Method, device, equipment and storage medium for three-classification of white blood cells

Abstract

The present application discloses a three-category processing method, device, equipment and storage medium for leukocyte classification, which includes: performing differential calculation on data points in a histogram to obtain at least one differential value; traversing at least one differential value, and counting the number of target differential values ​​that are the same as a preset differential value, thereby obtaining the number of troughs; determining the first classification point and the second classification point in the histogram according to the classification strategy corresponding to the number of troughs, and dividing the leukocytes into three categories through the first classification point and the second classification point. That is, performing differential calculation on data points in a histogram drawn by leukocyte pulse values, and determining all troughs in the histogram based on the preset differential value. And counting the number of troughs, according to different classification strategies corresponding to different numbers of troughs, accurate classification, while improving the classification accuracy, improves the efficiency of leukocyte classification.

Description

Technical Field

[0001] The present application relates to the technical field of blood cell analysis, and in particular to a method, device, equipment and storage medium for three-category white blood cell classification. Background Art

[0002] A blood cell analyzer is an instrument that can detect cells in the blood. It can count and classify white blood cells, red blood cells, platelets, etc. Among them, white blood cells in the blood are very important for the human immune function, and their classification provides important data support for clinical screening. Therefore, studying the white blood cell classification method and accurately classifying white blood cells is of great significance to improve the efficiency of disease detection and screening.

[0003] White blood cells are usually divided into three categories, namely lymphocytes, intermediate cells and neutrophils, and are usually detected using an artificial mirror or microscope. This method requires professionals to spend a lot of time on the detection and has a low detection efficiency. Summary of the invention

[0004] The main purpose of the present application is to provide a method, device, equipment and storage medium for three-category white blood cell classification, aiming to solve the technical problem of low efficiency of existing white blood cell classification.

[0005] To achieve the above objectives, the present application provides a method for three-classification of leukocytes, the method comprising:

[0006] Performing differential calculation on data points in a histogram to obtain at least one differential value, wherein the histogram is drawn based on the conversion of the white blood cell pulse value;

[0007] Traversing the at least one differential value, counting the number of target differential values ​​that are the same as a preset differential value, and obtaining the number of troughs, wherein the preset differential value represents the differential value of the data point corresponding to the trough position in the histogram;

[0008] According to the classification strategy corresponding to the number of troughs, a first classification point and a second classification point in the histogram are determined, and the white blood cells are divided into three categories by the first classification point and the second classification point.

[0009] Exemplarily, performing differential calculation on the data points in the histogram to obtain at least one differential value includes:

[0010] Calculating the first-order difference of the data points in the histogram to obtain a first-order difference value;

[0011] Based on the first-order difference value, the second-order difference of the data point in the histogram is calculated to obtain the at least one trough difference value.

[0012] Exemplarily, traversing the at least one differential value, counting the number of target differential values ​​that are the same as the preset differential value, and obtaining the number of troughs includes:

[0013] Comparing the actual at least one differential value with the preset differential value;

[0014] Determine a value of the at least one actual differential value that is the same as the preset differential value, and mark it as the target differential value;

[0015] The number of the target difference values ​​is counted to obtain the number of troughs.

[0016] Exemplarily, the determining the first classification point and the second classification point in the histogram according to the classification strategy corresponding to the number of troughs, and classifying the white blood cells into three categories by the first classification point and the second classification point, includes:

[0017] Get the preset target classification points;

[0018] comparing the number of troughs with the number of target classification points;

[0019] When the number of troughs is equal to the target number of classification points, the data points corresponding to the target differential value are determined to be the first classification point and the second classification point, and the white blood cells are divided into three categories by the first classification point and the second classification point.

[0020] Exemplarily, after comparing the number of troughs with the number of target classification points, the method further includes:

[0021] When the number of troughs is greater than the number of target classification points, traversing the data points in the histogram;

[0022] The two data points with the smallest ordinate values ​​in the histogram are determined to obtain the first classification point and the second classification point, and the white blood cells are divided into three categories according to the first classification point and the second classification point.

[0023] Exemplarily, after comparing the number of troughs with the number of target classification points, the method further includes:

[0024] When the number of troughs is less than the number of target classification points, determining the data point corresponding to the trough-target difference value as the first classification point;

[0025] Based on the first classification point, traverse the data points in the histogram in both forward and reverse directions, and use the traversed data points as the current point;

[0026] If the curve segment formed by at least one adjacent point before the current point is a downward trend curve segment, and the curve segment formed by at least one adjacent point after the current point is an upward trend curve segment, the current point is determined to be a second classification point, and the white blood cells are divided into three categories through the first classification point and the second classification point.

[0027] Exemplarily, before performing differential calculation on the data points in the histogram to obtain at least one differential value, the method includes:

[0028] The data points in the histogram are smoothed and / or normal distribution fitted.

