A cell localization and tracking method and related device

The blood cell localization and tracking method combining the YOLO model and the k-means algorithm solves the problem of blood cell localization offset under a microscope, and achieves accurate blood cell tracking and analysis.

CN116630968BActive Publication Date: 2026-11-17AVE SCI & TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202310589846.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-05-24
Publication Date
2026-11-17
Estimated Expiration
2043-05-24

AI Technical Summary

Technical Problem

When examining blood under a microscope, the displacement of blood cells leads to a decrease in analytical accuracy, and current technologies make it difficult to accurately track blood cells.

Method used

The YOLO model is used to locate blood cells in blood images, and morphological analysis and k-means algorithm are combined for clustering to predict the location of subsequent blood cells.

Benefits of technology

It enables precise location and tracking of blood cells, improving the accuracy and robustness of analysis and adapting to various complex scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116630968B_ABST
    Figure CN116630968B_ABST
Patent Text Reader

Abstract

This application discloses a cell localization and tracking method, relating to the field of microscopic examination technology, comprising: locating blood cells in a blood image to obtain the positional information of the blood cells; performing morphological analysis on the blood cells to extract their morphological features; performing cluster analysis on the blood cells based on their positional information and morphological features; selecting target blood cells from the blood cells based on the cluster analysis results; and predicting the positions of subsequent blood cells based on the position of the target blood cells. This cell localization and tracking method can accurately locate and track blood cells. This application also discloses a cell localization and tracking device, equipment, and computer-readable storage medium, all of which have the above-mentioned technical effects.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of microscopic examination technology, and in particular to a cell localization and tracking method; it also relates to a cell localization and tracking device, apparatus, and computer-readable storage medium. Background Technology

[0002] Microscopic examination refers to the examination under a microscope, which involves sampling and preparing a slide for observation, analysis, and judgment under a microscope. Human excrement, secretions, and blood are all objects of microscopic examination. During the acquisition of positive targets (white blood cells) in blood images, the high magnification of the microscope and the stringent requirements for mechanical precision can lead to misalignment of the positive targets during acquisition, thus affecting the accuracy of subsequent analysis. Therefore, locating and tracking blood cells has become a key research focus in automated blood microscopy, and providing a solution for accurately locating and tracking blood cells has become a pressing technical problem for those skilled in the art. Summary of the Invention

[0003] The purpose of this application is to provide a method for locating and tracking blood cells, which can accurately locate and track blood cells. Another purpose of this application is to provide a blood cell location and tracking device, apparatus, and computer-readable storage medium, all of which have the above-mentioned technical effects.

[0004] To address the aforementioned technical problems, this application provides a method for locating and tracking blood cells, comprising:

[0005] Locate blood cells in a blood image to obtain the location information of the blood cells;

[0006] The blood cells were subjected to morphological analysis to extract their morphological characteristics;

[0007] Cluster analysis was performed on the blood cells based on their location information and morphological characteristics.

[0008] Based on the cluster analysis results, target blood cells are selected from the blood cells.

[0009] Predict the location of subsequent blood cells based on the location of the target blood cells.

[0010] Optionally, locating blood cells in a blood image and obtaining the location information of the blood cells includes:

[0011] The blood image is divided into multiple blocks;

[0012] Locate the blood cells in each of the aforementioned blocks to obtain the location information of the blood cells in each of the aforementioned blocks.

[0013] Optionally, selecting target blood cells from the blood cells based on the cluster analysis results includes:

[0014] The blood cells with the highest percentage of the specified categories are selected as the target blood cells.

[0015] Optionally, predicting the location of subsequent blood cells based on the location of the target blood cell includes:

[0016] The location of the target blood cell is used as the predicted location of subsequent blood cells.

[0017] Optionally, locating blood cells in a blood image and obtaining the location information of the blood cells includes:

[0018] The location information of blood cells in a blood image is obtained by locating the blood cells using the YOLO model.

[0019] Optionally, the step of performing cluster analysis on the blood cells based on their location information and morphological characteristics includes:

[0020] Based on the location information and morphological features, the k-means algorithm is used to perform cluster analysis on the blood cells.

[0021] Optional, also includes:

[0022] Calculate the prediction success rate, and release a prediction success signal when the prediction success rate reaches a preset threshold.

