WATER-BASED CELL STAINING COMPOSITION, METHOD FOR PREPARING SAME AND USE THEREOF, CELL SAMPLE PREPARATION AND CELL COUNTING METHODS

Abstract

This invention patent relates, in a first aspect, to cell staining compositions comprising a mixture of dyes or salts and complexes thereof in an aqueous medium and optionally a cell lysis agent. Also disclosed are methods for preparing a staining composition and a lysis agent, besides the uses of the staining composition in methods for preparing cell samples. The compositions of the present invention make simultaneous cell staining and differentiation possible, allowing a direct differential count to be carried out, without requiring the stained sample to be smeared on a slide.

Claims

1-15. (canceled)

16. A cell dye composition comprising methyl green, pironin, and fluorescein, or salts and complexes thereof, in a predominantly aqueous medium.

17. The cell dye composition of claim 16, further comprising a cell lysing agent.

18. The cell dye composition of claim 16 comprising: 0.3 mM to 8 mM of methyl green; 0.3 mM to 8 mM of pironin; and 0.1 mM to 3 mM of fluorescein; wherein the methyl green, the pironin, and the fluorescein form a complex in a predominantly aqueous medium.

19. The cell dye composition of claim 17 comprising: 0.3 mM to 8 mM of methyl green; 0.3 mM to 8 mM of pironin; and 0.1 mM to 3 mM of fluorescein; wherein the methyl green, the pironin, and the fluorescein, and optionally the cell lysing agent, form a complex in a predominantly aqueous medium.

20. The cell dye composition of claim 16 comprising: an alcohol concentration of less than or equal to 35%.

21. The cell dye composition of claim 17 comprising: an alcohol concentration of less than or equal to 35%.

22. The cell dye composition of claim 17, wherein the cell lysing agent is a quaternary agent capable of disrupting membranes of red blood cells, previously diluted in aqueous medium, at concentrations from 25 g/L to 80 g/L and from 2 g/L to 9 g/L, respectively.

23. The cell dye composition of claim 19, wherein the cell lysing agent is a quaternary agent capable of disrupting the membranes of the red blood cells, previously diluted in aqueous medium, at the concentrations from 25 g/L to 80 g/L and from 2 g/L to 9 g/L, respectively.

24. A process for the preparation the cell dye composition of claim 16 comprising the steps of: a) preparing a Methyl Green-Pyronin solution comprising a concentration of from 0.3 mM to 8 mM of methyl green and from 0.3 mM to 8 mM of pironin, and having a pH of 3.8 to 5.5; and b) adding fluorescein to the Methyl Green-Pyronin solution in a ratio of from one part of the Methyl Green-Pyronin solution to two parts of fluorescein in a concentration of 0.1 mM to 3 mM to two and a half parts of Methyl Green-Pyronin solution to one part of fluorescein, obtaining a Methyl Green-Pyronin Fluorescein solution with a pH between 4.8 and 6.4.

25. A process for the preparation cell dye composition of claim 17 comprising the steps of: a) preparing a Methyl Green-Pyronin solution comprising a concentration of from 0.3 mM to 8 mM of methyl green and from 0.3 mM to 8 mM of pironin, and having a pH of 3.8 to 5.5; b) adding fluorescein to the Methyl Green-Pyronin solution in a ratio of from one part of the Methyl Green-Pyronin solution to two parts of fluorescein in a concentration of 0.1 mM to 3 mM to two and a half parts of Methyl Green-Pyronin solution to one part of fluorescein, obtaining a Methyl Green-Pyronin Fluorescein solution with a pH between 4.8 and 6.4; and c) adding a lysing agent in a ratio of from one part of the lysing agent to two parts of the Methyl Green-Pyronin Fluorescein solution to two parts of lysing agent to one part of the Methyl Green-Pyronin Fluorescein solution, wherein the final dyeing solution of Methyl Green-Pyronin Fluorescein with Lysing Agent remains at a pH between 5.5 and 6.6.

26. A method of preparing a sample of cells comprising: cell lysing, and differential staining of cell types in a single stage in aqueous medium.

27. The method of claim 26, comprising: mixing a sample of cells with a dye composition, wherein the dye composition includes methyl green, pironin, and fluorescein, or salts and complexes thereof, in a predominantly aqueous medium.

28. The method of claim 27, wherein permeabilization of the cell membrane by the dye composition occurs in the aqueous medium when mixing a sample of cells with the dye composition, enabling dye penetration and staining inside the cells.

29. The method of claim 27, wherein mixing a sample of cells with the dye composition is done at a ratio of one part of body fluid to 5 to 200 parts of the dye composition.

30. A method of counting cells comprising: a) mixing a sample of cells with the dye composition of claim 16; b) placing the sample mixture of cells with the dye composition in a cell counting apparatus; and c) counting of a total value of cell units, identifying and determining proportions of each cell subtype in a body fluid sample in a single step.

31. A method of counting cells comprising: a) mixing a sample of cells with the dye composition of claim 17; b) placing the sample mixture of cells with the dye composition in a cell counting apparatus; and c) counting of a total value of cell units, identifying and determining proportions of each cell subtype in a body fluid sample in a single step.

32. A cell counting method comprising: (i) imaging cells suspended in a dye composition; and (ii) global cell type counting and differential counting of cell subtypes in a single step.

33. The method of claim 30, wherein the counting is performed automatically by image processing, or non-automatically by an individual using a microscope.

34. The method of claim 30, further comprising a step of registering images of stained and lysed cells and storing the images for future evaluation.

35. A method of preparing a sample of cells for cell analysis comprising adding the dye composition according to claim 16.