[0029] Exemplarily, to achieve the above purpose, the present application also provides a leukocyte three-differentiation processing device, the leukocyte three-differentiation processing device comprising:

[0030] A differential calculation module, used for performing differential calculation on data points in a histogram to obtain at least one differential value, wherein the histogram is drawn based on the conversion of the white blood cell pulse value;

[0031] A statistical module, configured to traverse the at least one differential value, count the number of target differential values ​​that are the same as a preset differential value, and obtain the number of troughs, wherein the preset differential value represents a trough position in the histogram;

[0032] A classification module is used to determine a first classification point and a second classification point in the histogram according to a classification strategy corresponding to the number of troughs, and to classify the white blood cells into three categories through the first classification point and the second classification point.

[0033] Illustratively, to achieve the above-mentioned purpose, the present application also provides a white blood cell three-classification processing device, which includes a memory, a processor, and a white blood cell three-classification processing program stored in the memory and executable on the processor, and the white blood cell three-classification processing program implements the steps of the white blood cell three-classification processing method as described above when executed by the processor.

[0034] Illustratively, to achieve the above-mentioned purpose, the present application also provides a computer storage medium, on which a white blood cell three-classification processing program is stored, and when the white blood cell three-classification processing program is executed by a processor, the steps of the white blood cell three-classification processing method as described above are implemented.

[0035] Compared with the low accuracy of white blood cell classification in the prior art, the present application performs differential calculation on the data points in the histogram to obtain at least one differential value, and the histogram is drawn according to the conversion of the white blood cell pulse value; traverse the at least one differential value, count the number of target differential values ​​that are the same as the preset differential value, and obtain the number of troughs, and the preset differential value represents the differential value of the data points corresponding to the trough position in the histogram; according to the classification strategy corresponding to the number of troughs, determine the first classification point and the second classification point in the histogram, and divide the white blood cells into three categories through the first classification point and the second classification point; that is, perform differential calculation on the data points in the histogram drawn by the white blood cell pulse value, and determine all the troughs in the histogram based on the preset differential value. And count the number of troughs, and according to different classification strategies corresponding to different numbers of troughs, accurately classify, and improve the classification accuracy while improving the efficiency of white blood cell classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a schematic diagram of the process of the first embodiment of the white blood cell three-classification processing method of the present application;

[0037] Figure 2 This is a schematic diagram of the results of three-classification of white blood cells in the first embodiment of the three-classification processing method of white blood cells of the present application;

[0038] Figure 3 This is a functional module diagram of a preferred embodiment of the leukocyte three-classification processing device of the present application;

[0039] Figure 4 It is a structural diagram of the hardware operating environment involved in the embodiment of the present application.

[0040] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0041] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0042] The present application provides a method for three-classification of leukocytes, referring to Figure 1 , Figure 1 This is a schematic diagram of the process of the three-classification processing method of white blood cells in this application.

[0043] The present application also provides an embodiment of a three-classification processing method for white blood cells. It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that here. The three-classification processing method for white blood cells can be applied to a computer. For ease of description, the following omits the execution subject to describe the various steps of the three-classification processing method for white blood cells. The three-classification processing method for white blood cells includes:

[0044] Step S110, performing differential calculation on data points in the histogram to obtain at least one differential value, wherein the histogram is drawn based on the conversion of the white blood cell pulse value.

[0045] The shortcoming of the existing three-category white blood cell classification processing technology is that by taking the second-order derivative of the data points in the histogram to obtain the trough, multiple troughs will generally be obtained, and then the classification points are screened based on the set cell volume threshold to achieve three-category classification. If the threshold setting is unreasonable, the two trough points selected from the multiple troughs cannot become the correct classification points, resulting in low accuracy of white blood cell classification. In this application, the data points in the histogram drawn by the white blood cell pulse value are differentially calculated, and all the troughs in the histogram are determined based on the preset differential value. And the number of troughs is counted, and different classification strategies correspond to different numbers of troughs, and accurate classification is achieved to avoid relying on the set cell volume threshold for classification point determination, thereby improving the accuracy of white blood cell classification.

[0046] It should be noted that the blood cell instrument collects the measurement data of the white blood cell impedance channel and analyzes the data to determine the proportion of the three types of white blood cell classifications. Among them, through the principle of electrical impedance, white blood cells are regarded as a physical particle, which generates resistance or voltage when passing through the electric field, thereby generating a pulse wave. The number of pulse waves reflects the number of white blood cells. The pulse waves generated by white blood cells of different sizes are also different in size. Different white blood cells can be classified and counted according to the size of the pulse wave. Therefore, a histogram is drawn according to the white blood cell pulse value in the impedance channel, and the peaks and troughs are found through the ups and downs of the histogram for classification.

