Method and device for classifying biological particles in a sample

Abstract

Method for classifying biological particles, moving in a sample, the method comprising: acquiring images of the sample respectively at different measurement times; from the acquired images, monitoring the respective positions of different particles at the measurement times; then, at each measurement time, and for each particle,: from each acquired image, determining N characteristics of the particle, so as to obtain, at each measurement time, a vector of characteristics, of dimension N, N being an integer greater than 2; applying a first classification algorithm, so as to assign a temporal class to the particle, at the measurement time;the method being characterized in that it also comprises, from each temporal class assigned to the same particle, at the different measurement times, application of a classification algorithm so as to assign a typological class to said particle, the typological state class being representative of the evolution of the particle during the different measurement times.;

Description

Title of the invention: Method and device for classifying biological particles in a sample Technical field

[0001] The technical field of the invention is the characterization of biological particles, in particular cells, in a culture medium. PREVIOUS ART

[0002] In vitro live cell imaging is a technique used in cell biology and molecular biology to observe and record biological processes occurring inside cells in real time. Unlike fixed cell imaging, which captures static images, live cell imaging allows cellular responses to be observed in their natural, dynamic context. The main advantage of live cell imaging is that it allows the capture of spatiotemporal aspects of cellular responses that are not accessible through fixed cell imaging. This allows the evolution of cell development over time to be followed. This allows for the understanding of complex cellular mechanisms, such as cell division or cell migration.This also makes it possible to characterize individual changes affecting cellular development, whether stem cells, cells of the same lineage, or cells in response to pharmacological perturbations.

[0003] The Gordonov publication "Thne sériés modeling of live-cell shape dynamics for image-based phenotypic profiling” describes the imaging application of live cells of the MDA MB-231 type labeled with fluorescent markers. MDA MB-231 is a human breast cancer cell line. Different cells of the same line are arranged in different media, respectively containing different therapeutic actives. Cell images are performed by 2D epifluorescence microscopy, with xl0 magnification. Different images are acquired, allowing the identification and tracking of cells at different times. At each time, and for each cell, a large number of morphological characteristics are extracted. The extracted morphological characteristics, at each time, and for each cell, are subject to a dimension reduction by principal component analysis (PCA).Thus, for each cell, we can visualize a trajectory in a two-dimensional latent space, resulting from the PCA. The latent space is discretized into classes, each class corresponding to a morphological state of a cell. The publication shows that the cellular trajectory of a cell can pass, over time, through different morphological states, the probabilities of changes. state that can be modeled using an approach based on hidden Markov models. This approach makes it possible to establish temporal signatures representing the evolution of the morphology of cells of the same lineage subjected to different active ingredients.

[0004] However, such an approach requires individual modeling of the evolution of the morphology of each cell, which is not feasible if too many cells are available, or if the observation period is too long. In the aforementioned publication, 293 cells were taken into account, over 53 time intervals.

[0005] The publication Wang. “Live-cell imaging and analysis reveal cell phenotypic transition reveal cell phenotypic transition dynamics inherently missing in snapshot data” describes the application of live-cell imaging to A549 VIM-RFP cells, which are human lung carcinoma A549 cells genetically modified to express a red fluorescent protein (RFP). In this publication, morphological features of different cells are extracted as a function of time, and a trajectory of each cell in a latent space determined by PCA is determined, similarly to the previous publication.

[0006] However, it has been found that a classification of cells based on their trajectory is difficult, in particular when a large number of cells are available and when the trajectories are long, following an observation of several hours or even several days. First of all, each trajectory can be affected by noise linked to the acquisition of images, or to the extraction of morphological characteristics on each image. This makes a classification based on the trajectory difficult. In addition, a classification of trajectories is based on a calculation of distances between trajectories. However, the trajectories are established from a large number of images, typically more than 100, which makes the calculation of distances expensive in terms of computing time, in particular when dealing with trajectories of different lengths.Another difficulty related to the classification of trajectories comes from the fact that the lengths of the trajectories can be different from each other, because they are related to the lifetime of a cell in the analyzed environment.

[0007] The inventor proposes an improved method for imaging living cells, allowing characterization of the evolution of a large number of cells, typically several thousand, over long periods, typically several days. Statement of the invention

[0008] A first object of the invention is a method for classifying biological particles, moving in a sample, the method comprising: - a) acquisition of images of the sample respectively at different measurement times; - b) from the acquired images, monitoring of the respective positions of different particles at the measurement times;

[0009] then, at each measurement instant, and for each particle,: - c) from each acquired image, determination of characteristics of the particle, so as to obtain, at each measurement time, a vector of characteristics, of dimension N, N being an integer greater than or equal to 2; - d) application of a first classification algorithm, so as to assign a temporal class to the particle, at the measurement time;

[0010] the method being characterized in that it also comprises - e) from each temporal class assigned to the same particle, at different measurement times, application of a second classification algorithm so as to assign a typological class to said particle, the typological class being representative of the evolution of the particle during the different measurement times.

[0011] Step d) may include: - d-1) projection of the characteristic vector into a first latent space, whose dimension is less than N, so as to form a first latent vector, representative of the characteristics of the particle at the instant of measurement; - d-2) application of the first classification algorithm to the first vector latent in order to assign the temporal class to the particle, at the instant of measurement.

[0012] Step e) may comprise, for each particle: - e-1) constitution of a trajectory vector from the temporal classes assigned to said particle at each measurement time; - e-2) projection of the trajectory vector into a second latent space, of so as to form a second latent vector, representative of the particle; - e-3) application of the second classification algorithm to the second latent vector so as to assign the typological class to the particle.

[0013] The trajectory vector comprises a quantity or list of each time class to which the particle is assigned at each measurement time.

