200 of the present invention, the display output device 203 is configured to output the probability information directly in the form of probability values or indirectly in the form of probability grading or ranking related to the probability values .

In an embodiment of the cell analysis device 200 of the present invention, the image processing device 202 may be configured to: select the target type of the cell from the candidate cell types of the cell type information according to the probability information; And the display output device 203 may be configured to output a digital image of the cell and corresponding classification information based on the selected target type. For example, when the cell analysis device automatically selects eosinophils as the target type of a certain cell, a digital image showing the cell is output in the display area belonging to the eosinophil type of the display output device, and In the case of a digital image, all the cell types to be selected for the cell and their corresponding probability values are displayed in the corresponding area of the display output device, for example, to assist the user in re-examination. Further, the image processing device 202 may be specifically configured to select the cell type to be selected corresponding to the maximum probability value in the probability information as the target type of the cell.

In an embodiment of the cell analysis device 200 of the present invention, the image processing device 202 may be further configured to: sort the probability information included in the classification information of the cells; based on the sorting result of the probability information, The classification information of the cells is output in a preset output mode. Wherein, the outputting the classification information of the cell by a preset output method may include: outputting the classification information of the cell in a preset display area of the display output device, and further, it may also be used as an external operation device (such as a mouse) (Or stylus) when the pointer is placed on a cell, the virtual interface displays the classification information of the cell in response to the operation of the external operation device. Since different users have different usage habits, with the above technical solutions provided by the embodiments of the present invention, users can flexibly set the output display area of the cell classification information according to their own usage habits.

In an embodiment of the cell analysis apparatus 200 of the present invention, the image processing device 202 may be further configured to: in response to human-computer interaction and according to the probability information, adjust the target type of the selected cell. Since the cell analysis device classifies the cells based on the digital image of the cells in the blood sample, the classification information of the cells output may be wrong. Therefore, through human-computer interactive operations (that is, human-computer interactive commands , Including but not limited to: control instructions issued by external control devices, user's voice instructions), the user can instruct the cell analysis device to automatically adjust the target type of the selected cell according to the probability information or the user manually according to the probability information Adjust the target type of the selected cells to avoid misdetection of the cells to be tested. This embodiment is particularly advantageous when the probability information included in the classification information of the cells is sorted in ascending or descending order.

In an embodiment of the cell analysis apparatus 200 of the present invention, the image processing device 202 may be further configured to: before adjusting the target type of the selected cell, in response to human-computer interaction and according to the probability information, filter the current The cell with the probability value corresponding to the target type is smaller than the corresponding first threshold and outputs a digital image of the selected cell. The image processing device 202 may be further configured to: in response to human-computer interaction and based on the probability information, filter the difference between the probability value corresponding to the current target type and the probability value adjacent to the corresponding probability information less than the first Two threshold cells and output a digital image of the selected cells. Wherein, the first threshold may be set in association with the cell type to be selected.

In an embodiment of the cell analysis apparatus 200 of the present invention, the image processing device 202 may be further configured to: change another cell type to be selected from the cell type information of the selected cell to the target type of the selected cell; In the probability information of the selected cell, the probability value corresponding to the additional cell type to be selected is only smaller than the probability value corresponding to the target type before change. Therefore, by automatically changing the target type of the selected cell to the probability information of the selected cell through the cell analysis device, the corresponding probability information is only less than the probability information corresponding to the target type before the change. The operation efficiency of changing the target type of the selected cell is improved, and the automatic adjustment of the cell analysis device is realized. Further, when the user finds that the classification of the selected cell is wrong, the target type of the selected cell can be manually changed to the correct cell type to be selected through human-computer interaction, because the cell analysis device can The probability information included in the classification information is sorted in ascending or descending order. At this time, the probability information in the classification information of the cell can play a role in assisting the user to make a quick judgment, so as to achieve accurate classification of the cells and avoid misdetection.

