Classifications

• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
  • G01N15/10—Investigating individual particles
  • G01N15/14—Optical investigation techniques, e.g. flow cytometry
  • G01N15/1468—Optical investigation techniques, e.g. flow cytometry with spatial resolution of the texture or inner structure of the particle
  • G01N15/147—Optical investigation techniques, e.g. flow cytometry with spatial resolution of the texture or inner structure of the particle the analysis being performed on a sample stream
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
  • G01N15/01—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials specially adapted for biological cells, e.g. blood cells
  • G01N2015/016—White blood cells
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
  • G01N15/10—Investigating individual particles
  • G01N15/14—Optical investigation techniques, e.g. flow cytometry
  • G01N2015/1477—Multiparameters

Definitions

• the present invention relates to the general field of biological fluid analysis and more particularly to the field of automated systems for analyzing biological fluid.
• the invention relates to methods of classifying cell populations by counting and discriminating from a data processing as derived from a biological fluid analysis device. Such a method is intended to be used in an analysis automaton.
• Flow cytometry is one of the techniques adapted to the statistical study of cell populations because the cells are studied one by one on a sample of several hundreds or thousands of cells.
• a dye or a fluorescent agent more commonly known as "molecular probe”
• information relating to these cells is accessible to the biologist.
• molecular probes combining, according to the principle of surface antigens, an antibody and a luminophore makes it possible to reveal specific functions located on the surface of cell membranes.
• the principles of flow cytometry are as follows.
• the microscopic objects to be analyzed are transported by a liquid vein at the focusing point of a beam of light, usually a laser.
• Detectors are positioned along specific axes of sight for the purpose of collecting the light-matter interaction signals.
• a first detector placed in the vicinity of the axis of the incident laser beam, measures the diffraction at small angles: generally, it is sized to be sensitive to low spatial frequencies, ie to the volume of the particle and its index of refraction.
• the direct laser beam that has not interacted with this particle is blocked by a mask.
• detectors may be placed at 90 ° to the axis of the incident beam.
• the detected light is analyzed according to one or more spectral components corresponding to the fluorescence or diffraction lights.
• an electronic gate consists of passing each biological cell through a very small orifice. This orifice is traversed by a constant current whose intensity is modulated by the variation of electrical resistance induced by the transit of the particle in the said orifice. This signal is approximately proportional to the volume of the cell.
• the electronic gate can also be ac powered according to US 4, 791, 355.
• Hematology analyzers also include an optical channel for measuring the absorbance of a particle passing through the measurement vessel.
• a hematology analyzer for example, is to count the different cells present in a blood sample, to differentiate these cells and thus to be able to give a proportion of each of the cell classes with respect to the whole sample. .
• the interpretation of measurements from a blood cell counter operating on whole blood and allowing the counting and differentiation of cell populations requires, in the vast majority of cases, a graphical representation in the form of a matrix in two dimensions.
• the populations represented are identified according to two physical parameters which are either optical, electrical, or both.
• the LMNE matrix which is the standard representation of sub-leukocyte populations on HORIBA ABX (5 diff) instruments, the two measurements used are absorbance and resistivity.
• This matrix is called LMNE because it allows the differentiation and counting of Lymphocytes, Monocytes, Neutrophils and Eosinophils, which are the white blood cell populations, or leukocytes, normally present in the blood.
• visualization on a two-dimensional screen is traditionally performed by selecting two variables. This consists of making an orthogonal projection in terms of these two variables.
• the main purpose of the present invention is thus to overcome such drawbacks by proposing a method intended to be used in a biological fluid analysis automaton capable of detecting cells in the liquid and to determine a n-tuple comprising at least four parameters.
• physical (n> 3) for each detected cell said method being intended for classification by discrimination and counting in at least one set of cell classes and their representation, classification and representation being advantageously adapted to the detection of pathology signatures ( s), and comprising the steps of: a) storing, beforehand, a plurality of mathematical transformations T of a plurality of tuples in m-tuples, m ⁇ n, each transformation, associated with a particular classification of the elements n- uplets to within a predetermined set of cell classes and determined based on statistical knowledge on the cells constituting these cell classes, making it possible to place the cell classes of a biological fluid having the average statistical characteristics in distinct zones of the composite space at m, the plurality of stored transformations advantageously allowing the obtaining of various placements of the