[0047] Specifically, the process of drawing a histogram is:

[0048] The pulse value of white blood cells in the impedance channel is obtained, and the pulse value is converted into the white blood cell volume data to obtain the white blood cell volume data, and the number of white blood cell volume data at the same value is counted. Thus, a histogram is drawn with the white blood cell volume on the horizontal axis and the cell number on the vertical axis to obtain the specific number corresponding to different white blood cell volumes.

[0049] The process of converting the pulse value to the white blood cell volume data is as follows: if the maximum cell pulse value m and the maximum white blood cell volume data n are obtained, and the real-time cell pulse value obtained from the impedance channel is s, then the white blood cell volume data v corresponding to the real-time cell pulse value is:

[0050] v = n / m*s,

[0051] Therefore, all leukocyte pulse values ​​are plotted as a histogram with the leukocyte volume on the horizontal axis and the cell number on the vertical axis.

[0052] White blood cells are distributed in different calculation channels according to the size of cell volume pulses and divided into different subgroups. The first group (small cell area) is mainly composed of mature lymphocytes, including nucleated red blood cells and giant platelets under abnormal conditions; the second group (intermediate cell area) is mainly composed of monocytes, including eosinophils and basophils, and may also include primitive cells or immature cells under abnormal conditions; the third group (large cell area) is mainly composed of neutrophils, including band-shaped nuclear granulocytes and metamyelocytes. That is, white blood cells are divided into three categories: lymphocytes, monocytes, and neutrophils. The following description takes lymphocytes, monocytes, and neutrophils as examples.

[0053] Since the white blood cells in normal peripheral blood are mainly neutrophils, lymphocytes and monocytes, the white blood cell histogram is displayed as a smooth curve with three peaks, and there is a valley (i.e., classification point) at the junction of the distribution areas of these three groups of cells. When the white blood cell classification ratio is abnormal or abnormal cells appear, the height, number and valley characteristics of the white blood cell histogram curve peak will change. Therefore, accurate classification of white blood cells has an important data basis for determining whether there are abnormal cells.

[0054] In this embodiment, the data points in the histogram are differentially calculated. The differential calculation refers to the difference between two consecutive adjacent data in a discrete function, which is divided into first-order differential and higher-order differential. After differential calculation is performed on all data points in the histogram, at least one differential value is obtained, and the differential value is used to determine the data points corresponding to the trough positions in the histogram. The data points at these trough positions are likely to be classification points in the three classifications of white blood cells.

[0055] Step S120, traversing the at least one differential value, counting the number of target differential values ​​that are the same as a preset differential value, and obtaining the number of troughs, wherein the preset differential value represents the differential value of the data point corresponding to the trough position in the histogram.

[0056] In at least one differential value obtained by differential calculation of all data points in the histogram, the target differential value refers to a value that is consistent with the preset differential value. The preset differential value represents the differential value of the data point corresponding to the trough position in the histogram, that is, when the target differential value is consistent with the preset differential value, it means that the data point corresponding to the target differential value is at the trough position of the histogram.

[0057] As an example, the preset differential value is 2. Since first-order differential and second-order differential processing are performed when differential calculation is performed on data points in the histogram, the positions corresponding to the data points with a first-order differential value of 0 are the peaks and troughs of the curve in the histogram, and the positions corresponding to the data points with a second-order differential result (i.e., differential value) of 2 are the troughs of the curve in the histogram. Therefore, the preset differential value is determined to be 2.

[0058] Exemplarily, traversing the at least one differential value, counting the number of target differential values ​​that are the same as the preset differential value, and obtaining the number of troughs includes:

[0059] Step A1, comparing the actual at least one differential value with the preset differential value;

[0060] Step A2, determining a value of the at least one actual differential value that is the same as the preset differential value, and marking it as the target differential value;

[0061] Step A3, counting the number of the target difference values ​​to obtain the number of troughs.

[0062] Compare at least one differential value with the preset differential value, and select the target differential value that is the same as the preset differential value. That is, when the preset differential value is 2, traverse at least one differential value, use the differential value obtained by the traversal as the current value, and compare the current value with 2. If the current value is equal to 2, mark the current value as the target differential value and count 1. Continue to traverse the next differential value, use the next differential value as the current value, compare the current value with 2, and continue to determine whether the current value is consistent with 2. If they are consistent, mark the current value as the target differential value and add 1 to the previous count (1). If they are inconsistent, traverse the next differential value until at least one differential value is compared with 2, and the traversal is completed. And count the number of target differential values ​​equal to 2 in at least one differential value to obtain the number of troughs.

[0063] Step S130, determining a first classification point and a second classification point in the histogram according to a classification strategy corresponding to the number of troughs, and dividing the white blood cells into three categories by using the first classification point and the second classification point.