[0023] To address the aforementioned technical problems, this application also provides a cell positioning and tracking device, comprising:

[0024] The positioning module is used to locate blood cells in a blood image and obtain the position information of the blood cells;

[0025] The extraction module is used to perform morphological analysis on the blood cells and extract the morphological features of the blood cells;

[0026] The analysis module is used to perform cluster analysis on the blood cells based on their location information and morphological characteristics.

[0027] The selection module is used to select target blood cells from the blood cells based on the cluster analysis results;

[0028] The prediction module is used to predict the location of subsequent blood cells based on the location of the target blood cell.

[0029] To address the aforementioned technical problems, this application also provides a cell positioning and tracking device, comprising:

[0030] Memory, used to store computer programs;

[0031] A processor for executing the computer program to implement the steps of the cell localization and tracking method as described above.

[0032] To address the aforementioned technical problems, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the cell localization and tracking method described above.

[0033] The blood cell localization and tracking method provided in this application includes: locating blood cells in a blood image to obtain the location information of the blood cells; performing morphological analysis on the blood cells to extract their morphological features; performing cluster analysis on the blood cells based on their location information and morphological features; selecting target blood cells from the blood cells based on the cluster analysis results; and predicting the location of subsequent blood cells based on the location of the target blood cells.

[0034] As can be seen, the blood cell localization and tracking method provided in this application first obtains the location information and morphological features of blood cells in a set of blood images through localization and morphological feature extraction. Then, based on the location information and morphological features of blood cells, a clustering algorithm is used to analyze the distribution pattern of blood cells and predict the location of subsequent blood cells, thereby achieving accurate localization and tracking of blood cells. In addition, the blood cell localization and tracking method provided in this application has strong robustness and can adapt to various complex scenarios.

[0035] The blood cell localization and tracking device, equipment, and computer-readable storage medium provided in this application all have the aforementioned technical effects. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the prior art and embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart illustrating a blood cell localization and tracking method provided in an embodiment of this application.

[0038] Figure 2 This is a schematic diagram of a blood cell positioning and tracking device provided in an embodiment of this application;

[0039] Figure 3 This is a schematic diagram of a blood cell localization and tracking device provided in an embodiment of this application. Detailed Implementation

[0040] The core of this application is to provide a method for locating and tracking blood cells, which can accurately locate and track blood cells. Another core aspect of this application is to provide a blood cell location and tracking device, equipment, and computer-readable storage medium, all of which have the aforementioned technical effects.

[0041] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a blood cell localization and tracking method provided in an embodiment of this application. (Refer to...) Figure 1 As shown, the method includes:

[0043] S101: Locate blood cells in a blood image and obtain the location information of the blood cells;

[0044] Deep learning models can be used to identify blood cells in blood images, locate their positions, and obtain their location information. Blood images include red blood cells, white blood cells, platelets, etc., but white blood cells are usually the focus of analysis during blood microscopy; therefore, the blood cells being located and tracked are white blood cells.

[0045] In some embodiments, locating blood cells in a blood image and obtaining the location information of the blood cells includes:

[0046] The location information of blood cells in a blood image is obtained by locating the blood cells using the YOLO model.

[0047] In this embodiment, the deep learning model specifically adopts the YOLO model. The YOLO model has advantages such as high speed and good ability to avoid misidentification of the background. Therefore, using the YOLO model to locate blood cells in a blood image and obtain the position information of the blood cells can achieve fast and accurate identification and localization.

[0048] It is clear that using the YOLO model to locate blood cells in a blood image and obtain the location information of the blood cells is one implementation method provided in this application, and not the only one. In addition to the YOLO model, other deep learning models can also be used to locate blood cells.

[0049] In addition, in some embodiments, locating blood cells in a blood image and obtaining the location information of the blood cells includes:

[0050] The blood image is divided into multiple blocks;

[0051] Locate the blood cells in each of the aforementioned blocks to obtain the location information of the blood cells in each of the aforementioned blocks.

[0052] Compared to other components in blood, white blood cells are present in relatively small quantities. To quickly collect white blood cells, large fields of view are typically used for sampling, resulting in high-resolution blood images. For example, blood images may have a resolution of 2046×2040. This high resolution can affect processing speed. Therefore, to improve the processing speed of subsequent operations, this embodiment first divides the blood image into blocks and then locates the blood cells within each block.

