Sample analyzer and scatter diagram detection method and device

Abstract

The invention relates to a sample analyzer and a scatter diagram detection method and device. The method comprises the following steps: acquiring a first scatter diagram of a target sample; the first scatter diagram is used for representing distribution data of blood cells contained in the target sample; carrying out segmentation processing on the first scatter diagram according to a particle segmentation algorithm to obtain quantity indexes of various blood cells; based on the abnormal quantity index, obtaining first early warning information of the target sample; performing classification processing on the first scatter diagram according to a classification model to obtain second early warning information output by the classification model; under the condition that the first early warning information is consistent with the second early warning information, outputting first target early warning information; the first target early warning information is the first early warning information and / or the second early warning information. In this way, the accuracy of the first target early warning information can be guaranteed, and then the accuracy and reliability of the diagnosis result determined by the medical staff according to the first target early warning information are guaranteed.

Description

Technical Field

[0001] This application relates to the field of in vitro diagnostic technology, and in particular to a sample analyzer, a scatter plot detection method and apparatus. Background Technology

[0002] By analyzing the content of different cells in the body fluids of a target individual and comparing it with the range of cell content in normal body fluids, medical personnel can be provided with a basis for disease diagnosis; this basis is an important indicator of a person's health status. Traditional techniques use data statistics and template matching to analyze scatter plots to obtain the content of various cell types and output alarm information for cells with abnormal content. However, due to interference from various factors, the determined content of various cell types may not be accurate enough, leading to inaccurate alarm information and relatively low reliability.

[0003] Therefore, ensuring the accuracy and reliability of test results is an urgent problem to be solved. Summary of the Invention

[0004] Therefore, it is necessary to provide a sample analyzer, scatter plot detection method and device with high accuracy and reliability to address the above-mentioned technical problems.

[0005] In a first aspect, this application provides a sample analyzer, comprising:

[0006] A sampling component for obtaining target samples from a sample container containing samples;

[0007] A preparation component is used to mix the target sample with reagents to obtain a sample solution;

[0008] A detection component is used to detect the sample solution prepared by the preparation component to obtain a first scatter plot characterizing the distribution data of blood cells contained in the target sample;

[0009] A scatter plot detection device includes a processor and a memory storing processor-executable instructions; wherein the processor is used to acquire a first scatter plot; segment the first scatter plot according to a particle segmentation algorithm to obtain quantity indicators of various types of blood cells; obtain a first warning information for the target sample based on the abnormal quantity indicators; and classify the first scatter plot according to a classification model to obtain a second warning information output by the classification model.

[0010] An interactive component is used to output a first target warning information when the first warning information and the second warning information are consistent; the first target warning information is the first warning information and / or the second warning information.

[0011] Secondly, this application provides a scatter plot detection method, the method comprising:

[0012] Obtain a first scatter plot of the target sample; the first scatter plot is used to characterize the distribution data of blood cells contained in the target sample;

[0013] The first scatter plot is segmented using a particle segmentation algorithm to obtain the quantity indicators of various types of blood cells;

[0014] Based on the abnormal quantity index, the first warning information for the target sample is obtained;

[0015] The first scatter plot is classified according to the classification model to obtain the second early warning information output by the classification model.

[0016] If the first warning information and the second warning information are consistent, a first target warning information is output; the first target warning information is the first warning information and / or the second warning information.

[0017] In one embodiment, the method further includes:

[0018] If at least one abnormal indicator in the first warning information is inconsistent with the abnormal indicator in the second warning information, the abnormal indicator that is inconsistent between the first warning information and the second warning information shall be determined as the target abnormal indicator.

[0019] The first scatter plot is matched with the second scatter plot of the abnormal samples in the sample knowledge base to obtain the matching result;

[0020] Based on the matching result, a second target warning information is output; the second target warning information includes a third warning information corresponding to the matched second scatter plot and / or the second warning information corresponding to the first scatter plot.

[0021] In one embodiment, the step of matching the first scatter plot with a second scatter plot of abnormal samples in the sample knowledge base to obtain a matching result includes:

[0022] If the sample knowledge base meets the first matching condition, the matching result is determined based on the similarity obtained by matching the first scatter plot with each of the second scatter plots; the first matching condition includes that the number of types of abnormal indicators contained in the sample knowledge base is greater than a first threshold and the number of abnormal samples corresponding to each abnormal indicator is greater than a second threshold.

[0023] The step of outputting a second target warning message based on the matching result includes:

[0024] If the matching result is successful, the third warning information corresponding to the second scatter plot with the highest similarity is determined as the second target warning information;

[0025] If the matching result is unsuccessful, the second warning information corresponding to the first scatter plot is determined to be the second target warning information.

[0026] In one embodiment, the step of matching the first scatter plot with a second scatter plot in the sample knowledge base to obtain a matching result includes:

[0027] If the sample knowledge base does not meet the first matching condition, or if the sample knowledge base meets the first matching condition but the number of the target abnormal indicators is less than the third threshold; the first matching condition includes that the number of types of abnormal indicators contained in the sample knowledge base is greater than the first threshold and the number of abnormal samples corresponding to each abnormal indicator is greater than the second threshold.

[0028] For each of the target anomaly indicators, the first scatter plot is matched with each second scatter plot containing the target anomaly indicator to determine the matching result.

[0029] The step of outputting a second target warning message based on the matching result includes:

[0030] For a successfully matched target anomaly indicator, the third alarm information corresponding to the target anomaly indicator of the second scatter plot with the highest similarity is taken as the first type of alarm information.

[0031] For target anomaly indicators that fail to match, the second warning information corresponding to the target anomaly indicator will be used as the second type of alarm information;

[0032] The output includes the second target warning information containing the first type of alarm information and the second type of alarm information.

[0033] In one embodiment, the step of matching the first scatter plot with a second scatter plot of abnormal samples in the sample knowledge base to obtain a matching result includes:

[0034] If the maximum similarity between the second scatter plot and the first scatter plot in the sample knowledge base is greater than the first preset threshold, then the matching is determined to be successful;

[0035] If the similarity between the first scatter plot and the second scatter plot of any abnormal sample used for matching in the sample knowledge base is less than the first preset threshold, then the matching is determined to be unsuccessful.

[0036] In one embodiment, the step of classifying the first scatter plot according to the classification model to obtain the second warning information output by the classification model includes:

[0037] The first scatter plot is processed using the classification model to obtain the confidence levels of various anomaly indicators;

[0038] The abnormal indicators with a confidence level greater than or equal to the second preset threshold are output as the second warning information of the target sample.