[0064] In order to improve the accuracy of the three-classification of white blood cells, the corresponding classification strategy is selected for different numbers of troughs to determine the two classification points in the histogram, and three classifications of white blood cells are obtained, namely lymphocytes, monocytes, and neutrophils. Among them, the two classification points are the first classification point and the second classification point.

[0065] Exemplarily, the determining the first classification point and the second classification point in the histogram according to the classification strategy corresponding to the number of troughs, and classifying the white blood cells into three categories by the first classification point and the second classification point, includes:

[0066] Step B1, obtaining a preset target classification point;

[0067] Step B2, comparing the number of troughs with the number of target classification points;

[0068] Step B3, when the number of troughs is equal to the target classification point number, determine that the data points corresponding to the target differential value are the first classification point and the second classification point, and divide the white blood cells into three categories through the first classification point and the second classification point.

[0069] Get the preset target classification points. The target classification points refer to the classification points used for white blood cell classification, that is, the number of dividing points between different types of white blood cells. Usually, white blood cells are divided into three categories: lymphocytes, monocytes, and neutrophils. Therefore, the target classification points are 2. The curve in the histogram can be divided into three parts through two classification points, one for each type of white blood cell. Figure 2 .

[0070] Compare the number of troughs with the target classification points, and determine the corresponding classification strategy based on the classification results.

[0071] Specifically, if the target classification point number is 2, according to the classification strategy corresponding to the number of troughs, the process of three-classification of white blood cells is:

[0072] Classification strategy one, when the number of troughs is equal to 2, there are 2 target differential values, therefore, the data point corresponding to the target differential value is the classification point. It can be understood that the target differential value is one of at least one differential value, and the differential value is obtained by differential calculation of the data points in the histogram. According to the target differential value, the corresponding data points in the histogram can be determined, that is, the first classification point and the second classification point. At the same time, the horizontal coordinate value of the data point in the histogram (i.e., the white blood cell volume data) can be obtained. The division of the horizontal coordinate value divides the white blood cells into three categories, and combined with the corresponding numbers of different types of white blood cells, the proportions of the three types of white blood cells can be determined.

[0073] Classification strategy two, when the number of troughs is greater than the target number of classification points, traverse the data points in the histogram; determine the two data points with the smallest ordinate values ​​in the histogram, obtain the first classification point and the second classification point, and divide the white blood cells into three categories through the first classification point and the second classification point.

[0074] That is, when the number of troughs is greater than 2, the ordinate values ​​(cell volume data) of the data points corresponding to all target differential values ​​in the histogram are traversed to obtain the two smallest ordinate values. The data points corresponding to the two ordinate values ​​are the classification points, namely the first classification point and the second classification point.

[0075] Therefore, in this embodiment, relatively accurate classification of white blood cells can be obtained by traversing the data in the graph twice, and the time complexity is small, which greatly improves the speed and efficiency of the white blood cell classification process.

[0076] It should be noted that when the number of troughs is much greater than 2, that is, when there are multiple troughs in the histogram calculated by difference, the ordinate successive values ​​of the data points corresponding to the target differential values ​​within a certain range can be selected for traversal to improve the efficiency of the traversal. Among them, the certain range can be determined according to the cell classification. Specifically, white blood cells are divided into three categories: lymphocytes, monocytes, and neutrophils. Each type of white blood cell has a volume range under normal circumstances or in a normal state. Based on the volume range and the expansion degree, the expanded range, that is, the certain range, can be obtained. For example, the volume range of monocytes is (t1, t2). If it is necessary to expand or widen h or h%, the certain range is (t1-h, t2+h) or (t1+t1*h%, t2+t2*h%). It can be understood that the expansion degree can be obtained according to actual needs, expert experience or model training.

[0077] As an example, the historical expansion degree, historical classification points, and the number of historical troughs are combined into a training data set, and the conventional mathematical model is trained through the training data set to obtain an instantiated mathematical model. Therefore, when the number of troughs is known, the number of troughs and two classification points are input into the instantiated mathematical model, the expansion degree is calculated, and then a certain range is determined. The ordinate values ​​of the data points corresponding to the target differential values ​​within a certain range are traversed to obtain the two smallest ordinate values. The data points corresponding to the two ordinate values ​​are the classification points, that is, the first classification point and the second classification point.

[0078] Classification strategy three, when the number of troughs is less than the target classification points, determine the data point corresponding to the wave target differential value as the first classification point; based on the first classification point, traverse the data points in the histogram in both forward and reverse directions, and use the traversed data points as the current point; if the curve segment formed by at least three adjacent points before the current point is a downward trend curve segment, and the curve segment formed by at least three adjacent points after the current point is an upward trend curve segment, then determine that the current point is the second classification point, and divide the white blood cells into three categories through the first classification point and the second classification point.

[0079] That is, based on the differential calculation characteristics of the white blood cell data points in the histogram, when the number of troughs is less than 2, it can be determined that the number of troughs is 1. Therefore, the data point corresponding to the current unique target differential value is the first classification point, and based on the first classification point, the data points in the histogram are traversed in both forward and reverse directions to determine the second classification point.