Description

DESCRIPTION OF THE FIGURES

[0105] FIG. 1Agglutination of blood particles stained by dyes prepared in alcoholic medium, indicating (1) the lumps of blood stained in alcohol solution and (2) whole blood with EDTA in Guiemsa standard solution.

[0106] FIG. 2Formation of lumps of blood diluted and stained with Leishman in alcoholic medium and visualized in Neubauer chamber.

[0107] FIG. 3Absorbance spectra of the individual components of the proposed dye solution and the final solution.

[0108] FIG. 4Absorbance spectra of the final solution of MGPF(H) read on Sep. 19, 2014, six months after and fourteen months after that is when pironin degradation begins.

[0109] FIG. 5Eosinophils in blood distension stained with MGPF. (a) with magnification of 400; (b) with magnification of 1,000.

[0110] FIG. 6Basophil in blood distension stained with MGPF, at magnification of 1000.

[0111] FIG. 7Neutrophil samples, stained in liquid medium, stained in liquid medium with MGPF(H) and MGPF, obtained from whole blood and synovial fluids. Images obtained with magnification of 1000.

[0112] FIG. 8Lymphocyte samples, stained in liquid medium with MGPF(H) and MGPF, obtained from whole blood and synovial fluids. Images obtained with magnification of 1000.

[0113] FIG. 9Monocyte samples, stained in liquid medium, stained in liquid medium with MGPF(H) and MGPF, obtained from whole blood and synovial fluids. Images obtained with magnification of 1000.

[0114] FIG. 10Eosinophil samples, stained in liquid medium, stained in liquid medium with MGPF(H) and MGPF, obtained from whole blood and synovial fluids. Images obtained with magnification of 1000.

[0115] FIG. 11Basophil sample, stained in liquid medium with MGPF(H), obtained from whole blood. Images obtained with magnification of 1000.

[0116] FIG. 12Bland-Altman analysis comparing the global counts performed by automatic equipment in the laboratory of clinical analysis (LCA) and manual with MGPF(H).

[0117] FIG. 13Bland-Altman analysis comparing global counts in a Neubauer chamber with Turk's solution and MGPF(H)

[0118] FIG. 14Bland-Altman analysis comparing the neutrophil differentiating counts performed with an automatic equipment in the LCA and manual with MGPF(H).

[0119] FIG. 15Bland-Altman analysis comparing the differential counts of neutrophils in liquid medium stained with MGPF(H) and in blood distension stained with Leishman.

[0120] FIG. 16Bland-Altman analysis comparing the lymphocyte differentiating counts performed with an automatic equipment in the LCA and manual with MGPF(H).

[0121] FIG. 17Bland-Altman analysis comparing the differential counts of lymphocytes in liquid medium stained with MGPF(H) and in blood distension stained with Leishman.

[0122] FIG. 18Bland-Altman analysis comparing the monocyte differentiating counts performed with an automatic equipment in the LCA and manual with MGPF(H).

[0123] FIG. 19Bland-Altman analysis comparing the differential counts of monocytes in liquid medium stained with MGPF(H) and in blood distension stained with Leishman.

[0124] FIG. 20Bland-Altman analysis comparing the eosinophil differentiating counts performed with an automatic equipment in the LCA and manual with MGPF(H).

[0125] FIG. 21Bland-Altman analysis comparing the differential counts of eosinophils in liquid medium stained with MGPF(H) and in blood distension stained with Leishman.

[0126] FIG. 22RGB histograms obtained from neutrophil images. Wherein (A) are the histograms of R channel; (B) are the histograms of (G) channel; and (C) are the histograms of channel B.

[0127] FIG. 23RGB histograms obtained from lymphocyte images. Wherein (A) are the histograms of R channel; (B) are the histograms of (G) channel; and (C) are the histograms of channel B.

[0128] FIG. 24RGB histograms obtained from monocyte images. Wherein (A) are the histograms of R channel; (B) are the histograms of (G) channel; and (C) are the histograms of channel B.

[0129] FIG. 25RGB histograms obtained from eosinophil images. Wherein (A) are the histograms of R channel; (B) are the histograms of (G) channel; and (C) are the histograms of channel B.

[0130] FIG. 26RGB histogram obtained from basophil image. Wherein (A) are the histograms of R channel; (B) are the histograms of (G) channel; and (C) are the histograms of channel B.

EXAMPLES

[0131] These Examples are merely presented for illustration and should not, in any way, be construed as limiting the range and scope of the present invention.

Example 1. Preparation of the Dye Composition

[0132] In order to obtain a viable staining in liquid medium, the preparation of the dye solution and the lysing agent is divided into five steps. The purpose of splitting the preparation in stages is achieving a balanced dye solution that meets three requirements for achieving acceptable differential staining for diagnosis: 1) not producing coagulation or particle agglutination; 2) maintaining the integrity of the leukocytes for long enough in order to perform the characterizations and counts; 3) in the case of leukocyte staining from whole blood, allowing hemolysis, without the presence of important cell debris, which may disrupt the recognition and counting of leukocytes. The differential staining protocol in liquid medium is then applied:

[0133] Step 1: Preparation and Purification of MG:

[0134] In the first step, a buffer solution is prepared in partially isotonic aqueous medium so that it has a buffering power in the pH range of not less than 3.5 and not more than 5.5 in order to obtain the performance of the methyl green dye. The buffer solution should be adjusted for each body fluid type, so that the staining and counting requirements are met. The methyl green is then prepared in the partially isotonic aqueous buffer solution at a concentration of not less than 0.3 mM and not more than 8.0 mM, depending on the type of staining desired and the type of body fluid to be applied. Depending on the origin and formulation of the methyl green, it is necessary to carry out its purification. When purification is needed, the violet crystal contaminant, which is common in powder formulations of methyl green, should be removed. The purification protocol may vary according to the quality of the product that is provided by each manufacturer.

