CN117917997A

CN117917997A - Detection of cell aggregates using quantitative phase contrast microscopy

Info

Publication number: CN117917997A
Application number: CN202280051544.3A
Authority: CN
Inventors: 奥利弗·海登; 乔安娜·埃伯; 塞巴斯蒂安·拉施; 托比亚斯·拉迈尔; 斯特凡·罗伊霍; 克里斯蒂安·克林克
Original assignee: Technische Universitaet Muenchen
Current assignee: Technische Universitaet Muenchen
Priority date: 2021-07-29
Filing date: 2022-07-29
Publication date: 2024-04-23
Also published as: CN117751281A

Abstract

Methods of detecting cell aggregates of biological cells using quantitative phase-contrast microscopy and devices for detecting cell aggregates of biological cells using the methods are disclosed. The method includes preparing a suspension including biological cells from a sample. A flow of the suspension is generated along a microfluidic channel to viscoelastically and/or hydrodynamically focus cell aggregates in the suspension in a focal plane of the quantitative phase-contrast microscope. One or more phase-shifted images of the biological cells in the suspension are taken using the quantitative phase-contrast microscope. Cell aggregates in the one or more phase shift images are identified. The sample is a whole blood sample or a blood component sample, and identifying cell aggregates in the one or more phase-shifted images includes identifying platelet aggregates in the one or more phase-shifted images.

Description

Detection of cell aggregates using quantitative phase contrast microscopy

Technical Field

The present invention relates to the field of biomedical research and clinical diagnostics. In particular, the present invention relates to a method for detecting cell aggregates of biological cells using a quantitative phase-contrast microscope and an apparatus for detecting cell aggregates of biological cells.

Background

Conventional cytometry determines parameters such as the number of red blood cells, platelets, white blood cells, and subtypes thereof in a patient's blood sample, which can be used to diagnose many different diseases. Recent studies have shown that aggregates of blood cells, such as leukocyte-platelet aggregates and leukocyte aggregates, can also be used as useful biological indicators of a variety of pathological conditions, such as cardiovascular disease and bacterial or viral infections, see, e.g., m. 29 (7) 677-685,J.G.Burel et al, eLife, 2019;8:e46045, and Michelson et al, circulation.2001;104:1533-1537.

However, reliable detection of blood cell aggregates is challenging because aggregates are fragile objects and may easily disintegrate. This prevents analysis of blood counts such as Mie scattering or fluorescence-based flow cytometry using conventional methods, which may require complex and time-consuming sample preparation, including selective lysis of red blood cells, staining of cellular components, and/or fluorescent labeling. Sample preparation and measurement itself can affect cell morphology and can lead to disintegration of cell aggregates, for example due to mechanical forces applied to them in centrifugation or high flow (typically 1-10 m/s) flow cytometry, a necessary condition for adequate statistics of individual cells.

Digital holographic microscopy uses interference between an imaging beam and a reference beam to obtain phase and amplitude information of light transmitted by a sample and allows, for example, the reconstruction of quantitative phase-shifted images of the sample, see, for example, EP 1524491 A1 and EP 2357539 A1. In recent years, digital holographic microscopy has been successfully used in biomedical applications, such as living cell imaging. The phase shift image of the cells can be used to reliably identify cell types based on analysis of morphological parameters and/or using a machine learning classifier. In combination with a microfluidic system, this allows for example high throughput label-free blood sample analysis, such as blood cell counting, see e.g. US2019/0195774 A1, to aid in the diagnosis of diseases such as malaria, leukemia and myeloproliferative neoplasms, see e.g. m.ugele et al, adv.sci.1800761 (2018), WO 2019/063548 A1 and m.ugele et al, proc.spie 11060, optical methods IV for biomaterial detection, characterization and imaging (Op-tical Methods for Inspection, characterization, AND IMAGING of Biomaterials IV), 110600V (2019). Furthermore, digital holographic microscopy has been used to conduct quantitative analysis of platelet aggregates adhering to surfaces, see for example WO2016/170180A1. However, the currently known methods are either unable to reliably detect cell aggregates or are not suitable for automated high-throughput analysis in a clinical setting. Furthermore, these methods provide only limited information about the composition of the cell aggregates, and thus may not be suitable for obtaining clinically relevant information, which may for example require analysis of both individual platelets and large platelet-platelet aggregates.