[0080] Specifically, based on the first classification point, the data points on the right side of the first classification point are traversed in the forward direction, and the data points on the left side of the first classification point are traversed in the reverse direction to find the downward trend of the histogram. After a certain number of data points are found to be continuously decreasing, it is considered that there is a downward trend. After the downward trend, if a certain number of data points are continuously increasing, the initial point of the increase is considered to be a trough, and the trough point is the second classification point. That is, the traversed data point is used as the current point; if the curve segment formed by at least one adjacent point before the current point is a downward trend curve segment, and the curve segment formed by at least one adjacent point after the current point is an upward trend curve segment, then the current point is determined to be the second classification point, and the white blood cells are divided into three categories through the first classification point and the second classification point.

[0081] Therefore, in this embodiment, relatively accurate classification of white blood cells can be obtained by traversing the data in the graph twice, and the time complexity is small, which greatly improves the speed and efficiency of the white blood cell classification process.

[0082] It should be noted that the number of at least three adjacent points is determined according to the calculation model. Specifically, by testing different numbers of adjacent points, the corresponding trough data points are determined. Thereafter, the target data points are determined by reviewing with a standard machine or microscope for three-category classification of white blood cells. Based on the target data points, the accuracy of the trough data points corresponding to the current number of adjacent points is calculated. If the accuracy meets the expected value, the current number of adjacent points is the number of at least one adjacent point. It can be understood that if the number of at least one adjacent point is 5, the 5 data points continuously decrease, and the curve segment formed is a downward trend curve segment, which is determined to be a downward trend. The 5 data points continuously increase, and the curve segment formed is an upward trend curve segment, which is determined to be an upward trend. As a result, the determined trough point, that is, the second classification point is more accurate, which improves the accuracy of the three-category classification of white blood cells.

[0083] As an example, if the number of at least three adjacent points is 5, the 5 adjacent points may decrease continuously or have a certain decrease ratio. It can be understood that if the decrease ratio is 80%, it means that among the 5 adjacent data points, 4 data points have a downward trend, and it can be determined that the 5 adjacent data points have a downward trend. It should be noted that the 4 data points may be continuous or discontinuous.

[0084] After the first classification point and the second classification point are determined in the histogram, the white blood cell volume data of the histogram is divided into three categories according to the abscissa of the first classification point and the abscissa of the second classification point, and the number of white blood cells corresponding to each white blood cell volume data in each category is counted, and the number of white blood cells in each category is summed to obtain the total number of cells in each category. Thus, the total number of cells in each category is summed to obtain the total number, thereby determining the proportion of the total number of cells in each category in the total number. Therefore, the proportions of the three types of cells, lymphocytes, monocytes, and neutrophils, in the white blood cells are determined.

[0085] Compared with the low accuracy of white blood cell classification in the prior art, the present application performs differential calculation on the data points in the histogram to obtain at least one differential value, and the histogram is drawn according to the conversion of the white blood cell pulse value; traverse the at least one differential value, count the number of target differential values ​​that are the same as the preset differential value, and obtain the number of troughs, and the preset differential value represents the differential value of the data points corresponding to the trough position in the histogram; according to the classification strategy corresponding to the number of troughs, determine the first classification point and the second classification point in the histogram, and divide the white blood cells into three categories through the first classification point and the second classification point. That is, the data points in the histogram drawn by the white blood cell pulse value are differentially calculated, and all the troughs in the histogram are determined based on the preset differential value. And the number of troughs is counted, and different classification strategies are corresponding to different numbers of troughs, and accurate classification is performed, which improves the classification accuracy while improving the efficiency of white blood cell classification.

[0086] Based on the above first embodiment, a second embodiment of the three-classification processing method of white blood cells of the present application is proposed. In this embodiment, the differential calculation of the data points in the histogram to obtain at least one differential value includes:

[0087] Step S130, calculating the first-order difference of the data points in the histogram to obtain a first-order difference value;

[0088] Step S140: Calculate the second order difference of the data points in the histogram based on the first order difference value to obtain the at least one trough difference value.

[0089] In this embodiment, when the independent variable changes from x to x+1, the change Δy of the function y=y(x) is x =(y(x+1)-y(x)) / Δx, (x=0, 1, 2, ...) is called the first-order difference of the function y(x) at point x, denoted by Δy x =Δy x =(y(x+1)-y(x)) / Δx, (x=0, 1, 2,...).

[0090] The horizontal axis of the histogram is the white blood cell volume and the vertical axis is the number of cells. When the first-order difference is performed on the data points in the histogram, x is the white blood cell volume and y is the number of cells. Thus, the first-order difference value is calculated by the above first-order difference formula.