[0135] To carry out the purification of methyl green, successive washes of the solution in chloroform are made. Washing is carried out by adding chloroform to the methyl green solution. The violet crystals dissolve in chloroform by stirring. By leaving the solution still, the chloroform with violet crystals are separated in a phase, which is separated in a separating funnel. At the end, a purified solution having a pH of not less than 3.8 and not more than 5.5 should be obtained.

[0136] Step 2: Pironin Addition

[0137] In the second step, the addition of the dye pironin in the MG solution already purified in concentration of not less than 0.3 mM and not greater than 7.0 mM is carried out. After homogenization, the solution must be filtered to obtain a solution (MGP) with a pH of not less than 3.8 and not higher than 5.5.

[0138] Step 3: Fluorescein Preparation

[0139] Fluorescein is prepared in the third step, in a hot alcoholic medium, in a concentration of not less than 0.1 mM and not greater than 3.0 mM. After filtration, a solution having a pH of not less than 9 and not more than 10 is obtained.

[0140] Step 4: Preparation of the MGPF Solution with Addition of Fluorescein

[0141] In the fourth step the MGPF solution is obtained by mixing the MGP solution with the fluorescein solution in the ratio of not less than one part of MGP to two parts of fluorescein and not more than two and one-half parts of MGP to one part of fluorescein, thus obtaining the MGPF solution with a pH of not less than 4.8 and not more than 6.4. The mixing of the MGP solution with the fluorescein solution should be balanced in order to safely prevent coagulation or the formation of particulates agglutinated in the medium. The balance of the solution should be optimized for the body fluid in which it will be used. This solution is used for staining cells in blood distension or in liquid medium, but without haemolysis. At this step, the dye solution is already viable since it is possible to stain the leukocytes present in most body fluids that do not have too many red blood cells, such as urine, cerebrospinal fluid, cavity liquids, such as ascites, pleural, peritoneal and synovial; and food fluids, such as milk.

[0142] Step 5: Preparation of the Final MGPF(H) Solution with Addition of the lysing AGENT

[0143] Addition of the lysing agent is an alternative for leukocyte staining in whole blood. The lysing agent can be quaternary ammonium compound in the ratio of 5 g/L to 80 g/L associated with a buffer solution and diluted in aqueous medium. This solution is incorporated into the MGPF solution, yielding the MGPF(H) solution. The ratio of MGPF to the lysing agent may be carried out in the range of one part of the lysing agent to two parts of the staining solution up to the ratio of two parts of lysing agent to one part of the staining solution. The ratio should guarantee hemolysis without damage to the leukocytes and allowing their staining. The final dye solution of MGPF(H) should remain at pH not less than 5.5 and not higher than 6.6.

Example 2: Method of Preparing and Reading the Cells

[0144] As recommended by the usual technique of reading the amounts of leukocytes in liquid medium, a sample of whole blood is diluted in the dye solution in the ratio of one part of blood to 20 parts of the dye solution. By placing this solution in the Neubauer chamber, the stained cells are differently counted. Thus, counting occurs for both the overall value (total leucocytes per cubic millimeter of blood) and for the ratios of each cell type.

[0145] The complete protocol can be carried out in less than two minutes by collecting a precise volume of blood by digital puncture, and presents the same results as the counts performed by using an automated counter and through manual methodology, which takes up to 60 minutes until the result of each exam.

[0146] The same coloring solution, without the addition of the lysing agent (MGPF) also performs the blood distension staining in one step, but with time of action in the range of 15 to 30 minutes. The time of action depends on the conditions of pH, temperature and fixation technique that were used.

Example 3. Characterization and Evaluation of the Stability of Dye Solutions

[0147] The dye solutions which gave rise to the invention, which is the final solution named MGPF(H), were evaluated by spectrophotometry on two dates. The first assessment took place the day after its initial preparation on Sep. 19, 2014. The same solution was evaluated again, under the same conditions, six months, nine months, twelve months and fifteen months after its preparation. The chemical stability of the solutions, as well as their action on cell staining, remained unchanged until the twelfth month.

[0148] The chemical stability of the dye solutions was verified by absorption spectrophotometry, in two steps. In the first, all individual solutions of each dye were read in the spectrophotometer, in order to evaluate the individual composition of each component. Absorptions of the intermediate dye solutions and the final solution in two MGPF versions (without lysing agent) and MGPF(H) (with lysing agent) were also read. In order to know whether individual solutions react with each other and form other substances. In the second step the chemical stability over time was evaluated, that is, if there is degradation of the final dye solution after long periods of time. In the graph of FIG. 3, relative values of the absorptions of each solution, and of their combinations, as a function of the wavelength (between 350 nm and 750 nm) are presented. The relative values of the absorptions vary as a function of the dilutions, which were necessary for the spectrophotometer to operate within its working range. Therefore, the absorptions presented are in relative values. Analyzing the absorption curves shown in FIG. 3, it is shown that the MG in acetate buffer solution has an absorption band between 590 nm and 680 nm, with a peak at 640 nm. Pironin in the buffer solution absorbs in the band between 480 nm and 570 nm, with a peak at 505 nm. Alcoholic fluorescein absorbs in the range of 440 nm to 520 nm, with a peak at 490nm. However, it should be noted that to obtain this spectrum, both MG and pironin were diluted 25 fold, whereas fluorescein did not require dilution. When analyzing the absorbance of the combination of the two MG and P (MGP) cationic dyes, it is observed that there are absorption peaks at 640 nm (corresponding to the wavelength of the MG) and subsidiary peaks at 550 and 505 nm (corresponding to the P length wave). The same happens when we observe the combination of MG, P and F (MGPF) dyes. However, when the two dyes are associated with fluorescein, this should not be evident. The association of the three dyes was read in the presence of the lysing agent (MGPF (H) line) and without the presence of the lysing agent (MGPF line, which is under the MGPF (H) line). Comparing the MGPF and MGPF(H) lines, it is noted that there are no differences between them, which indicates that the presence of the lysing agent does not interfere with the solution. Thus, it was found that the staining activity of each dye alone was maintained and that each dye reacts with the cell structure or with the molecules with which it has affinity.