In an embodiment of the cell analysis apparatus 200 of the present invention, the image processing device 202 may also be configured to record the adjustment process of the target type of each cell. Further, the cell analysis device can also save the recorded adjustment process of the target type of each cell for subsequent follow-up inspection. Further, the image processing device 202 may also be configured to mark the adjusted cells, for example, different marks may be applied to the cells that have been adjusted different times, so as to display the adjustment process of the cells for the user.

In an embodiment of the cell analysis device 200 of the present invention, the image processing device 202 may be further configured to: according to the recorded adjustment process, when the target type is the same type of cells to be selected, the number of adjustments exceeds the adjustment cumulatively At the threshold, the first threshold for the same cell type to be selected is increased.

In an embodiment of the cell analysis apparatus 200 of the present invention, the image processing device 202 may be further configured to perform single adjustment or batch adjustment on the selected cells. Through the technical solution shown in this embodiment, the user can flexibly select the adjustment method according to the number of cells to be detected.

In an embodiment of the cell analysis device 200 of the present invention, the image processing device 202 may be further configured to sort the probability information corresponding to the same target type of different cells belonging to the same target type in ascending or descending order Sorting; and the display output device 203 may be further configured to: output the differences belonging to the same target type based on the sorting results of the probability information corresponding to the same target type of the different cells belonging to the same target type Digital image of cells. .

In an embodiment of the cell analysis apparatus 200 of the present invention, the image processing device 202 may be further configured to: sort the probability information corresponding to the same target type based on the different cells belonging to the same target type As a result, the target types of the different cells belonging to the same target type are adjusted, thereby improving the retest efficiency.

In an embodiment of the cell analysis apparatus 200 of the present invention, the image processing device 202 may be further configured to uniformly adjust the target type of the cell with a probability value corresponding to the same target type less than a predetermined threshold. Thus, the user can quickly and intuitively know the cells that may be misclassified.

In an embodiment of the cell analysis device 200 of the present invention, the image processing device 202 may be further configured to classify the acquired digital image of the cell through a neural network algorithm, especially a deep neural network algorithm, To form classification information of the cells. Through the technical solution shown in this embodiment, the cell analysis device can automatically classify the digital images of the cells in the blood sample, so as to increase the speed of the digital image processing of the cells in the blood sample by the cell analysis device .

In an embodiment of the cell analysis device 200 of the present invention, the cell analysis device further includes an information transceiving device configured to receive a count control instruction. The image processing device 202 may therefore be configured to: in response to the counting control instruction, count the number of cells in the selected area of the push piece to be tested, which is composed of all digital images in the blood sample, and record the selected area The number of target type cells. Since all the digital images in the blood sample can be combined into a push piece to be tested, the technical solution shown in this embodiment can realize the statistics of the number of cells in the push piece to be tested, so that the user can easily analyze the blood sample The proportion of mutated cells in all digital images is counted.

In an embodiment of the cell analysis device 200 of the present invention, the display output device 203 may include a display output interface configured to output a signal corresponding to the image to an external device. Further, the display output device 203 may further include a display device connected to the display output interface, which is configured to receive a signal output by the display output interface and correspondingly display an image.

6 is a schematic diagram of a cell analysis device provided by an embodiment of the present invention. The cell analysis device includes: an image acquisition device (not shown in the figure) configured to acquire a digital image of cells in a blood sample; an image processing device (Not shown in the figure), configured to classify the acquired digital images of the cells to form classification information of each of the cells, wherein the classification information of each of the cells includes cell type information and probability, respectively Information, the cell type information is used to characterize at least two predefined candidate types of the cell, and the probability information is used to characterize the probability that the corresponding cell type to be selected is the target type of the cell; display 603, configuration To output digital images of the cells and classification information of the cells. The representation form of the image processing device may be one or more application specific integrated circuits (ASIC, Application Integrated Circuit), DSP, programmable logic device (PLD, Programmable Logic Device), complex programmable logic device (CPLD , Complex Programmable Logic Device, Field Programmable Gate Array (FPGA, Field-Programmable Gate Array), general-purpose processor, controller, microcontroller (MCU, Micro Controller), microprocessor (Microprocessor), or other electronic element.

