• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention relates to a method for identifying biological particles using a stack of holographic images obtained by means of an optical system. A stack of image blocks centered on the biological particle to be analyzed is extracted from the image stack and a reference block corresponding to the focusing plane is determined. A characteristic quantity is calculated on each block of the stack and the profile of this characteristic quantity along the optical axis of the system is compared to a plurality of standard profiles relating to known particle types. Alternatively blocks of the stack for predetermined defocus deviations are extracted from the stack of blocks and the blocks thus extracted are compared to standard blocks relating to known particle types.
    公开号:FR3030749A1
    申请号:FR1462998
    申请日:2014-12-19
    公开日:2016-06-24
    发明作者:Francois Perraut；Pierre Joly；Quentin Josso；Meike Kloster-Landsber；Alice Douet
    申请人:Biomerieux SA；Commissariat a lEnergie Atomique CEA；Commissariat a lEnergie Atomique et aux Energies Alternatives CEA；
    IPC主号:
    专利说明:

    [0001] TECHNICAL FIELD The present invention relates generally to the field of optical analysis of biological particles. It is particularly applicable to microbiological diagnosis, more particularly to the identification of microorganisms and / or their state in response to stress. It can also be applied to monitoring cell cultures. STATE OF THE PRIOR ART Digital holographic microscopy or DHM (Digital Holographic Microscopy) is a known imaging technique that makes it possible to overcome the depth-of-field constraints of conventional optical microscopy.
    [0002] Schematically, it consists in recording a hologram formed by the interference between the light waves diffracted by the observed object and a reference wave having a spatial coherence. A general overview of digital holographic microscopy can be found in Myung K.Kim's review article "Principles and techniques of digital holographic microscopy" published in SPIE Reviews Vol. 1, No. 1, January 2010. Recently, it has been proposed to use digital holographic microscopy to identify microorganisms of automated material. Thus, the article by N. Wu et al. entitled "Three-dimensional identification of microorganisms using a digital holographic microscope" published in Computational and Mathematical Methods in Medicine, Vol. 2013, art. ID No. 162105 discloses a method of identifying different types of bacteria in the volume to be analyzed by digital propagation to the particle's focusing plane. Images developed at different depths are used to reconstruct a three-dimensional representation of microorganisms. These are then classified by means of a nonlinear 3D filtering.
    [0003] Similarly, Ahmed El Mallahi's article entitled "Automated Threedimensional Detection and Classification of Living Environments Using Digital Holography Microscopy with Partial Space Coherent Source: Application to Monitoring Drinking Water Resources" published in Applied Optics, Vol. 52 No. 1, January 2013, describes a method comprising a first step of detecting the position of the bacteria in the volume to be analyzed, a focusing step at different depths of the volume by means of digital propagation, then a classification of the bacteria from their morphological characteristics. The aforementioned identification methods are however complex insofar as they require focusing in successive planes of focus. On the other hand, focusing in a single focus plane, in other words at a single depth of analysis, is generally not sufficient to identify a type of microorganism with a low rate of false detection. The object of the present invention is therefore to provide a method for identifying organic particles by digital holographic microscopy which allows to obtain a low rate of false detection while being simple and robust. DISCLOSURE OF THE INVENTION The present invention is defined by a method for identifying biological particles from a stack of holographic images obtained by means of an optical system, in which: said holographic images are obtained for a plurality defocusing deviations from a focusing plane, the defocusing deviations being taken along the optical axis of the optical system; a reference holographic image is selected from the holographic image stack as being the closest to a focus plane; for a biological particle of interest, a stack of image blocks comprising said biological particle of interest is extracted from the stack of holographic images; the type of the particle of interest is identified from the stack of image blocks thus extracted.
    [0004] According to a first variant embodiment, the holographic images are acquired by the optical system for a plurality of positions along the optical axis. According to a second variant embodiment, a holographic image is acquired by the optical system and the other holographic images of the holographic image stack are calculated from the first holographic image by means of a digital propagation model. In this case, the first holographic image is advantageously taken with a non-zero defocus deviation with respect to the focus plane. Said reference image may be chosen from the stack of holographic images such as that maximizing a predetermined contrast criterion. A reference block is then selected in the stack of image blocks as the image block centered on the particle of interest belonging to the reference image, the position of the reference block on the optical axis then being chosen. as an origin for defocus deviations.
