Method and device for detecting parameters of blood leukocytes and neutrophile multilobal nucleogranulocytes

Abstract

In the method and the device for detecting the parameters of the blood leukocytes and the neutrophile multilobal nucleogranulocytes, a blood sample is pretreated to obtain a microscopic examination sample; tiling a microscopic examination sample; shooting the tiled microscopic examination sample to obtain a microscopic examination sample image; identifying the microscopic examination sample image by using an AI identification algorithm, identifying neutral multi-leaf nucleogranulocytes in a selected area S1 of the image, and obtaining the total number NUMS1 of the neutral multi-leaf nucleogranulocytes in the selected image; the above AI recognition algorithm recognizes the neutral multi-leaf nucleogranulocytes according to the neutral multi-leaf nucleogranulocyte feature data set. According to the method, the concentrations of the neutral multi-leaf nucleogranulocytes with different leaf numbers can be accurately obtained, the neutral multi-leaf nucleogranulocytes can be accurately and quantitatively analyzed, and the classification of the neutral multi-leaf nucleogranulocytes is improved to the level capable of specifically identifying the leaf numbers, so that the classification of leukocytes is further refined, and the classification efficiency of leukocytes is improved. Particularly, the classification of the neutrophil enters a new level.

Description

Technical Field

[0001] The present application belongs to the technical field of analysis of tangible elements based on microscopically magnified images, and particularly relates to a method for detecting and calculating parameters of neutrophil multisegmented granulocytes, a data storage device, and a neutrophil multisegmented granulocyte parameter acquisition device. Background Art

[0002] White blood cells are the body's "guardians" in the fight against disease. When pathogens invade the body, white blood cells can deform and pass through the capillary walls, concentrating at the site of invasion to surround and engulf the bacteria. White blood cells are colorless, spherical, and have a nucleus. They are larger than red blood cells, with a diameter between 7 and 20 μm (micrometers). The normal adult range is (4.0-10.0) × 10^9 cells / L. A variety of factors can cause an increase or decrease in white blood cells. Pathological increases or decreases in white blood cells are very important and serve as an important reference for clinical disease diagnosis. For example, doctors often use changes in white blood cell count to preliminarily determine whether the above clinical symptoms are caused by bacterial infection.

[0003] The number of white blood cells in normal adult blood is (4.0~10.0)×10^9 / L. Figure 1 As shown, neutrophils can be divided into five types. Counting these five types individually, either instrumentally or manually, is known as a five-part differential white blood cell count. Neutrophils (neutrophils) account for 50% to 70%, eosinophils 0.5% to 5%, basophils 0% to 1%, monocytes 3% to 8%, and lymphocytes 20% to 40%. In Wright-stained blood smears, neutrophil cytoplasm appears colorless or very pale red, with numerous diffusely distributed, tiny (0.2 to 0.4 micrometers) characteristic light red or light purple granules.

[0004] Neutrophils are the primary component of white blood cells and the most abundant type of granulocyte, accounting for 40% to 70% of all white blood cells in humans. As the host's first line of defense against invading pathogens, neutrophils possess inherent phagocytic abilities, capable of absorbing nanoparticles and engulfing dying red blood cells. Upon activation, they eliminate foreign pathogens and load targets. Neutrophils can also migrate across blood vessels to adjacent tissues, making them natural candidates for in vivo medical missions. Neutrophils possess chemotactic, phagocytic, and bactericidal properties. Neutrophils originate from the bone marrow and possess lobed or rod-shaped nuclei. Their cytoplasm contains numerous fine, neutrophil-like granules that are neither basophilic nor acidophilic. These granules are primarily lysosomes, containing a rich array of enzymes such as myeloperoxidase, lysozyme, alkaline phosphatase, and acid hydrolases, which are involved in the cell's phagocytic and digestive functions. Neutrophils can be classified into various types based on their life cycle.

[0005] In normal human neutrophils, the nucleus is rod-shaped or 2-5 lobed, with lobes connected by thin filaments. Neutrophils are typically classified according to the degree of nuclear differentiation, typically into rod-shaped neutrophils and lobed neutrophils. Due to the limitations of existing technology, neutrophil classification is typically limited to this level.

[0006] In some cases, for a more in-depth analysis of pathology, further refinement of the 50% to 70% of neutrophils is often performed based on blood smear microscopy. Blood smear microscopy is often affected by manual smearing techniques, and due to uneven front-to-back smears and the high counting workload, the reproducibility of data from multiple smears is poor. Full smear counts are typically not performed; only qualitative analysis is performed to determine if particularly abnormal neutrophils with multisegmented nuclei are present. This is also significantly affected by the ability of the interpreter, and identification can sometimes be inaccurate. Manual identification of neutrophil segmentation using smears is inaccurate and inefficient, making efficient quantitative analysis impossible. Efficient classification and quantitative analysis of neutrophil segmentation remains a technical challenge.

[0007] In some pathological conditions, the segmentation of neutrophils often indicates pathological features. The number of neutrophils with different numbers of lobes, the quantitative relationship between neutrophils with different numbers of lobes, and the presence and number of neutrophils with more lobes can serve as references for some disease states. Further in-depth and detailed classification of neutrophil segmentation is highly necessary.