[0014] Step e-1) may comprise: - creation of a prediction neural network, configured to predict each temporal class assigned to a particle from the characteristics determined at each measurement time, the prediction neural network comprising: • an input layer, formed by the characteristics determined for a particle; • an output layer, formed by the temporal classes successively assigned to the particle at each measurement instant; • at least one intermediate layer, between the input layer and the output layer; - learning the neural network from the characteristics determined for different particles, as well as with the temporal classes successively assigned to each of said particles; - formation of the trajectory vector from an intermediate layer of the neural network.

[0015] During step c), each characteristic can be chosen from: - at least one morphological characteristic; - and / or at least one dynamic characteristic, corresponding to a variation of a morphological characteristic as a function of time or a speed of movement; - and / or at least one neighborhood characteristic, representing a particle density in the neighborhood of the particle; - and / or all or part of a segmentation mask, the segmentation mask corresponding to a region of interest extracted from the image, comprising the particle.

[0016] Each step c) may include a determination of at least two characteristics of the particle.

[0017] According to one possibility, each sub-step dl) comprises: - d-li) partition of the feature vector into K elementary vectors, K being an integer less than N; - d-lü) projection of each elementary vector into an elementary latent space, so as to form, in each elementary latent space, an elementary latent vector; - d-liii) projection of each elementary latent vector resulting from d-lii) into the first latent space.

[0018] During step a), the sample may be placed facing an image sensor, with no image-forming optics being placed between the sample and the image sensor.

[0019] During step a), the sample may be placed facing an image sensor, an optical system, of the lens or objective type, being placed between the sample and the image sensor, the optical system defining an object plane and an image plane. According to one possibility, the object plane is offset relative to the sample and / or the image plane. is offset from the image sensor. Alternatively, the optical system is focused on the sample.

[0020] The biological particle may be a cell or a microorganism.

[0021] According to one possibility, the method comprises, during step e) - Among the particles: • identification of particles detected on the image acquired at the first or last measurement instant, for which the trajectory, defined by the time classes successively assigned during the different measurement instants, is incomplete; • identification of particles detected after the image acquired at the first instant and not detected at the last instant of measurement, for which the trajectory, defined by the time classes successively assigned during the different instants of measurement, is complete; - for each particle whose trajectory is incomplete, identification of at least one model particle, with a complete trajectory, the start or end of whose trajectory corresponds to the incomplete trajectory of said particle; - typological classification of each particle with an incomplete trajectory based on the typological classification of at least one identified model particle.

[0022] The first latent space and / or the second latent space are defined during a learning phase. The same applies to the classes defined by the first classification algorithm and the second classification algorithm. The latter are defined, preferably in an unsupervised manner, during the learning phase.

[0023] A second object of the invention is a device for observing a sample, comprising: - an image sensor, configured to be placed facing the sample; - a processing unit, configured to collect images formed by the image sensor at different times, and programmed to implement steps c) to e) of a method according to the first subject of the invention.

[0024] The device may include the characteristics described in connection with the first subject of the invention.

[0025] A third object of the invention is a computer program, programmed to implement steps c) to e) of a method according to the first object of the invention.

[0026] The invention will be better understood upon reading the description of the exemplary embodiments presented in the remainder of the description, in conjunction with the figures listed below. FIGURES

[0027] [Fig.1A] shows schematically an example of a device according to the invention.

[0028] [Fig.lB] shows schematically another example of a device according to the invention.

[0029] [Fig.2] illustrates the main steps of a method for classifying biological particles according to the invention.

[0030] [Fig.3A] shows an image acquired by an image sensor, placed in a lensless configuration.

[0031] [Fig.3B] shows an image obtained by implementing a holographic reconstruction algorithm to the image shown in [Fig.3A].

[0032] Figures 4 and 5 illustrate a segmentation of the reconstructed image, so as to extract segmentation masks, each segmentation mask being representative of a cell, and preferably of a single cell.

[0033] [Fig.6] shows a temporal distribution of different characteristics.

[0034] [Fig.7A] shows a formation of a first latent vector from a vector of characteristics.

[0035] [Fig.7B] shows a trajectory of a cell in a first latent space.

[0036] [Fig.7C] shows a projection of a trajectory vector into a second latent space, different from the first latent space.

[0037] [Fig.7D] shows obtaining second latent vectors of different cells in the second latent space.

[0038] [Fig.8A] shows a distribution of a silhouette score of different temporal classes of cells.

[0039] [Fig.8B] represents the cell classes discussed in connection with [Fig.8A].

[0040] [Fig.9A] shows a distribution of cells in different temporal classes.

[0041] [Fig.9B] shows, for each cell, a proportion of classification in different temporal classes.

[0042] [Fig. 10] represents a distribution of cell characteristics, classified into different typological classes in a second latent space.

[0043] [Fig. 11 A] represents variations in characteristics of cells classified in different typological classes, in a second latent space different from the first latent space.

[0044] Figures 1 IB and 1 IC represent variations of characteristics of cells classified into different typological classes, in the first latent space.

[0045] [Fig. 12] schematizes a neural network, from which an intermediate layer can be extracted to form the trajectory vector of a cell.

[0046] Figures 13A to 13C illustrate alternative embodiments. PRESENTATION OF SPECIAL EMBODIMENTS

[0047] [Fig. 1A] represents an example of a device according to the invention. A light source 11 is configured to emit a light wave propagating towards a sample 10, along a propagation axis Z. The light wave is emitted according to a spectral band AX.

[0048] The sample 10 is a sample comprising biological particles 12, in particular cells or microorganisms, which it is desired to characterize. It may also be spores, bacteria, or even micro-algae.

[0049] In the example described, the particles 12 are cells of the same cell line (cells of the “fibroblast” type originating from genetically modified mice whose “KER” gene regulating the circadian cycle has been inhibited, bathed in a 10m culture medium. Although the cells are of the same cell line, they can evolve or develop differently from each other, in the culture medium, depending on the time. It is thus possible to determine sub-phenotypes, corresponding to different cell typologies, having different evolutions over time.