    [0005] The reference block can then be updated by searching among the image blocks centered on the particle of interest and belonging to the neighboring images of the reference image in the stack of holographic images, the one maximizing a predetermined contrast criterion. According to a first embodiment, for each block of the image block stack, the value of at least one characteristic quantity is calculated on this block and a profile of said characteristic quantity is obtained along the length of the block. optical axis from the values of the characteristic quantity thus calculated. The profile of said characteristic quantity is then compared with a threshold and it is deduced that the biological particle of interest is of a first type if this profile crosses this threshold and of a second type if it does not cross it. Alternatively, by means of a similarity criterion, the profile of said characteristic quantity is compared with a plurality of standard profiles obtained for various known types of biological particles and the type of the biological particle is deduced from the standard profile presenting the greatest similarity with the profile of said characteristic quantity.
    [0006] The similarity criterion can be chosen from an intercorrelation, a Pearson coefficient, a quadratic difference. Alternatively, the profile of said characteristic quantity is classified by means of a supervised learning method among a plurality of classes of profiles, each class corresponding to a given type of biological particle. According to a second embodiment, a plurality of blocks corresponding to predetermined defocus deviations are selected from said stack of image blocks. Comparing, using a similarity criterion, said plurality of selected blocks to the same pluralities of standard blocks, each plurality of standard blocks corresponding to a given type of biological particle, the type of the biological particle of the interest being derived from the plurality of type blocks having the greatest similarity with said plurality of selected blocks. The similarity criterion can again be an intercorrelation, a quadratic difference, a quadratic difference after spatial Fourier transform, a criterion based on a principal component analysis. BRIEF DESCRIPTION OF THE DRAWINGS Other characteristics and advantages of the invention will appear on reading preferred embodiments with reference to the appended figures in which: FIG. 1 schematically shows a holographic image recording device useful for the method of identifying biological particles according to the invention; Fig. 2 schematically represents a flowchart of the biological particle identification method according to a first embodiment of the invention; Fig. 3 schematically shows the extraction of a stack of image blocks from an image stack; Fig. 4 schematically illustrates the search for a reference block in a stack of blocks; Figs. 5A and 5B show axial profiles of a characteristic magnitude respectively two known types of biological particles; Fig. 6 schematically represents a flowchart of the biological particle identification method according to a second embodiment of the invention; Fig. Figure 7 gives an example of a m-tuple of standard blocks for various known types of biological particles. DETAILED DESCRIPTION OF PARTICULAR EMBODIMENTS The method for identifying biological particles according to the invention uses a holographic image recording device such as that described with reference to FIG. 1. This device, 100, comprises a light source, preferably temporally coherent, 110, of small spectral width, for example less than 200 nm or even 100 nm or even 25 nm. This light source may be in particular a laser diode or a light emitting diode. The light source is preferably spatially coherent. The light beam emitted by the source is conveyed by means of an optical fiber under the sample to be analyzed, 120. This sample consists of a liquid such as water, a buffer solution, a culture medium or a reagent medium. , in which are the biological particles, P, to be identified. Alternatively, the sample may be in the form of a solid medium, preferably a transparent medium, such as an agar, in which the particles in question are located. The sample may also be a gaseous medium. The biological particles can be located inside the medium or on the surface of the sample.
    [0007] The biological particles to be identified may be microorganisms such as bacteria or yeasts, for example. It can also be cells, multicellular organisms, or any other particle of pollutant particle type, dust. The size of the particles observed varies between 100 nm to several hundreds of um, even a few millimeters.
    [0008] The sample is enclosed in an analysis chamber 130, delimited vertically by a lower blade 131, for example a conventional microscope slide, and an upper blade 132. The analysis chamber is delimited laterally by an adhesive 135 or by any other waterproof material. The lower and upper blades are transparent to the wavelength of the light source. The biological particles are immobilized in the chamber, either because the medium in which they are is solid (agar), or because the medium is fluidic but the particles are fixed on the inner face 132i of the upper blade. The particles may be mobile provided that their speed of movement is sufficiently low so that the particles can be considered as immobile during the measurement time. The device 100 further comprises an optical system, 140, consisting for example of a microscope objective 145 and a tube lens 141. The optical system is optionally equipped with a filter 143 which can be located in front of the lens or well between the lens and the tube lens. The optical system 140 is characterized inter alia by its optical axis A, its object plane Ho, also called the focusing plane, at a distance D from the lens, and its image plane q, conjugated from the object plane by the optical system . In other words, to an object located in the object plane, flo, corresponds to a sharp image of this object in the image plane, q. The object and image planes are orthogonal to the optical axis A. An image sensor, 150, for example a CCD or CMOS sensor, is located in or near the image plane q. Thus, the sensor 150 acquires a transmission image of a portion of the focusing plane.