7 is a schematic diagram of a cell analysis system provided by an embodiment of the present invention. The cell analysis device of the present invention is applied to the cell analysis system. The cell analysis system includes a cell analysis device 701, a cell analysis device 702, and a display device 703 . Wherein, the cell analysis device 701 includes: an image acquisition device (not shown in the figure) configured to acquire a digital image of cells in the blood sample; and an image processing device (not shown in the figure) configured to analyze the acquired The digital images of the cells are classified to form classification information of each of the cells respectively; the display output device 7013 is configured to output the digital images of the cells and the classification information of the cells. Wherein, when the digital image of the cells in the blood sample is selected, the classification information of the selected cells can be displayed on the display interface of the cell analysis device, wherein the classification information of the cells includes cell type information and probability, respectively Information, the cell type information is used to characterize at least two predefined candidate types of the cell, and the probability information is used to characterize the probability that the corresponding cell type to be selected is the target type of the cell.

The cell analysis device 701 has the same structure as the cell analysis device 702 and includes: an image acquisition device (not shown in the figure), an image processing device (not shown in the figure), and a display output device 7023. The cell analysis system shown in this embodiment is a cluster application of the cell analysis equipment disclosed in the present invention. The number of the cell analysis equipment and the number of display equipment are not limited by the present invention.

The display device 703 may output the cell classification information processed by the cell analysis device 701 and / or the cell analysis device 702.

8 is a schematic structural diagram of an optional cell analysis device provided by an embodiment of the present invention. As shown in FIG. 8, the cell analysis device 800 may be a medical device including a cell image processing function, a portable analyzer, or the like. The cell analysis device 800 shown in FIG. 8 includes: at least one processor 801, a memory 802, at least one network interface 804, and a user interface 803. The various components in the cell analysis device 800 are coupled together via a bus system 805. It can be understood that the bus system 805 is used to implement connection and communication between these components. In addition to the data bus, the bus system 805 also includes a power bus, a control bus, and a status signal bus. However, for clarity, various buses are marked as the bus system 805 in FIG. 8.

The user interface 803 may include a display, a keyboard, a mouse, a trackball, a click wheel, buttons, buttons, a touch panel, or a touch screen.

It can be understood that the memory 802 may be a volatile memory or a non-volatile memory, and may also include both volatile and non-volatile memory. Among them, the non-volatile memory can be read-only memory (ROM, Read Only Memory), programmable read-only memory (PROM, Programmable Read-Only Memory), erasable programmable read-only memory (EPROM, Erasable Programmable Read- Only Memory), Electrically Erasable Programmable Read Only Memory (EEPROM, Electrically Erasable Programmable Read-Only Memory), Magnetic Random Access Memory (FRAM, Ferromagnetic Random Access Memory), Flash Memory (Flash) Memory, Magnetic Surface Memory , Compact disc, or read-only compact disc (CD-ROM, Compact, Read-Only Memory); the magnetic surface memory can be a disk storage or a tape storage. The volatile memory may be a random access memory (RAM, Random Access Memory), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM, Static Random Access Memory), synchronous static random access memory (SSRAM, Synchronous Static Random Access Memory), dynamic random access Memory (DRAM, Dynamic Random Access), synchronous dynamic random access memory (SDRAM, Synchronous Dynamic Random Access Memory), double data rate synchronous dynamic random access memory (DDRSDRAM, Double Data Rate, Synchronous Dynamic Random Access Random Access Memory), enhanced Type synchronous dynamic random access memory (ESDRAM, Enhanced Synchronous Dynamic Random Access Memory), synchronous connection dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), direct memory bus random access memory (DRRAM, Direct Rambus Random Access Ram ). The memory 802 described in the embodiments of the present invention is intended to include these and any other suitable types of memory.