• display medium is intended to mean a computer screen, a paper medium, or any other means of visual representation, whether it is an integral part of the device or a deportee.
• tag is understood to mean a color, a pictogram or any other graphic element making it possible to dissociate visually the corresponding n-tuples from distinct classes or cell classes.
• space dynamic 3D
• steps f) to h) corresponds to a sequence of several steps of a so-called "classification" algorithm, each of these steps relating to a set of classes considered, and comprising at least one transformation that defines a composite space and a filter of discrimination and reclassification of the elements observed in at least two classes.
• Each algorithm step one for each pair of cell classes to be discriminated, updates the class of each selected particle as the corresponding n-tuple.
• the method according to the invention offers a more precise discrimination result by integrating all the physical parameters measured on the analyzer, for example the following parameters FSC, SSC, FL1, RES.
• the compensation allows to subtract in the intensity of a given fluorescence, the influence of other fluorescences. This is traditionally used only in the case of fluorescence.
• the method according to the invention makes it possible to remove or add signal proportions in an algorithmic manner to obtain a better discrimination of the populations of interest.
• this characteristic can be applied in particular to the results of morphological measurements (SSC, FSC, RES ...) -
• the invention makes possible an improved representation of subpopulations of white blood cells in the form of a two-dimensional matrix.
• the invention identifies the cells according to their maturation and their physical characteristics, nucleus and cytoplasm, with the assurance that all the cell populations are distinct from each other in order to classify them in cell classes and that the recovery phenomena are weak, especially in the absence of pathologies. This therefore allows an easy interpretation for those skilled in the art.
• the n input variables are merged to create a m-dimensional composite space. In this space, it is then possible to make two-dimensional projections relevant for what we want to show, possibly after making relevant rotations. An area of interest can then be formatted, in particular by zoom and offset functions. Composite space and formatting are suitable for viewing on a projection or for automatic discrimination of classes of items covered by the classification. Discrimination can be achieved by looking for peaks or valleys on a histogram.
• the method according to the invention has already made it possible to develop automatic classification algorithms for leukocytes and has provided powerful visualization means for all the leucocyte classes or certain cellular features, for example in view of detection of several subpopulations of lymphocytes.
• the invention uses statistics on the physical parameters of the relevant cellular classes based on upstream variance / covariance matrices. Stored transformations are linear or nonlinear.
• the physical parameters are values RES, FSC, FL1, SSC.
• value is meant the maximum height of the pulse on the RES, FSC, FL1 or SSC channel.
• a functional transformation Tn-> m linear or nonlinear, which consists in transforming the measurement vector of dimension n (n> 3) into a composite vector of dimension m ⁇ n by: F: R n -> R m
• x is the initial measure vector of dimension n, otherwise denoted xi (Xl ... Xn) and y is the image vector which represents l element yi (Yl ... Ym) in the vector space composite norm of dimension m.
• the transformation is therefore characterized by these two matrices A and C and these two vectors B and D. If the matrix C is zero and all the elements of D are non-zero, then we are in a linear case.
• This characteristic makes it possible to ensure a distribution of the different cell classes into distinct zones of the composite space since all the factors in the matrices A and C and in the vectors B and D are determined using statistical knowledge on the cell classes observed in a normal biological fluid.
• the transformations in m-dimensional space are then such that, in the presence of normal blood, the point clouds of the different cellular classes are located in distinct zones. It is thus clearly understood that, once a biological fluid has an extraordinary composition, for example during pathology, the distribution into distinct and / or bounded zones will not be obtained and the biological disorder can be demonstrated.
• the application of the same transformation followed by the application of the same filter can be repeated to refine the discrimination.
• such a repetition is advantageously carried out following other distinct transformations which will have allowed one or more discriminations according to criteria other than those of said repeated transformation.
• the series of transformations used is associated with a particular classification of a predetermined set of cell classes revealing a pathology, this series being determined according to statistical knowledge on the cells constituting these cellular classes.
• the pathology transformation for placing the cell classes of a biological fluid having the average statistical characteristics of the pathology in distinct areas of the m-dimensional composite space, the pathology transformation for dissociating a normal biological fluid from a biological fluid having a particular pathology.
• the series of transformations used in steps f) and g) and the transformation, in the two-dimensional or three-dimensional space is such that the cell classes are classified by degree of maturity.