[0053] S102: Perform morphological analysis on the blood cells and extract the morphological characteristics of the blood cells;

[0054] Based on the location of blood cells, morphological algorithms can be used to perform morphological analysis on the located blood cells, extracting their morphological features. These features can include the size, shape, color, and outline of the blood cells. Morphological algorithms can include OpenCV algorithms, among others.

[0055] The purpose of performing morphological analysis on blood cells is to reduce the influence of background and impurities. The extracted morphological features of blood cells are used for cluster analysis.

[0056] In addition, mechanical parameters can be acquired. These parameters are used to convert image pixels to motor steps, enabling the tracking of blood cells based on these parameters. For example, a 1000-pixel shift to the left in a blood image corresponds to a 10-step movement of the motor to the left.

[0057] S103: Based on the location information and morphological characteristics of the blood cells, perform cluster analysis on the blood cells;

[0058] After extracting a certain amount of location information and morphological features of blood cells, a clustering algorithm is used to perform cluster analysis on the blood cells, and then the location of the next blood cell in the microscope field of view is predicted based on the cluster analysis results.

[0059] Blood images show a dense distribution of blood cells, with three or more white blood cells present simultaneously in a single microscopic field of view. Clustering algorithms are used to analyze and identify the true targets to be tracked, thereby reducing the risk of target loss.

[0060] In some embodiments, the clustering analysis of the blood cells based on their location information and morphological characteristics includes:

[0061] Based on the location information and morphological features, the k-means algorithm is used to perform cluster analysis on the blood cells.

[0062] In this embodiment, the k-means algorithm is specifically used for clustering. The k-means algorithm is simple to implement and provides accurate clustering results. Therefore, using the k-means algorithm to perform clustering analysis on blood cells can quickly and accurately obtain clustering analysis results.

[0063] It is also clear that using the k-means algorithm to perform cluster analysis on blood cells is one implementation method provided in this application, and not the only one. In addition to the k-means algorithm, other clustering algorithms can also be used for cluster analysis.

[0064] S104: Select target blood cells from the blood cells based on the cluster analysis results;

[0065] Based on the cluster analysis of the currently located blood cells, target blood cells are selected from the cluster analysis results.

[0066] In some embodiments, selecting target blood cells from the blood cells based on cluster analysis results includes:

[0067] The blood cells with the highest percentage of the specified categories are selected as the target blood cells.

[0068] Through cluster analysis, this embodiment selects the blood cells with the highest category proportion as target blood cells, and uses the position of the blood cells with the highest category proportion as the basis for predicting the position of subsequent blood cells, which can more accurately and reliably predict the position of subsequent blood cells.

[0069] S105: Predict the location of subsequent blood cells based on the location of the target blood cells.

[0070] After selecting the target blood cell, the position of the next blood cell in the microscope field of view is predicted based on the position of the target blood cell, so as to track the next blood cell and acquire an image of the next blood cell at the predicted position.

[0071] In some embodiments, predicting the location of subsequent blood cells based on the location of the target blood cell includes:

[0072] The location of the target blood cell is used as the predicted location of subsequent blood cells.

[0073] This embodiment uses the location of the target blood cell as the predicted location of the next blood cell to quickly obtain the predicted location of the blood cell. Alternatively, the location of the target blood cell can be expanded to a certain size area, and this area can then be used as the predicted location of the next blood cell.

[0074] After predicting the location of blood cells, the predicted location information can be used to relocate new blood cells, and then a new round of analysis and prediction can be performed to obtain new predicted locations. This cycle continues until the blood cell collection process is completed.

[0075] To ensure the accuracy of the prediction, some embodiments also include:

[0076] Calculate the prediction success rate, and release a prediction success signal when the prediction success rate reaches a preset threshold.

[0077] The predicted location information may contain errors. Therefore, this embodiment summarizes and aggregates the prediction results, and uses a prediction success rate reaching a preset threshold as the condition for initiating a new round of prediction. If the prediction success rate reaches the preset threshold, a signal indicating successful positioning is released to the instrument, and a new round of analysis is initiated until all blood cell images are acquired.