The memory 802 in the embodiment of the present invention includes but is not limited to: tri-state content addressable memory, static random access memory capable of storing various types of data such as received cell images to support the operation of the cell analysis device 800. Examples of these data include: any computer program for operating on the cell analysis device 800, such as an operating system 8021 and application program 8022, storing image data, classification information, and the like. Among them, the operating system 8021 contains various system programs, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks. The application program 8022 may include various application programs, such as a client or an application program with a cell analysis function, etc., which are used to implement: including: acquiring a digital image of cells in a blood sample; classifying the acquired digital images of the cells To form classification information of each of the cells separately; output various application services including digital images of the cells and classification information of the cells. The program for realizing the corresponding operation of classifying the cell image according to the embodiment of the present invention may be included in the application program 8022.

The method disclosed in the foregoing embodiment of the present invention may be implemented by the processor 801. The processor 801 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method may be completed by operations in the form of hardware integrated logic circuits or software in the processor 801. The aforementioned processor 801 may be a general-purpose processor, a digital signal processor (DSP, Digital Processor), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, and the like. The processor 801 may implement or execute the disclosed methods, steps, and logical block diagrams in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of the present invention may be directly implemented and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium. The storage medium is located in the memory 802. The processor 801 reads the information in the memory 802 and completes the foregoing corresponding steps in combination with its hardware.

In an exemplary embodiment, the cell image processing system 800 may be implemented by one or more application specific integrated circuits (ASIC, Application Specific Integrated Circuit), DSP, programmable logic device (PLD, Programmable Logic Device), complex programmable logic device (CPLD, Complex Programmable Logic Device), Field Programmable Gate Array (FPGA, Field-Programmable Gate Array), general-purpose processor, controller, microcontroller (MCU, Micro Controller), microprocessor (Microprocessor), or Other electronic components are implemented and configured to perform the cell image processing method.

In an exemplary embodiment, an embodiment of the present invention further provides a computer-readable storage medium, such as a memory 802 including a computer program, which can be executed by the processor 801 of the cell image processing system 800 to complete the aforementioned method Various steps. The computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disk, or CD-ROM; it may also be various devices including one or any combination of the above memories, such as Portable analyzers, etc.

An embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it executes: adjusting the relative position of the digital imaging device and the smear based on a mechanical transmission; Digital images of cells in the blood sample; classifying the acquired digital images of the cells to form classification information of each of the cells, wherein the classification information of each of the cells includes cell type information and probability information, The cell type information is used to characterize at least two predefined candidate types of the cell, and the probability information is used to characterize the probability that the corresponding cell type to be selected is the target type of the cell; the number of the cell is output Images and classification information of the cells.

Those skilled in the art should understand that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the embodiments of the present invention may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Moreover, the embodiments of the present invention may take the form of a computer program product implemented on one or more computer usable storage media (including disk storage and optical storage, etc.) containing computer usable program code.

Embodiments of the present invention are described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram and a combination of the flow and / or block in the flowchart and / or block diagram can be implemented by a computer program operation. These computer programs can be provided to operate on the processor of a general-purpose computer, special-purpose computer, embedded processing machine, or other programmable data processing device to produce a machine that allows operations performed by the processor of the computer or other programmable data processing device to produce An apparatus for realizing the functions specified in one block or multiple blocks of one flow or multiple flows of a flowchart and / or one block or multiple blocks of a block diagram.

These computer program operations can also be stored in a computer readable memory that can guide a computer or other programmable data processing device to work in a particular manner, so that the operations stored in the computer readable memory produce a manufactured article including an operating device, the operation The device implements the functions specified in one block or multiple blocks of the flowchart one flow or multiple flows and / or block diagrams.

These computer program operations can also be loaded onto a computer or other programmable data processing device, so that a series of operating steps are performed on the computer or other programmable device to produce computer-implemented processing, which is executed on the computer or other programmable device The operations provide steps for implementing the functions specified in the flow chart flow or flows and / or the block diagram block or blocks.

The above are only preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modification, equivalent replacement and improvement made within the spirit and principle of the present invention should be included in Within the protection scope of the present invention.