[0039] In one embodiment, the particle segmentation algorithm includes a three-dimensional point cloud particle segmentation algorithm; the first scatter plot is three-dimensional point cloud data; the step of segmenting the first scatter plot according to the particle segmentation algorithm to obtain the quantity indicators of various blood cells includes:

[0040] The three-dimensional point cloud data is segmented according to the three-dimensional point cloud particle segmentation algorithm to obtain the quantity index of various types of blood cells; the various types of blood cells include specific types of cells and non-specific types of cells; the specific types of cells include at least one of primitive cells, abnormal lymphocytes, atypical lymphocytes and plasma cells.

[0041] The method further includes:

[0042] For the specific type of cell, the first target warning information is output based on the first warning information corresponding to the specific type of cell;

[0043] For the non-specific cell type, the first target warning information is output based on the first warning information and the second warning information corresponding to the non-specific cell type, and / or, the second target warning information is output based on the second warning information and the third warning information corresponding to the non-specific cell type.

[0044] In one embodiment, outputting the second target warning information includes:

[0045] The second target warning information is displayed on the interactive interface, and an anomaly identifier is added to the second target warning information corresponding to the target anomaly indicator; the anomaly identifier is used to indicate that the second target warning information is suspicious.

[0046] Thirdly, this application provides a scatter plot detection device, the device comprising:

[0047] The acquisition module is used to acquire a first scatter plot of the target sample; the first scatter plot is used to characterize the distribution data of blood cells contained in the target sample;

[0048] The first processing module is used to segment the first scatter plot according to the particle segmentation algorithm to obtain the quantity indicators of various types of blood cells;

[0049] The first processing module is used to obtain a first warning message for the target sample based on the abnormal quantity index;

[0050] The second processing module is used to classify the first scatter plot according to the classification model to obtain the second warning information output by the classification model.

[0051] The analysis module is used to output a first target warning information when the first warning information and the second warning information are consistent; the first target warning information is the first warning information and / or the second warning information.

[0052] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the scatter plot detection method described in any embodiment of this application.

[0053] The aforementioned sample analyzer, scatter plot detection method, and device process the first scatter plot in different ways, match the obtained warning information, and determine that the accuracy and reliability of the matched first and / or second warning information are relatively high, and can be used as the first target warning information output. This ensures the accuracy of the first target warning information, thereby guaranteeing the accuracy and reliability of the diagnostic results determined by medical personnel based on the first target warning information. Attached Figure Description

[0054] Figure 1 This is an application environment diagram illustrating a scatter plot detection method according to an exemplary embodiment;

[0055] Figure 2 This is a flowchart illustrating a scatter plot detection method according to an exemplary embodiment;

[0056] Figure 3 This is a schematic diagram illustrating a classification model according to an exemplary embodiment;

[0057] Figure 4 This is a schematic diagram illustrating a scatter plot matching method according to an exemplary embodiment;

[0058] Figure 5 This is a schematic diagram illustrating a scatter plot matching method according to an exemplary embodiment;

[0059] Figure 6 This is a schematic diagram illustrating a three-dimensional point cloud particle segmentation algorithm according to an exemplary embodiment;

[0060] Figure 7 This is a schematic diagram illustrating the output of target warning information according to an exemplary embodiment;

[0061] Figure 8 This is a flowchart illustrating a scatter plot detection method according to an exemplary embodiment;

[0062] Figure 9 This is a flowchart illustrating a scatter plot detection method according to an exemplary embodiment;

[0063] Figure 10 This is a structural block diagram of a scatter plot detection device according to an exemplary embodiment;

[0064] Figure 11 This is an internal structural diagram of an electronic device according to an exemplary embodiment. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0066] The terms "first," "second," and "third" used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this application, "at least one" is used to indicate one or more; "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0067] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0068] The method described in this application embodiment can be applied to an electronic device; the electronic device can be any mobile terminal or fixed terminal with scatter plot detection function.

[0069] In some embodiments, this application provides a sample analyzer, including:

[0070] A sampling component for obtaining target samples from a sample container containing samples;

[0071] A preparation component is used to mix the target sample with reagents to obtain a sample solution;

[0072] A detection component is used to detect the sample solution prepared by the preparation component to obtain a first scatter plot characterizing the distribution data of blood cells contained in the target sample;

[0073] A scatter plot detection device includes a processor and a memory storing processor-executable instructions; wherein the processor is used to acquire a first scatter plot; segment the first scatter plot according to a particle segmentation algorithm to obtain quantity indicators of various types of blood cells; obtain a first warning information for the target sample based on the abnormal quantity indicators; and classify the first scatter plot according to a classification model to obtain a second warning information output by the classification model.

[0074] An interactive component is used to output a first target warning information when the first warning information and the second warning information are consistent; the first target warning information is the first warning information and / or the second warning information.

[0075] In this embodiment, the sample analyzer is applied in the field of in vitro diagnostics, such as a human blood analyzer, a veterinary blood analyzer, or a veterinary blood analysis point-of-care testing device; the function of the sample analyzer is to test the sample. In related technologies, the sample analyzer is equipped with interactive components, such as a touch screen, or a non-touch screen, keyboard, and mouse.

[0076] Optionally, the scatter plot detection method in this application embodiment can be applied to a software program in a sample analyzer or blood cell analyzer. For example, the software program can be, but is not limited to, at least one of various applications (APPs), quick apps, and mini-programs.

[0077] Optionally, such as Figure 1As shown, the method described in this embodiment can be applied to a third-party terminal 101. Data transmission between the third-party terminal and the sample analyzer 102 occurs via a communication network. For example, the third-party terminal 101 can be a device that provides voice and / or data connectivity to a user. For instance, the terminal can be an IoT terminal, such as a sensor device, a mobile phone (or "cellular" phone and computer with IoT terminal); or it can be a fixed, portable, pocket-sized, handheld, or computer-embedded device. Portable devices can be smartwatches, smart bracelets, head-mounted devices, etc.

[0078] In some embodiments, such as Figure 2 As shown, a scatter plot detection method is provided, including the following steps:

[0079] S201, Obtain a first scatter plot of the target sample; the first scatter plot is used to characterize the distribution data of blood cells contained in the target sample.

[0080] In this embodiment, the target sample can be a bodily fluid sample from any organism susceptible to disease. Optionally, the bodily fluid sample includes, but is not limited to, at least one of blood samples, urine samples, and cerebrospinal fluid samples. For example, the target sample can be at least one of blood samples from humans, mammals, birds, and amphibians.

[0081] In this embodiment, the scatter plot can be image data characterizing the distribution features of various types of blood cells. The blood cells may include, but are not limited to, at least one of red blood cells, white blood cells, and platelets. White blood cells may include neutrophils, lymphocytes, monocytes, eosinophils, and basophils.

[0082] In this embodiment of the application, the first scatter plot may include, but is not limited to, at least one of the WDF (DIFF) channel scatter plot, WNR channel scatter plot, and WPC channel scatter plot.