[0008] When neutrophils have excessive segmentation, they may be divided into four or even five to six lobes or more. If the ratio of five lobes exceeds 0.05, this is also called a right shift in the neutrophil nucleus. This is caused by a deficiency of hematopoietic substances, resulting in impaired DNA synthesis or decreased hematopoietic function. This phenomenon is primarily seen in megaloblastic anemia, pernicious anemia, after treatment with antimetabolite drugs, and during the recovery period from infection. Transient right shift can also occur.

[0009] An increased proportion of neutrophils with segmented nuclei may indicate an acute infection, including localized infections such as carbuncles and furuncles. It can also be caused by systemic infections, including pneumonia, sepsis, purulent tonsillitis, purulent meningitis, epidemic cerebrospinal meningitis, rheumatic fever, acute appendicitis, diphtheria, and acute cholecystitis. It can also be caused by poisoning with lead, mercury, sleeping pills, digitalis, antipyrine, arsenic poisoning, insect poisoning, or foreign protein absorption. It can also be caused by metabolic uremia, diabetic ketoacidosis, gout, or pregnancy poisoning. Hematologic diseases include acute massive hemorrhage, acute hemolysis, acute and chronic myeloid leukemia, Hodgkin's disease, and polycythemia vera. Tissue necrosis can occur in myocardial infarction, gangrene, neoplastic necrosis, and burns. Other conditions, such as cerebral hemorrhage, brain tumors, postoperative conditions, Cushing's syndrome, myxedema, malignant tumors, and cachexia, can also contribute to an increased proportion of neutrophils with segmented nuclei.

[0010] A decrease in the proportion of neutrophils with segmented nuclei can be seen in severe infections such as severe sepsis and miliary tuberculosis. Special infections include typhoid fever, paratyphoid fever, viral infections, rickettsial infections, and protozoal diseases (malaria, kala-azar, and protozoal infections). It can also be caused by blood diseases such as aplastic anemia, pernicious anemia, leukopenic leukemia, acute and chronic lymphocytic leukemia, malignant lymphoma, multiple myeloma, and paroxysmal nocturnal hemoglobinuria. It can also be caused by medications such as organolytic enzymes, anti-tumor drugs, anti-thyroid drugs, anticonvulsants, sulfonamides, phenothiazines, chloramphenicol, and synmycin. Radiation exposure: radiation sickness, etc. Systemic lupus erythematosus, Banti syndrome, cirrhosis, and Felty syndrome can also cause a decrease in the proportion of neutrophils with segmented nuclei. Congenital diseases include cyclic neutropenia, congenital neutropenia, and familial benign eosinophilia. Others may include cachexia, hemodialysis, etc.

[0011] In the existing technology, blood analysis equipment usually only performs five classifications of white blood cells, namely neutrophils, eosinophils, basophils, monocytes, and lymphocytes; classification of neutrophils is usually further divided into neutrophil bands and neutrophils; it is difficult to perform more in-depth and accurate classification and quantitative analysis of neutrophils.

[0012] In the prior art, visible element analysis based on microscopically magnified images can generally only be used to identify specific white blood cell types. Since the volume of the sample corresponding to the image cannot be obtained, only qualitative analysis can usually be performed, and quantitative analysis is difficult.

[0013] The applicant filed a series of Chinese patents, such as

[0014] 1. CN2020112669290, “Cell analysis methods and systems and quantitative methods and systems”;

[0015] 2. CN2020112669182, “Cell Suspension Sample Imaging Method, System, and Kit”;

[0016] 3. CN2022104799126, “Fast focusing method for microscopic image acquisition device and microscopic image measurement method”;

[0017] 4. CN2023116834572, “Blood White Blood Cell Identification AI Training Method and Computational Processing and Storage Device”

[0018] A novel technical solution is used to measure the cell content in the blood, including the analysis, identification, and counting of white blood cells. Previously, the applicant's technical solution only included the analysis, identification, and counting of five types of white blood cells: neutrophils, eosinophils, basophils, monocytes, and lymphocytes. This identification and counting was based on these categories, without further subdivision of neutrophils and neutrophils.

[0019] How to achieve further subdivision of neutrophil segmentation based on the above technical framework is a technical problem to be solved.

[0020] Glossary:

[0021] Neutrophil multisegmented granulocytes refer to neutrophil multisegmented granulocytes whose nuclei are differentiated into at least two lobes.

[0022] The hemolytic agent SDS is: sodium dodecyl sulfate; the hemolytic agent TWEEN20 is: Tween 20 a commonly used surfactant; the hemolytic agent TWEEN80 is: Tween 80 a commonly used surfactant. Summary of the Invention

[0023] The applicant conducted a large number of experiments and collected blood samples from various animals. After long-term research and experimental verification, he proposed a liquid-based blood suspension microscopy imaging and established a quantitative analysis mechanism for neutrophil multisegmented granulocytes. It can not only accurately quantitatively analyze neutrophil multisegmented granulocytes, but also improve the classification of neutrophil multisegmented granulocytes to a level that can specifically identify the number of lobes, that is, it can identify neutrophil bilobed granulocytes, neutrophil trilobed granulocytes, neutrophil tetralobed granulocytes, neutrophil pentalobed granulocytes, and neutrophil sextlobed granulocytes, and can accurately obtain the concentration of neutrophil multisegmented granulocytes with various different numbers of lobes, making the classification of white blood cells more refined, especially the classification of neutrophils to a new level. Based on this, the proportion of different neutrophil multisegmented granulocytes relative to neutrophils and white blood cells can be provided. This information can not only be used for qualitative analysis, but also for accurate quantitative analysis, which has elevated the research and exploration of neutrophils to a higher level.