• the invention also relates to a device for classification by discrimination and counting of at least one set of three cellular classes intended to be connected to a biological fluid analysis automaton capable of detecting cells in the liquid and able to determine a number of cells.
• a biological fluid analysis automaton capable of detecting cells in the liquid and able to determine a number of cells.
• -uplet comprising at least four physical parameters (n> 3) for each detected cell, said device comprising:
• data processing means for associating a particular tag with m-tuples according to their class; data processing means for applying the transformation of the plurality of tuples to 3-tuples, 2-tuples, or 1-tuples;
• the various steps of the method according to the invention are determined by instructions of computer programs.
• the invention also relates to a computer program on an information medium, this program being capable of being implemented in a computer, hardware-type processing means, such as an FPGA ("field programmable gate array") or any other type of programmable electronics, this program comprising instructions adapted to the implementation of the steps of the method according to the invention.
• This program can use any programming language, and be in the form of source code, machine code, or intermediate code between source code and machine code, such as in a partially compiled form, or in any other desirable shape.
• the invention also relates to a computer-readable information medium, comprising instructions of a computer program as mentioned above.
• the information carrier may be any entity or device capable of storing the program.
• the medium may comprise a storage means, such as a ROM, for example a CD, or a DVD whether rewritable or not, ... or a microelectronic circuit ROM, or a means of magnetic recording, for example a floppy disk, a hard disk, a non-volatile memory (flash memory for example such as a USB key etc).
• the information medium may be a transmissible medium such as an electrical or optical signal, which may be conveyed via an electrical or optical cable, by radio or by other means.
• the program according to the invention can in particular be downloaded on a network of the type
• the information carrier may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the method in question.
• FIG. 1 schematically represents a device according to the invention
• Figures 2a to 2c show results obtained by the method according to the invention for normal blood
• Figure 3 shows in block form the expected positions of the populations of Figure 2b
• FIGS. 4a to 4j show results obtained by the method according to the invention for normal blood (5a) and pathological bloods (5b to 5j).
• FIG. 1 schematically illustrates a device for implementing the invention.
• This device comprises receiving means 9 for receiving data from a biological fluid analyzer 1 for determining n physical parameters X1 to Xn, n> 3, per xi cell detected. These n parameters define a tuple or vector measure xi (Xl ... Xn).
• Xl FSC in the following
• X2 SSC
• X3 FL1
• RES resistivity
• Thiazole Orange which binds to intracellular nucleic acids, thus making it possible to demonstrate the nucleated cells, in particular here, the white blood cells, is used.
• the device comprises means 11 for selecting a group of n-tuples belonging to a subset of so-called input classes used in the step Csi.
• the device comprises a memory 18 in which software elements are stored allowing transformations T of the received data set xi (X1 ... Xn, Csi) to spaces having a number of dimensions. strictly less than n and, in the example proposed, the implementation of filters A of discrimination of subsets of output cell classes affected in the step Cdi.
• the device further comprises means 12 for selecting a transformation Tn-> m of a constellation of data in n-dimensional space into a constellation in a space with m-dimensions.
• each transformation may be associated with one or more particular classifications of a predetermined set of cell classes and determined according to statistical knowledge on the cells constituting the cell populations corresponding to these cellular classes.
• classification is meant any set of cellular classes in which the cells detected will be classified. Depending on the purpose of the analysis, these classifications are different.
• the first objective is to know if the analyzed blood has a composition in the ranges of normality.
• the invention provides access to this information by ensuring a dissociation of the cell classes to be discriminated in the case of normal blood and it also allows its display during a display on a screen.
• Another objective will be to confirm the typicity of a pathological blood.
• the typical elements of such blood for a particular pathology are also known statistically, the associated transformation Tn-> m will dissociate the corresponding cell classes during a display on a screen. Moreover, the calculations obtained will make it possible to give the result on the pathological state of the studied blood.
• Each transformation thus makes it possible to place the cell classes of the particular classification sought of a biological fluid having average statistical characteristics in distinct zones of the m-dimensional composite space.
• Processing means 13 implement the selected transformation Tn-> m on the set of data xi (X1 ... Xn) in order to produce a set of data yi (Yl ... Ym) in a space with m-dimensions .
• This functional transformation Tn-> m linear or non-linear, consists in transforming the vector of measurements of dimension n (n> 3) into a composite vector of dimension m ⁇ n by a functional transformation:
• x is the initial measure vector of dimension n, otherwise denoted xi (Xl .. .Xn)
• y is the image vector which represents the element yi (Yl ... Ym) in the standard composite vector space of dimension m.
• the transformation is therefore characterized by the two matrices A and C and the two vectors B and D. If the matrix C is zero and all the elements of D are non-zero, then we are in a linear case.
• the cell classes that are desired to be discriminated are then placed in the m-dimensional space as a plurality of dissociated constellations.
• Selection means 14 of the discrimination filters then make it possible to select the filter A for the current step.
• each step is defined by a selection filter - a subset of classes -, a transformation, and a discrimination filter that will reclassify the m-tuples in the composite space into a subset of classes.
• the selection can be manual, performed by a user, or preprogrammed and therefore automatic.
• these filters A are then used for the observation of the constellations allowing the best discrimination between the cells of several distinct populations. A number of cell discriminations are then performed in the m-dimensional space.
• the choice of the filter is associated, for each step, with the choice of the transformation Tn-> m which was carried out previously.
• One or more filters can be used with the same transformation. Similarly, the same filter can be implemented after two distinct transformations. The same filters and transformations can also act on subsets of distinct selected tuples.
• the set formed by the selection of a subset of n-tuples (or n + 1 tuples since a class is associated with the tuplet proper), the selection and the application of a transformation and the selection and the application of one or more filter (s) constitutes a step of the discrimination algorithm used according to the invention.
• the set of steps each implementing a selection of tuples, a transformation and at least one filter constitutes the actual discrimination algorithm after which all the cells are associated with one of the classes that one wishes. discriminate.
• a principle of repetition of population correction steps can be implemented.
• a first initial classification is performed by a predefined area selection process.
• a discrimination filter to a subset of classes is then applied. Only vectors having an initial classification belonging to the subset of selected vectors are considered.
• the boundaries of separation, or reclassification are previously fixed, thus stored or determined using histograms according to previously established criteria and statistics.
• a control module 16 verified at the end of the application of the current algorithm step if all the planned steps have been made. If it is not the case, another algorithm step is made. If this is the case then all the tuples are sent to processing means 17.
• the classification carried out by the discriminations associates a different graphic distinction COLk with each cell class Ck to be discriminated, for example a point color or a point shape. We thus obtain a set of points yi (Yl ... Ym, COLk) in the m-dimensional space to which it corresponds a set of points xi (Xl ... Xn, COLk) in n-dimensional space.
• these processing means 17 apply a transformation Tn-> 1, Tn-> 2 or Tn-> 3 to the set of points xi (X1 ... Xn, COLk) in a set of composite vectors Ei (Zl, Z2, C0Lk) in a two-dimensional or three-dimensional space. The result is then a distribution of the cell classes in colored planar or three-dimensional constellations.
• Tn-> 1, Tn-> 2 or Tn-> 3 is such that, in the case where the biological fluid has the average characteristics of a normal blood, the point clouds have little or no overlap when the Ei points (Zl, Z2, C0Lk) are displayed on display means 19 is flat, in particular, in the case of a dimension, we can add the channel density to represent the histogram, or with a dynamic 3D space. This is not the case for orthogonal two-dimensional projections generally used.
• This two-dimensional display on a screen will allow the user to obtain information very quickly since the transformation in a space with m-dimensions and then the transformation to a two-dimensional space will have been selected in relation to the question to which we wanted to answer: for example, is it normal blood? is it a blood revealing such pathology? etc.
• An essential point of the invention is that the plurality of transformations stored according to the invention in a memory of the device according to the invention makes it possible to obtain matrices more appropriate to each question envisaged.
• the first step 1 consists in applying a transformation of the measured data X1, X2, X3, X4 constituting the n-tuple associated with each cell in order to obtain for each cell coordinates Y1, Y2 and Y3 function of the measured values of X1, X2, X3 and X4 and of three constants dependent on the calibration of the measurement banks and the acquisition system.
• a filter is applied and we will select the subset of the points belonging to the classes that will be displayed.
• the third step consists in applying a transformation of the n-tuples to a uni, bi or three-dimensional space, here two-dimensional for display. For each of these points then denoted Ei, two coordinates Z1 and Z2 are then associated with each cell with, for example, a COLk color corresponding to the class Ck determined by the classification.
• FIG. 2 represents three examples of results of such transformations towards a two-dimensional space allowing a specific visualization of the cell classes as a function of the analysis needs.