[0050] The sample 10 is, in this example, contained in a fluidic chamber 15 of thickness e. The thickness e of the sample 10, along the propagation axis Z, typically varies between 10 μm and 1 cm, and is preferably between 20 μm and 500 μm. The sample 10 extends along a plane, called the sample plane, preferably perpendicular to the propagation axis Z. It is held on a support 10s at a distance d from an image sensor 16.

[0051] The distance D between the light source 11 and the sample 10 is preferably greater than 1 cm. It is preferably between 2 and 30 cm. Advantageously, the light source, seen by the sample, is considered to be point-like. The light source 11 may be a light-emitting diode as shown in [Fig.1A]. It may be associated with a spatial diaphragm or filter or an optical fiber. The light source 11 may be a laser source, such as a laser diode.

[0052] Preferably, the spectral emission band AX of the light wave emitted by the source has a width of less than 100 nm. By spectral bandwidth is meant a width at half-height of said spectral band.

[0053] The sample 10 is arranged between the light source 11 and the image sensor 16 previously mentioned. The latter preferably extends parallel, or substantially parallel to the sample. The term substantially parallel means that the two elements may not be strictly parallel, an angular tolerance of a few degrees, less than 20° or 10° being allowed. The image sensor 16 is capable of forming an image I according to a detection plane Pq. In the example shown, it is an image sensor comprising a matrix of pixels, CCD or CMOS type. The detection plane preferably extends perpendicular to the Z propagation axis of the incident light source.

[0054] The configuration shown in FIG. 1A is a lensless imaging configuration, in which there is no imaging optics between the sample and the image sensor. This does not prevent the possible presence of focusing microlenses at each pixel of the image sensor 16, the latter having no function of magnifying the image acquired by the image sensor. The distance between the sample 10 and the pixel matrix of the image sensor 16 is advantageously between 50 μm and 2 cm, preferably between 100 μm and 2 mm.

[0055] The device shown in [Fig. 1 A] may be of the Cytonote type (supplier Iprasense), described in Allier C. et al, “CNN-based cell analysis: from image to quantitative representation”, Frontiers in Physics, 01 / 25 / 2022.

[0056] In [Fig. 1B], another configuration is shown, according to which the device comprises an optical system 17, of the lens 17 type, or combination of lenses. The optical system can be configured to form a focused image of the sample on the detection plane of the image sensor. Alternatively, the optical system can be configured to form a defocused image of the sample. The optical system comprises an object plane and an image plane. A defocused image is obtained when the object plane and / or the image plane are slightly offset respectively relative to the sample or the detection plane. By slightly offset, we mean by an offset distance of a few tens of μm.

[0057] The use of lensless imaging ([Fig.lA]) or defocused imaging ([Fig.lB]) is particularly suitable for images of transparent objects or objects with low contrast with the ambient environment. This makes it possible to observe cells without labeling with an exogenous colored or fluorescent marker. The evolution of the cells is thus observed, without the influence of a marker. The use of lensless imaging makes it possible to form images according to a high field of observation, which allows simultaneous observation of a large population of cells.

[0058] When the particles are not considered transparent, the image can be performed by implementing an image sensor coupled to a focused optical system. Preferably, the optical system allows the sample to be observed at a high depth of field. This type of configuration may be suitable for the observation of particles such as spermatozoa or microalgae.

[0059] The device comprises a processing unit 20, programmed to implement processing or calculation operations described below, in connection with [Fig. 2]. The instructions followed by the processing unit are stored in a memory 22, connected to the processing unit by wired or wireless connection. The processing unit 20 may for example include a microprocessor. The processing unit may be connected to a 24 screen.

[0060] [Fig.2] describes the main steps of the invention.

[0061] Step 100: acquisition of images of the sample at different times.

[0062] The images of the sample are acquired according to a lensless imaging configuration, at different measurement times f, with 1 <f< T.f est un entier désignant chaque instant de mesure. Dans l’image acquise, chaque cellule apparaît sous la forme d’une figure de diffraction, comme représenté sur la [Fig.3A]. Une telle image n’est généralement pas directement exploitable, sans le recours à un algorithme de reconstruction holographique tel que décrit dans l’étape 110. Le pas d’échantillonnage temporel peut être de quelques minutes ou quelques dizaines de minutes, par exemple 10 ou 20 minutes.

[0063] Step 110: application of a reconstruction algorithm.

[0064] By using a holographic reconstruction algorithm, a more usable image of the sample can be formed, allowing the latter to be observed. This may, for example, be a so-called phase image of the sample. In the phase image, each particle appears in the form of a spot, reflecting a variation in the optical path of the light wave, induced by the particle. The image is called a phase image, because the variation in the optical path is related to the phase shift of the light wave by the particle. Such a phase image can be obtained by implementing known reconstruction algorithms. Examples of reconstruction algorithms are described in US 10816454 or in the publication Hervé et al “Alternation of inverse problem approach and deep learning for lens-free microscopy image reconstruction”, Scientific Reports, 10(1), 20207. [Fig.3B] shows a phase image of each cell.It is understood that the phase image has sufficient quality to detect and delineate each cell. The application of the reconstruction algorithm is not necessary when the image is acquired with an image sensor coupled to an optical system focused on the sample.