[0078] The blood cell localization and tracking method provided in this application can be implemented based on multiple blood images showing consecutively appearing white blood cells. The number of blood images can be determined based on the analysis speed and accuracy. The more blood images analyzed, the more accurate the tracking and localization of subsequent blood cells. For example, after obtaining five blood images showing consecutively appearing white blood cells, the blood cell localization and tracking method provided in this application is initiated. For each blood image, the blood cells are located and their morphological features are extracted. Based on the location information and morphological features obtained, a clustering analysis algorithm is used to perform cluster analysis, and based on the cluster analysis results, the predicted location of subsequent blood cells is obtained.

[0079] In summary, the blood cell localization and tracking method provided in this application includes: locating blood cells in a blood image to obtain the location information of the blood cells; performing morphological analysis on the blood cells to extract their morphological features; performing cluster analysis on the blood cells based on their location information and morphological features; selecting target blood cells from the blood cells based on the cluster analysis results; and predicting the location of subsequent blood cells based on the location of the target blood cells. This blood cell localization and tracking method first obtains a set of location information and morphological features of blood cells in a blood image through localization and morphological feature extraction. Then, based on the location information and morphological features of the blood cells, a clustering algorithm is used to analyze the distribution pattern of the blood cells and predict the location of subsequent blood cells, thereby achieving accurate localization and tracking of blood cells. Furthermore, the blood cell localization and tracking method provided in this application is robust and can adapt to various complex scenarios.

[0080] This application also provides a blood cell localization and tracking device, which is described below and can be referred to in conjunction with the method described above. Please refer to... Figure 2 , Figure 2 This is a schematic diagram of a blood cell localization and tracking device provided in an embodiment of this application, combined with... Figure 2 As shown, the device includes:

[0081] The positioning module 10 is used to locate blood cells in a blood image and obtain the position information of the blood cells;

[0082] Extraction module 20 is used to perform morphological analysis on the blood cells and extract the morphological features of the blood cells;

[0083] The analysis module 30 is used to perform cluster analysis on the blood cells based on their location information and morphological characteristics.

[0084] The selection module 40 is used to select target blood cells from the blood cells based on the cluster analysis results;

[0085] The prediction module 50 is used to predict the location of subsequent blood cells based on the location of the target blood cell.

[0086] Based on the above embodiments, as a specific implementation method, the positioning module 10 includes:

[0087] A segmentation unit is used to divide the blood image into multiple blocks;

[0088] The positioning unit is used to locate the blood cells in each of the blocks and obtain the position information of the blood cells in each of the blocks.

[0089] Based on the above embodiments, as a specific implementation method, module 40 is specifically used for:

[0090] The blood cells with the highest percentage of the specified categories are selected as the target blood cells.

[0091] Based on the above embodiments, as a specific implementation method, the prediction module 50 is specifically used for:

[0092] The location of the target blood cell is used as the predicted location of subsequent blood cells.

[0093] Based on the above embodiments, as a specific implementation method, the positioning module 10 is specifically used for:

[0094] The location information of blood cells in a blood image is obtained by locating the blood cells using the YOLO model.

[0095] Based on the above embodiments, as a specific implementation method, the analysis module 30 is specifically used for:

[0096] Based on the location information and morphological features, the k-means algorithm is used to perform cluster analysis on the blood cells.

[0097] Based on the above embodiments, as a specific implementation method, it further includes:

[0098] The calculation module is used to calculate the prediction success rate and release a prediction success signal when the prediction success rate reaches a preset threshold.

[0099] The blood cell localization and tracking device provided in this application first obtains the positional information and morphological features of blood cells in a set of blood images through localization and morphological feature extraction. Then, based on the positional information and morphological features of the blood cells, a clustering algorithm is used to analyze the distribution pattern of blood cells and predict the position of subsequent blood cells, thereby achieving precise localization and tracking of blood cells. In addition, the blood cell localization and tracking device provided in this application can adapt to various complex scenarios.

[0100] This application also provides a blood cell localization and tracking device, referenced... Figure 3 As shown, the device includes a memory 1 and a processor 2.