[0024] In this application, the inventors propose a method for detecting parameters of neutrophil multisegmented granulocytes. A blood sample is preprocessed to obtain a microscopic sample; the microscopic sample is flattened; the flattened microscopic sample is photographed to obtain a microscopic sample image; an AI recognition algorithm is used to identify the above-mentioned microscopic sample image, and the neutrophil multisegmented granulocytes in the selected area S1 of the image are identified, and the total number NUMS1 of neutrophil multisegmented granulocytes in the selected image is obtained; the above-mentioned AI recognition algorithm identifies neutrophil multisegmented granulocytes based on the neutrophil multisegmented granulocyte feature data set.

[0025] The above-mentioned neutrophil multisegmented granulocyte feature dataset is obtained by training after manually annotating neutrophil multisegmented granulocyte images. The above-mentioned neutrophil multisegmented granulocytes are neutrophils with a lobe number greater than M or equal to M, and the minimum value of M is 2; it includes any of the following technical features: TC1: the target neutrophil multisegmented granulocytes are neutrophil multisegmented granulocytes from human blood, M=6; TC2: the target neutrophil multisegmented granulocytes are neutrophil multisegmented granulocytes from non-mammalian blood, M=5; TC3: M is any natural number from 2 to 8; TC4: M is any natural number from 2 to 8; the number of cells corresponding to each neutrophil M-segmented granulocyte is counted respectively.

[0026] The above-mentioned neutrophil multisegmented nuclear granulocyte parameter is the number of neutrophil multisegmented nuclear granulocytes per unit volume; it includes any one of the following technical features: TCC1: the above-mentioned blood sample is pre-treated by dilution and diluted N times; the tile height of the above-mentioned microscopic sample is H; the volume V corresponding to the total number of neutrophil multisegmented nuclear granulocytes NUMS1 is = the selected area S1×H; the number of neutrophil multisegmented nuclear granulocytes per unit volume is = N×the total number of neutrophil multisegmented nuclear granulocytes NUMS1 / blood sample volume V; TCC2: the above-mentioned blood sample is pre-treated by concentration and concentrated N times; the tile height of the above-mentioned microscopic sample is H; the volume V corresponding to the total number of neutrophil multisegmented nuclear granulocytes NUMS1 is = the selected area S1×H; the number of neutrophil multisegmented nuclear granulocytes per unit volume is = the total number of neutrophil multisegmented nuclear granulocytes NUMS1 / (blood sample volume V×N).

[0027] The above-mentioned neutrophil multisegmented granulocyte parameter is the proportion of neutrophil multisegmented granulocytes in white blood cells, and also includes using an AI recognition algorithm to identify the above-mentioned microscopic sample image and identify the number of white blood cells NUMW1 in the selected area S1 of the image. The above-mentioned AI recognition algorithm identifies white blood cells based on the white blood cell feature data set; the proportion of neutrophil multisegmented granulocytes in white blood cells = the total number of neutrophil multisegmented granulocytes NUMS1 / the number of white blood cells NUMW1.

[0028] The above-mentioned white blood cell feature dataset is obtained by manually annotating white blood cell images and then training.

[0029] The above-mentioned neutrophil multisegmented nuclear granulocyte parameter is the proportion of neutrophil multisegmented nuclear granulocytes in neutrophils, and also includes using an AI recognition algorithm to identify the above-mentioned microscopic sample image, identifying the neutrophils NUMW2 in the selected area S1 of the image, and the above-mentioned AI recognition algorithm identifies white blood cells based on the neutrophil characteristic data set; the proportion of neutrophil multisegmented nuclear granulocytes in neutrophils = the total number of neutrophil multisegmented nuclear granulocytes NUMS1 / the number of neutrophils NUMW2.

[0030] The above neutrophil feature dataset is obtained by manually annotating neutrophil images and then training.

[0031] The above-mentioned blood sample is pretreated, including any of the following technical features: TA1: adding a dye to the blood sample to stain the white blood cells; TA2: diluting the blood sample, the dilution range is 1:50 to 1:500; TA3: adding a liquid dye to the blood sample, the liquid dye dilutes the blood sample and stains the white blood cell nuclei; TA4: centrifuging the blood sample, letting it settle, taking the layer where the white blood cells are located to obtain an enriched sample, diluting and staining the enriched sample to obtain a microscopic sample; TA5: adding a hemolytic agent to the blood sample to dissolve the red blood cells, enriching the white blood cells, taking the layer where the white blood cells are located to obtain an enriched sample, diluting and staining the enriched sample to obtain a microscopic sample; TA6: filtering the blood sample to remove the red blood cells and obtain the white blood cells.

[0032] In technical feature TA4, the centrifugation treatment of the blood sample includes: low-speed centrifugation at 100g to 1000g for 3 to 10 minutes; and the static sedimentation is static sedimentation for more than 20 minutes.

[0033] In technical feature TA5, the above-mentioned hemolytic agent includes SDS, TWEEN20 and / or TWEEN80, and the above-mentioned enriched leukocytes are sedimentation enrichment or centrifugation enrichment.