[0065] Step 120: Preprocessing - segmentation

[0066] During this step, the reconstructed image is segmented so as to isolate each cell. The segmentation may be preceded by preprocessing steps, for example of the thresholding type, so as to facilitate the segmentation. The segmentation consists of extracting a segmentation mask from each reconstructed image corresponding to each cell. Preferably, a segmentation mask corresponds to a single cell. The segmentation is for example carried out with the Cellpose algorithm, described in Wang et al “Cellpose: a generalist algorithm for cellular segmentation” Nature methods, 18(1), 100-106. The algorithm Segmentation has preferably been trained, or partially re-trained on the culture of cells of interest, in this example PER (PERIOD)-TKO (Triple Knock Out) type cells. Re-training can also be carried out with a few cells from the sample, for example a few dozen. [Fig.4] shows a part of the observation image of the sample (phase image), after segmentation. Each clear part corresponds to a segmentation mask that can be easily extracted from the image.

[0067] [Fig.5] shows different thumbnails, each thumbnail representing a part of the extracted image, corresponding to a cell.

[0068] Step 130: Tracking

[0069] After segmentation, a tracking algorithm is implemented to track each cell at each measurement time t. The tracking algorithm is for example Trackpy, described in https: / / soft-matter.github.io / trackpy / v0.6.2 / . The tracking algorithm follows a cell in the sample until it divides. The division corresponds to a significant and rapid variation in morphological characteristics, generally a significant drop in dry mass coupled with a significant increase in thickness.

[0070] Step 140: Extraction of morphological characteristics

[0071] At each measurement time, segmentation makes it possible to extract two-dimensional morphological characteristics of the cell. These include, for example: - the area of ​​the segmentation mask; - the perimeter of the mask; - or the eccentricity of the mask. The eccentricity of the mask corresponds to the eccentricity of an ellipse fitted to the segmentation mask; - dry mass: this is a difference between the mass of the cell and the mass of the surrounding medium occupying the same volume of the cell. The dry mass can be estimated as described in the previously cited “Allier” publication, in particular in paragraph 2.1 of this publication. More precisely, the dry mass can be estimated as follows:

[0072] Sir designates a two-dimensional coordinate of a point of the cell, the latter induces a phase variation Ô(p{r ) = (p(r) - (p where Pq is the phase shift of the incident light wave due to the culture medium 10, and ^(r) is the phase shift of the incident light wave due to the cell. The quantity 5(p{r ), corresponds to the phase shift of the incident light wave due to the cell, relative to the medium.

[0073] The variation in optical path induced at each point of the cell, measured directly from the phase image, and noted OPD^r) is such that I0074! OPD(r) = (1)

[0075] Where 2 is the wavelength of the incident light wave.

[0076] By integrating expression (1) on the surface $ of the segmentation mask, we can estimate a difference in optical volume OVD induced by the cell, with

[0077] QVD = \DPD ( r} dr (2)

[0078] The dry mass CDM is estimated according to

[0079] CDM- (3), a is cell dependent. In mammalian cells, a = 0.18 pm3pg'.

[0080] Another morphological characteristic is the thickness of each cell, considered as homogeneous, established according to where OPJJ^r) is obtained, on the phase image, at a coordinate r corresponding to the cell. 5 is representative of a variation of the refractive index between the cell and the culture medium. <5 is for example equal to 0.02.

[0081] Advantageously, characteristics other than morphological ones are extracted, so as to enrich the information relating to each cell. These may be characteristics linked to the neighborhood of each cell, for example a density of cells around the cell considered. This is the number of cells identified in a predetermined neighborhood of the cell, for example in a region of interest of 100 pm x 100 pm centered on the cell.

[0082] It is also advantageous to take into account so-called dynamic characteristics, linked to the movement or evolution of the cell between different measurement times. The dynamic characteristics may include: - cell speed, i.e. the speed between two different measurement times, for example two consecutive measurement times: the speed between two consecutive measurement times is an instantaneous speed. It is also possible to take into account an average speed, between non-consecutive measurement times, for example measurement times offset by 3 time increments. - instantaneous or average variation of the instantaneous dry mass: this is the variation of the dry mass, as defined in (3), between two consecutive measurement times (instantaneous variation) or between non-consecutive measurement times (average variation). The average variation can, for example, be determined by considering measurement times offset by 3 time increments.

[0083] The characteristics can also concern the life of the cell, between its birth, that is to say its appearance on one of the acquired images and its division or its death. This can be: - lifetime: number of time increments between the first and last images of the sample on which the cell appears, multiplied by the acquisition time step (here 20 minutes), so as to have a duration in minutes; - the average speed during the life of the cell: distance traveled by the cell between its birth and division (or death) divided by the lifespan; - the average growth rate of the cell, from its birth to its division.

[0084] An important aspect of the invention is not to be limited to morphological characteristics, and to take into account the characteristics linked to the neighborhood and / or the dynamic characteristics and / or the characteristics linked to the lifespan of the cell as previously described.

[0085] Thus, at each instant, a cell can be described by a set of N characteristics, N being for example equal to thirteen. The characteristics are thus: - five morphological characteristics: area, perimeter, thickness, dry mass, eccentricity; - a neighborhood characteristic: density of cells in the neighborhood of the cell; - four dynamic characteristics: instantaneous speed, average speed (calculated over 3 time increments), instantaneous dry mass variation, average dry mass variation (calculated over 3 time increments); - three characteristics linked to the lifetime of the cell: lifetime, average speed calculated according to the lifetime, variation in average dry mass calculated according to the lifetime.

[0086] [Fig.6] shows an evolution of six characteristics averaged over all the cells in the sample, over a period of 3500 minutes (x-axis). The y-axes correspond respectively, from left to right and from top to bottom, to averages, for all the cells, of the following characteristics: area, dry mass, thickness, perimeter, eccentricity and instantaneous velocity. The vertical lines represent the confidence interval at ± lo.

[0087] Step 150: Filtering

[0088] During this step, cells considered to be anomalies are deleted. These may, for example, be cells leaving or entering the field of observation, or cells whose lifespan is too short, for example less than 4 hours, or immobile particles, which may not be cells, for example dust.