[0091] Based on the first-order difference value, calculate the second-order difference of the data point in the histogram. The second-order difference refers to the difference of the first-order difference when the independent variable x changes to x+1:

[0092] Δ(Δy)=Δy x+1 -Δy x =(y x+2 -y x+1 )-(y x+1 -y x )=y x+2 -2y x+1 +y x ,

[0093] It is called the second-order difference of the function y = y(x), denoted by Δ 2 y x ,Right now:

[0094] Δ 2 y x =(y x+2 -2y x+1 +y x ) / Δx,

[0095] Therefore, based on the first-order difference value, the first-order difference value and the second-order difference of the data point in the histogram are calculated to obtain at least one difference value.

[0096] It should be noted that when the first-order difference and second-order difference are performed on the data points in the histogram, the data points with a first-order difference value of 0 correspond to the peaks and troughs of the curve in the histogram, and the data points with a second-order difference result (i.e., a difference value) of 2 correspond to the troughs of the curve in the histogram. Among them, 2 is the preset difference value.

[0097] In this embodiment, by calculating the first-order difference of all data points in the histogram, the data points corresponding to all the peaks and troughs in the histogram are obtained, and by taking the second-order difference of the first-order difference of the data points in the histogram, all the trough points in the histogram are obtained. These trough points serve as the basis for determining the classification points.

[0098] That is, the first classification point and the second classification point are obtained by screening these trough points. Therefore, by performing differential calculation on the data points in the histogram, all the trough points in the histogram can be found out more quickly, thereby improving the efficiency of the three-classification processing of white blood cells.

[0099] Based on the above first embodiment or second embodiment, a third embodiment of the three-classification processing method of leukocytes of the present application is proposed. In this embodiment, before performing differential calculation on the data points in the histogram to obtain at least one differential value, the method includes:

[0100] Step S150, performing smoothing processing and / or normal distribution fitting on the data points in the histogram.

[0101] In addition, a conventional three-group classification method for white blood cells is used for detection by a blood cell analyzer. The classification method is to draw a histogram based on the impedance channel data, and to find the peaks and valleys of the histogram for classification. In practical applications, this method may easily lead to classification errors because the actual impedance data transformation histogram may not be smooth enough, and the three classifications are not obvious. In other words, the selection of the smoothness of the image in the histogram will affect the effect of white blood cell classification. Therefore, before performing differential calculation on the data points in the histogram to obtain at least one differential value, the data points in the histogram are smoothed and / or normally distributed, and the graphic burrs of the histogram are filtered to obtain a smooth graphic. Thereby improving the accuracy and efficiency of the white blood cell classification process.

[0102] As an example, the process of smoothing the data points in the histogram is:

[0103] Determine that the data point to be smoothed is the first data point, the data point adjacent to the first data point in the reverse direction is the second data point, and the data point adjacent to the first data point in the forward direction is the third data point. Obtain a preset smoothness, and based on the smoothness, adjust the second data point and the third data point to obtain a new second data point and a new third data point. Average the new second data point and the new third data point to obtain a new first data point, and the new first data point is the adjusted data of the first data point. Adjust the first data point in the histogram to the new first data point to complete the smoothing of the first data point. The other data points in the histogram are basically the same as the above processing process, and new other data points after other data points are processed based on the smoothness are obtained, and the graph in the histogram is drawn based on the new other data points to complete the smoothing process. It should be noted that the data points to be smoothed are all original data in the histogram. That is, when the third data point is smoothed again, the data point adjacent to the third data point in the positive direction is the fourth data point, and the data point adjacent to the third data point in the reverse direction is the first data point. The smoothing is performed based on the fourth data point, the first data point and the smoothness, rather than based on the new first data point, thereby ensuring the accuracy of the data.

[0104] That is, smoothness is a constant that can be adjusted according to the classification effect. For example, if the smoothness is 3, three adjacent points are taken as smoothed data. The data of any point in the histogram replaces the data point with the average of the sum of the three data points on the left and the three data points on the right. According to the classification effect obtained during the test, a generally suitable smoothness value can be obtained.

[0105] Among them, smoothness is a constant (smoothness is the number of selected adjacent points). According to historical tests, the smoothness is the same in the entire histogram.

[0106] As an example, the process of fitting a normal distribution to the data points in a histogram is:

[0107] Input the horizontal and vertical coordinate values ​​of the data point in the histogram into the normal distribution function to determine whether it conforms to the normal distribution. If so, the data point is not adjusted; if not, the horizontal and vertical coordinate values ​​of the data point are adjusted based on the normal distribution function to conform to the normal distribution, and the corresponding vertical coordinate value is obtained. Based on the normal adjusted vertical coordinate value of the data point, the data point in the histogram is adjusted. As a result, the other data points in the histogram are basically the same as the above processing process, and the same normal distribution fitting process is performed on other data points, and other data points in the histogram are adjusted to obtain a histogram with smooth graphics.