    [0009] The relative position of the optical system 140 relative to the analysis chamber 130 is vertically adjustable. For example, the lens is secured to a lens holder that can move along a vertical slide. Thus, it is possible to focus on one or more biological particles of interest.
    [0010] The image formed on the image sensor is a holographic image insofar as it results from the interference between a wave diffracted by the biological particles and a reference wave that has passed through the sample without having interacted with it.
    [0011] Alternatively, it is possible to divide the light beam into two components, for example by means of a semi-transparent plate (not shown). The first component then serves as a reference wave and the second component is diffracted by the sample, the image in the image plane of the optical system resulting from the interference between the diffracted wave and the reference wave.
    [0012] The images of the sample thus acquired are then processed by the identification method according to the invention. Fig. 2 schematically represents a flowchart of the biological particle identification method according to a first embodiment of the invention. In step 210, a plurality (or stack) of holographic images of the biological particles of interest within the sample is obtained. According to a first variant, these images are acquired experimentally by the imaging system, each of these images corresponding to different distances to the analysis chamber, taken along the optical axis A. According to a second variant, a first image is acquired at a first distance from the analysis chamber, this distance does not necessarily correspond to focusing conditions on these particles. In particular, it should be noted that when the biological particles to be analyzed are transparent to the wavelength of the source, this first image will not be taken under the focusing conditions but, at a predetermined distance from the position of focus. This distance is preferably less than 2 mm, preferably less than 1 mm, 500 μm. The first image thus acquired is then called defocused. Additional images are computed from a defocused initial image, using a numerical propagation model as explained below.
    [0013] The additional images calculated are those that would be observed at different distances between the optical system and the analysis chamber, that is to say in different axial positions of the imaging system. The digital propagation image calculation method, as explained in the article by Sang-Hyuk Lee et al. entitled "Holographic microscopy of holographically trapped three-dimensional structures" published in Optics Express, Vol. 15; No. 4, February 19, 2007, pp. 1505-1512. More precisely, if we denote h (r) the Rayleigh-Sommerfeld propagation function, let: lae '(1) h, (r) = 27-t-az R where z is the defocusing height, in other words the deviation from the plane of focus, r = (x, y) is the position in the plane of the image, R2 _ r2 ± z2 ... ei k = 25-n / 2 is the number of As a wave relative to the propagation medium, the wave in the ordinate plane z may be expressed as: a (r, z) = la (r, z) lexp (iv (r, z)) (2- 1) 1 +00 20a (r, z) = i-471-2 B (q) 11 (q) exP (iqr) d2q (2-2) -00 Where B (q) is the Fourier transform of b ( r), intensity of the diffracted wave in the focal plane (the intensity of the reference wave is here assumed to be constant), H (q) is the Fourier transform of hz (r) and q is the dual variable of r in the Fourier transform. It is thus understood that it is possible to construct an image stack / 1, ... 4 for ordinates Zp ..., ZN along the optical axis, the origin of the ordinates being taken at the axial position focusing, each image Ir, being defined by a complex amplitude a (r, z).
    [0014] In step 220, in the image stack, / 1, ..., 4, obtained in the preceding step, a reference image I ref is selected. This reference image is the one that best corresponds to the ideal conditions of development on the biological particles of interest. Indeed, under ideal experimental conditions, the particles are located on the inner face 132i of the upper blade 132 and this face is perpendicular to the optical axis A. The ideal focusing conditions are then those in which the focus plane is merged with the aforementioned inner face. In practice, when the biological particles are not transparent, the selection of the reference image can be done according to a maximum contrast criterion applied to a zone containing the biological particles. This maximum contrast criterion may for example be a maximum standard deviation or a maximum average gradient value in this zone. When the biological particles are transparent, an optical system having a spherical aberration will advantageously be used so as to avoid a complete disappearance of the signal at the focusing position.
    [0015] In any case, the axial position of the reference image is then taken as the reference position (z = 0), the positions of the other images of the stack being calculated with respect to this reference position. In step 230, at least one particle to be analyzed is selected in the reference image. This selection can be automated and carried out for example on the basis of morphological and / or photometric criteria. For each particle to be analyzed, its position (x, y) is determined in the reference image and a region of interest is determined in the form of an image block, in short, centered on this position. Note that this block may have a size larger or smaller than that of the particle to be analyzed. This block of image, in short, is extracted from the reference image / ref.