Disclosure of Invention

It is therefore an object of the present invention to provide a method that allows for rapid and reliable detection of biological cells and cell aggregates and is suitable for automated high throughput analysis in a clinical setting.

This object is achieved by a method for detecting cell aggregates of biological cells using a quantitative phase contrast microscope according to the first aspect of the invention and a device for detecting cell aggregates of biological cells according to the first aspect of the invention.

The method for detecting cell aggregates of biological cells using a quantitative phase contrast microscope according to the first aspect of the present invention comprises preparing a suspension comprising biological cells from a sample and a viscoelastic fluid. The viscoelastic fluid comprises a shear-thinning polymer having a molecular weight of from 2MDa to 10MDa, wherein the mass fraction of the shear-thinning polymer in the suspension is less than 0.2%. The method further includes generating a flow of the suspension along the microfluidic channel to viscoelastically focus the cell aggregates in the suspension in a focal plane of the quantitative phase-contrast microscope. One or more phase-shifted images of biological cells in suspension are captured using a quantitative phase-contrast microscope, and cell aggregates in the one or more phase-shifted images are identified.

The method may be performed, for example, using an apparatus for detecting cell aggregates of biological cells according to any of the embodiments described below. The sample may be a sample extracted from a patient, in particular a liquid sample, such as a blood sample. Thus, the biological cells may be or comprise, for example, blood cells, such as erythrocytes (erythrocytes), leukocytes (leukocytes), and/or platelets (thrombocytes), and/or rare cells, such as circulating tumor cells and/or circulating endothelial cells.

Suspensions may be prepared by adding viscoelastic fluids to the sample or cells extracted from the sample, and vice versa. Viscoelastic fluids are fluids that have both viscous and elastic properties, i.e., that exhibit the properties of a viscous fluid as well as the properties of an elastic solid. The viscoelastic fluid may be a non-newtonian fluid exhibiting a viscosity dependent upon the applied shear rate, in particular a shear-thinning fluid exhibiting a viscosity that decreases with the applied shear rate. The viscoelastic properties of the viscoelastic fluid may be at least partially produced by a shear-thinning polymer contained therein. The viscoelastic fluid may, for example, be an aqueous solution comprising a shear-thinning polymer, such as a solution consisting of water or phosphate buffered saline and a shear-thinning polymer.

When a suspension flow is generated, the viscoelastic properties of the viscoelastic fluid may result in viscoelastic focusing of objects (e.g., cells and/or cell aggregates) contained in the suspension. The flowing object may for example migrate towards a region of low shear rate, such as a central region of suspension flow, where the suspension has the highest flow rate. In a microfluidic channel, this may be, for example, near a center plane between two opposing sidewalls of the microfluidic channel, or near a center line between two pairs of opposing sidewalls of the microfluidic channel.

The viscoelastic focusing of cells and/or cell aggregates in the suspension flow is adjusted such that the cells and/or cell aggregates in the suspension are focused in the focal plane of the quantitative phase-contrast microscope. For example, a viscoelastic fluid may result in viscoelastic focusing of cells and/or cell aggregates at the center of suspension flow, e.g., near the center plane or centerline of a microfluidic channel. In some embodiments, the center plane and/or the center line of the microfluidic channel may be located in the focal plane of the quantitative phase-contrast microscope. The cell aggregates in the suspension may be focused such that the center of the cell aggregates is confined within a limit in a direction perpendicular to the focal plane of the microscope. For example, at least 90% of the cell aggregates, in one example at least 95% of the cell aggregates, may be limited to a restricted range. Preferably, the limit is less than 20 μm, in some examples less than 10 μm, in one example less than 5 μm. The limit range may be in particular equal to or less than twice the depth of field of the microscope, in one example equal to or less than the depth of field of the microscope. In a preferred embodiment, the individual cells are also focused in the focal plane of the microscope, e.g., such that at least 80% of the individual cells, in some examples at least 90% of the individual cells, in one example at least 95% of the individual cells are confined within a limit.