[0149] In addition, the MGPF (H) line is the overlap of 25 diluted MG lines with 25 diluted pironin (P). And as expected, the fluorescein line (F) (undiluted) cannot be observed in the graph overlap. However, no other peak appears on the absorption line of MGPF (H), indicating that no other substance was formed in the association. The individual absorption spectra and peaks are similar to those obtained by Tycho et al. (1976) and Horobim (2008). The absorbance of the lysing agent is negligible in this range of the spectrum, as observed in FIG. 3, it does not react with the other components of the solution, so it does not appear in the spectrum of the final solution.

[0150] The absorbance spectra of the first step were obtained on Sep. 19, 2014. For the second step, a reading was performed in the same spectrophotometer, under the same conditions of temperature and relative humidity on Mar. 6, 2015, of the final solution. The spectra comparing the final solution read on Sep. 19, 2014, with the same solution read on Mar. 6, 2015, are shown in FIG. 4.

Example 4. Application of the Dyeing Solution and Method of Preparation and Counting

[0151] Examples of the staining results are shown in FIGS. 5 and 6, where it is also possible to observe that a great advantage of this dye solution is the absence of artifacts, fragments or lumps of lysed erythrocytes, which could impair the counting by diverting attention from the professional, or in the case of image processing, simplification in the segmentation step. It is noteworthy that, in liquid medium, the staining did not exceed more than one minute, besides the time inherent to making the slide or the chamber, totaling two minutes.

Example 5. Evaluation of Cells and Cell Structures in the Images Obtained in Blood Distentions Stained with MGPF

[0152] Eosinophils

[0153] Eosinophils were stained with the MGPF dye solution, as observed in FIGS. 5(a) and 5(b). The slides containing the blood distensions were previously fixed in methanol, which facilitates the entry of fluorescein dye into the cells and the staining of the granules characteristic of this cell type.

[0154] As reported by Stankiewicz et al. (1996) the available methods do not allow visualization of the eosinophilic granules in viable cells. In most techniques, eosinophils are stained after being fixed and treated by fixatives that can change the appearance and cell function. Thus, the ability to observe viable eosinophils with their stained granules can provide important information about the activities of these cells. The complete recognition of eosinophils depends on good staining by fluorescein.

[0155] Basophils

[0156] When basophils are found in MGP stained distension, as it is shown in FIG. 6, the granules stain pink due to their histamine content and mainly due to the presence of heparin, a glycosaminoglycan-type heteropolysaccharide (Andra Filho, 2009), since they have affinity with pironin (TYCHO et al, 1976).

[0157] This morphological and staining behavior also occurs in liquid medium, since the pironin component of the MGPF dye solution showed that it satisfactorily penetrates the cell, for example, demonstrating the RNA content in the lymphocytes stained in liquid medium. Thus, it marks the heparin content of basophils by their affinity.

Example 6. Evaluation of Cells and Cell Structures in the Images Obtained in Liquid Medium

[0158] To exemplify the results obtained with this invention, peripheral blood leukocytes were identified, classified and counted in five morphological types: neutrophils, lymphocytes, monocytes, eosinophils and basophils.

[0159] To exemplify the results obtained with this invention, leukocytes present in other body fluids: urine, liquor (cerebrospinal fluid), cavity liquids, such as ascites, pleural, peritoneal and synovial, were also characterized and categorized by means of their images. For purposes of characterization, classification and counting, the images of the cells contained in these liquids present the same characteristics and information as the images of the cells contained in the blood. In order to obtain adequate staining in these liquids, it was necessary to adjust the pH and/or concentration of the final solution as well as centrifugation of the liquids for concentration of the cell volume. The staining occurs in the presence of the two solutions: MGPF and MGPF(H).

[0160] The observation and description of the cells below corresponds to that found in the liquid readings between slide and coverslip, stained with MGPF(H) or MGPF using a 1000 magnification.

[0161] Cells were visualized in their characteristics spherical shape and size, showing no rupture of cell membranes. The images presented in FIGS. 7 to 11 had their relative sizes preserved, i.e., considering that all images had the same magnification (1000), the areas cropped around the cell were always the same (500500 pixels). Therefore, the cell area is proportional to the size of the cutting window.

[0162] Neutrophils

[0163] In this cell, the cytoplasm was light, clear and without granulation. The nucleus appeared in its characteristic segmentation form. Both rods and polymorphonuclear cells were classified as neutrophils. Nuclei with 2 to 4 segmentations could be observed, as it can be observed in the examples presented in FIG. 7. Due to the affinity of the MG component with DNA, the nuclear segments became bluish-green.