• Figure 2A is one of these representations, it will be described later.
• Figure 2B corresponds to another transformation allowing a privileged visualization of the maturation states of cell lines, it will also be described later.
• FIG. 2C allows a privileged visualization of pathologies of the lymphoid lineage, in particular chronic lymphocytic leukemia (CLL).
• CLL chronic lymphocytic leukemia
• the available transformations are advantageously used successively on the same set of points Xi (X1 ... Xn).
• a first transformation can answer a basic question: is the blood normal or not? If the user observes that the point clouds are not dissociated, he may think of a technical problem or a pathology. He also has access in parallel to a count of the cells of each cell class, as well as their relative proportion.
• the cell classes for example marked with different colors
• overlap in the final representation obtained it is possible to conclude on the blood.
• the classification was carried out by means of different transformations, in spaces that are not those displayed in 2D. The separation can therefore be exact even if the cell populations are projected on one another in the visualization plane.
• the process can be restarted with a transformation associated with a particular classification corresponding for example to a pathology particular or at a particular age of the patient.
• the new transformation generates a set of points in a m-dimensional space that may be similar or different from that useful for the first transformation used.
• this transformation corresponding to a particular classification, the cell populations to look at more precisely are similar or, more generally, different from those whose classification was aimed at during the first transformation.
• the transformations include linear and / or nonlinear calculations which make it possible to obtain the best possible representation of the results according to a favorable angle for access to the information sought. This angle is associated with the pathology to be highlighted from the raw data collected.
• the invention also makes it possible to implement interactive exploration means.
• transformation Tn-> m and the filter subset are suitable for observation, for example of a cell line, a family of pathologies or other.
• Xl, X2, X3, X4 respectively correspond to the small angle diffraction, to the 90 ° diffusion, to a fluorescence channel with orange thiazole and resistivity as reagent.
• the constants CIi and C2i are defined according to the characteristics of the analyzer, in particular those of the optical bench.
• FIG. 5 represents, in the form of two-dimensional surfaces, the different populations observed in normal type blood and in pathological blood with the transformation as explained above.
• the invention makes possible the visualization of five leukocyte populations, including basophils.
• the types of representations presented in these figures are examples to give an idea of the disposition of the different leukocyte populations according to the invention as a function of pathological or non-pathological blood samples.
• Figure 4A corresponds to normal blood.
• Figure 4B is a blood signaling hairy cell leukemia.
• Figure 4C corresponds to a blood signaling a myeloma.
• Figure 4D corresponds to a blood signaling Sezary syndrome.
• FIGS. 4E and 4F correspond to bloods that signify leukocyte leukemia of the type LAL (Acute Lymphoid Leukemia) B2.
• FIG. 4G corresponds to a blood signaling leukemia LAL Burkitt / LAL B3.
• FIG. 4H corresponds to a blood signing LAL T leukemia.
• FIG. 41 corresponds to a blood signing AML leukemia (Acute Myeloid Leukemia).
• Figure 4J corresponds to a blood signaling an LLC pathology (Chronic Lymphocytic Leukemia).
• the result of such a transformation will then be a representation such as that represented in FIG. 2B.
• the second step 2 consists of a graphical transformation containing a translation and a rotation matrix (Euler matrix) such that the angles of rotation in our case are 130 °, 51 ° and 209 ° around a center located at (2048). , 2048, 2048).
• Euler matrix a rotation matrix
• the third step 3 consists of a graphical adaptation to optimize the visualization of the families of cells in a 4096 X 4096 graph such as:
• the representation of the cell classes obtained in this matrix form makes it possible to see all of the leukocyte sub-populations present in the whole blood sample at one time.
• the sub-populations are visible and well separated and there is little overlap between the different populations except in the case of abnormal blood.
• the abscissae and the ordinates do not have a concrete meaning.
• the disposition of populations on the matrix is important.
• the positions on the two-dimensional space of the cell classes are ordered along the vertical (ordinate) starting with the most immature cells followed by the mature cells and then decomposing in the horizontal direction ( abscissae) between cells having a mono-nucleated structure and those having a poly-nucleated structure.
• This representation is close to the schematic view of the tree of classification of blood cells from stem cells. It allows an easier and more direct interpretation of the different leukocyte subpopulations by the practitioner or the biologist.
• Figure 3 schematically represents distinct areas in which it is possible to place the different leukocyte populations that can be expected to be found in the two-dimensional matrix described above. The presence or absence of populations on the screen depends on analyzed blood, normal or abnormal / pathological.