[0083] S202, the first scatter plot is segmented according to the particle segmentation algorithm to obtain the quantity indicators of various types of blood cells.

[0084] In some embodiments, depending on the data type processed by the particle segmentation algorithm, the particle segmentation algorithm can be divided into two-dimensional or three-dimensional particle segmentation algorithms. For example, if the data type is three-dimensional (3D) point cloud data, the particle segmentation algorithm can be a 3D point cloud particle segmentation algorithm.

[0085] In some embodiments, the first scatter plot is a DIFF scatter plot, and the various types of blood cells may include neutrophils, lymphocytes, monocytes, eosinophils, and basophils.

[0086] In this embodiment of the application, the quantity indicator may include at least one of quantity and quantity percentage.

[0087] S203, based on the abnormal quantity index, obtain the first warning information for the target sample.

[0088] In some embodiments, obtaining the first early warning information for the target sample based on the abnormal quantity index includes:

[0089] The abnormal quantity index and the abnormal index corresponding to the abnormal quantity index are determined by comparing the quantity index of each type of blood cell with the third preset threshold corresponding to the type of blood cell.

[0090] The first warning information is to identify the number of abnormal indicators and the corresponding abnormal indicators.

[0091] In this embodiment, the abnormal quantity index is used to indicate whether the quantity index is relatively high or relatively low. For example, blood cells are white blood cells, and the normal range for the white blood cell count is 4 × 10⁻⁶. 9 / L to 10×10 9 / L, if the white blood cell count is 11×10 9 If the white blood cell count is / L, it indicates that the white blood cell count is too high, which is an abnormal count.

[0092] In this embodiment of the application, the abnormal indicators are used to indicate the abnormal types of blood cells present; optionally, the abnormal indicators may include, but are not limited to, at least one of the following: presence of immature cells (juvenile cells), left shift of the nucleus, platelet aggregation, erythrocyte agglutination, and presence of nucleated erythrocytes.

[0093] In this embodiment, the first warning information and the second and third warning information in the following embodiments all include M types of blood cells and N abnormal indicators; M and N are both positive integers; M and N are both greater than 1. The first warning information, the second warning information, and the third warning information may also include, but are not limited to, at least one of N quantitative indicators and N confidence levels.

[0094] For example, the electronic device processes the first scatter plot according to the particle segmentation algorithm to obtain the proportion of white blood cells, neutrophils, lymphocytes and monocytes; if the proportion of neutrophils is 7% and the third preset threshold is 5%, and 7% is greater than 5%, then the abnormal indicator of neutrophils can be determined as nuclear left shift, and the proportion of neutrophils and the abnormal indicator are used as the first warning information.

[0095] S204, classify the first scatter plot according to the classification model to obtain the second warning information output by the classification model.

[0096] In this embodiment of the application, the classification model can be any model that classifies the information contained in the image data.

[0097] Optionally, the classification model may include, but is not limited to, at least one of a neural network model, a classifier model (Support Vector Machine, SVM), and a random forest model.

[0098] For example, the classification model can be a convolutional neural network model, such as... Figure 3 As shown, Figure 3 This is a schematic diagram of a classification model. The classification model consists of three convolutional layers, three pooling layers, a flattened layer, a fully connected layer (Dense), and an activation function (Sigmoid).

[0099] For example, a classification model could be a random forest model. Random forest models can be used to handle both classification and regression problems. Compared to a single decision tree model, random forest models offer higher prediction accuracy and stability.

[0100] In some embodiments, classifying the first scatter plot according to a classification model to obtain the second warning information output by the classification model includes:

[0101] The first scatter plot is processed using the classification model to obtain the confidence levels of various anomaly indicators;

[0102] The abnormal indicators with a confidence level greater than or equal to the second preset threshold are output as the second warning information of the target sample.

[0103] In this embodiment, the confidence level is used to characterize the probability that the i-th cell type will exhibit the j-th abnormal indicator; i and j are positive integers, and i and j can be the same. For example, the confidence level can be a probability value.

[0104] In some embodiments, the second preset threshold is 0.5. The electronic device processes the first scatter plot according to a pre-trained classification model to obtain a first abnormal indicator of left shift of the nucleus with a first confidence level of 0.78, and a second abnormal indicator of platelet aggregation with a second confidence level of 0.3. Since the confidence level of the first abnormal indicator is 0.78, which is greater than 0.5, left shift of the nucleus is determined to be positive, indicating that the target sample exhibits left shift of the nucleus, and the left shift of the nucleus and the confidence level of 0.78 are output as the second warning information. Since the confidence level of the second abnormal indicator is 0.3, which is less than 0.5, platelet aggregation is determined to be negative, indicating that the target sample does not exhibit platelet aggregation.

[0105] S205, if the first warning information and the second warning information are consistent, output the first target warning information; the first target warning information is the first warning information and / or the second warning information.

[0106] For example, if the first abnormal indicator in the first warning information is a left shift of the nuclear nucleus in the target sample and the quantity indicator is the percentage of neutrophils, and the first abnormal indicator in the second warning information is a left shift of the nuclear nucleus in the target sample and the confidence level of the left shift is 0.67, which is greater than 0.5, then the left shift of the nuclear nucleus is positive; then it is determined that the first abnormal indicator in the first warning information and the second warning information are consistent, and the first warning information or the second warning information corresponding to the first abnormal indicator is determined to be the first target warning information and output.

[0107] For example, if the first warning information includes a first abnormal indicator and a second abnormal indicator; and the second warning information includes a first abnormal indicator, a second abnormal indicator, and a third abnormal indicator; and it can be determined that the first abnormal indicator and the second abnormal indicator in the first warning information and the second warning information are consistent, then the first abnormal indicator and the second abnormal indicator, as well as their corresponding cell types and quantity percentages, can be determined as the first target warning information and output.

[0108] In some embodiments, when the first warning information and the second warning information are consistent, outputting a first target warning information includes:

[0109] Output the first target warning information and normal indicators of normal cells; normal indicators include the number of normal cells of each type.

[0110] In some embodiments, the first target warning information and the second target warning information in the following embodiments may be output in tabular, list, graphical or voice form.

[0111] In this embodiment, the first scatter plot is processed in different ways, and the resulting warning information is matched. This allows for the determination that the accuracy and reliability of the matching first and / or second warning information are relatively high, and thus it can be output as the first target warning information. In this way, the accuracy of the first target warning information can be guaranteed, thereby ensuring the accuracy and reliability of the diagnostic results determined by medical personnel based on the first target warning information.

[0112] In some embodiments, the method further includes:

[0113] If at least one abnormal indicator in the first warning information is inconsistent with the abnormal indicator in the second warning information, the abnormal indicator that is inconsistent between the first warning information and the second warning information shall be determined as the target abnormal indicator.