[0108] In this embodiment, the data points in the histogram are smoothed and / or normal distribution fitted to filter out the burrs in the histogram to obtain a smooth graph, thereby improving the accuracy and efficiency of the leukocyte classification process.

[0109] For example, Figure 3 As shown, the present application also provides a three-classification of white blood cells, and the three-classification of white blood cells includes:

[0110] A differential calculation module 10 is used to perform differential calculation on data points in a histogram to obtain at least one differential value, wherein the histogram is drawn based on the conversion of the white blood cell pulse value;

[0111] A statistical module 20, configured to traverse the at least one differential value, count the number of target differential values ​​that are the same as a preset differential value, and obtain the number of troughs, wherein the preset differential value represents the differential value of the data point corresponding to the trough position in the histogram;

[0112] The classification module 30 is used to determine the first classification point and the second classification point in the histogram according to the classification strategy corresponding to the number of troughs, and classify the white blood cells into three categories through the first classification point and the second classification point.

[0113] And / or, the difference calculation module includes:

[0114] A first calculation submodule, used to calculate the first-order difference of the data points in the histogram to obtain a first-order difference value;

[0115] The second calculation submodule is used to calculate the second order difference of the data points in the histogram based on the first order difference value to obtain the at least one trough difference value.

[0116] And / or, the statistical module includes:

[0117] A first comparison submodule, used for comparing the actual at least one difference value with the preset difference value;

[0118] a marking submodule, configured to determine a value in the at least one actual differential value that is the same as the preset differential value, and mark it as the target differential value;

[0119] The statistical submodule is used to count the number of the target difference values ​​to obtain the number of troughs.

[0120] Wherein, the classification module includes:

[0121] The acquisition submodule is used to obtain the preset target classification points;

[0122] A second comparison submodule, used for comparing the number of troughs with the number of target classification points;

[0123] The first classification submodule is used to determine that the data points corresponding to the target differential value are the first classification points and the second classification points when the number of troughs is equal to the target classification points, and to classify the white blood cells into three categories through the first classification points and the second classification points.

[0124] And / or, the classification module further includes:

[0125] A first traversal submodule, used for traversing the data points in the histogram when the number of troughs is greater than the number of target classification points;

[0126] The second classification submodule is used to determine the two data points with the smallest ordinate values ​​in the histogram, obtain the first classification point and the second classification point, and classify the white blood cells into three categories based on the first classification point and the second classification point.

[0127] Wherein, the acquisition module also includes:

[0128] A determination submodule, configured to determine, when the number of troughs is less than the number of target classification points, the data point corresponding to the trough-target difference value as the first classification point;

[0129] A second traversal submodule is used to traverse the data points in the histogram in both forward and reverse directions based on the first classification point, and use the traversed data point as the current point;

[0130] The third classification submodule is used to determine that the current point is a second classification point if the curve segment formed by at least one adjacent point before the current point is a downward trend curve segment, and the curve segment formed by at least one adjacent point after the current point is an upward trend curve segment, and the white blood cells are divided into three categories through the first classification point and the second classification point.

[0131] Wherein, the device further comprises:

[0132] A fitting module is used to perform smoothing and / or normal distribution fitting on the data points in the histogram.

[0133] The specific implementation method of the three-classification of white blood cells in the present application is basically the same as the various embodiments of the three-classification processing method of white blood cells described above, and will not be repeated here.

[0134] In addition, the present application also provides a three-classification processing device for white blood cells. Figure 4 As shown, Figure 4 It is a structural diagram of the hardware operating environment involved in the embodiment of the present application.

[0135] In one possible implementation, Figure 4 This is a structural diagram of the hardware operating environment of the white blood cell three-classification processing equipment.

[0136] like Figure 4 As shown, the white blood cell three-classification processing device may include a processor 701, a communication interface 702, a memory 703 and a communication bus 704, wherein the processor 701, the communication interface 702 and the memory 703 communicate with each other via the communication bus 704, the memory 703 is used to store computer programs; the processor 701 is used to implement the steps of the white blood cell three-classification processing method when executing the program stored in the memory 703.

[0137] The communication bus 704 mentioned in the above-mentioned leukocyte three-classification processing device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus 704 can be divided into an address bus, a data bus, and a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0138] The communication interface 702 is used for communication between the above-mentioned leukocyte three-classification processing device and other devices.

[0139] The memory 703 may include a random access memory (Random Access Memory, RMD), or a non-volatile memory (Non-Volatile Memory, NM), such as at least one disk memory. Optionally, the memory 703 may also be at least one storage device located away from the processor 701.

[0140] The above-mentioned processor 701 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0141] The specific implementation of the leukocyte three-classification processing device of the present application is basically the same as the various embodiments of the above-mentioned leukocyte three-classification processing method, and will not be repeated here.