    [0016] In step 240, other images of the stack are extracted, the image blocks corresponding to the same position as the block extracted from the reference image. A stack of image blocks is thus obtained, that is to say a restriction of the image stack to the zone of interest centered on the particle to be analyzed. It will be understood that each block of the stack corresponds to a different axial position and therefore to different defocusing conditions with respect to the short reference block. Fig. 3 schematically represents an image stack / 1, ..., 4 and a stack of image blocks centered on a particle of interest P with coordinates (x, y). The blocks here are square in shape, but it will be understood that other block shapes can be envisaged without departing from the scope of the present invention. Optionally, in step 250, the selection of the reference image, and therefore of the reference image block for the particle to be analyzed, is refined. Indeed, when all the particles are not in the same plane orthogonal to the optical axis, which may be particularly the case when the inner face of the upper blade is not perfectly orthogonal to the optical axis, the plane focus may differ from particle to particle. One then searches among the blocks of the stack, and advantageously among the blocks of images on either side of the reference image / ref selected in step 220, that which best corresponds to the focusing conditions. This selection can be performed, as in step 220, using a maximum contrast criterion, but this time on the blocks of the stack. The maximum contrast criterion may again be a maximum standard deviation or a maximum average gradient value on the block. The block thus selected then becomes the new reference block, Brief, of the stack. Fig. 4 represents an example of searching for a reference block in a stack of image blocks of a biological particle. The indices of the blocks are represented as abscissae, each block Br corresponding to an axial position Zr, different, and the ordinate the standard deviation of the intensity of the pixels in the block. Searching the reference image, / ref, in step 220 and extracting the centered block on the particle of interest in step 230 gives a designated first reference block (in the figure by Bi). f. The finer tuning of the tuning, at the optional step 250 (2 gives a second reference block B) corresponding to the maximum of the rejection curve the standard deviation. When this step is performed, it is therefore this second reference block that will be taken as a reference block, in short, for the rest of the identification method.
    [0017] In step 260, it is estimated that at least one characteristic value a z ,, for each block Br, of the stack of blocks, and an axial profile of this characteristic quantity is thus deduced along the optical axis. More precisely, if the blocks correspond to axial positions Zp ..., ZN, the axial block profile consists of the sequence aZi), G (Z2), ..., G (ZN).
    [0018] This characteristic quantity may in particular be a statistical quantity relative to the block, for example an average, a median, a standard deviation of the intensity of the pixels on the block. In addition, when the images of the stack have been obtained by digital propagation and not acquired experimentally, a complex value can be associated with each pixel of the block. The characteristic quantity can then relate to the real part or the imaginary part of these complex values. For example, the squared mean of the imaginary part of the complex amplitude on the block can be taken as the characteristic value, ie for block 4: G (z, i) = ELP (a (r, z, i) ) 121 (3) where E (.) Means the mean with respect to re 13n. If necessary, the discrete values G (;), G (Z2),..., G (ZN) can be interpolated to obtain a finer resolution of the profile of the characteristic quantity. In step 270, the type of the biological particle is identified from the axial profile thus obtained. According to a first variant, which is particularly simple, one can compare the axial profile G (z) with a threshold value and discriminate between two types of particles. Figs. 5A and 5B represent the axial profiles of a characteristic quantity (in this case, the average squared value of the imaginary part of the amplitude in the block) for two types of biological particles, namely for Staphylococcus epidermidis (FIG. 5A) and for Staphylococcus aureus. We see here that a threshold value, Th, such that 0.4 makes it possible to effectively discriminate the two types of biological particles (here two types of staphylococci). If the axial profile crosses this threshold value, the biological particle will be identified as S. epidermidis whereas if it does not cross it it will be identified as S. aureus. Note that this first variant does not require to determine the reference block Briefly with precision. However, the reference image I ref should be sought to select the particle to be analyzed in step 230.