Viscoelastic focusing with a shear-thinning polymer having a molecular weight of 2MDa to 10MDa, with a mass fraction in suspension of less than 0.2%, can allow for reliable focusing of cell aggregates as well as individual cells, while reducing mechanical stress on the cell aggregates and preventing polymer-induced cell aggregation. Focusing cell aggregates, particularly combinations of cell aggregates and individual cells, is challenging due to the different sizes that cell aggregates and cells may have. The size of the cell aggregates may be, for example, 1 μm to 50 μm, and the size of the cells may be 1 μm to 20 μm. For example, human platelets typically have a size of 1 μm to 3 μm, while leukocytes typically have a size of 7 μm to 15 μm. The force acting on an object in a viscoelastic fluid may depend on the size of the object, such that objects of different sizes may be focused at different points/locations, or objects of a certain size may not be focused at all. At the same time, shear stresses in the fluid flow may act on the cell aggregates and may cause them to break apart, especially at higher flow rates. This may limit the range of flow rates that can be used, making focusing the cell aggregates even more difficult. In addition, the shear-thinning polymer can affect cell morphology, see, e.g., j. Gonzalez-Molina et al, sci Rep 9,8505 (2019), and in some cases can even induce the formation of "artificial" cell aggregates, e.g., for erythrocytes, as described in j. K. Armstrong et al, biophysical Journal 87 (2004), 4259-4270. Surprisingly, the inventors of the present invention have found that adding a shear-thinning polymer having a molecular weight of 2MDa to 10MDa at a mass fraction of less than 0.2% induces viscoelasticity in the suspension, which enables the use of viscoelastic focusing to sufficiently restrict cell aggregates in the suspension at a reduced flow rate to prevent disintegration of the cell aggregates while not inducing the formation of "artificial" cell aggregates. Thus, the present invention allows the investigation of cell aggregates, in particular blood cell aggregates, by quantitative phase-contrast microscopy without the need for cell fixation and erythrocyte lysis. Furthermore, aggregate testing may be performed using standard blood collection procedures for conventional blood collection tubes.

To detect the cell aggregates, the method further comprises capturing one or more phase-shifted images of the biological cells in suspension using a quantitative phase-contrast microscope. The quantitative phase contrast microscope may be, for example, a stacked imaging device or a digital holographic microscope, for example as described in detail below for the device according to the first aspect of the invention. Capturing one or more phase shifted images may, for example, include capturing a sequence of images of a measurement volume along the microfluidic channel as the suspension flows along the microfluidic channel. As used herein, a phase shift image may encode the phase shift of light of one or more wavelengths as a function of position, e.g., the phase shift of light reflected or propagating through an imaging sample, e.g., the flow of a suspension as a function of position in the imaging sample. In a preferred arrangement, as opposed to conventional flow cytometry with continuous cell measurements, multiple cells and/or cell aggregates are imaged in parallel to compensate for reduced throughput due to lower flow rates.

One or more of the phase shift images may be analyzed to identify cell aggregates therein, for example, to distinguish cell aggregates from individual cells. Cell aggregates can be identified, for example, based on one or more morphologically-related parameters related to their size, shape, and/or structure, such as average diameter (equivalent diameter) and/or phase shift (optical height), for example, by defining one or more thresholds for each parameter. Additionally or alternatively, cell aggregates can also be identified using classical and/or artificial intelligence (AI-based) based computer vision techniques, for example using neural network based classifiers. Identifying the cell aggregates in the one or more phase shift images may particularly comprise determining a total number or fraction of cell aggregates in the one or more phase shift images, wherein the fraction of cell aggregates may be, for example, a ratio of the total number of cell aggregates to the total number of individual cells and cell aggregates.