[0164] Lymphocytes

[0165] It can be observed in this cell the characteristic of high nucleus cytoplasm ratio, with evident nuclear observation stained in intense blue, as it is seen in FIG. 8. During the counts, it was possible to observe sharp nucleoli in some cells, as seen in FIG. 8(d). The nucleoli presented a reddish coloration that outlined the nucleus, which can be related to the affinity of the pironin for RNA, present in such structure. In slides of Giemsa stained blood distensions, this finding is not common (BEUTLER, 2001). Cytoplasmic bands stained with different pink intensities confirmed the affinity of the pironin component of the MGPF(H) or MGPF solution for RNA, characterizing differences in cell activity. No cytoplasmic granules were observed.

[0166] Monocytes

[0167] These cells were shown with a wide variety of nuclear and cytoplasmic forms, as it is shown in FIG. 9. Generally, they appeared as large cells, confirming their description as read in RAWAT et al. (2015). The nuclei were visualized in several forms, ranging from rounded, reniform, oval, lobed and even folded. Nucleoli were not evidenced. The cytoplasm sometimes appeared vacuolated and with a fine granulation stained in loose reddish-pink, by the affinity with pironin, since it represents its RNA content. This staining tends to be less intense than in reactive lymphocytes, which may aid in the visual differentiation between monocytes and these lymphocytes, corroborating the descriptions presented by Beutler (Beutler, 2001).

[0168] Eosinophils

[0169] The identification of these cells took into account some of their normally reported characteristics, such as cell size, considered to be medium between granulocytes: neutrophils and basophils, the low nucleus cytoplasm ratio, the nucleus location, usually eccentric, with bilobular segmentation most often and absence of visible nucleoli, as observed in FIG. 10. The yellowish cytoplasmic staining, produced by fluorescein, and the bluish-green staining produced by MG, weaker than the nucleus, differentiates these cells from the neutrophils.

[0170] Basophils

[0171] The description of the morphological and staining characteristics of this cell type, stained with MGPF(H) or MGPF in liquid medium, are very similar to those found in the classic literature, as shown in FIG. 11. Due to the large amount of heparin rich cytoplasmic granules, which the basophil has and the affinity of the pironin component for the glycosaminoglycan, the cells referenced are richly stained in intense pink, almost disappearing with the nucleus, which is slightly bluish under the substantially colored cytoplasm. In a more detailed testing, it can be seen that the cytoplasm staining refers to the large granules, which are not as dark as in the classical staining of Romanvisk. The diversity of characteristics in relation to other cells and their scarcity, less than 1% of total leukocytes, contributes to this classification.

Example 7: Evaluation of Dye Solution by Comparing Counts of Cells with Established Protocols and with the New Dye

[0172] In order to evaluate the applicability of the MGPF(H) dye solution as a differential leukocyte detection instrument, it is necessary to determine if the counts performed are compatible with the counts performed by the established methodologies. Then, global counting and differential counting were evaluated by ANOVA variance analysis and by Blend-Altman agreement analysis.

[0173] Global Count

[0174] For each blood sample, three global counts were performed, one by the Laboratory of Clinical Analysis (LAC), the second made by the inventors in a Neubauer chamber with Turk's solution and the third made by the inventors using MGPF(H).

[0175] The results of these counts are presented in table 1. Next to the table the results obtained by ANOVA analysis are presented. FIGS. 12 and 13 show the Blend-Altman analyzes for MGPF(H)LCA and MGPF(H)Turk's solution. No significant statistical differences were found between counting techniques, through ANOVA and Blend-Altman, thus disclosing that they are equivalent.

TABLE-US-00001 TABLE 1 Result of the global leukocyte count by three techniques: Automatic, performed by an independent Clinical Analysis Laboratory. Manual in Neubauer chamber with Turk's solution. Manual in Neubauer chamber with MGPF(H). Values expressed in cells/mm.sup.3. Sample LCA turk MGPF (H) Patient 1 7000 7750 7400 Patient 2 5000 5075 4875 Patient 3 7300 8150 8200 Patient 4 7000 7825 6100 Patient 5 6800 6725 6800 Patient 6 6800 6588 6867 Patient 7 5600 6750 6650 Patient 8 5800 7300 7350 Patient 9 6100 5825 4400 Patient 10 6800 6450 7650 Patient 11 5600 5450 5425 Patient 12 5200 5500 6250 Patient 13 5500 5550 6125 Patient 14 7000 6900 7725 Patient 15 6700 6900 6700

[0176] The results of the one-way ANOVA test were: p=0.62, with critical F greater than 3.35, and F was 0.50. According to the test, if the p value is greater than 0.05 and F is smaller than the critical F, the counts are not statistically different, that is, they can be considered equivalent.

[0177] Differential Counting

[0178] For each blood sample, differential counts of leukocytes were performed by three techniques, one automated by LCA, the second by the inventors, manually using MGPF (H) and the third by the inventors, carried out manually, in Leishman-stained blood distension.

[0179] Neutrophils

[0180] The results of these counts are presented in table 2. Next to the table the results obtained by ANOVA test are presented. FIGS. 14 and 15 show the Blend-Altman analyzes for LCAMGPF(H) and LeishmanMGPF(H). No significant statistical differences were found between counting techniques, through ANOVA and Blend-Altman, thus disclosing that they are equivalent.