Claims

A method for analyzing cells, applied to cell analysis equipment, the method includes:

Obtain digital images of cells in blood samples;

Classify the acquired digital images of the cells to form classification information of each of the cells, wherein the classification information of each of the cells includes cell type information and corresponding probability information;

The digital image of the cell and the classification information of the cell are output.
The method according to claim 1, wherein

The cell type information includes at least two predefined cell types to be selected, and the probability information includes a probability value that the corresponding cell type to be selected is a target type of the cell.
The method according to claim 1 or 2, wherein the outputting the digital image of the cell and the classification information of the cell includes:

The probability information is output directly in the form of probability values or indirectly in the form of probability ranking or ranking related to the probability values.
The method according to any one of claims 1 to 3, wherein the outputting the digital image of the cell and the classification information of the cell includes:

Select the target type of the cell from the candidate cell types of the cell type information according to the probability information;

Based on the selected target type, a digital image of the cell and corresponding classification information are output.
The method according to claim 4, wherein the selecting the target type of the cell from the candidate cell types of the type information according to the probability information includes:

The cell type to be selected corresponding to the maximum probability value in the probability information is selected as the target type of the cell.
The method according to claim 4 or 5, wherein the method further comprises:

In response to human-computer interaction and according to the probability information, the target type of the selected cell is adjusted.
The method according to claim 6, wherein, before adjusting the target type of the selected cell, the method further comprises:

In response to human-computer interaction and according to the probability information, screen cells with a probability value corresponding to the current target type less than the first threshold and output digital images and classification information of the screened cells, and / or screen corresponding to the current target type The difference between the probability value of and the probability value adjacent to it in the corresponding probability information is less than the second threshold value and outputs a digital image and classification information of the selected cells.
The method according to claim 7, wherein the first threshold is set in association with the cell type to be selected.
The method according to any one of claims 6 to 8, wherein the adjusting the target type of the selected cells includes:

Change another cell type to be selected from the cell type information of the selected cell to the target type of the selected cell, and in the probability information of the selected cell, the probability value corresponding to the other cell type to be selected is only less than before the change The probability value corresponding to the target type.
The method according to any one of claims 6 to 9, wherein the method further comprises:

Record the adjustment process of the target type of each cell.
The method of claim 10, wherein the method further comprises:

According to the recorded adjustment process, when the number of times of adjustment of cells whose target type is the same candidate cell type exceeds the adjustment threshold, the first threshold for the same candidate cell type is increased.
The method according to any one of claims 6 to 11, wherein the method further comprises:

Label the adjusted cells.
The method according to any one of claims 1 to 12, wherein the outputting the digital image of the cell and the classification information of the cell includes:

Sort the probability information included in the classification information of the cells;

Based on the sorting result of the probability information, the classification information of the cell is output through a preset output mode.
The method according to claim 13, wherein the sorting of the probability information included in the classification information of the cells includes:

Sort the probability information included in the classification information of the cells in ascending or descending order.
The method according to any one of claims 3 to 14, wherein the method further comprises:

For different cells whose target type is the same cell type to be selected, sort the probability values of the different cells corresponding to the same cell type to be sorted in ascending or descending order;

Based on the sorting result, a digital image of the different cells is output.
The method according to any one of claims 6 to 15, wherein the adjusting the target type of the selected cell comprises:

Single adjustment or batch adjustment of selected cells.
The method according to any one of claims 1 to 16, wherein the classifying the acquired digital image of the cell to form classification information of the cell includes:

A neural network algorithm is used to classify the acquired digital images of the cells to form classification information of the cells.
The method according to any one of claims 1 to 17, wherein the method further comprises:

Receive count control commands,

In response to the counting control instruction, the number of cells in the selected area of the digital image of the blood sample is counted, and the number of target type cells in the selected area is recorded.
A cell analysis device, the cell analysis device includes:

A control device configured to adjust the relative position of the digital imaging device and the blood sample;

Digital imaging device, including lens group and digital camera;

An image acquisition device configured to acquire a digital image of cells in a blood sample;

An image processing device configured to classify the acquired digital images of the cells to form classification information of each of the cells, wherein the classification information of each of the cells includes cell type information and corresponding probability information;