[0101] Memory 1 is used to store computer programs;

[0102] Processor 2 is used to execute computer programs to perform the following steps:

[0103] Locate blood cells in a blood image to obtain their positional information; perform morphological analysis on the blood cells to extract their morphological features; perform cluster analysis on the blood cells based on their positional information and morphological features; select target blood cells from the blood cells based on the cluster analysis results; and predict the positions of subsequent blood cells based on the positions of the target blood cells.

[0104] The blood cell localization and tracking device provided in this application first obtains the location information and morphological features of blood cells in a set of blood images through localization and morphological feature extraction. Then, based on the location information and morphological features of the blood cells, a clustering algorithm is used to analyze the distribution pattern of blood cells and predict the location of subsequent blood cells, thereby achieving precise localization and tracking of blood cells. In addition, the blood cell localization and tracking device provided in this application can adapt to various complex scenarios.

[0105] For a description of the equipment provided in this application, please refer to the above method embodiments; further details will not be provided here.

[0106] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the following steps:

[0107] Locate blood cells in a blood image to obtain their positional information; perform morphological analysis on the blood cells to extract their morphological features; perform cluster analysis on the blood cells based on their positional information and morphological features; select target blood cells from the blood cells based on the cluster analysis results; and predict the positions of subsequent blood cells based on the positions of the target blood cells.

[0108] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0109] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments; further details will not be repeated here.

[0110] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatuses, devices, and computer-readable storage media disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant details can be found in the method section.

[0111] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0112] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0113] The above provides a detailed description of the blood cell localization and tracking method, apparatus, device, and computer-readable storage medium provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A method of cell localization tracking, the method comprising: include: Locate blood cells in a blood image to obtain the location information of the blood cells; The blood cells mentioned are white blood cells; The blood cells were subjected to morphological analysis to extract their morphological characteristics; Cluster analysis was performed on the blood cells based on their location information and morphological characteristics. Based on the cluster analysis results, target blood cells are selected from the blood cells. Predict the location of subsequent blood cells based on the location of the target blood cells; The step of selecting target blood cells from the blood cells based on the cluster analysis results includes: The blood cells with the highest percentage of a particular category are selected as the target blood cells. The step of predicting the location of subsequent blood cells based on the location of the target blood cell includes: The location of the target blood cell is used as the predicted location of subsequent blood cells; The process of locating blood cells in a blood image and obtaining the location information of the blood cells includes: The blood image is divided into multiple blocks; Locate the blood cells in each of the aforementioned blocks to obtain the location information of the blood cells in each of the aforementioned blocks.

2. The cell localization tracking method according to claim 1, wherein, The process of locating blood cells in a blood image and obtaining the location information of the blood cells includes: The location information of blood cells in a blood image is obtained by locating the blood cells using the YOLO model.

3. The cell localization tracking method of claim 1, wherein, The step of performing cluster analysis on the blood cells based on their location information and morphological characteristics includes: Based on the location information and morphological features, the k-means algorithm is used to perform cluster analysis on the blood cells.

4. The cell localization tracking method of claim 1, wherein, Also includes: Calculate the prediction success rate, and release a prediction success signal when the prediction success rate reaches a preset threshold.

5. A cell localization tracking apparatus, characterized by, include: The positioning module is used to locate blood cells in a blood image and obtain the position information of the blood cells; The blood cells mentioned are white blood cells; The extraction module is used to perform morphological analysis on the blood cells and extract the morphological features of the blood cells; The analysis module is used to perform cluster analysis on the blood cells based on their location information and morphological characteristics. The selection module is used to select target blood cells from the blood cells based on the cluster analysis results; The prediction module is used to predict the location of subsequent blood cells based on the location of the target blood cell; The selection module is specifically used to: select the blood cells with the highest percentage of each category from the blood cells as the target blood cells; The prediction module is specifically used to: use the location of the target blood cell as the predicted location of subsequent blood cells; The positioning module includes: A segmentation unit is used to divide the blood image into multiple blocks; The positioning unit is used to locate the blood cells in each of the blocks and obtain the position information of the blood cells in each of the blocks.

6. A cell positioning tracking apparatus, characterized by, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the cell localization and tracking method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the cell localization and tracking method as described in any one of claims 1 to 4.

Citation Information

Patent Citations

  • System and method for tracking the movement of biological materials

    CN101611411A