    [0019] According to a second variant, the axial profile of the characteristic quantity is compared, by means of a similarity criterion, with axial profiles of this same characteristic quantity previously obtained for biological particles of known types, q (Z), .. ., GK (Z), stored for example in a database. The profiles q (Z), ..., GK (Z) are called typical profiles. Using a common origin (reference block) makes comparison easier. The similarity criterion may be for example an intercorrelation value or a Pearson coefficient. The index LE {1, ..., K} for which the type profile q (Z) is the closest to the profile G (z) gives the type of the biological particle. According to a third variant of identification, in step 270, several characteristic quantities Ge (zr,), f = i,..., L are calculated for each of the blocks Br, of the stack. A plurality of axial Ge (Z) profiles characterizing the particle are thus obtained. This plurality of profiles can be compared by means of a similarity criterion to the same plurality of profiles obtained for each of the K particles of known type, ie Gek (Z), f = 1, ..., L, k = 1 , ..., K. Then, using a similarity criterion, the plurality of Gek (z) type axial profiles closest to the measured profiles is determined. As in the first variant, the index k. gives the type of biological particle analyzed.
    [0020] According to a fourth variant, the identification of the particle uses a supervised learning method. This variant assumes the prior acquisition of axial profiles for a plurality K of biological particle types. The biological particles are then classified into K classes by taking as their descriptor the values of the axial profile or preferably a characteristic of this profile, each class being able to be represented by a group of points in a space with L dimensions. From the axial profile of a biological particle to be analyzed, we determine the corresponding point in space at L dimensions and we search for the group of points, that is to say the class to which it belongs.
    [0021] For example, if one takes as a characteristic parameter of a block the one defined by the expression (3) and as characteristic of the profile its maximum value, it has been shown that it was possible to efficiently classify the Acinetobacter johnsonii bacteria, Enterobacter aerogenes, Escherichia neck, and Staphylococcus epidermidis. Thus on samples of known populations, we obtained the following confusion matrix: CM = (139 24 52 4 23 188 91 16 16 54 260 1 7 8 4 174 where the different lines correspond to populations of different types and the columns correspond to the classes predicted by the identification method It is noted that the confusion matrix has a predominant diagonal and therefore the identification error rate is relatively low.
    [0022] Those skilled in the art will understand that the usual techniques of supervised learning such as Bayesian classification techniques or support vector machines may be used to classify biological particles from their axial profiles.
    [0023] Fig. 6 schematically represents a flowchart of the biological particle identification method according to a second embodiment of the invention. The second identification method according to the invention is also based on a stack of image blocks centered on the biological particle of interest. The blocks are extracted from a stack of images acquired or calculated as previously described. More specifically, steps 610 to 650 are respectively identical to steps 610 to 650 of the identification method according to the first embodiment of the invention. Their description will not be repeated here.
    [0024] However, in step 660, no characteristic quantity is calculated, but a sub-plurality M of blocks located at predetermined deviations (in) z) from the reference block is selected in the stack of blocks. In short (z = O). In practice, the images of the stack being acquired or calculated (by digital propagation) at regular intervals, the selection in the stack of blocks will relate to predetermined indices with respect to the index of the reference block.
    [0025] In step 670, the M-tuple of blocks thus selected is compared by means of a similarity criterion with M-tuples of standard blocks, each M-tuple of standard blocks being relative to a biological particle of type known, the blocks of each M-tuple having themselves been obtained at the predetermined distances mentioned above. The similarity criterion may be a spatial correlation, a Pearson coefficient, a quadratic difference between a Fourier transform of the blocks of the particle to be analyzed and a transformation of the standard blocks, or even be based on a Principal Component Analysis (PCA). In the latter case, it is possible to determine for each block of the M-tuple of blocks of the particle to be analyzed and each block of the M-tuple of typical blocks the main axes of the distribution of the pixels and to compare (for example by means of a scalar product of director vectors) the alignment of the main axes of the blocks of the particle to be analyzed with the main axes of the standard blocks. Whatever the similarity criterion used, the M-tuple of the type blocks closest to the M-tuple of blocks of the particle to be analyzed gives in 680 the type of the biological particle. Fig. Figure 7 gives an example of a M-tuple of typical blocks for different biological particles. In the illustrated case M = 3 and the predetermined defocusing deviations are respectively 15 μm (image blocks of the Pre column), 6 μm (image blocks of the 2'd 'column) and 0 μm (blocks of the column). image of the 3rd column in the focusing plane), the differences being counted positively in the direction of propagation of the beam. The standard blocks of the first line are for the E. coun species, the standard blocks for the second line are for the A. johnsonii species, and the standard blocks for the third line are for the S. epidermermidis species. .