TABLE-US-00002 TABLE 2 Result of the neutrophil differential counting by three techniques: Automatic, performed by an independent Clinical Analysis Laboratory. Manual on Leishman-stained blood distension. Manual with MGPF(H). Values expressed in total number of cells/mm3. Sample LCA MGPF (H) Leish. Patient 1 4200 4292 4844 Patient 2 2700 2876 2538 Patient 3 3600 4510 3994 Patient 4 4100 3386 4108 Patient 5 3400 3706 3127 Patient 6 3400 3948 2866 Patient 7 2800 3126 3476 Patient 8 3500 4778 4782 Patient 9 2700 2244 2825 Patient 10 4000 4093 3838 Patient 11 3000 3038 2616 Patient 12 2300 2875 2420 Patient 13 3200 3246 3053 Patient 14 4500 4790 4002 Patient 15 3900 3618 3864

[0181] The results of the one-way ANOVA test were: p=0.77, with critical F greater than 2.6, and F was 0.27. The p value much greater than 0.05 and F lower than the critical F indicates that the counts do not present statistical differences, that is, they can be considered equivalent.

[0182] Lymphocytes

[0183] The results of these counts are presented in table 3. Next to the table the results obtained by ANOVA test are presented. FIGS. 16 and 17 show the Blend-Altman analyzes for LCAMGPF(H) and LeishmanMGPF(H). No significant statistical differences were found between counting techniques, through ANOVA and Blend-Altman, thus disclosing that they are equivalent.

TABLE-US-00003 TABLE 3 Result of the lymphocyte differential counting by three techniques: Automatic, performed by an independent Clinical Analysis Laboratory. Manual on Leishman-stained blood distension. Manual with MGPF(H). Values expressed in total number of cells/mm3. Sample LCA MGPF (H) Leish. patient 1 2100 2331 2170 patient 2 1600 1487 1802 patient 3 2700 2952 2893 patient 4 2200 2044 2465 patient 5 2400 2482 2354 patient 6 2700 2575 2734 patient 7 1900 2727 2329 patient 8 1700 1985 2044 patient 9 2500 1716 2126 patient 10 2100 2525 1871 patient 11 1900 1845 2071 patient 12 2400 2813 2475 patient 13 1700 2450 1943 patient 14 1800 2240 2139 patient 15 2000 2345 2415

[0184] The results of the one-way ANOVA test were: p=0.51, with critical F greater than 2.6, and F was 0.73. The p value much greater than 0.05 and F lower than the critical F indicates that the counts do not present statistical differences, that is, they can be considered equivalent.

[0185] Monocytes

[0186] The results of these counts are presented in table 4. Next to the table the results obtained by ANOVA test are presented. FIGS. 18 and 19 show the Blend-Altman analyzes for LCAMGPF(H) and LeishmanMGPF(H). No significant statistical differences were found between counting techniques, through ANOVA and Blend-Altman, thus disclosing that they are equivalent.

TABLE-US-00004 TABLE 4 Result of the monocyte differential counting by three techniques: Automatic, performed by an independent Clinical Analysis Laboratory. Leishman-stained blood distension. Manual stained with MGPF(H). Values expressed in total number of cells/mm3. Sample LCA MGPF (H) Leish. Patient 1 500 703 698 Patient 2 400 439 431 Patient 3 600 738 774 Patient 4 400 458 861 Patient 5 600 612 807 Patient 6 400 240 725 Patient 7 600 632 776 Patient 8 500 441 402 Patient 9 600 308 583 Patient 10 600 995 613 Patient 11 600 488 654 Patient 12 500 563 550 Patient 13 400 429 444 Patient 14 500 541 690 Patient 15 400 603 345

[0187] The ANOVA test results were: p=0.10, with critical F greater than 2.45, and the obtained F was 2.3. The p value much greater than 0.05 and F lower than the critical F indicates that the counts do not present statistical differences, that is, they can be considered equivalent.

[0188] According to Koepke (1977) apud England (1979), monocyte counts show a high variability, due in large part to an expected variation, but mainly due to differences in counting techniques, since normal or reactive lymphocytes may be erroneously identified as monocytes, making the monocyte count sometimes unsatisfactory. Another source of interference in the statistical analysis are the results presented by the LCA, which occur in multiples of 100. Assuming that a reading error of about 50 cells may occur and given that this error represents a significant percentage of the total cells, the statistical results tend to diverge from the expected values.

[0189] Eosinophils

[0190] The results of these counts are presented in table 5. FIGS. 20 and 21 show the Bland-Altman analyzes for LCAMGPF(H) and LeishmanMGPF(H). However, due to the low number of cells available per patient and the fact that the LCA maintains the count in multiples of 100, it is not possible to perform an appropriate statistical analysis. Then, the counts data are presented based on the characterization and visual recognition of the cells by experienced hematologists.

TABLE-US-00005 TABLE 5 Result of the eosinophil differential counting by three techniques: Automatic, performed by an independent Clinical Analysis Laboratory. Leishman-stained blood distension. Manual with MGPF(H). Values expressed in total number of cells/mm3. Sample LCA MGPF (H) Leish. Patient 1 200 37 38,8 Patient 2 300 73 304 Patient 3 300 200 448 Patient 4 300 213 352 Patient 5 400 350 403 Patient 6 300 103 197 Patient 7 300 166 168 Patient 8 100 147 73 Patient 9 300 132 291 Patient 10 100 38 129 Patient 11 100 54 109 Patient 12 100 22 55 Patient 13 100 64 111 Patient 14 200 154 69 Patient 15 200 134 276

[0191] Basophils

[0192] Basophils are rare cells, and their count usually results in null values in most blood counts. As LCA maintains the count in multiples of 100, the result for most patients was null for absolute values, and non-zero values were presented in the relative results (as a percentage of total leukocytes), as shown in table 6. Thus, Bland-Altman's statistical analysis or concordance analysis is meaningless. Then, the counts data are presented based on the characterization and visual recognition of the cells by experienced hematologists.