[0142] In addition, an embodiment of the present application further proposes a computer storage medium, on which a white blood cell three-classification processing program is stored. When the white blood cell three-classification processing program is executed by a processor, the steps of the white blood cell three-classification processing method described above are implemented.

[0143] The specific implementation of the computer storage medium of the present application is basically the same as the embodiments of the above-mentioned white blood cell three-classification processing method, and will not be repeated here.

[0144] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0145] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0146] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, device, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0147] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for three-classification of leukocytes, characterized in that: The method comprises: Performing differential calculation on data points in a histogram to obtain at least one differential value, wherein the histogram is drawn based on the conversion of the white blood cell pulse value; Traversing the at least one differential value, counting the number of target differential values ​​that are the same as a preset differential value, and obtaining the number of troughs, wherein the preset differential value represents the differential value of the data point corresponding to the trough position in the histogram; Determining a first classification point and a second classification point in the histogram according to a classification strategy corresponding to the number of troughs, and dividing the white blood cells into three categories by using the first classification point and the second classification point; The step of determining a first classification point and a second classification point in the histogram according to a classification strategy corresponding to the number of troughs, and classifying the white blood cells into three categories by using the first classification point and the second classification point, comprises: Get the preset target classification points; Comparing the number of troughs with the number of target classification points; When the number of troughs is equal to the target number of classification points, the data points corresponding to the target differential value are determined to be the first classification point and the second classification point, and the white blood cells are divided into three categories by the first classification point and the second classification point.

2. The method according to claim 1, characterized in that The step of performing differential calculation on the data points in the histogram to obtain at least one differential value includes: Calculating the first-order difference of the data points in the histogram to obtain a first-order difference value; Based on the first-order difference value, the second-order difference of the data point in the histogram is calculated to obtain the at least one trough difference value.

3. The method according to claim 1, characterized in that The traversing the at least one differential value, counting the number of target differential values ​​that are the same as the preset differential value, and obtaining the number of troughs includes: Comparing the actual at least one difference value with the preset difference value; Determine a value of the at least one actual differential value that is the same as the preset differential value, and mark it as the target differential value; The number of the target difference values ​​is counted to obtain the number of troughs.

4. The method according to claim 1, characterized in that After comparing the number of troughs with the number of target classification points, the method further comprises: When the number of troughs is greater than the number of target classification points, traversing the data points in the histogram; The two data points with the smallest ordinate values ​​in the histogram are determined to obtain the first classification point and the second classification point, and the white blood cells are divided into three categories according to the first classification point and the second classification point.

5. The method according to claim 1, characterized in that After comparing the number of troughs with the number of target classification points, the method further includes: When the number of troughs is less than the target number of classification points, determining the data point corresponding to the target difference value as the first classification point; Based on the first classification point, traverse the data points in the histogram in both forward and reverse directions, and use the traversed data points as the current point; If the curve segment formed by at least one adjacent point before the current point is a downward trend curve segment, and the curve segment formed by at least one adjacent point after the current point is an upward trend curve segment, the current point is determined to be a second classification point, and the white blood cells are divided into three categories through the first classification point and the second classification point.

6. The method according to claim 1, characterized in that Before performing differential calculation on the data points in the histogram to obtain at least one differential value, the method includes: The data points in the histogram are smoothed and / or normal distribution fitted.

7. A three-classification processing device for white blood cells, characterized in that: The device comprises: A differential calculation module, used for performing differential calculation on data points in a histogram to obtain at least one differential value, wherein the histogram is drawn based on the conversion of the white blood cell pulse value; A statistical module, configured to traverse the at least one differential value, count the number of target differential values ​​that are the same as a preset differential value, and obtain the number of troughs, wherein the preset differential value represents a trough position in the histogram; A classification module, used to determine a first classification point and a second classification point in the histogram according to a classification strategy corresponding to the number of troughs, and to classify the white blood cells into three categories by using the first classification point and the second classification point; The classification module is used to implement: Get the preset target classification points; Comparing the number of troughs with the number of target classification points; When the number of troughs is equal to the target number of classification points, the data points corresponding to the target differential value are determined to be the first classification point and the second classification point, and the white blood cells are divided into three categories by the first classification point and the second classification point.

8. A three-classification processing device for white blood cells, characterized in that: The white blood cell three-classification processing device includes a memory, a processor, and a white blood cell three-classification processing program stored in the memory and executable on the processor. When the white blood cell three-classification processing program is executed by the processor, the steps of the white blood cell three-classification processing method according to any one of claims 1 to 6 are implemented.

9. A computer storage medium, characterized in that: The computer storage medium stores a white blood cell three-classification processing program, which, when executed by a processor, implements the steps of the white blood cell three-classification processing method according to any one of claims 1 to 6.

Images