    [0026] It is observed that the type blocks of the E. coli and A. johnsonii species have structures very close to 15 μm whereas the typical S. epidermermidis block has a very different structure. Similarly, the type blocks of A. johnsonii and S. epidermermidis have very similar structures at 0 um, whereas E. coli has a very different structure.
    [0027] The use of a triplet of image blocks thus makes it possible to remove the ambiguity of identification.
    权利要求:
    Claims (15)
    [0001]
    REVENDICATIONS1. A method of identifying biological particles from a stack of holographic images obtained by means of an optical system, characterized in that: - said holographic images are obtained (210,610) for a plurality of defocus deviations relative to a focusing plane, the defocusing deviations being taken along the optical axis of the optical system; a reference holographic image is selected (220, 620) in the holographic image stack as being the closest to a focus plane; for a biological particle of interest, a stack of image blocks comprising said biological particle of interest (240.640) is extracted from the stack of holographic images; the type of the particle of interest is identified (270, 680) from the stack of image blocks thus extracted.
    [0002]
    The biological particle identification method according to claim 1, characterized in that the holographic images are acquired by the optical system for a plurality of positions along the optical axis.
    [0003]
    The method of identifying biological particles according to claim 1, characterized in that a first holographic image is acquired by the optical system and the other holographic images of the holographic image stack are calculated from the first image. holographic using a numerical propagation model.
    [0004]
    4. Method of identification of biological particles according to claim 3, characterized in that the first holographic image is taken with a defocus deviation non-zero with respect to the plane of focus.
    [0005]
    5. Method for identifying biological particles according to one of the preceding claims, characterized in that said reference image is chosen from the stack of holographic images such as that maximizing a predetermined contrast criterion.
    [0006]
    6. Method for identifying biological particles according to claim 5, characterized in that a reference block is selected from the image block of the image block centered on the particle of interest belonging to the reference image, the position of the reference block on the optical axis then being chosen as the origin for defocus deviations.
    [0007]
    7. A method for identifying biological particles according to claim 6, characterized in that the reference block is updated by searching among the image blocks centered on the particle of interest and belonging to the neighboring images of the reference image. in the stack of holographic images, the one maximizing a predetermined contrast criterion.
    [0008]
    8. Method for identifying biological particles according to one of the preceding claims, characterized in that for each block of the image block stack, the value of at least one characteristic quantity on this block is calculated (260). and that a profile of said characteristic magnitude along the optical axis is obtained from the values of the characteristic quantity thus calculated.
    [0009]
    9. Method for identifying biological particles according to claim 8, characterized in that the profile of said characteristic quantity is compared with a threshold and it is deduced that the biological particle of interest is of a first type. if this profile crosses this threshold and a second type if it does not cross it.
    [0010]
    Biological particle identification method according to claim 8, characterized in that, by means of a similarity criterion, the profile of said characteristic quantity is compared to a plurality of standard profiles obtained for different known types of biological particles and deducing the type of the biological particle of the typical profile having the greatest similarity with the profile of said characteristic quantity.
    [0011]
    11. Method of identifying biological particles according to claim 10, characterized in that the similarity criterion is selected from an intercorrelation, a Pearson coefficient, a quadratic difference.
    [0012]
    The method for identifying biological particles according to claim 8, characterized in that the profile of said characteristic quantity is classified by means of a supervised learning method among a plurality of classes of profiles, each class corresponding to a type given biological particle.
    [0013]
    13. A method of identifying biological particles according to claim 6, characterized in that one selects (660) in said stack of image blocks a plurality of blocks corresponding to predetermined defocus deviations.
    [0014]
    14. A method for identifying biological particles according to claim 13, characterized in that one compares (670), using a similarity criterion, said plurality of blocks selected to the same pluralities of standard blocks, each plurality of standard blocks corresponding to a given type of biological particle, the type of the biological particle of interest being derived (680) from the plurality of type blocks having the greatest similarity with said plurality of selected blocks.