[0034] In technical feature TA6, a leukocyte filter membrane is used to filter the blood sample, the pore size of the leukocyte filter membrane is larger than that of red blood cells, and the pore size of the leukocyte filter membrane is smaller than that of leukocytes.

[0035] In the present application, the inventors also propose a computing and processing device, which is used to execute the neutrophil multisegmented granulocyte parameter detection method including all or part of the above-mentioned method.

[0036] In the present application, the inventors also propose a data storage device, in which the data storage device contains program codes for executing all or part of the above-mentioned method for detecting parameters of polymorphonuclear granulocytes.

[0037] In the present application, the inventors also proposed a device for acquiring parameters of neutrophil multisegmented granulocytes, which runs part or all of the above-mentioned neutrophil multisegmented granulocyte parameter detection method.

[0038] In the present application, the inventors also proposed a blood leukocyte parameter for evaluating the number of neutrophils in the blood. The number of neutrophils per unit volume = the number of neutrophils in the blood sample volume V / blood sample volume V.

[0039] In this application, the inventors also proposed a blood leukocyte parameter for evaluating the ratio of the number of neutrophils and multisegmented granulocytes in the blood to the number of leukocytes. The ratio of neutrophils and multisegmented granulocytes to leukocytes = the number of neutrophils and multisegmented granulocytes / the number of leukocytes.

[0040] In this application, the inventors also proposed a blood leukocyte parameter for evaluating the ratio of the number of polymorphic neutrophils to neutrophils in the blood. The ratio of polymorphic neutrophils to neutrophils = the number of polymorphic neutrophils / the number of neutrophils.

[0041] The above-mentioned blood white blood cell parameters include any one of the following technical features: TB1: the number of lobes M of the cell nucleus in the above-mentioned neutrophil multisegmented granulocytes is greater than 5 lobes or equal to 5 lobes; TB2: the number of lobes M of the cell nucleus in the above-mentioned neutrophil multisegmented granulocytes is greater than 2 lobes or equal to 2 lobes; TB3: the number of lobes M of the cell nucleus in the above-mentioned neutrophil multisegmented granulocytes is greater than 6 lobes or equal to 6 lobes.

[0042] One of the technical effects of the above technical solution is: using AI recognition algorithms to identify neutrophil multisegmented granulocytes, obtaining the total number of neutrophil multisegmented granulocytes NUMS1 in the selected image, and upgrading the classification of neutrophils from the two classifications of rod-shaped nuclei and lobed nuclei to a higher level, capable of classifying neutrophil multisegmented granulocytes with different numbers of lobes. By performing more detailed classification of neutrophil lobes and leveraging the computing power and recognition advantages of AI recognition, the accuracy and efficiency of neutrophil multisegmented granulocyte classification are greatly improved. This method is applicable not only to suspended liquid microscopic specimens, but also to smear microscopic specimens.

[0043] One of the technical effects of the above technical solution is that the neutrophil multisegmented granulocyte feature dataset is obtained by training after manually annotating neutrophil multisegmented granulocyte images, thereby improving the accuracy of neutrophil multisegmented granulocyte identification.

[0044] One of the technical effects of the above technical solution is that the neutrophil multisegmented granulocyte parameter is the number of neutrophils per unit volume, and the number of neutrophils per unit volume = the number of neutrophils in the blood sample volume V / the blood sample volume V. Based on classification recognition, combined with image-based sample quantification, accurate classification and counting are achieved, taking the precise quantitative analysis of neutrophils in white blood cells a step further, reaching the level of accurate quantitative analysis of neutrophils.

[0045] One of the technical benefits of the above technical solution is that the neutrophil multisegmented granulocyte parameter is the proportion of neutrophil multisegmented granulocytes in white blood cells. The neutrophil multisegmented granulocyte proportion in white blood cells = the number of neutrophil multisegmented granulocytes / the number of white blood cells. Based on the accurate counting of neutrophil multisegmented granulocytes, the proportion of neutrophil multisegmented granulocytes in white blood cells can be provided, enriching the white blood cell related parameters and providing more abundant diagnostic reference data for clinical use.

[0046] One of the technical effects of the above technical solution is that the neutrophil multisegmented granulocyte parameter is the proportion of neutrophil multisegmented granulocytes in neutrophils, and the proportion of neutrophil multisegmented granulocytes in neutrophils = the number of neutrophil multisegmented granulocytes / the number of neutrophils. Based on the accurate counting of neutrophil multisegmented granulocytes, information on the proportion of neutrophil multisegmented granulocytes in neutrophils can be provided, enriching neutrophil-related parameters and providing more abundant diagnostic reference data for clinical use.

[0047] One of the technical effects of the above technical solution is that the white blood cell feature dataset is obtained by training after manually annotating white blood cell images, thereby improving the accuracy of white blood cell recognition.

[0048] One of the technical effects of the above technical solution is that the neutrophil feature dataset is obtained by training after manually annotating neutrophil images, thereby improving the accuracy of neutrophil recognition.

[0049] One of the technical effects of the above technical solution is that white blood cells are stained so that the nuclei of neutrophils can be stained, thereby enabling more accurate identification of neutrophils with multisegmented nuclei.

[0050] One of the technical benefits of the above-mentioned technical solution is that blood sample dilution, ranging from 1:50 to 1:500, enables analysis and identification of neutrophils (i.e., white blood cells) and red blood cells based on the same microscopically magnified image, thereby improving the overall efficiency of blood cell analysis.