TABLE-US-00006 TABLE 6 Result of the monocyte differential counting by three techniques: Automatic, performed by an independent Clinical Analysis Laboratory. Leishman-stained blood distension. Manual with MGPF (H). Values expressed as a percentage of total leukocytes. Sample LCA MGPF (H) Leish. Patient 1 0.4 0.5 0.0 Patient 2 0.6 0.2 0.0 Patient 3 0.5 0.3 0.5 Patient 4 0.3 0.1 0.5 Patient 5 0.2 0.1 0.5 Patient 6 0.2 0.0 1.0 Patient 7 0.3 0.2 0.3 Patient 8 0.4 0.3 0.2 Patient 9 0.4 0.2 0.3 Patient 10 0.4 0.2 0.0 Patient 11 0.5 0.3 0.3 Patient 12 0.3 0.2 0.2 Patient 13 0.3 0.3 0.3 Patient 14 0.2 0.1 0.2 Patient 15 0.6 0.5 0.5

Example 8: Evaluation of the Effectiveness of the Method of Cell Classification and Counting by Image Processing

[0193] In order to evaluate the effectiveness of this automatic leukocyte classification method, 226 cell images were processed, containing samples of the five leukocyte cell types and 20 non-cell artifacts.

[0194] The histogram metrics, such as: width of the bases of intensities; relative position of frequency peaks; order of occurrence of the first and last valid frequencies in Red (Red), Green (Green) and Blue (Blue) (RGB), that is, greater than 0 (Zero); peak shift to the right or left, allow the characterization of the cell types.

[0195] The same images were evaluated by experienced biomedicians in leukocyte classification. The comparative results of classification and automatic counting with biomedical classification and counting are presented in table 7, whose R2 correlation between biomedical and software classification is equal to 0.998, showing that the counts are equivalent.

TABLE-US-00007 TABLE 7 Comparison of leukocyte counts performed by biomedical and image processing software. Linear correlation (R.sup.2) equal to 0.998. Biomedical Software Classification Classification Neutrophil 97 96 Lymphocyte 65 63 Monocyte 35 38 Eosinophils 8 7 Basophils 2 2 NC 19 20 Total 226 226

[0196] FIG. 20 shows the uniformity of neutrophil histogram behavior, which is different from the uniformity of the histograms obtained from lymphocyte images (FIG. 21), from monocyte images (FIG. 22), from eosinophil images (FIG. 23) and from basophil images (FIG. 24).

REFERENCES

[0197] ADOLLAH, R.; MASHOR, M. Y.; NASIR, N. F. M.; ROSLINE, H.; MAHSIN, H.; ADILAH, H. Blood cell image segmentation: a review. In: 4th Kuala Lumpur International Conference on Biomedical Engineering 2008. Springer Berlin Heidelberg, 2008. pages 141-144.

[0198] ANDRO FILHO, Newton. Challenges in the quality of heparin. Revista Brasileira de Hematologia e Hemoterapia, v. 31, n. 5, pages 306-307, 2009.

[0199] BAIN B. J. Clulas Sanguineas: Um Guia Prtico. 4th ed. Porto Alegre: Artmed, 2007. 487 p.

[0200] BANCROFT, J. D.; SUVARNA, S. K.; LAYTON, C. Bancroft's Theory and Practice of Histological Techniques. 7th ed. Elsevier, 2013. 654 p.

[0201] BASTEN, Antony; BOYER, Markley H.; BEESON, Paul B. Mechanism of eosinophilia I. Factors affecting the eosinophil response of rats to Trichinella spiralis. The Journal of experimental medicine, v. 131, n. 6, pages 1271-1287, 1970.

[0202] BEUTLER Ernest, LICHTMAN Marshall A., COLLER Barry S., KIPPS Thomas J., SELIGSOHN Uri. Willians Hematology. 6th ed. McGraw-Hill, Medical Publishing Division. 2001. 2047 p

[0203] CARRERAS, Maria Cecilia et al. Kinetics of nitric oxide and hydrogen peroxide production and formation of peroxynitrite during the respiratory burst of human neutrophils. FEES letters, v. 341, n. 1, p. 65-68, 1994.

[0204] DE LA RUE, Mario Luiz. Eosinofilia devida a parasitas. RBAC, v. 33, n. 4, p. 221-223, 2001.

[0205] DUCREST, S. et al. Flowcytometric analysis of basophil counts in human blood and inaccuracy of hematology analyzers. Allergy, v. 60, n. 11, p. 1446-1450, 2005.

[0206] ENGLAND, J. M. Prospects for automated differential leucocyte counting in the routine laboratory. Clinical & Laboratory Haematology, v. 1, n. 4, p. 263-273, 1979.

[0207] FALCONE, Franco H.; HAAS, Helmut; GIBBS, Bernhard F. The human basophil: a new appreciation of its role in immune responses. Blood, v. 96, n. 13, p. 4028-4038, 2000.

[0208] GROTTO, Helena Z W. O hemograma: importncia para a interpretao da bipsia. Rev. bras. hematol. hemoter, v. 31, n. 3, p. 178-182, 2009.

[0209] HOROBIN, R. W.; KIERNAN, J. A. CONN'S Biological Stains. 10. ed. New York: Taylor&Francis, 2008. 555 p.

[0210] KIERNAN, John Alan. Histological and histochemical methods: theory and practice. Shock, v. 12, n. 6, pages 479, 1999.

[0211] KOEPKE, J. A. A delineation of performance criteria for the differentiation of leukocytes. American journal of clinical pathology, v. 68, n. 1 Suppl, p. 202-206, 1977. apud ENGLAND, J. M. Prospects for automated differential leucocyte counting in the routine laboratory. Clinical & Laboratory Haematology, v. 1, n. 4, p. 263-273, 1979.