Landscapes

• Chemical & Material Sciences (AREA)
• Dispersion Chemistry (AREA)
• Physics & Mathematics (AREA)
• Health & Medical Sciences (AREA)
• Life Sciences & Earth Sciences (AREA)
• Analytical Chemistry (AREA)
• Biochemistry (AREA)
• General Health & Medical Sciences (AREA)
• General Physics & Mathematics (AREA)
• Immunology (AREA)
• Pathology (AREA)
• Investigating Or Analysing Biological Materials (AREA)

Priority Applications (5)

Applications Claiming Priority (2)

Publications (1)

Family

ID=40651741

Family Applications (1)

Country Status (7)

Families Citing this family (5)

Citations (11)

Family Cites Families (23)

• 2008
  • 2008-09-05 FR FR0855986A patent/FR2935802B1/fr not_active Expired - Fee Related
• 2009
  • 2009-08-05 WO PCT/FR2009/051559 patent/WO2010026328A1/fr not_active Ceased
  • 2009-08-05 JP JP2011525592A patent/JP5314145B2/ja not_active Expired - Fee Related
  • 2009-08-05 EP EP09740436A patent/EP2318820B1/fr active Active
  • 2009-08-05 ES ES09740436T patent/ES2401223T3/es active Active
  • 2009-08-05 US US13/060,501 patent/US20110167029A1/en not_active Abandoned
  • 2009-08-05 CN CN200980134783.XA patent/CN102144153B/zh active Active

Patent Citations (11)

Also Published As

Similar Documents

Legal Events