[0089] Step 1 55: Normalization

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097]

[0098]

[0099] Following step 140, each cell is assigned a feature vector xjf, of which each coordinate corresponds to a characteristic. n is an integer such that 1 < n < N, corresponding to a rank assigned to each characteristic. The index j designates the cell. The characteristics can be standardized, so that: XiÀn Or: - Xt ( n ) is the average of the nth characteristics for all the cells j at time t. - oixt( n ) ) is the standard deviation of the nth characteristics for all cells j at time t) The normalization step is advantageous, but optional, in which each variable is normalized. This allows each variable to have the same weight for the following steps. Step 1 6 0: Dimension reduction - Projection into a first latent space Following step 140, we have a set of characteristics for each cell, and this in each time increment. One objective of the method is to perform a temporal classification of each cell, based on the characteristics. The classification is called temporal because it is performed, for each cell, at each instant. At each instant, the number of characteristics is high, typically greater than 5, or even greater than 10. This is due to the consideration of characteristics other than morphological, for example, neighborhood characteristics of the cell or dynamic characteristics or characteristics related to the lifetime of the cell. Given the number N of characteristics, the classification can advantageously be performed in a first latent space L1, whose dimension M is less than N. The first latent space L1 is defined during a learning phase, by applying a dimension reduction algorithm to the characteristics. In Figure 7A, a cell is represented, from which a feature vector xj-t is extracted, and this at each instant t. The feature vector is projected into the first latent space LL. In the example shown, the first latent space Ll has a dimension M = 2, being defined by the axes XI and Yl. Tests have shown that the dimension of the first latent space Ll can be equal to 2 or 3, while allowing a satisfactory classification of the cells at each instant. Different dimensionality reduction methods are available. Generally speaking, a dimensionality reduction method is used to represent original coordinates, in a feature space, of dimension N, into coordinates in a latent space of dimension M, with M < N. The dimension of the latent space depends on the number of features and their complexity. Preferably, it is less than 10, or even less than 5 so as to facilitate classification. The transition from the feature space to the latent space is carried out by a projection function f applied to the feature vector. The projection of the vector xjJt into the first latent space forms a first latent vector vjf, of dimension M.

[0100] Several dimension reduction methods are known to those skilled in the art. For example, principal component analysis (PCA) is an unsupervised and linear dimensionality reduction method. Nonlinear methods can be implemented, for example the UMAP (Uniform Manifold Approximation and Projection) method.

[0101] This dimension reduction step, although often advantageous for high-dimensional or correlated data, is however optional.

[0102] Step 170: Temporal classification.

[0103] During this step, each cell is classified according to the characteristics projected into the first latent space, forming the first latent vector vji.

[0104] Classes Kl / , have previously been defined in the first latent space LL The index p is an integer such that 1 < p < P, P designating the number of classes defined in the first latent space. The classes are identical for all the measurement instants Thus, at each instant a class is assigned to each particle, according to the first latent vector PjJ resulting from a projection of the characteristic vector XP in the first latent space LL

[0105] In [Fig.7A], 4 classes Kl,l, Kl,2, Kl,3 and Kl,4 are represented, each class being represented by a dotted circular area. The classes were defined in an unsupervised manner, on the basis of unannotated cells. This type of unsupervised classification is usually referred to by the Anglo-Saxon term clustering, each cluster representing a class established in an unsupervised manner.

[0106] The temporal classification of each cell is carried out at each instant, hence the designation “temporal” classification. It is thus possible to determine a trajectory of each cell between the classes Kl / over time, as shown in Figure 7B. In Figure 7B, a trajectory of a cell has been materialized between a first instant t=l and a last instant t = T. The trajectory designates the succession of classes Kl / successively assigned to each cell over time.

[0107] Step 180: Formation of the trajectory vector

[0108] Considering a total acquisition duration of two days, and a time sampling step of 20 minutes, we obtain for each cell, a series

[0109]

[0110] of classes of the size corresponding to the lifetime of the cell divided by the time sampling step, knowing that the number of cells can exceed several thousand. Step 170, implemented for each cell and at each instant, makes it possible to obtain as many trajectories, in the first latent space, as there are cells, and as many points per trajectory as there are measurement instants at which the cell is detected on the image. A classification of the trajectories in the first latent space is complex to implement, and does not allow sufficient separation between the cells. During this step, the trajectory does not correspond to a sequence of coordinates in the first latent space, but to a chronological sequence of temporal classes defined following step 170. Taking classes into account, and not successive coordinates of each cell in the first latent space, makes it possible to limit the influence of noise resulting from image acquisition or the various operations preceding classification, in particular segmentation. In addition, the assignment of classes to each cell is an approach consistent with the evolution of each cell during its development: each class can be considered as a state in which the cell finds itself, during its development. Chronologically assigning different temporal classes to a cell corresponds to a biological reality, reflecting the evolution of the state of the cell over time. During this step, we form a trajectory vector Cj, each term of which çi LÀ is an integer between 1 and p, corresponding to a temporal class assigned to the particle at each instant f. The trajectory vector Cj constitutes a signature of the evolution of the 12j particle in the sample. This signature makes it possible to trace how the cell develops in the medium. The dimension of the trajectory vector corresponds to the number of time increments T. Other types of trajectory vectors can be considered, for example based on a proportion of classes assigned to each cell during its trajectory, as described later.

[0111] When the dimension reduction step (step 170) is not implemented, the classification is performed in a space where each axis corresponds to a measured characteristic. As previously described, this is an unsupervised classification, the classes being defined with unannotated cells.

[0112] According to one possibility, the classes are defined in an unsupervised manner using a part of the cells in the sample, forming a training population. The other part of the sample is then made up of so-called test cells, which can be clustered into the classes defined with the training population.