In a preferred embodiment, identifying the cell aggregates in the one or more phase shift images comprises determining the number of cells in the individual cell aggregates and/or the cell type of some or all of the cells in the individual cell aggregates. This may include, for example, performing image segmentation on a portion of the phase shift image associated with the cell aggregate (e.g., a region of interest that contains only the cell aggregate and no other individual cells or cell aggregates) to identify the composition of the cell aggregate. Image segmentation may be performed, for example, using a thresholding algorithm, such as one or more thresholds based on phase shift, to assign portions of the phase shifted image to individual components, and/or using a watershed algorithm, such as by interpreting the phase shift as a terrain elevation, and identifying "basins" within the resulting terrain as components of cell aggregation. Additionally or alternatively, the image segmentation may also be performed using edge-based methods, such as geodesic active contour methods, see e.g. p.marquez-Neila, l.baumarela and l.alvarez, "morphological methods "(A morphology ap-proach to curve and surface),In:IEEE Transactions on Pattern Analysis and Machine Intelligence 36.1(2013),pp.2–17, based on curvature and surface evolution, such as Chan-Vese algorithm, see e.g. t.chan and l.vese," an edge-free active contour model "(An active contour model without edges),In:Inter-national Conference on Scale-Space Theories in Computer Vi-sion,Springer,1999,pp.141–151, and/or graph-based methods, such as Felzenszwalb algorithm, see e.g. p.f. felzenszwalb and d.p. huttenlocher, "efficient graph-based image segmentation "(Efficient graph-based image segmenta-tion),In:International Journal of computer vision 59.2(2004),pp.167-181. may additionally or alternatively be performed using AI-based computer vision techniques, such as using neural networks such as U-Net, see e.g. O.Ronneberger, P.Fischer and t.brox," U-Net: convolutional network "(U-Net:Convolutional networks for biomedical images segmentation),In:Interna-tional Conference on Medical images computing and computer ssisted intervention,Springer,2015,pp.234–241,Mask R-CNN, of biomedical image segmentation, see e.g. K.He, G.Gkioxari, P.Dolljr, and R.Girshick,"XRCNN",In：Proceedings of the IEEE International conference on com-puter vision,2017,pp.2961–2969, and/or pulse coupled neural network, see e.g. m.chen, x.yu, y. Huttenlocher, "the overall depth of view may be adjusted according to the overall depth of view, and/or alternatively, the overall depth of view may be adjusted according to the overall depth of view, the traffic, the overall depth of the microscope, and the overall depth of the microscope may be adjusted according to the required for the overall depth of the quantitative analysis. A histogram of aggregate size distribution and/or cell composition may be drawn. For example in the case of assays using activated substances or detailed analysis of patient samples.

The method may further comprise determining one or more morphological parameters for some or all components of the cell aggregate, for example from the segmented image. The one or more morphological parameters may include, for example, a minimum diameter, an average diameter, a maximum diameter, a perimeter, an aspect ratio, a minimum phase shift, an average phase shift, a maximum phase shift, a variation or standard deviation of a phase shift, and/or a correlation length of a phase shift. The one or more morphological parameters may include, among other things, one or more texture features, such as entropy or uniformity, energy, one or more features extracted from a co-occurrence matrix, such as a gray co-occurrence matrix (GLCM), and/or one or more Haralick features. In some embodiments, one or more morphological parameters may be extracted using AI-based computer vision techniques, such as extracting features using a neural network.

The method may further comprise determining the cell type of some or all components of the cell aggregate, e.g., based on one or more morphological parameters. For example, one or more thresholds for various morphological parameters may be used to determine the cell type. Additionally or alternatively, regression analysis, linear discriminant analysis, decision tree classification, random forest classification, support Vector Machines (SVMs), quadratic discriminant analysis, K-means clustering, logistic regression, and/or naive bayes classifier may be used to determine cell types. In some embodiments, AI-based computer vision techniques may additionally or alternatively be used, such as using neural network-based classifiers to determine cell types.

In preferred embodiments, the method further may include identifying individual cells in the one or more phase shift images, e.g., similar to the identification of cell aggregates described above. Individual cells and cell aggregates can be distinguished, for example, based on one or more morphologically-related parameters related to their size, shape, and/or structure, and/or can be distinguished using classical and/or AI-based computer vision techniques. The method may further comprise determining a cell type of the individual cells, e.g. a cell type similar to the determination of the composition of the cell aggregates described above.