[0114] The first scatter plot is matched with the second scatter plot of the abnormal samples in the sample knowledge base to obtain the matching result;

[0115] Based on the matching result, a second target warning information is output; the second target warning information includes a third warning information corresponding to the matched second scatter plot and / or the second warning information corresponding to the first scatter plot.

[0116] For example, if the first warning information includes a first abnormal indicator and a second abnormal indicator; and the second warning information includes a first abnormal indicator, a second abnormal indicator, and a third abnormal indicator; and it can be determined that the third abnormal indicator in the second warning information is inconsistent with the abnormal indicators in the first warning information, then the third abnormal indicator can be determined as the target abnormal indicator.

[0117] In this embodiment of the application, the sample knowledge base includes multiple historical abnormal sample data; each abnormal sample may include a sample identifier, a second scatter plot and an abnormal index; each abnormal sample may also include, but is not limited to, at least one of cell type, quantity index, confidence level and microscopic examination results.

[0118] In this embodiment of the application, when the first warning information and the second warning information are not completely consistent, it can be assumed that there is interference from anomalies in the first scatter plot. By matching the first scatter plot with the second scatter plot of the abnormal samples in the sample knowledge base, a relatively accurate second target warning information is obtained based on the matching result, thereby reducing the interference caused by anomalies and ensuring the accuracy of the second target warning information.

[0119] In some embodiments, the step of matching the first scatter plot with a second scatter plot of abnormal samples in the sample knowledge base to obtain a matching result includes:

[0120] If there exists a second scatter plot in the sample knowledge base that has a similarity greater than a first preset threshold with the first scatter plot, then the matching result is determined to be a successful match.

[0121] If the similarity between the first scatter plot and the second scatter plot of any abnormal sample used for matching in the sample knowledge base is less than the first preset threshold, then the matching result is determined to be an unsuccessful match.

[0122] For example, such as Figure 4As shown, the first preset threshold is 0.8. The first scatter plot is matched with the second scatter plots of the first to Kth samples, where K is a positive integer. For example, the similarity between the second scatter plot of the first sample and the first scatter plot is 0.9, the similarity between the second scatter plot of the second sample and the first scatter plot is 0.7, and so on, with the similarity between the second scatter plot of the Kth sample and the first scatter plot being 0.1. If the second scatter plot with the highest similarity to the first scatter plot in the sample knowledge base is the second scatter plot of the first sample, and the similarity of the second scatter plot of the first sample (0.9) is greater than 0.8, then the matching result is determined to be a successful match.

[0123] In some embodiments, the step of matching the first scatter plot with a second scatter plot of abnormal samples in the sample knowledge base to obtain a matching result includes:

[0124] If the sample knowledge base meets the first matching condition, the matching result is determined based on the similarity obtained by matching the first scatter plot with each of the second scatter plots; the first matching condition includes that the number of types of abnormal indicators contained in the sample knowledge base is greater than a first threshold and the number of abnormal samples corresponding to each abnormal indicator is greater than a second threshold.

[0125] The step of outputting a second target warning message based on the matching result includes:

[0126] If the matching result is successful, the third warning information corresponding to the second scatter plot with the highest similarity is determined as the second target warning information;

[0127] If the matching result is unsuccessful, the second warning information corresponding to the first scatter plot is determined to be the second target warning information.

[0128] For example, the first threshold can be 6, 7 or 10, etc.; the second threshold can be 800, 1000 or 1050, etc.

[0129] In some embodiments, the second preset threshold is 6 types, the second preset threshold is 1000, and the sample knowledge base includes K samples, where K is a positive integer and K is greater than 1000. The sample knowledge base contains 8 types of abnormal indicators, and the number of abnormal samples containing each abnormal indicator is greater than 1000. It can be considered that the historical abnormal sample data of the sample knowledge base is relatively large and the abnormal indicator types are comprehensively covered. Figure 4 As shown, the electronic device compares the similarity of the first scatter plot with all the second scatter plots in the sample knowledge base to determine the matching result.

[0130] In some embodiments, if the matching result is successful, the electronic device determines that the third warning information corresponding to the second scatter plot with the highest similarity includes N abnormal indicators, where N is a positive integer; the electronic device determines the third warning information corresponding to the N abnormal indicators (i.e., all the third warning information corresponding to the second scatter plot with the highest similarity) as the second target warning information. If the matching result is unsuccessful, the electronic device determines that the second warning information corresponding to the abnormal indicators with a confidence level greater than or equal to a second preset threshold is the second target warning information.

[0131] In this embodiment, due to the large volume of historical abnormal sample data in the sample knowledge base and its comprehensive coverage of abnormal indicators, the accuracy of the matching similarity can be guaranteed to be high. By matching the second scatter plot of each abnormal sample with the first scatter plot for similarity, the second scatter plot with the highest similarity can be selected. When the match is successful, the second target warning information obtained based on the third warning information corresponding to the second scatter plot with the highest similarity has high accuracy. When the first warning information and the second warning information are inconsistent and the first scatter plot and the second scatter plot fail to match, the second warning information with relatively high accuracy is selected as the second target warning information output, which can relatively guarantee the accuracy and reliability of the second target warning information.

[0132] In some embodiments, the step of matching the first scatter plot with a second scatter plot in the sample knowledge base to obtain a matching result includes:

[0133] If the sample knowledge base does not meet the first matching condition, or if the sample knowledge base meets the first matching condition but the number of the target abnormal indicators is less than the third threshold; the first matching condition includes that the number of types of abnormal indicators contained in the sample knowledge base is greater than the first threshold and the number of abnormal samples corresponding to each abnormal indicator is greater than the second threshold.

[0134] For each of the target anomaly indicators, the matching result is determined by performing similarity matching between the first scatter plot and each second scatter plot containing the target anomaly indicator.

[0135] The step of outputting a second target warning message based on the matching result includes:

[0136] For a successfully matched target anomaly indicator, the third alarm information corresponding to the target anomaly indicator of the second scatter plot with the highest similarity is taken as the first type of alarm information.

[0137] For target anomaly indicators that fail to match, the second warning information corresponding to the target anomaly indicator will be used as the second type of alarm information;

[0138] The output includes the second target warning information containing the first type of alarm information and the second type of alarm information.

[0139] For example, the third threshold is 2, 3, or 5, etc.

[0140] In one embodiment, if the number of types of abnormal indicators contained in the sample knowledge base is greater than a first threshold and the number of abnormal samples corresponding to each abnormal indicator is greater than a second threshold, it can be considered that the historical abnormal sample data of the sample knowledge base is relatively large and the types of abnormal indicators are comprehensively covered; however, if the abnormal indicators that are inconsistent between the first warning information and the second warning information are, i.e., the target abnormal indicator is only the first abnormal indicator, and the number of target abnormal indicators in the first warning information and the second warning information is 1, which is less than the third preset threshold 2, it can be determined that there is no need to match the first scatter plot with all the second scatter plots in the sample knowledge base, but to match the first scatter plot only with the second scatter plot containing the abnormal samples of the first abnormal indicator, and determine the matching result corresponding to the first abnormal indicator.