[0051] One of the technical effects of the above technical solution is that the liquid dye dilutes and dyes the blood sample simultaneously, thereby improving the efficiency of blood sample pretreatment.

[0052] One of the technical effects of the above technical solution is: diluting and staining the leukocyte-enriched sample to obtain a microscopic sample, improving the efficiency of obtaining leukocytes after enrichment, and obtaining enough leukocyte samples in as few pictures as possible.

[0053] One of the technical effects of the above technical solution is that the hemolytic agent dissolves red blood cells or filters out red blood cells and then enriches white blood cells, which not only improves the efficiency of obtaining white blood cells, but also eliminates the influence of red blood cells on white blood cell identification, and further improves the identification and counting efficiency of neutrophils with multisegmented nuclear granulocytes.

[0054] One of the technical effects of the above technical solution is that the hemolytic agent is easy to obtain, the method of filtering red blood cells is simple, and it is an efficient and low-cost enrichment method. After enrichment, the probability of the appearance of white blood cells in the image can be greatly increased. There is no need to consider the tiling of red blood cells for image acquisition, which greatly improves the efficiency of image acquisition and can improve the recognition efficiency of neutrophil multisegmented granulocytes.

[0055] One of the technical effects of the above technical scheme is: for some animals, normal neutrophil multisegmented granulocytes will have 1 to 4 lobes, neutrophil monosegmented granulocytes are neutrophil rod-shaped granulocytes; in neutrophil bisegmented granulocytes, the nucleus is divided into 2 lobes; in neutrophil trisegmented granulocytes, the nucleus is divided into 3 lobes; in neutrophil tetrasegmented granulocytes, the nucleus is divided into 4 lobes; in neutrophil pentasegmented granulocytes, the nucleus is divided into 5 lobes; when neutrophil pentasegmented granulocytes appear in the blood of some animals, it represents some obvious disease states. When the neutrophil pentasegmented granulocytes are counted in detail and the concentration, that is, the number of neutrophil sexegmented granulocytes per unit volume, is obtained, further accurate quantitative evaluation can be carried out.

[0056] One of the technical effects of the above technical solution is that, in humans, normal multisegmented neutrophils have 1 to 5 lobes. Monosegmented neutrophils are neutrophil rods; bisegmented neutrophils have 2 lobes; trisegmented neutrophils have 3 lobes; tetrasegmented neutrophils have 4 lobes; pentasegmented neutrophils have 5 lobes; and sexegmented neutrophils have 6 lobes. When sexegmented neutrophils appear in human blood, it indicates a distinct disease state. By counting sexegmented neutrophils in detail and obtaining their concentration (i.e., the number of sexegmented neutrophils per unit volume), accurate quantitative assessment can be performed.

[0057] One of the technical effects of the above technical solution is that, whether for humans or animals, normal neutrophil multisegmented granulocytes will have more than two lobes. The precise identification and technology of neutrophil multisegmented granulocytes with different lobes provide clinical with deeper and more multi-dimensional neutrophil multisegmented granulocyte information.

[0058] One of the technical effects of the above technical solution is that by identifying neutrophil multisegmented granulocytes with different numbers of lobes, the distribution status of neutrophil multisegmented granulocytes with different numbers of lobes can be accurately counted, providing a research means for pathological research. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is one of the schematic diagrams of the method for detecting parameters of neutrophil multisegmented granulocytes;

[0060] Figure 2 This is the second schematic diagram of the method for detecting parameters of neutrophil multisegmented granulocytes;

[0061] Figure 3 This is the third schematic diagram of the method for detecting parameters of neutrophil multisegmented granulocytes;

[0062] Figure 4 This is one of the schematic diagrams of the blood sample pretreatment method;

[0063] Figure 5 This is the second diagram of the blood sample pretreatment method;

[0064] Figure 6 This is the third diagram of the blood sample pretreatment method;

[0065] Figure 7 This is the fourth diagram of the blood sample pretreatment method;

[0066] Figure 8 is a schematic diagram of a neutrophil; Figure 8 There are 6 types of neutrophils included;

[0067] Figure 9 This is a magnified microscopic image of a neutrophil with multisegmented nuclei to be identified;

[0068] Figure 10 It is a magnified microscopic image that identifies neutrophils with multisegmented nuclei;

[0069] Figure 11 yes Figure 10 In the middle, identify the neutrophil with multisegmented nuclei. DETAILED DESCRIPTION

[0070] The contents of this application are further described in detail below in conjunction with the accompanying drawings. It should be noted that the following is a description of the preferred embodiments of the present invention and does not constitute any limitation to the present invention. The description of the preferred embodiments of the present invention is only an illustration of the general principles of the present invention. The numbers such as "first", "second" and "A" and "B" involved in the present invention are only for the convenience of explanation and do not represent the order relationship in time or space. The combination of letters and numbers "TA", "TB" and "H" involved in the present invention are only for the convenience of explanation, and the specific meaning is determined by the specific words referred to.