The shear-thinning polymer may in some embodiments have a molecular weight of 3MDa to 6MDa, preferably 3.5MDa to 4.5MDa, in one example 4.0MDa, where Da is the uniform atomic mass unit (u). Increasing the molecular weight of the polymer may promote viscoelastic focusing of the subject, but may also result in an increase in the rate of polymer-induced "artificial" cell aggregate formation and an increase in mechanical stress on the cell aggregate due to interactions with the polymer in suspension. The inventors have found that the use of shear thinning polymers having molecular weights in these ranges is particularly advantageous for achieving sufficient confinement of both cell aggregates and individual cells in suspension even at low flow rates and not inducing the formation of "artificial" cell aggregates. Preferably, the shear thinning polymer is a linear polymer, for example a polymer comprising a single unbranched linear chain.

In some embodiments, the mass fraction of the shear-thinning polymer in the suspension may be 0.03% to 0.12%, preferably 0.04% to 0.06%, in one example 0.05%. The mass fraction of a given component of the suspension may for example be defined as the ratio of the total mass of the components in the suspension to the total mass of the suspension, i.e. the sum of the total masses of the components of the suspension. Increasing the mass fraction of the shear-thinning polymer may promote viscoelastic focusing of the subject, but may also result in an increase in the rate of polymer-induced "artificial" cell aggregate formation and an increase in mechanical stress on the cell aggregate due to interactions with the polymer in suspension. The inventors have found that the use of a mass fraction of a shear-thinning polymer within these ranges is particularly advantageous for achieving a sufficient confinement of both cell aggregates and individual cells in suspension even at low flow rates and not inducing the formation of cell aggregates.

Preferably, the shear thinning polymer is a water soluble polymer. The shear thinning polymer may be selected from, for example, poly (ethylene oxide) (PEO), poly (vinyl pyrrolidone) (PVP), hyaluronic Acid (HA), and Polyacrylamide (PAA). Preferably, the shear thinning polymer is poly (ethylene oxide) or poly (vinyl pyrrolidone). In one example, the shear-thinning polymer is 4MDa water soluble linear polymer PEO, which has a mass fraction in suspension of 0.05%. In another example, the shear-thinning polymer is 4MDa water soluble linear polymer PEO, which has a mass fraction in suspension of 0.2%. In some embodiments, the viscoelastic fluid may comprise one or more additional shear-thinning polymers, wherein the additional shear-thinning polymers may, for example, comprise the same type of polymer but having a different molecular weight (e.g., PEO with a molecular weight distribution of 3MDa to 6 MDa) and/or different types of polymers of the same or different molecular weights (e.g., PVP with a molecular weight of 2MDa in addition to PEO with a molecular weight of 4 MDa). In such embodiments, the total mass fraction of all the shear-thinning polymers in the suspension may be less than 0.2%, preferably 0.03% to 0.12%, in one example 0.04% to 0.06%, and/or the molecular weight of some or all of the shear-thinning polymers may be from 2MDa to 10MDa, preferably from 3MDa to 6MDa. In one example, 3.5MDa to 4.5MDa.

In a preferred embodiment, the flow rate of the suspension along the microfluidic channel is chosen such that the shear stress within the flow is below 50Pa, preferably below 10Pa, in one example below 5Pa. The shear rate, e.g., the gradient of the flow rate, of the suspension in the flow in the microfluidic channel may be, for example, less than 10,000s ^-1, in some examples less than 5,000s ^-1, in one example less than 2,000s ^-1. The flow rate of the suspension may be, for example, 1mm/s to 1.0m/s, preferably 1mm/s to 250mm/s, in some examples 5mm/s to 100mm/s, in one example 8mm/s to 64mm/s.

In some embodiments, the length from the entrance of the microfluidic channel to the focal point of the quantitative phase-contrast microscope is 30mm to 60mm, preferably 35mm to 50mm. The inlet of the microfluidic channel may for example be an input port for providing a suspension to the microfluidic channel, or a junction between two or more channels incorporated into the microfluidic channel, such as a hydrodynamic focus junction. In some examples, the microfluidic channel may be straight or substantially straight, wherein the inlet of the microfluidic channel may also be, for example, a curved channel portion at the beginning of the microfluidic channel. The microfluidic channel length is chosen within the above-mentioned range, for example, to promote stable (viscoelastic) focusing even of smaller objects, such as individual cells, especially platelets.