[0113] According to another possibility, the classes were previously defined, during a previous learning session, with cells considered to be identical or representative of the cells of the analyzed sample.

[0114] Step 190: Typological classification

[0115] In order to perform a classification allowing a better separation of cells having different signatures, the trajectory vector of each cell is subject to a dimension reduction, in a second latent space L2, as shown in [Fig.7C]. The second latent space L2 is different from the first latent space L1, but the same dimension reduction methods as in step 160 apply.

[0116] In Figure 7C, a second latent space L2 of dimension 2 is represented, defined by the axes X2 and Y2. Typological classes K2 / / are defined in the second latent space. The index 9 is an integer such that 1 <q<Q, Q désignant le nombre de classes définies dans le deuxième espace latent. On assigne une seule classe typologique à chaque trajectoire. Ainsi, la classification typologique correspond à une classification de trajectoires.

[0117] The projection of the vector C j into the second latent space L2 makes it possible to form a second projected vector WJ, called a typological vector, to which a class K2,q is assigned. This is a so-called typological vector, because it depends on the type of cell, and in particular on the evolution of the cell during the measurement times.

[0118] Each typological class K2,q is representative of the typology of the cell in the sample. While the temporal classification, in the first latent space L1, allows to determine the state of a cell at each instant, the typological classification, in the second latent space L2, allows a discrimination, between the cells, according to their development. This is a classification intrinsic to each cell.

[0119] According to one possibility, during step 180, the trajectory vector represents a number of times at which the cell is assigned to each temporal class. In this case, the trajectory vector Cj has a dimension equal to the number of classes P, and each term Cj(p) corresponds to a quantity, possibly normalized, of times at which the cell is classified in the class Kl,p. This makes it possible to reduce the dimension of the trajectory vector, at the expense of a loss of information regarding the chronology of the temporal classification. Taking into account a trajectory vector in which each term represents a quantity of times at which the cell has been classified in each class makes it possible to overcome the number of temporal increments in which the cell is observed and classified. This makes it possible to obtain, for each cell, a trajectory vector of the same length, regardless of the cell's lifetime.

[0120] Figure 7D schematizes a typological classification of different cells, each point corresponding to a typological vector WJ assigned to a cell. Each typological class is representative of a particular development of the cell. One or more typological classes may be representative of an abnormal development of cells, or of a development disturbed by the presence of an active principle in the culture medium.

[0121] The steps previously described were implemented on a total number of 6434 cells, with 180 time increments spaced 20 minutes apart (confirm). The cells were segmented into a training set (5148 cells) and a test set (1286 cells). The training set made it possible to determine the first latent space, the second latent space and to define the classes in each of them.

[0122] After a reduction of dimensions to a 2D space using the UMAP algorithm, classical unsupervised classification algorithms were implemented such as Kmeans, Birch (Balance iterative reducing and clustering using hierarchies) or GMM (Gaussian Mixture Models), the number of classes P varying between 2 and 6. The classification performances were quantified by a silhouette score, the use of which is classical to determine the performance of an unsupervised classifier, and to choose the number of classes.

[0123] Figures 8A and 8B illustrate the performance of a classifier using a silhouette score. In [Fig.8B], a classification of cells from the test set is shown, based on their characteristics as previously described, in a first two-dimensional latent space. The classification was performed by implementing the Birch classification algorithm with 4 classes.

[0124] The silhouette score s is defined by:

[0125] s = (4) max(«ï,W

[0126] where

[0127] a = average intra-class distance

[0128] b = distance between a point of the class and another nearest class.

[0129] In [Fig.8A], we have plotted, for 4 classes represented, the silhouette score of each point, in the form of a histogram. The y-axis of [Fig.8A] corresponds to each class, identified by an integer (0, 1, 2 and 3) and the x-axis corresponds to the silhouette score. For each class, we obtain a histogram, that is to say a number of points of the class presenting the score.

[0130] The dotted line corresponds to an average of the silhouette score for all points. The closer this average is to 1, the better the classification performance. Other classification scores can be considered, for example the Davies-Bouldin score or the Calinski-Harabats score.

[0131] In [Fig.8B], we distinguish 4 classes, which correspond to a typology of particular cells: - Class 0: large cells (area, perimeter), more eccentric than the average, and presenting a movement speed higher than the average of the other cells. - Classes 1 and 2: small cells, differentiated by the density of cells in their vicinity; - Class 3: cells with a longer lifespan, thicker and whose movement speed is lower than the average of the other cells.

[0132] It is observed that considering dynamic characteristics (speed of movement) or those relating to lifespan makes it possible to discriminate between cells based on these criteria, and not only on taking into account morphological criteria.

[0133] [Fig.9A] shows the number of different classes assigned to each trajectory for the first step of temporal classification, for all 6434 cells. We observe that a majority of cells are successively assigned to different classes, and in particular 3 or 4 classes. This demonstrates the relevance of the temporal classification of cells, according to the measured characteristics, possibly after dimension reduction. The number of cells whose trajectory is limited to a single class is very small.

[0134] [Fig.9B] shows, for 1250 cells (y-axis), a proportion of classification in each class (x-axis). The sum of the proportions, on the same line, is equal to 1.

[0135] The trajectory vectors as represented in [Fig.9B], of dimension 2, comprise 4 classes. [Fig. 10] shows, for each class of the second classification step, a histogram of the values ​​of each characteristic. From top to bottom, the characteristics are: thickness, area, perimeter, density, lifetime. We observe that:

[0136] Class 0 of the second latent space (K2,0) is distinguished by the distribution of area, perimeter and density in the vicinity of the cell: larger and less dense cells.

[0137] Class 3 of the second latent space (K2,3) is distinguished by thickness and lifetime.