    [0015]
    15. The method of identification of biological particles according to claim 14, characterized in that the similarity criterion is an intercorrelation, a quadratic difference, a quadratic difference after spatial Fourier transform, a criterion based on a principal component analysis.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3234550B1|2022-01-05|Method for identifying biological particles using stacks of defocused holographic images
    EP3274689B1|2022-02-16|Method and apparatsu for analysing particles
    EP3274694B1|2019-12-18|Method for determining the state of a cell
    EP3433679B1|2020-11-04|Method for observing a sample, by calculation of a complex image
    EP3218769B1|2019-03-20|Analysis method including the holographic determination of a position of a biological particle
    EP3433678A1|2019-01-30|Holographic method for characterising a particle in a sample
    FR3056749A1|2018-03-30|METHOD FOR NUMBERING LEUKOCYTES IN A SAMPLE
    EP3559631B1|2021-01-06|Method for counting particles in a sample by means of lensless imaging
    EP3631416B1|2021-12-29|Method for analysing microorganisms
    EP3343201A1|2018-07-04|Method for counting and characterising particles in a moving fluid
    WO2017207184A1|2017-12-07|Device and method for acquiring a particle present in a sample
    EP3462381A1|2019-04-03|Method for detecting microorganisms in a sample
    EP3545362B1|2022-03-09|Method for forming a high resolution image by lensless imaging
    EP3637194B1|2021-11-03|Method for determining parameters of a particle
    WO2019243725A1|2019-12-26|Method and device for counting thrombocytes in a sample
    FR3082944A1|2019-12-27|METHOD FOR OBSERVING A SAMPLE WITH LENS-FREE IMAGING, TAKING INTO ACCOUNT A SPATIAL DISPERSION IN THE SAMPLE
    FR3075370A1|2019-06-21|METHOD FOR CALIBRATING AN ANALYZING DEVICE AND ASSOCIATED DEVICE
    FR3090107A1|2020-06-19|Method for characterizing a particle from a hologram.
    同族专利:
    公开号 | 公开日
    US10458897B2|2019-10-29|
    JP2018507392A|2018-03-15|
    CN107110762B|2020-09-01|
    CN107110762A|2017-08-29|
    EP3234550B1|2022-01-05|
    FR3030749B1|2020-01-03|
    US20170284926A1|2017-10-05|
    WO2016097092A1|2016-06-23|
    JP6644073B2|2020-02-12|
    EP3234550A1|2017-10-25|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    DE102005036326A1|2005-07-29|2007-02-01|P.A.L.M. Microlaser Technologies Ag|Method for analyzing a biological object, e.g. single cell, comprises recording a digital holographic micro-interferometric image of the object and analyzing the object by evaluating the image|
    WO2013041951A1|2011-09-22|2013-03-28|Foce Technology International Bv|Optical platelet counter method|
    KR100303608B1|1997-05-22|2001-11-22|박호군|Method and device for automatically recognizing blood cell|
    WO2000003308A2|1998-07-10|2000-01-20|Kansas State University Research Foundation|Particle image velocimetry apparatus and methods|
    US7948632B2|2005-12-22|2011-05-24|Phase Holographic Imaging Phi Ab|Method and apparatus for analysis of a sample of cells|
    WO2011149525A1|2010-05-25|2011-12-01|Arryx, Inc.|Holographic fluctuation microscopy apparatus and method for determining mobility of particle and/or cell dispersions|
    CN103238120B|2010-11-12|2016-08-10|布鲁塞尔大学|For characterizing the optical means of transparent grain|
    CN102014240B|2010-12-01|2013-07-31|深圳市蓝韵实业有限公司|Real-time medical video image denoising method|
    EP2661603A4|2011-01-06|2014-07-23|Univ California|Lens-free tomographic imaging devices and methods|
    HU229591B1|2011-05-03|2014-02-28|Mta Szamitastech Autom Kutato|Device for color three dimensional image creating|
    CA2842377C|2011-07-19|2019-08-27|Ovizio Imaging Systems N.V.|A method and system for detecting and/or classifying cancerous cells in a cell sample|
    EP2657792B1|2012-04-24|2017-11-15|Imec|Device and method for holographic reflection imaging|
    JP6265364B2|2012-07-24|2018-01-24|国立大学法人電気通信大学|Cell identification device, cell identification method, cell identification method program, and recording medium recording the program|