Preferably, the suspension has a flow height in a direction perpendicular to the focal plane of the quantitative phase-contrast microscope of 30 μm to 100 μm, in some examples 30 μm to 70 μm, in one example 40 μm to 60 μm, for example 50 μm. The height of the suspension flow may be measured, for example, at the focal point of a quantitative phase-contrast microscope, and may be, for example, the distance between the sidewalls of a microfluidic channel in contact with the suspension flow, the distance between the suspension flow and the interface between one or more additional flows surrounding the suspension flow in the microfluidic channel, or a combination thereof. The flow height of the suspension may for example be determined by the height of the microfluidic channel, which may for example be within the ranges described above. Additionally or alternatively, the height of the suspension flow may also be controlled by hydrodynamic focusing by creating one or more sheath flows along the microfluidic channel, for example as described below. This may allow a flow of suspension having a height within the above-described range to be generated even in microfluidic channels having a larger height. The choice of the flow height of the suspension within the ranges described above may for example be advantageous in ensuring a stable viscoelastic focusing of the cells and/or cell aggregates within the suspension.

In a preferred embodiment, the method further comprises generating two or more sheath flows along the microfluidic channel to hydrodynamically focus the flow of the suspension such that the cell aggregates in the suspension are focused in the focal plane of the quantitative phase contrast microscope. In other words, in addition to viscoelastic focusing of cells and/or cell aggregates within a suspension flow, the flow of the suspension itself may be hydrodynamically focused by creating two or more sheath flows along the microfluidic channel. For example, the sheath flow may be generated such that the sheath flow flows between the suspension flow and respective sidewalls of the microfluidic channel, e.g., such that the suspension flow is sandwiched between a pair of sheath flows flowing along opposite sides of the microfluidic channel. The sheath flow may be configured to restrict the flow of the suspension in one or more directions, e.g., a direction perpendicular to the focal plane of the microscope and/or a direction parallel to the focal plane of the microscope. The flow of the hydrodynamically focused suspension may, for example, allow for reducing the height of the flow of the suspension in the microfluidic channel and/or prevent objects within the suspension flow, such as individual cells and/or cell aggregates, from contacting the sidewalls of the microfluidic channel. For example, hydrodynamic focusing may be used if the height of the microfluidic channel is greater than 100 μm, preferably if the height of the microfluidic channel is greater than 70 μm, in some examples if the height of the microfluidic channel is greater than 60 μm, in one example if the height of the microfluidic channel is greater than 50 μm. Additionally or alternatively, hydrodynamic focusing may also be used to control the position of the flow of the suspension within the microfluidic channel, for example to move or offset the flow of the suspension from a central plane or centerline of the microfluidic channel.

In combination, the viscoelastic focusing of the cells and/or cell aggregates in the suspension flow and the hydrodynamic focusing of the suspension flow can be adjusted such that the cells and/or cell aggregates in the suspension are focused in the focal plane of the quantitative phase-contrast microscope. For example, the flow rates and/or flows of the two or more sheath flows may be selected such that the flow of the suspension is hydrodynamically focused at a central region of the microfluidic channel, and the viscoelastic fluid may cause viscoelastic focusing of cells and/or cell aggregates at the center of the suspension flow, e.g., near a central plane or centerline of the microfluidic channel. In some embodiments, the center plane and/or the center line of the microfluidic channel may be located in the focal plane of the quantitative phase-contrast microscope. In other embodiments, the cells and/or cell aggregates in suspension may be focused in different regions of the microfluidic channel, e.g., near a plane or line that is offset from the center plane or centerline of the microfluidic channel, e.g., by selecting asymmetric flow rates or flows of two or more sheath flows.

In some embodiments, some or all of the two or more sheath flows may include a viscoelastic fluid, particularly a viscoelastic fluid that includes the same shear-thinning polymer as the suspension. The mass fraction of the shear-thinning polymer in each sheath flow may, for example, be equal to or less than the mass fraction of the shear-thinning polymer in the suspension.