[0141] In some real-time examples, the target anomaly indicators are anomaly indicators A, B, and C. The first scatter plot is matched with a second scatter plot containing anomaly indicators A, B, or C, respectively. The second scatter plot with the highest similarity corresponding to anomaly indicator A is designated as the X scatter plot, the second scatter plot with the highest similarity corresponding to anomaly indicator B as the Y scatter plot, and the second scatter plot with the highest similarity corresponding to anomaly indicator C as the Z scatter plot. If anomaly indicators A and C match successfully, but anomaly indicator B does not match successfully, then the third warning information corresponding to anomaly indicator A in the X scatter plot and the third warning information corresponding to anomaly indicator C in the Z scatter plot are designated as first-type alarm information; the second warning information corresponding to anomaly indicator B is designated as second-type alarm information. The first and second-type alarm information are then output as the second target warning information.

[0142] In one embodiment, such as Figure 5 As shown, the second threshold is 860. If the number of abnormal samples corresponding to at least one abnormal indicator in the sample knowledge base is less than 860, it can be considered that the historical abnormal sample data in the sample knowledge base is relatively small. The target abnormal indicators in the first and second warning messages are the fourth and fifth abnormal indicators. For the fourth abnormal indicator, the first scatter plot is matched with the second scatter plot containing the abnormal samples of the fourth abnormal indicator to determine the matching result corresponding to the fourth abnormal indicator. For the fifth abnormal indicator, the first scatter plot is matched with the second scatter plot containing the abnormal samples of the fifth abnormal indicator to determine the matching result corresponding to the fifth abnormal indicator.

[0143] In one embodiment, such as Figure 5As shown, the target anomaly indicator is the fourth anomaly indicator. The maximum similarity of the second scatter plot matching the fourth anomaly indicator is 0.9, and the first preset threshold is 0.7. Since 0.9 is greater than 0.7, the matching result corresponding to the fourth anomaly indicator is determined to be a successful match. The second scatter plot corresponding to the maximum similarity of the fourth anomaly indicator is the X scatter plot. The third warning information corresponding to the X scatter plot includes the first to seventh anomaly indicators. Only the third warning information corresponding to the fourth anomaly indicator is determined to be the first type of alarm information. The first type of alarm information is output as the second target warning information.

[0144] In one embodiment, such as Figure 5 As shown, the target anomaly indicator includes the fifth anomaly indicator. The maximum similarity of the second scatter plot matching the fifth anomaly indicator is 0.5, and the first preset threshold is 0.7. Since 0.5 is less than 0.7, the matching result corresponding to the fifth anomaly indicator is determined to be an unsuccessful match. The second warning information corresponding to the first scatter plot includes the first to fifth anomaly indicators. Only the second warning information corresponding to the fifth anomaly indicator is determined as the second type of alarm information. The second type of alarm information is output as the second target warning information.

[0145] In this embodiment, when the historical abnormal sample data in the sample knowledge base is relatively small, the types of abnormal indicators covered are incomplete, or the number of target abnormal indicators is small, compared to matching the first scatter plot with all second scatter plots in the sample knowledge base, matching only the second scatter plots corresponding to abnormal samples containing the target abnormal indicators can speed up the matching efficiency and improve the matching performance while ensuring the relative accuracy of the second target warning information. Furthermore, by outputting the second target warning information containing both first and second type alarm information, the accuracy and reliability of the abnormal indicator information in the second target warning information can be guaranteed to be high, ensuring the comprehensiveness and reliability of the second target warning information, thereby ensuring the accuracy and reliability of the diagnostic results determined by medical personnel based on the second target warning information.

[0146] In some embodiments, the particle segmentation algorithm includes a three-dimensional point cloud particle segmentation algorithm; the first scatter plot is three-dimensional point cloud data; the step of segmenting the first scatter plot according to the particle segmentation algorithm to obtain the quantity indicators of various blood cells includes:

[0147] The three-dimensional point cloud data is segmented according to the three-dimensional point cloud particle segmentation algorithm to obtain the quantity index of various types of blood cells; the various types of blood cells include specific types of cells and non-specific types of cells; the specific types of cells include at least one of primitive cells, abnormal lymphocytes, atypical lymphocytes and plasma cells.

[0148] The method further includes:

[0149] For the specific type of cell, the first target warning information is output based on the first warning information corresponding to the specific type of cell;

[0150] For the non-specific cell type, the first target warning information is output based on the first warning information and the second warning information corresponding to the non-specific cell type, and / or, the second target warning information is output based on the second warning information and the third warning information corresponding to the non-specific cell type.

[0151] For example, such as Figure 6 As shown, the 3D point cloud particle segmentation algorithm includes group sampling, convolutional layers, pooling layers, flattening layers, upsampling, and an output layer (Softmax).

[0152] In this embodiment, since the first scatter plot is processed using a 3D point cloud particle segmentation algorithm, the first warning information for specific cell types is relatively accurate. Therefore, the first warning information corresponding to specific cell types can be directly output as the first target warning information. Furthermore, by dividing cells into specific and non-specific cell types, matching the first warning information corresponding to non-specific cell types with the second warning information, and outputting the first target warning information and / or second target warning information corresponding to non-specific cell types based on the matching results, it is unnecessary to match all cell types, thus further saving computational power, improving matching efficiency, and enhancing matching performance.

[0153] In some embodiments, outputting the second target warning information includes:

[0154] The second target warning information is displayed on the interactive interface, and an anomaly identifier is added to the second target warning information corresponding to the target anomaly indicator; the anomaly identifier is used to indicate that the second target warning information is suspicious.

[0155] In this embodiment, the anomaly identifier may be in the form of strings, letters, graphics, and colors, among others. The anomaly identifier is used to characterize the second target warning information corresponding to the target anomaly indicator as having a relatively moderate level of reliability.

[0156] In some embodiments, the interactive page of the interactive component includes a result display area; the result display area is used to display the first target warning information, the unmarked second target warning information, the marked second target warning information and / or the normal indicators of normal cells; the normal indicators include the number of normal cells of each type.

[0157] In this embodiment of the application, the interactive page can be a software application (User Interface, UI) interface located in the electronic device; or it can be a third-party platform interface connected to the electronic device. For example, if the electronic device is a blood cell analyzer, the third-party platform interface can be the software interface of a mobile terminal connected to the blood cell analyzer.