[0071] How to further subdivide neutrophil segmentation is not yet addressed in existing technologies. Manual identification of segmentation in neutrophil segmentation is inefficient, and the efficiency of experience accumulation and sharing is also low. In human blood, neutrophil segmentation normally includes bilobed, trilobed, quadrilobed, and pentalobed neutrophils; the presence of hexalobed neutrophils indicates certain pathological signs. In the blood of some animals, bilobed, trilobed, and quadrilobed neutrophils normally include pentalobed neutrophils; the presence of pentalobed neutrophils indicates certain pathological signs. These more in-depth classifications will also bring the analysis of neutrophil segmentation into a more in-depth realm. Even though medical research has yet to delve deeper into the concentrations and ratios of neutrophils bifurcated, trifurcated, quadrifurcated, and quintuplet-like granulocytes, providing this information for medical research has scientific value. However, the potential value of being ahead of medical research as a device manufacturer is currently unforeseen.

[0072] like Figure 1 As shown, in an embodiment of a method for detecting parameters of neutrophils, a blood sample is preprocessed to obtain a microscopic sample; the microscopic sample is flattened; the flattened microscopic sample is photographed to obtain a microscopic sample image; an AI recognition algorithm is used to recognize the microscopic sample image, identify neutrophils in a selected area S1 of the image, and obtain the total number of neutrophils in the selected image, NUMS1; the AI ​​recognition algorithm recognizes neutrophils based on a neutrophil feature dataset. The neutrophil feature dataset is obtained by training after manually annotating neutrophil images.

[0073] like Figure 1 As shown, the above-mentioned neutrophil multisegmented granulocyte parameter is the number of neutrophil multisegmented granulocytes per unit volume; the above-mentioned blood sample is diluted N times after dilution pretreatment; the tile height of the above-mentioned microscopic sample is H; the volume V corresponding to the total number of neutrophil multisegmented granulocytes NUMS1 is = the selected area S1×H; the number of neutrophil multisegmented granulocytes per unit volume is = N×the total number of neutrophil multisegmented granulocytes NUMS1 / the volume of the blood sample V.

[0074] In other embodiments, the blood sample is pre-treated for concentration and concentrated N times; the tile height of the microscopic sample is H; the volume V corresponding to the total number of neutrophils NUMS1 is equal to the selected area S1×H; the number of neutrophils per unit volume is equal to the total number of neutrophils NUMS1 / (blood sample volume V×N).

[0075] like Figure 2 As shown, in an embodiment of a method for detecting a neutrophil multisegmented granulocyte parameter, the neutrophil multisegmented granulocyte parameter is the ratio of neutrophils to white blood cells. The method further includes using an AI recognition algorithm to identify the microscopic sample image and the number of white blood cells NUMW1 in a selected area S1 of the image. The AI ​​recognition algorithm identifies white blood cells based on a white blood cell feature dataset; the ratio of neutrophils to white blood cells = the total number of neutrophils NUMS1 / the number of white blood cells NUMW1. The white blood cell feature dataset is obtained by training after manually annotating white blood cell images.

[0076] like Figure 3 As shown, in an embodiment of a method for detecting a neutrophil multisegmented nuclear granulocyte parameter, the neutrophil multisegmented nuclear granulocyte parameter is the ratio of neutrophils multisegmented nuclear granulocytes to neutrophils, and the method further includes using an AI recognition algorithm to identify the microscopic sample image and neutrophils NUMW2 in a selected area S1 of the image. The AI ​​recognition algorithm identifies white blood cells based on a neutrophil feature dataset; the ratio of neutrophils multisegmented nuclear granulocytes to neutrophils = the total number of neutrophils multisegmented nuclear granulocytes NUMS1 / the number of neutrophils NUMW2. The neutrophil feature dataset is obtained by training after manually annotating neutrophil images.

[0077] like Figure 4 In an embodiment of a method for detecting polymorphonuclear neutrophil parameters, the blood sample is pre-treated, and a dye is added to the blood sample to stain the white blood cells. In some embodiments, the blood sample is diluted in a dilution range of 1:50 to 1:500. TA3: A liquid dye is added to the blood sample, which dilutes the blood sample and stains the white blood cell nuclei.

[0078] like Figure 5 In an embodiment of a method for detecting polymorphonuclear neutrophil parameters, a blood sample is centrifuged, allowed to settle, and an enriched sample is obtained from the leukocyte layer. The enriched sample is then diluted and stained to obtain a microscopic specimen. The centrifugation of the blood sample includes low-speed centrifugation at 100g to 1000g for 3 to 10 minutes; the settling is performed for at least 20 minutes.

[0079] like Figure 6 In an embodiment of a method for detecting parameters of polymorphonuclear neutrophils, a hemolytic agent is added to a blood sample to lyse red blood cells, enrich white blood cells, and obtain an enriched sample from the white blood cell layer. The enriched sample is diluted and stained to obtain a microscopic specimen. The hemolytic agent includes SDS, TWEEN20, and / or TWEEN80, and the white blood cell enrichment is performed by sedimentation enrichment or centrifugation enrichment.

[0080] like Figure 7 In an embodiment of a method for detecting polymorphonuclear neutrophil parameters, TA6: filtering a blood sample to remove red blood cells and obtain white blood cells. The blood sample is filtered using a white blood cell filter membrane, wherein the pore size of the white blood cell filter membrane is larger than that of red blood cells and the pore size of the white blood cell filter membrane is smaller than that of white blood cells.