    EP2954309B1|2013-02-05|2019-08-28|Massachusetts Institute of Technology|3-d holographic imaging flow cytometry|
    WO2015195642A1|2014-06-16|2015-12-23|Siemens Healthcare Diagnostics Inc.|Analyzing digital holographic microscopy data for hematology applications|
    US10871745B2|2014-08-01|2020-12-22|The Regents Of The University Of California|Device and method for iterative phase recovery based on pixel super-resolved on-chip holography|AU2015212758B2|2014-01-30|2019-11-21|Bd Kiestra B.V.|A system and method for image acquisition using supervised high quality imaging|
    FR3028616A1|2014-11-13|2016-05-20|Commissariat Energie Atomique|ANALYSIS METHOD COMPRISING THE DETERMINATION OF A POSITION OF A BIOLOGICAL PARTICLE|
    FR3033359B1|2015-03-02|2017-04-07|Snecma|MONOBLOC DRAWING DISK HAVING A HUB HAVING AN EVIDENCE FACED BY A BODY COMPRISING SAME|
    FR3034196B1|2015-03-24|2019-05-31|Commissariat A L'energie Atomique Et Aux Energies Alternatives|PARTICLE ANALYSIS METHOD|
    CA2985848A1|2015-04-23|2016-10-27|Bd Kiestra B.V.|Colony contrast gathering|
    US10696938B2|2015-04-23|2020-06-30|Bd Kiestra B. V.|Method and system for automated microbial colony counting from streaked sample on plated media|
    FR3044415B1|2015-11-27|2017-12-01|Biomerieux Sa|METHOD FOR DETERMINING THE REACTION OF A MICROORGANISM TO ITS EXPOSURE TO AN ANTIBIOTICS|
    US10386289B2|2016-12-23|2019-08-20|miDiagnostics NV|Method and system for determining features of objects in a suspension|
    FR3066503B1|2017-05-22|2021-05-07|Commissariat Energie Atomique|MICROORGANISMS ANALYSIS PROCESS|
    FR3071609B1|2017-09-27|2019-10-04|Commissariat A L'energie Atomique Et Aux Energies Alternatives|METHOD FOR DETECTING MICROORGANISMS IN A SAMPLE|
    FR3073047B1|2017-11-02|2021-01-29|Commissariat Energie Atomique|OPTICAL PROCESS FOR ESTIMATING A REPRESENTATIVE VOLUME OF PARTICLES PRESENT IN A SAMPLE|
    CN108180867B|2018-01-09|2020-11-03|深圳大学|Quantitative phase measurement method, device and system|
    FR3081552B1|2018-05-23|2020-05-29|Commissariat A L'energie Atomique Et Aux Energies Alternatives|DEVICE AND METHOD FOR OBSERVING A FLUORESCENT SAMPLE BY DEFOCALIZED IMAGING|
    FR3087009B1|2018-10-09|2020-10-09|Commissariat Energie Atomique|PROCESS FOR DETERMINING PARAMETERS OF A PARTICLE|
    WO2020261826A1|2019-06-28|2020-12-30|富士フイルム株式会社|Image processing device, evaluation system, image processing program, and image processing method|
    法律状态:
    2015-12-31| PLFP| Fee payment|Year of fee payment: 2 |
    2016-06-24| PLSC| Publication of the preliminary search report|Effective date: 20160624 |
    2016-12-29| PLFP| Fee payment|Year of fee payment: 3 |
    2018-01-02| PLFP| Fee payment|Year of fee payment: 4 |
    2018-12-31| PLFP| Fee payment|Year of fee payment: 5 |
    2019-12-31| PLFP| Fee payment|Year of fee payment: 6 |
    2020-12-28| PLFP| Fee payment|Year of fee payment: 7 |
    2021-12-31| PLFP| Fee payment|Year of fee payment: 8 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1462998|2014-12-19|
    FR1462998A|FR3030749B1|2014-12-19|2014-12-19|METHOD OF IDENTIFYING BIOLOGICAL PARTICLES BY STACKS OF DEFOCALIZED HOLOGRAPHIC IMAGES|FR1462998A| FR3030749B1|2014-12-19|2014-12-19|METHOD OF IDENTIFYING BIOLOGICAL PARTICLES BY STACKS OF DEFOCALIZED HOLOGRAPHIC IMAGES|
    CN201580068881.3A| CN107110762B|2014-12-19|2015-12-17|Method for identifying biological particles using a stack of defocused holographic images|
    PCT/EP2015/080148| WO2016097092A1|2014-12-19|2015-12-17|Method for identifying biological particles using stacks of defocused holographic images|
    EP15820468.5A| EP3234550B1|2014-12-19|2015-12-17|Method for identifying biological particles using stacks of defocused holographic images|
    JP2017532858A| JP6644073B2|2014-12-19|2015-12-17|A method for identifying biological particles using a stack of defocused holographic images.|
    US15/536,507| US10458897B2|2014-12-19|2015-12-17|Method for identifying biological particles using stacks of defocused holographic images|
    [返回顶部]