[0158] In this embodiment of the application, the result display area can be the entire area or a part of the interactive page.

[0159] For example, such as Figure 7 As shown, Figure 7 This is a schematic diagram of the target warning information output. The items include the first target warning information, the unlabeled second target warning information, the labeled second target warning information, and normal indicators of normal cells. Abnormality is indicated by the letter R.

[0160] In this embodiment of the application, by adding an anomaly identifier to the second target warning information corresponding to the target anomaly indicator, medical staff can be reminded to independently confirm whether the second target warning information corresponding to the target anomaly identifier is credible.

[0161] The following provides specific examples in conjunction with any of the above embodiments:

[0162] Example 1: Figure 8 This is an example illustrating a scatter plot detection method; such as Figure 8 As shown, the method is performed by an electronic device, and the method includes:

[0163] S801, obtain the first scatter plot of the target sample.

[0164] In an optional embodiment, the first scatter plot may be a DIFF scatter plot; the first scatter plot is used to characterize the distribution data of blood cells contained in the target sample.

[0165] S802, the first scatter plot is processed according to the particle segmentation algorithm to obtain the quantity indicators of various blood cells.

[0166] In one alternative embodiment, the first scatter plot is segmented according to a particle segmentation algorithm to obtain the percentage of each type of blood cell.

[0167] S803, based on the number of anomalies, obtains the first warning information for the target sample.

[0168] In one optional embodiment, an abnormal quantity index is determined by comparing the percentage of each type of cell with a third preset threshold corresponding to the cell type, and the abnormal quantity index and the abnormal index corresponding to the abnormal quantity index are determined as the first warning information.

[0169] S804. The first scatter plot is processed according to the classification model to obtain the confidence level of the anomaly index.

[0170] S805 outputs the abnormal indicators with a confidence level greater than or equal to the second preset threshold as the second early warning information of the target sample.

[0171] In one optional embodiment, the confidence level is a probability value; the probability value of each abnormal indicator is compared with a second preset threshold corresponding to the type, and the abnormal indicator with a confidence level greater than or equal to the second preset threshold is determined as the second warning information.

[0172] S806, determine whether the first warning information and the second warning information are consistent.

[0173] In an alternative embodiment, if yes, proceed to S807; otherwise, proceed to S808.

[0174] S807 outputs the first target warning information.

[0175] In one optional embodiment, a first warning message or a second warning message is determined as the first target warning message, and the first target warning message and the percentage of normal cells of each type are output; the first target warning message includes abnormal indicators and their corresponding cell types and percentages.

[0176] S808, match the first scatter plot with the second scatter plot of the abnormal samples in the sample knowledge base to obtain the matching result.

[0177] In an optional embodiment, if at least one abnormal indicator in the first warning information is inconsistent with the abnormal indicator in the second warning information, the inconsistent abnormal indicator in the first warning information and the second warning information is determined as the target abnormal indicator; and the first scatter plot is matched with the second scatter plot of abnormal samples in the sample knowledge base to obtain the matching result.

[0178] S809, confirm whether the match was successful.

[0179] In an optional embodiment, if yes, proceed to S810; if no, proceed to S811.

[0180] S810 outputs the second target warning information based on the third warning information corresponding to the second scatter plot.

[0181] In an optional embodiment, if the matching result is successful, the second target warning information is determined based on the third warning information corresponding to the second scatter plot with the highest similarity; an anomaly label is added to the second target warning information corresponding to the target anomaly index, and the unlabeled second target warning information, the labeled second target warning information, and the percentage of normal cells of each type are output.

[0182] S811, determine the second early warning information corresponding to the first scatter plot as the second target early warning information.

[0183] In an optional embodiment, if the matching result is unsuccessful, the second warning information corresponding to the first scatter plot is determined as the second target warning information; anomaly markers are added to the second target warning information corresponding to the target abnormality index, and the unmarked second target warning information, the marked second target warning information, and the percentage of normal cells of each type are output.

[0184] Example 2: Figure 9 This is an example illustrating a scatter plot detection method; such as Figure 9 As shown, the method is performed by an electronic device, and the method includes:

[0185] S901, obtain the first scatter plot of the target sample.

[0186] S902, the first scatter plot is processed according to the particle segmentation algorithm to obtain the quantity indicators of various blood cells.

[0187] In one optional embodiment, the particle segmentation algorithm is a three-dimensional point cloud particle segmentation algorithm; the first scatter plot is three-dimensional point cloud data; the three-dimensional point cloud data is segmented according to the three-dimensional point cloud particle segmentation algorithm to obtain the percentage of various types of blood cells, including specific types of cells and non-specific types of cells. Specific types of cells proceed to S903; non-specific types of cells proceed to S904.

[0188] S903 outputs the first target warning information based on the first warning information corresponding to a specific type of cell.

[0189] In one optional embodiment, a first warning message for a specific type of cell is determined based on the percentage of cells of a specific type and abnormal indicators; the first warning message for the specific type of cell is output as a first target warning message.

[0190] S904, based on the abnormal number of non-specific cell types, obtains the first warning information for non-specific cell types.

[0191] In one optional embodiment, the abnormal number index of non-specific type cells is determined by comparing the percentage of non-specific type cells with a third preset threshold corresponding to the cell type, and the abnormal number index and the abnormal index corresponding to the abnormal number index are determined as the first warning information for non-specific type cells.

[0192] S905, the first scatter plot is processed according to the classification model to obtain the confidence level of the anomaly index.

[0193] S906, output the abnormal indicators with a confidence level greater than or equal to the second preset threshold as the second early warning information of the target sample.

[0194] In one optional embodiment, the confidence level is a probability value; the probability value of each abnormal indicator is compared with a second preset threshold corresponding to the type, and the abnormal indicator with a confidence level greater than or equal to the second preset threshold is determined as the second warning information.

[0195] S907 determines whether the first warning information for non-specific cell types and the second warning information for non-specific cell types are consistent.

[0196] In an optional embodiment, if yes, proceed to S908; if no, proceed to S909.

[0197] S908 outputs first-target warning information for non-specific cell types.

[0198] In an optional embodiment, a first warning message or a second warning message for non-specific cell types is determined as the first target warning message, and the first target warning message for non-specific cell types and the percentage of normal cells of each type in non-specific cell types are output.

[0199] S909, match the first scatter plot with the second scatter plot of the abnormal samples in the sample knowledge base to obtain the matching result.

[0200] In an optional embodiment, if at least one abnormal indicator in the first warning information for non-specific cell types is inconsistent with the abnormal indicator in the second warning information, the inconsistent abnormal indicator in the first warning information and the second warning information is determined as the target abnormal indicator; and the first scatter plot is matched with the second scatter plot of abnormal samples in the sample knowledge base to obtain the matching result.

[0201] S910, confirm whether the match was successful.