[0081] In the present application, the inventors also proposed a blood leukocyte parameter for evaluating the number of neutrophils in the blood. The number of neutrophils per unit volume = the number of neutrophils in the blood sample volume V / blood sample volume V.

[0082] In this application, the inventors also proposed a blood leukocyte parameter for evaluating the ratio of the number of neutrophils and multisegmented granulocytes in the blood to the number of leukocytes. The ratio of neutrophils and multisegmented granulocytes to leukocytes = the number of neutrophils and multisegmented granulocytes / the number of leukocytes.

[0083] In this application, the inventors also proposed a blood leukocyte parameter for evaluating the ratio of the number of polymorphic neutrophils to neutrophils in the blood. The ratio of polymorphic neutrophils to neutrophils = the number of polymorphic neutrophils / the number of neutrophils.

[0084] The above-mentioned blood white blood cell parameters include any one of the following technical features: TB1: the number of lobes M of the cell nucleus in the above-mentioned neutrophil multisegmented granulocytes is greater than 5 lobes or equal to 5 lobes; TB2: the number of lobes M of the cell nucleus in the above-mentioned neutrophil multisegmented granulocytes is greater than 2 lobes or equal to 2 lobes; TB3: the number of lobes M of the cell nucleus in the above-mentioned neutrophil multisegmented granulocytes is greater than 6 lobes or equal to 6 lobes.

[0085] like Figure 8 Shown is a schematic diagram of a neutrophil; Figure 8 There are six types of neutrophils included in the data; A, B, C, D, E, and F are neutrophil band cells, neutrophil bilobed neutrophils, neutrophil sextelomeric neutrophils, neutrophil trilobed neutrophils, neutrophil tetralobed neutrophils, and neutrophil pentalobed neutrophils, respectively. The method of the present application can be used to count any of these types to obtain a single neutrophil multisegmented neutrophil parameter, or to aggregate the parameters to form a total neutrophil multisegmented neutrophil parameter.

[0086] like Figure 9 As shown in FIG, it is a microscopic magnified image of a neutrophil with multisegmented nuclei to be identified; Figure 10 Shown is a magnified microscopic image identifying a neutrophil with multisegmented nuclei; Figure 11 yes Figure 10Based on the microscopic magnified image, the identification and classification of neutrophils can be realized. It can not only perform qualitative analysis to know whether specific neutrophils are present, but also provide accurate information on the number of neutrophils per unit volume, the proportion of neutrophils in white blood cells, and the proportion of neutrophils in neutrophils, providing clinicians with more in-depth and richer information on neutrophils, breaking through the limitations of existing technologies and entering a more in-depth field of neutrophil analysis and identification.

[0087] In the present application, the inventors also propose a computing and processing device, which is used to execute the neutrophil multisegmented granulocyte parameter detection method including all or part of the above-mentioned method.

[0088] In the present application, the inventors also propose a data storage device, in which the data storage device contains program codes for executing all or part of the above-mentioned method for detecting parameters of polymorphonuclear granulocytes.

[0089] In the present application, the inventors also proposed a device for acquiring parameters of neutrophil multisegmented granulocytes, which runs part or all of the above-mentioned neutrophil multisegmented granulocyte parameter detection method.

[0090] Although the present invention is illustrated and described based on the preferred embodiment and several alternatives, the invention is not limited by the specific description in this specification. Other additional replacement or equivalent components can also be used to practice the present invention.

Claims

1. A method for detecting parameters of neutrophils with multisegmented nuclei, characterized in that: Blood samples were pre-processed to obtain microscopic specimens; Tiling of samples for microscopic examination; photographing the flattened microscopic sample to obtain an image of the microscopic sample; An AI recognition algorithm is used to identify the microscopic sample image, identify the neutrophil multisegmented granulocytes in the selected area S1 of the image, and obtain the total number NUMS1 of neutrophil multisegmented granulocytes in the selected image; the AI ​​recognition algorithm identifies the neutrophil multisegmented granulocytes based on the neutrophil multisegmented granulocyte feature data set.

2. The method for detecting neutrophil multisegmented granulocyte parameters according to claim 1, characterized in that: The neutrophil multi-segmented granulocyte feature dataset is obtained by manually annotating neutrophil multi-segmented granulocyte images and then training; the neutrophil multi-segmented granulocyte is a neutrophil with a lobed number greater than M or equal to M, and the minimum value of M is 2; Include any one of the following technical features: TC1: The target neutrophils are neutrophils from human blood, M=6; TC2: The target neutrophils are non-mammalian blood neutrophils, M=5; TC3: M is any natural number from 2 to 8; TC4: M is any natural number from 2 to 8; the number of cells corresponding to each neutrophil M segmented granulocyte is counted respectively.

3. The method for detecting neutrophil multisegmented granulocyte parameters according to claim 1, characterized in that: The neutrophil multisegmented granulocyte parameter is the number of neutrophil multisegmented granulocytes per unit volume; Include any one of the following technical features: TCC1: The blood sample is pre-treated by dilution and diluted N times; the microscopic sample tile height is H; the volume V corresponding to the total number of neutrophils NUMS1 = selected area S1 × H; the number of neutrophils per unit volume = N × the total number of neutrophils NUMS1 / blood sample volume V; TCC2: The blood sample is pre-treated for concentration and concentrated N times; the microscopic sample is tiled at a height of H; the volume V corresponding to the total number of neutrophils NUMS1 is equal to the selected area S1×H; the number of neutrophils per unit volume is equal to the total number of neutrophils NUMS1 / (blood sample volume V×N).