[0138] The different typologies of cells, corresponding respectively to each class of the second latent space, are represented in [Fig. 11A]. In [Fig. 11A], the average in each class over the average of the entire population (6434 cells) is represented for different characteristics. [Fig. 11A] should be compared with figures 11B and 11C, which represent classifications of cell trajectories carried out in the first latent space, using standard unsupervised classification methods: kmeans ([Fig. 11B]) with the distance of time series "dynamic time wrapping", kmeans with a classic Euclidean distance carried out on trajectories having the same length, the adaptation of each length being carried out by linear interpolation (see [Fig. 11B]). These methods do not allow grouping by classes with significant differences in certain characteristics between the different classes.

[0139] [Fig. 12] illustrates a variant of forming a trajectory vector, representative of the classes successively assigned to each cell during the temporal classification. We have seen previously that this vector can be formed from a list of each temporal class successively assigned to a cell, or from a proportion of assignments of each class to a cell. According to this variant, a neural network is implemented, as shown schematically in [Fig. 12], comprising an input layer, several intermediate layers and an output layer. The input data of the input layer are the set of characteristics successively established for a cell. The output layer corresponds to a chronology of the classification of the cell, obtained by implementing the chosen classification algorithm, for example BIRCH.The neural network is trained using the classes assigned chronologically to each cell. It is possible to take into account an intermediate layer, and to form the trajectory vector from the value of the nodes of the intermediate layer. The value of the nodes of each intermediate layer corresponds to a representation of the temporal classification carried out for the cell. This is a less direct representation than that based on a chronological list or a proportion of classes.

[0140] Figures 13A to 13C illustrate variations of the steps previously described.

[0141] In Figure 13A, a variant is shown according to which the vector of characteristic xjJ is partitioned, according to the characteristics, so as to form several elementary characteristic vectors, the number of elementary vectors being equal to K. For example, the morphological characteristics form a first elementary vector xj£ï, and the dynamic characteristics form a second elementary vector xjj2. Each elementary vector is projected into an elementary latent space: the first elementary vector xj.t\ is projected into a first elementary latent space Ll,l, to form a first elementary latent vector The first elementary latent space is defined by the axes X 1,1 and Y 1,1. The second elementary vector vjjA is projected into a second elementary latent space Ll,2, to form a second elementary latent vector vjt2. The second elementary latent space is defined by the axes XI,2 and Y 1,2. The elementary latent vectors vj / 2 and vjt2 are then projected into the first latent space LL

[0142] [Fig.13B] illustrates the possibility of using thumbnails, as represented in [Fig.5], each thumbnail corresponding to a cell, in the feature vector: the intensity assigned to each pixel in the cell is taken into account. As described in connection with [Fig.13A], the segmentation mask can be projected into a first elementary latent space while the other features are projected into another, or even other, elementary latent space(s).

[0143] As previously indicated, the invention takes advantage of taking into account characteristics different from the morphological characteristics of the prior art. These include dynamic characteristics (speed, variation in dry mass), or characteristics related to the lifetime of the cell (lifetime, average speed during the lifetime of the cell, variation in dry mass during the lifetime of the cell), or characteristics related to the neighborhood (cell density in the neighborhood of the cell). These characteristics can be taken into account to establish a characteristic vector, the latter being projected into the first latent space, possibly through intermediate projections into an elementary latent space (see Figures 13A and 13B). A trajectory vector Cj is thus obtained, representative of the successive temporal classifications of the particle in the first latent space.

[0144] According to the possibility illustrated in [Fig.l3C], certain characteristics, in particular the characteristics related to the lifetime of the cell, are added to the trajectory vector, before the projection into the second latent space.

[0145] A difficulty in classifying trajectories is related to the variability of the life of the cells in the sample. During the first image acquisition, some cells are already living, and divide or die during the image acquisition period. During the last image acquisition, some cells, born during the image acquisition period, are still living. Thus, the cells detected on the first or last image are cells whose life began before the first image, or whose life continues after the last image. For these cells, the trajectory, defined by the successively assigned temporal classes, is said to be incomplete, because it does not take into account the entire cell life.

[0146] We also have cells born and dead (or dividing) between the first and last image. For these cells, the trajectory is complete, because it takes into account the entire cell life.

[0147] Finally, the quantity of cells whose life begins and ends between the first image and the last image may be in the minority. This results in a diversity of the length of the trajectory of the different cells, the length of the trajectory depending on the number of images on which the cell appears.

[0148] Temporal classification is performed on all cells for all measurement times, except for the filtering step. However, cells with incomplete trajectories are preferably classified differently than cells with complete trajectories. The latter are easily detectable, since they are cells that appear and disappear during the acquisition period, i.e. between the first and last measurement times.

[0149] According to one possibility, only cells whose trajectory is complete are subject to a typological classification as previously described. The shorter trajectories correspond to cells which appear or disappear during the acquisition period.

[0150] Cells with incomplete trajectories contain usable information that can be used. To do this, each cell with an incomplete trajectory is compared: - either at the start of the complete trajectories, to address the incomplete trajectories corresponding to the cells born during the acquisition period; - either at the end of the complete trajectories, to address the incomplete trajectories corresponding to the cells which began their life before the acquisition period.

[0151] For each cell whose trajectory is incomplete, the comparison makes it possible to identify one or more complete “model” trajectories, the first or last time classes of which are closest to the time classes assigned to the incomplete trajectory considered. Thus, for each cell whose trajectory is incomplete, one or more complete model trajectories are determined. The proximity between an incomplete trajectory and a complete trajectory can be established by calculating a Levenstein distance between the sequence of time classes respectively assigned to each trajectory.

[0152] When only one model trajectory is determined, the incomplete trajectory is assigned to the typological class assigned to the model trajectory.

[0153] When several model trajectories are determined for the same incomplete trajectory, and these model trajectories are in the same typological class, the incomplete trajectory is assigned to said typological class.