[0202] In an optional embodiment, if yes, proceed to S911; if no, proceed to S912.

[0203] S911, based on the third warning information corresponding to non-specific cell types in the second scatter plot, output the second target warning information.

[0204] In an optional embodiment, if the matching result is successful, the second target warning information is determined based on the third warning information corresponding to the non-specific type cells in the second scatter plot with the highest similarity; an anomaly label is added to the second target warning information corresponding to the target anomaly index, and the unlabeled second target warning information, the labeled second target warning information, and the percentage of normal cells of each type in the non-specific type cells are output.

[0205] S912, the second warning information corresponding to non-specific cell types is identified as the second target warning information.

[0206] In an optional embodiment, if the matching result is unsuccessful, the second target warning information is determined based on the second warning information corresponding to non-specific cell types in the first scatter plot; anomaly markers are added to the second target warning information corresponding to the target abnormality index, and the unmarked second target warning information, the marked second target warning information, and the percentage of normal cells of each type in non-specific cell types are output.

[0207] The aforementioned sample analyzer, scatter plot detection method, and device process the first scatter plot in different ways, match the obtained warning information, and determine that the accuracy and reliability of the matched first and / or second warning information are relatively high, and can be used as the first target warning information output. This ensures the accuracy of the first target warning information, thereby guaranteeing the accuracy and reliability of the diagnostic results determined by medical personnel based on the first target warning information.

[0208] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0209] Based on the same inventive concept, this application also provides a scatter plot detection device for implementing the scatter plot detection method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more scatter plot detection device embodiments provided below can be found in the limitations of the scatter plot detection method described above, and will not be repeated here.

[0210] In some embodiments, such as Figure 10 As shown, a scatter plot detection device is provided, the device comprising:

[0211] The acquisition module 110 is used to acquire a first scatter plot of the target sample; the first scatter plot is used to characterize the distribution data of blood cells contained in the target sample;

[0212] The first processing module 120 is used to segment the first scatter plot according to the particle segmentation algorithm to obtain the quantity indicators of various blood cells;

[0213] The first processing module 120 is used to obtain a first warning message for the target sample based on the abnormal quantity index;

[0214] The second processing module 130 is used to classify the first scatter plot according to the classification model to obtain the second warning information output by the classification model;

[0215] The analysis module 140 is used to output a first target warning information when the first warning information and the second warning information are consistent; the first target warning information is the first warning information and / or the second warning information.

[0216] In some embodiments, the apparatus further includes:

[0217] The determination module is used to determine the abnormal indicator that is inconsistent between the first warning information and the second warning information as the target abnormal indicator when at least one abnormal indicator in the first warning information is inconsistent with the abnormal indicator in the second warning information.

[0218] The matching module is used to match the first scatter plot with the second scatter plot of abnormal samples in the sample knowledge base to obtain the matching result;

[0219] The analysis module 140 is used to output second target warning information based on the matching result; the second target warning information includes third warning information corresponding to the matched second scatter plot and / or the second warning information corresponding to the first scatter plot.

[0220] In some embodiments, the matching module is configured to determine the matching result based on the similarity obtained by matching the first scatter plot with each of the second scatter plots, provided that the sample knowledge base meets the first matching condition; the first matching condition includes that the number of types of abnormal indicators contained in the sample knowledge base is greater than a first threshold and the number of abnormal samples corresponding to each abnormal indicator is greater than a second threshold.

[0221] The analysis module 140 is used to determine the third warning information corresponding to the second scatter plot with the highest similarity as the second target warning information when the matching result is successful.

[0222] The analysis module 140 is used to determine the second warning information corresponding to the first scatter plot as the second target warning information when the matching result is unsuccessful.

[0223] In some embodiments, the matching module is configured to perform the following steps:

[0224] If the sample knowledge base does not meet the first matching condition, or if the sample knowledge base meets the first matching condition but the number of the target abnormal indicators is less than the third threshold; the first matching condition includes that the number of types of abnormal indicators contained in the sample knowledge base is greater than the first threshold and the number of abnormal samples corresponding to each abnormal indicator is greater than the second threshold.

[0225] For each of the target anomaly indicators, the first scatter plot is matched with each second scatter plot containing the target anomaly indicator to determine the matching result.

[0226] The analysis module 140 is used to perform the following steps:

[0227] For a successfully matched target anomaly indicator, the third alarm information corresponding to the target anomaly indicator of the second scatter plot with the highest similarity is taken as the first type of alarm information.

[0228] For target anomaly indicators that fail to match, the second warning information corresponding to the target anomaly indicator will be used as the second type of alarm information;

[0229] The output includes the second target warning information containing the first type of alarm information and the second type of alarm information.

[0230] In some embodiments, the matching module is configured to determine that the matching is successful if the maximum similarity between the second scatter plot and the first scatter plot in the sample knowledge base is greater than a first preset threshold.

[0231] The matching module is configured to determine that the matching is unsuccessful if the similarity between the first scatter plot and the second scatter plot of any abnormal sample in the sample knowledge base used for matching is less than the first preset threshold.

[0232] In some embodiments, the second processing module 130 is configured to perform the following steps:

[0233] The first scatter plot is processed using the classification model to obtain the confidence levels of various anomaly indicators;

[0234] The abnormal indicators with a confidence level greater than or equal to the second preset threshold are output as the second warning information of the target sample.

[0235] In some embodiments, the particle segmentation algorithm includes a three-dimensional point cloud particle segmentation algorithm; the first scatter plot is three-dimensional point cloud data; the first processing module 120 is used to segment the three-dimensional point cloud data according to the three-dimensional point cloud particle segmentation algorithm to obtain the quantity index of various types of blood cells; the various types of blood cells include specific types of cells and non-specific types of cells; the specific types of cells include at least one of primitive cells, abnormal lymphocytes, atypical lymphocytes, and plasma cells.

[0236] The analysis module 140 is used to output the first target warning information based on the first warning information corresponding to the specific type of cell for the specific type of cell.

[0237] The analysis module 140 is used to output the first target warning information based on the first warning information and the second warning information corresponding to the non-specific type of cell, and / or to output the second target warning information based on the second warning information and the third warning information corresponding to the non-specific type of cell.

[0238] In some embodiments, the analysis module 140 is used to display the second target warning information on the interactive interface and add an anomaly identifier to the second target warning information corresponding to the target anomaly indicator; the anomaly identifier is used to indicate that the second target warning information is suspicious information.

[0239] Each module in the aforementioned scatter plot detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the electronic device, or stored in the memory of the electronic device as software, so that the processor can call and execute the operations corresponding to each module. In one embodiment, an electronic device is provided, which can be a terminal, and its internal structure diagram can be as follows: Figure 11As shown, the electronic device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a scatter plot detection method. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0240] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0241] In one embodiment, an electronic device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: the computer program performs the steps in the above-described method embodiments.