The display output device is configured to output a digital image of the cell and classification information of the cell.
The cell analysis apparatus according to claim 19, wherein

The cell type information includes at least two predefined cell types to be selected, and the probability information includes a probability value that the corresponding cell type to be selected is a target type of the cell.
The cell analysis device according to claim 19 or 20, wherein

The display output device is configured to output the probability information directly in the form of probability values or indirectly in the form of probability grading or ranking related to the probability values.
The cell analysis apparatus according to claim 20 or 21, wherein

The image processing device is configured to: select the target type of the cell from the candidate cell types of the cell type information according to the probability information;

The display output device is configured to output a digital image of the cell and corresponding classification information based on the selected target type.
The cell analysis device according to claim 22, wherein

The image processing device is configured to select the cell type to be selected corresponding to the maximum probability value in the probability information as the target type of the cell.
The cell analysis device according to claim 22 or 23, wherein

The image processing device is further configured to adjust the target type of the selected cell in response to human-computer interaction and according to the probability information.
The cell analysis device according to claim 24, wherein

The image processing device is further configured to: in response to human-computer interaction and according to the probability information, screen cells with a probability value corresponding to the current target type less than the first threshold and output digital images and classification information of the screened cells; and / or,

The image processing device is further configured to: filter cells whose difference between the probability value corresponding to the current target type and the probability value adjacent to the corresponding probability information is less than the second threshold and output digital images and classification information of the selected cells .
The cell analysis apparatus according to claim 25, wherein

The first threshold is set in association with the cell type to be selected.
The cell analysis device according to any one of claims 24-26, wherein

The image processing device is configured to: change another cell type to be selected from the cell type information of the selected cell to the target type of the selected cell, and in the probability information of the selected cell, the other cell type to be selected The corresponding probability value is only smaller than the probability value corresponding to the target type before the change.
The cell analysis apparatus according to any one of claims 24-27, wherein

The image processing device is further configured to record the adjustment process of the target type of each cell.
The cell analysis device according to claim 27, wherein

The image processing device is further configured to: according to the recorded adjustment process, when the number of times the cells whose target type is the same to-be-selected cell type is adjusted exceeds an adjustment threshold, increase the first threshold for the same to-be-selected cell type.
The cell analysis device according to any one of claims 24-29, wherein

The image processing device is further configured to mark the adjusted cells.
The cell analysis device according to any one of claims 20 to 30, wherein

The image processing device is further configured to sort the probability information included in the classification information of the cells;

The display output device is further configured to output the classification information of the cell in a preset output mode based on the sorting result of the probability information.
The cell analysis device according to claim 31, wherein

The image processing device is further configured to sort the probability information included in the classification information of the cells in ascending or descending order.
The cell analysis device according to any one of claims 24-32, wherein

The image processing device is further configured to sort the probability values of the different cells corresponding to the same cell type to be sorted in ascending or descending order for different cells whose target type is the same cell type to be selected;

The display output device is further configured to output digital images of the different cells based on the sorting result.
The cell analysis device according to any one of claims 24-33, wherein

The image processing device is further configured to perform single adjustment or batch adjustment on the selected cells.
The cell analysis device according to any one of claims 20 to 34, wherein

The image processing device is further configured to classify the acquired digital image of the cell through a neural network algorithm to form classification information of the cell.
The cell analysis device according to any one of claims 20 to 34, wherein the cell analysis device further comprises

Information receiving and sending device, configured to receive counting control instructions;

The image processing device is further configured to: in response to the counting control instruction, count the number of cells in the selected area of the digital image of the blood sample, and record the number of target type cells in the selected area.
A cell analysis device, the cell analysis device includes:

Memory, configured to store executable instructions;

The processor is configured to execute the method for analyzing cells according to any one of claims 1 to 18 when executing executable instructions stored in the memory.
A computer-readable storage medium storing executable instructions configured to cause a processor to execute the executable instructions to implement the method for analyzing cells according to any one of claims 1 to 18.