[0154] When several model trajectories are determined for the same incomplete trajectory of a cell, and these model trajectories are in different classes typological, we carry out an identification of the model trajectory closest to the incomplete trajectory, by carrying out a comparison, at each instant of the trajectory, between: the feature vector of the cell associated with the incomplete trajectory; and the cell feature vector associated with each model trajectory.

[0155] At each instant, the comparison between the two vectors can be carried out by taking into account a Euclidean distance.

[0156] The comparison makes it possible to select the cell, with a complete trajectory, whose characteristic vectors, over time, are closest to the characteristic vectors of the cell with an incomplete trajectory. The cell with an incomplete trajectory is assigned to the class of the selected cell with a complete trajectory.

[0157] Although described in connection with cells, the invention can be applied to a classification of the evolution of microorganisms, for example bacteria or microalgae, or to the classification of mobile cells, for example spermatozoa.

Claims

Claims

1. Method for classifying biological particles (120, moving in a sample (10), the method comprising: - a) acquiring images of the sample respectively at different measurement times ('); - b) from the acquired images, monitoring the respective positions of different particles at the measurement times; then, at each measurement time, and for each particle,: - c) from each acquired image, determining characteristics of the particle, so as to obtain, at each measurement time, a characteristics vector (XM), of dimension N, N being an integer greater than or equal to 2; - d) applying a first classification algorithm, so as to assign a temporal class (Kl,^) to the particle, at the measurement time;the method being characterized in that it also comprises - e) from each temporal class assigned to the same particle, at different measurement times, application of a second classification algorithm so as to assign a typological class (K2,?) to said particle, the typological class being representative of the evolution of the particle during the different measurement times.;

2. Method according to claim 1, in which step d) comprises: - d-1) projection of the characteristic vector into a first latent space (L1), the dimension of which is less than N, so as to form a first latent vector (vj7), representative of the characteristics of the particle at the measurement instant; - d-2) application of the first classification algorithm to the first latent vector so as to assign the temporal class to the particle, at the measurement instant.

3. A method according to any one of claims 1 to 5, wherein step e) comprises, for each particle: - e-1) constitution of a trajectory vector from the temporal classes assigned to said particle at each measurement instant; - e-2) projection of the trajectory vector into a second latent space (L2), so as to form a second latent vector representative of the particle; - e-3) application of the second classification algorithm to the second latent vector so as to assign the typological class to the particle.

4. The method of claim 3, wherein the trajectory vector comprises a quantity or list of each time class to which the particle is assigned at each measurement time.

5. Method according to claim 3, in which step e-1) comprises: - constituting a prediction neural network, configured to predict each temporal class assigned to a particle from the characteristics determined at each measurement instant, the prediction neural network comprising: • an input layer, formed by the characteristics determined for a particle; • an output layer, formed by the temporal classes successively assigned to the particle at each measurement instant; • at least one intermediate layer, between the input layer and the output layer; - training the neural network from the characteristics determined for different particles, as well as with the temporal classes successively assigned to each of said particles; - forming the trajectory vector from an intermediate layer of the neural network.

6. Method according to any one of the preceding claims, wherein during step c), each characteristic is chosen from: - at least one morphological characteristic; - and / or at least one dynamic characteristic, corresponding to a variation of a characteristic morphological as a function of time or a speed of movement; - and / or at least one neighborhood characteristic, representing a density of particles in the neighborhood of the particle; - and / or all or part of a segmentation mask, the segmentation mask corresponding to a region of interest extracted from the image, comprising the particle.

7. A method according to any preceding claim, wherein each step c) comprises determining at least two characteristics of the particle.

8. Method according to any one of the preceding claims and claim 2, in which each step d) comprises: - di) partitioning the characteristic vector (xj^ into K elementary vectors, K being an integer less than N; - d-ii) projection of each elementary vector into an elementary latent space (Ll,l, Ll,2), so as to form, in each elementary latent space, an elementary latent vector - d-iii) projection of each elementary latent vector resulting from d-ii) into the first latent space.

9. A method according to any preceding claim, wherein in step a), the sample is arranged facing an image sensor (16), with no imaging optics being arranged between the sample and the image sensor.

10. Method according to any one of claims 1 to 8, wherein - during step a), the sample is arranged facing an image sensor (16), an optical system (17), of the lens or objective type, being arranged between the sample and the image sensor, the optical system defining an object plane and an image plane; - the object plane is offset relative to the sample and / or the image plane is offset relative to the image sensor.

11. Method according to any one of claims 1 to 8, wherein during step a), the sample is placed facing an image sensor, an optical system, of the lens or objective type, being placed between the sample and the image sensor, with the optical system focused on the sample.

12. A method according to any preceding claim, wherein the biological particle is a cell or a microorganism.

13. Method according to any one of the preceding claims, comprising, during step e): - Among the particles: • identification of particles detected on the image acquired at the first or last measurement instant, for which the trajectory, defined by the temporal classes successively assigned during the different measurement instants, is incomplete; • identification of particles detected after the image acquired at the first instant and not detected at the last measurement instant, for which the trajectory, defined by the temporal classes successively assigned during the different measurement instants, is complete; - For each particle whose trajectory is incomplete, identification of at least one model particle, with a complete trajectory, the beginning or end of the trajectory of which corresponds to the incomplete trajectory of said particle;- Typological classification of each particle with an incomplete trajectory based on the typological classification of at least one identified model particle.;

14. Device for observing a sample, comprising: - an image sensor, configured to be arranged facing the sample; - a processing unit, configured to collect images formed by the image sensor at different times, and programmed to implement steps c) to e) of a method according to any one of the preceding claims.

15. A computer program, programmed to implement steps c) to e) of a method according to any one of claims 1 to 13 when said program is executed on a computer.

Images