[0242] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0243] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and / or features involved in this application are all information, data and / or features authorized by the user or fully authorized by all parties.

[0244] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0245] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0246] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A sample analyzer characterized by, The method comprises: a sampling component configured to obtain a target sample from a sample container containing the sample; a preparation component configured to mix the target sample with a reagent to obtain a sample liquid; a detection component configured to detect the sample liquid prepared by the preparation component to obtain a first scatter plot representing distribution data of blood cells contained in the target sample; a scatter plot detection device comprising a processor and a memory storing processor-executable instructions; wherein the processor is configured to obtain the first scatter plot; perform segmentation processing on the first scatter plot according to a particle segmentation algorithm to obtain quantity indicators of various types of blood cells; obtain first warning information of the target sample based on abnormal quantity indicators; perform classification processing on the first scatter plot according to a classification model to obtain second warning information output by the classification model; an interaction component configured to output first target warning information when the first warning information and the second warning information are consistent; the first target warning information is the first warning information and / or the second warning information.

2. A scatter plot detection method, characterized by, The method comprises: obtaining a first scatter plot of a target sample; the first scatter plot is used to represent distribution data of blood cells contained in the target sample; performing segmentation processing on the first scatter plot according to a particle segmentation algorithm to obtain quantity indicators of various types of blood cells; obtaining first warning information of the target sample based on abnormal quantity indicators; performing classification processing on the first scatter plot according to a classification model to obtain second warning information output by the classification model; outputting first target warning information when the first warning information and the second warning information are consistent; the first target warning information is the first warning information and / or the second warning information.

3. The method of claim 2, wherein, The method further comprises: in the case that at least one abnormal indicator in the first warning information and an abnormal indicator in the second warning information are inconsistent, determining the abnormal indicator inconsistent in the first warning information and the second warning information as a target abnormal indicator; matching the first scatter plot with second scatter plots of abnormal samples in a sample knowledge base to obtain a matching result; outputting second target warning information according to the matching result; the second target warning information comprises third warning information corresponding to the matched second scatter plot and / or the second warning information corresponding to the first scatter plot.

4. The method of claim 3, wherein, The matching of the first scatter plot with second scatter plots of abnormal samples in a sample knowledge base to obtain a matching result comprises: in the case that the sample knowledge base satisfies a first matching condition, determining the matching result according to a similarity obtained by matching the first scatter plot with each second scatter plot; the first matching condition comprises that the number of types of abnormal indicators contained in the sample knowledge base is greater than a first threshold and the number of abnormal samples corresponding to each abnormal indicator is greater than a second threshold; the outputting of second target warning information according to the matching result comprises: in the case that the matching result is successful, determining third warning information corresponding to the second scatter plot with the greatest similarity as the second target warning information; In a case where the matching result is unsuccessful, it is determined that the second early warning information corresponding to the first scatter plot is the second target early warning information.

5. The method of claim 3, wherein, The matching of the first scatter plot with the second scatter plots in the sample knowledge base comprises: In a case where the sample knowledge base does not satisfy a first matching condition, or in a case where the sample knowledge base satisfies the first matching condition but the number of target abnormal indicators is less than a third threshold value, the first matching condition comprises that the number of types of abnormal indicators included in the sample knowledge base is greater than a first threshold value and the number of abnormal samples corresponding to each type of abnormal indicator is greater than a second threshold value; For each type of target abnormal indicator, the first scatter plot is subjected to similarity matching with each second scatter plot containing the target abnormal indicator to determine the matching result; The output of the second target early warning information according to the matching result comprises: For a target abnormal indicator with a successful match, the target abnormal indicator of the second scatter plot with the greatest similarity is taken as first-class alarm information; For a target abnormal indicator with an unsuccessful match, the second early warning information corresponding to the target abnormal indicator is taken as second-class alarm information; The second target early warning information containing the first-class alarm information and the second-class alarm information is output.

6. The method of claim 3, wherein, The matching of the first scatter plot with the second scatter plots in the sample knowledge base comprises: If the maximum similarity of the second scatter plots in the sample knowledge base to the first scatter plot is greater than a first preset threshold value, it is determined that the matching is successful; If the similarity of the first scatter plot to the second scatter plot of any abnormal sample in the sample knowledge base for matching is less than the first preset threshold value, it is determined that the matching is unsuccessful.

7. The method of claim 2, wherein, The classification processing of the first scatter plot by the classification model to obtain the second early warning information output by the classification model comprises: The processing of the first scatter plot by the classification model to obtain the confidence of each type of abnormal indicator; An abnormal indicator with a confidence greater than or equal to a second preset threshold value is output as the second early warning information of the target sample.

8. The method according to any one of claims 2 to 7, characterized in that, The particle segmentation algorithm comprises a three-dimensional point cloud particle segmentation algorithm; the first scatter plot is three-dimensional point cloud data; and the segmentation processing of the first scatter plot by the particle segmentation algorithm to obtain the quantity indicators of each type of blood cell comprises: The segmentation processing of the three-dimensional point cloud data by the three-dimensional point cloud particle segmentation algorithm to obtain the quantity indicators of each type of blood cell; the each type of blood cell comprises at least one of a specific type of cell and a non-specific type of cell; and the specific type of cell comprises at least one of a primitive cell, an abnormal lymphocyte, a dysplastic lymphocyte, and a plasma cell. The method further comprises: For the specific type of cell, the first target early warning information is output according to the first early warning information corresponding to the specific type of cell. The first target early warning information is output according to the first early warning information and the second early warning information corresponding to the non-specific type of cells, and / or the second target early warning information is output according to the second early warning information and the third early warning information corresponding to the non-specific type of cells.

9. The method of claim 3, wherein, The outputting the second target early warning information further includes: The second target early warning information is displayed on an interactive interface, and an abnormality identifier is added to the second target early warning information corresponding to the target abnormal index; the abnormality identifier is used to indicate that the second target early warning information is suspicious information.

10. A scatter plot detection apparatus, characterized by comprising: The device includes: An acquisition module is configured to acquire a first scatter plot of a target sample; the first scatter plot is used to represent distribution data of blood cells contained in the target sample; A first processing module is configured to perform segmentation processing on the first scatter plot according to a particle segmentation algorithm to obtain quantity indexes of various types of blood cells; The first processing module is configured to obtain first early warning information of the target sample based on the quantity indexes of the abnormalities; A second processing module is configured to perform classification processing on the first scatter plot according to a classification model to obtain second early warning information output by the classification model; An analysis module is configured to output first target early warning information in a case where the first early warning information and the second early warning information are consistent; the first target early warning information is the first early warning information and / or the second early warning information.

Images