4. The method for detecting neutrophil multisegmented granulocyte parameters according to claim 1, wherein: The neutrophil multisegmented granulocyte parameter is the proportion of neutrophil multisegmented granulocytes in white blood cells, and also includes using an AI recognition algorithm to identify the microscopic sample image and identify the number of white blood cells NUMW1 in the selected area S1 of the image. The AI ​​recognition algorithm identifies white blood cells based on a white blood cell feature data set; the proportion of neutrophil multisegmented granulocytes in white blood cells = the total number of neutrophil multisegmented granulocytes NUMS1 / the number of white blood cells NUMW1.

5. The method for detecting neutrophil multisegmented granulocyte parameters according to claim 4, characterized in that: The white blood cell feature dataset is obtained by manually annotating white blood cell images and then training.

6. The method for detecting neutrophil multisegmented granulocyte parameters according to claim 1, characterized in that: The neutrophil multisegmented nuclear granulocyte parameter is the proportion of neutrophil multisegmented nuclear granulocytes in neutrophils, and also includes using an AI recognition algorithm to identify the microscopic sample image, identifying the neutrophils NUMW2 in the selected area S1 of the image, and the AI ​​recognition algorithm identifies white blood cells based on the neutrophil characteristic data set; the proportion of neutrophil multisegmented nuclear granulocytes in neutrophils = the total number of neutrophil multisegmented nuclear granulocytes NUMS1 / the number of neutrophils NUMW2.

7. The method for detecting neutrophil multisegmented granulocyte parameters according to claim 6, characterized in that: The neutrophil feature dataset is obtained by manually annotating neutrophil images and then training.

8. The method for detecting neutrophil multisegmented granulocyte parameters according to claim 1, characterized in that: The blood sample is pre-processed, including any one of the following technical features: TA1: Add dye to the blood sample to stain the white blood cells; TA2: dilute blood samples in the range of 1:50 to 1:500; TA3: Adding a liquid dye to the blood sample dilutes the blood sample and stains the white blood cell nuclei; TA4: Centrifuge the blood sample, let it settle, take the leukocyte layer to obtain an enriched sample, dilute and stain the enriched sample to obtain a microscopic examination sample; TA5: Add hemolytic agent to the blood sample to dissolve red blood cells, enrich white blood cells, obtain the enriched sample from the layer containing white blood cells, dilute and stain the enriched sample to obtain a sample for microscopic examination; TA6: Filter the blood sample to remove red blood cells and obtain white blood cells.

9. The method for detecting neutrophil multisegmented granulocyte parameters according to claim 8, characterized in that: In technical feature TA4, the centrifugation treatment of the blood sample includes: low-speed centrifugation at 100g to 1000g for 3 to 10 minutes; and the static sedimentation is static sedimentation for more than 20 minutes.

10. The method for detecting neutrophil multisegmented granulocyte parameters according to claim 8, characterized in that: In technical feature TA5, the hemolytic agent includes SDS, TWEEN20 and / or TWEEN80, and the enriched leukocytes are enriched by precipitation or centrifugation.

11. The method for detecting parameters of neutrophils with multisegmented nuclei according to claim 8, characterized in that: In technical feature TA6, a leukocyte filter membrane is used to filter the blood sample, wherein the pore size of the leukocyte filter membrane is larger than that of the red blood cells, and the pore size of the leukocyte filter membrane is smaller than that of the leukocytes.

12. A computing and processing device, characterized in that: Used to perform a method for detecting neutrophil multisegmented granulocyte parameters comprising all or part of any one of claims 1 to 11.

13. A data storage device, characterized in that: The data storage includes a program code for executing all or part of the method for detecting neutrophil multisegmented granulocyte parameters according to any one of claims 1 to 11.

14. A device for acquiring parameters of neutrophils with multisegmented nuclei, characterized in that: Used to run part or all of the method for detecting neutrophil multisegmented granulocyte parameters according to any one of claims 1 to 11.

15. A blood leukocyte parameter, characterized in that: Used to assess the number of neutrophils in the blood. The number of neutrophils per unit volume = the number of neutrophils in the blood sample volume V / blood sample volume V.

16. A blood leukocyte parameter, characterized in that: Used to evaluate the ratio of the number of neutrophils and multisegmented granulocytes in the blood to the number of white blood cells. The ratio of neutrophils and multisegmented granulocytes to white blood cells = the number of neutrophils and multisegmented granulocytes / the number of white blood cells.

17. A blood leukocyte parameter, characterized in that: Used to evaluate the ratio of the number of polymorphic neutrophils to neutrophils in the blood. The ratio of polymorphic neutrophils to neutrophils = the number of polymorphic neutrophils / the number of neutrophils.

18. The blood leukocyte parameter according to any one of claims 15 to 17, characterized in that: Including any one of the following technical features, TB1: The number of lobes M of the nucleus in the neutrophil multisegmented granulocyte is greater than or equal to 5 lobes; TB2: The number of lobes M of the nucleus in the neutrophil multisegmented granulocyte is greater than or equal to 2 lobes; TB3: The number of lobes M of the nucleus in the neutrophil multisegmented granulocyte is greater than 6 lobes or equal to 6 lobes.

Images