[0212] KURIMOTO, Yoshiyuki; DE WECK, A. L.; DAHINDEN, C. A. Interleukin 3-dependent mediator release in basophils triggered by C5a. The Journal of experimental medicine, v. 170, n. 2, p. 467-479, 1989.

[0213] LEISHMAN, W. B. Note on a simple and rapid method of producing Romanowsky staining in malarial and other blood films. British medical journal, v. 2, n. 2125, p. 757, 1901.

[0214] LIMA, Adriana; et al. Efeito do Solvente sobre as Propriedades Espectroscpicas do Azul de Metileno. In: XI Encontro Latino Americano de Iniciao Cientfica e VII Encontro Latino Americano de Ps-GraduaoUniversidade do Vale do Paraba. 2007

[0215] MAHAJAN, S.; GOLATI, S. S.; MESHRAM, A.; JICHLKAN, N. Review: Detection of Types of Acute Leukemia. International Journal of Computer Science and Mobile Computing, v.3, n. 3, p. 104-111, March 2014.

[0216] MARIEB, E. N. Essentials of human anatomy and physiology, ed 10, New York: Pearson, 2011.

[0217] MARSHALL, P. N.; BENTLEY, S. A.; LEWIS, S. M. A standardized Romanowsky stain prepared from purified dyes. Journal of clinical pathology, v. 28, n. 11, p. 920-923, 1975.

[0218] McEwen, B. J. Eosinophils: A review. Vet. Res. Communic. 16: 11-42, 1992. apud Mrio Luiz de la Rue; RBAC, vol. 33(4): 221-223, 2001)

[0219] MIRI, Stanislav; JORGOVANOVI, Nikola. Automatic classification of leukocytes. Journal of automatic control, v. 16, n. 1, p. 29-32, 2006.

[0220] NASCIMENTO, V. A. 0. ; MORAES T. F.; OLIVEIRA H. J. Q.; Colorao de leuccitos em meio lquido para aquisio e diferenciao por processamento de imagens. In: XXIV Congresso Brasileiro de Engenharia BiomdicaCBEB UberlndiaMG. 2014

[0221] OLIVEIRA, Raimundo Antonio Gomes. Hemograma: como fazer e interpretar. So Paulo: LPM, 2007. 505 p.

[0222] ORNSTEIN, LEONARD; ANSLEY, HUDSON R. Spectral matching of classical cytochemistry to automated cytology. Journal of Histochemistry & Cytochemistry, v. 22, n. 7, p. 453-469, 1974.

[0223] OVALLE, W.; NAHIRNEY, P. C. Netter Bases da Histologia 2nd ed. Elsevier, 2014. 536 p

[0224] PATERNITI, Irene et al. PDE 7 inhibitors: new potential drugs for the therapy of spinal cord injury. PLoS One, v. 6, n. 1, p. e15937, 2011.

[0225] PERRY, SEYMOUR; REYNOLDS, JOHN; BAKER, MARY. Methyl-green-pyronin as a differential nucleic acid stain for peripheral blood smears. Blood, v. 11, n. 12, p. 1132-1139, 1956.

[0226] PIURI, Vincenzo; SCOTTI, Fabio. Morphological Classification of Blood Leucocytes by Microscope Image. In: CIMSA 2004-IEEE International Conference on Computational Intelligence for Measurement Systems and Applications. 2004.

[0227] POGGERE, Paula Andreia et al. Azul de Metileno: Propriedades e Tratamentos. Anais do III ENDICT-Encontro de Divulgao Cientfica e Tecnolgica. Universidade Tecnolgica Federal do Paran-Cmpus Toledo, 2011. www.utfpr.edu.br/toledo

[0228] PYRL, S.; Indicators of Inflammation in the Diagnosis of Mastitis. Veterinary Research, v.34, p. 565-578, 2003.

[0229] RAWAT, J.; BHADAURIA, H. S.; SINGH, A.; VIRMANI, J. Review of Leukocyte Classification Techniques for Microscopic Blood Images. In:, 2015 2nd International Conference on Computing for Sustainable Global Development (INDIA Com) p 1948-1954, 2015

[0230] ROSENFELD, Ricardo. Complete blood count. J. bras. patol. med. lab, v. 48, n. 4, p. 244-244, 2012.

[0231] ROWLEY, Anne H. et al. RNA-containing cytoplasmic inclusion bodies in ciliated bronchial epithelium months to years after acute Kawasaki disease. PLoS ONE 3(2):. 2008.

[0232] STANKIEWICZ, M. et al. Supravital staining of eosinophils. International journal for parasitology, v. 26, n. 4, p. 445-446, 1996.

[0233] SPRY, C. J. F. Synthesis and secretion of eosinophil granule substances. Immunology today, v. 6, n. 11, p. 332-335, 1985. apud De La Rue; RBAC, vol. 33(4): 221-223, 2001.

[0234] SPRY, C. J. F. Eosinophyl leucocytes p.329-332. In: MCGEE, James O.'D.; ISAACSON, Peter G.; WRIGHT, Nicholas A. (Ed.).Oxford Textbook of Pathology: v. 2a,b. Pathology of systems. Oxford University Press, 1992.

[0235] TYCKO, D. H. et al. Automatic leukocyte classification using cytochemically stained smears. Journal of Histochemistry & Cytochemistry, v. 24, n. 1, p. 178-194, 1976.

[0236] WITTEKIND, D. On the nature of Romanowsky dyes and the Romanowsky-Giemsa effect. Clinical & Laboratory Haematology, v. 1, n. 4, p. 247-262, 1979.