In some embodiments, the flow of the suspension is hydrodynamically focused by creating a pair of transverse sheath flows that pinch the flow of the suspension in a first direction and a pair of perpendicular sheath flows that pinch the flow of the suspension in a second direction perpendicular to the first direction, e.g., to restrict the flow of the suspension in the first and second directions. Each lateral sheath flow may, for example, flow between a suspension flow and a respective vertical sidewall of the microfluidic channel. Each vertical sheath flow may, for example, flow between the suspension flow and the bottom and top walls of the microfluidic channel, respectively. In other examples, the flow of the suspension may be hydrodynamically focused using only a pair of transverse sheath flows or using only a pair of perpendicular sheath flows, e.g., to restrict the flow of the suspension in a first direction or in a second direction by hydrodynamically focusing, while the restriction of the cells and/or cell aggregates in the other direction may be achieved by viscoelastic focusing, for example. The second direction may for example be perpendicular to the focal plane of the quantitative phase-contrast microscope and may for example be parallel to the imaging axis of the quantitative phase-contrast microscope.

In a preferred embodiment, the sample is a whole blood sample, such as unmodified blood extracted from a patient, or a blood component sample, such as a sample comprising one or more components of a whole blood sample, such as plasma or a portion thereof, a buffy coat comprising white blood cells and platelets, and/or red blood cells. In other examples, the sample may also be a sample of a different body fluid or a tissue sample extracted from a patient, in particular a tissue sample that is lysed into individual cells and/or cell aggregates. The sample may be or comprise, for example, a human sample, such as urine, exudate, lavage fluid, or sputum, for performing a cell aggregation test thereon. In some examples, one or more coagulation inhibiting substances, such as ethylenediamine tetraacetic acid (EDTA), heparin, or citrate, may be added to the whole blood sample or blood component sample, respectively, to prevent coagulation.

Identifying the cell aggregates in the one or more phase shift images may include identifying cell aggregates comprising one or more predetermined types of blood cells in the one or more phase shift images. Identifying cell aggregates in the one or more phase shift images may particularly include identifying platelet aggregates, i.e. aggregates composed of platelets, leukocyte-platelet aggregates, i.e. aggregates composed of one or more platelets and one or more leukocytes, and/or leukocyte aggregates, i.e. aggregates composed of leukocytes. This may include, for example, identifying components of the cell aggregates in one or more phase shift images, determining one or more morphological parameters of some or all of the components of the respective cell aggregates, and determining cell types of the respective components as described above. Additionally or alternatively, identifying cell aggregates in the one or more phase shift images may also include identifying cell aggregates comprising tumor cells, in particular cell aggregates comprising tumor cells and blood cells, such as cell aggregates consisting of tumor cells with platelets and/or white blood cells.

Additionally or alternatively, the method may further comprise determining a number of cell aggregates comprising at least a predetermined number of cells of one or more specific types of cells, e.g. an aggregate comprising at least a first number of cells of a first type or an aggregate comprising at least a first number of cells of a first type and at least a second number of cells of a second type. In a preferred embodiment, the method comprises determining the number of leukocyte-platelet aggregates comprising at least a predetermined number of leukocytes, in particular two or more leukocytes or three or more leukocytes. Additionally or alternatively, the method may further comprise, for example, determining a number of white blood cell aggregates and white blood cell-platelet aggregates comprising two or more white blood cells and/or white blood cell-platelet aggregates comprising two or more white blood cells and two or more platelets. The method may further comprise determining a number of cell aggregates consisting of a predetermined number of cells of one or more specific types of cells, e.g. an aggregate consisting of a first number of cells of a first type, an aggregate consisting of a first number of cells of a first type and a second number of cells of a second type, and/or an aggregate consisting of a first number of cells of a first type and a second number of cells of at least a second type, e.g. a number of leukocyte-platelet aggregates consisting of two leukocytes and one or more platelets or a number of leukocyte-platelet aggregates consisting of three leukocytes and one or more platelets. Additionally or alternatively, the method may further comprise determining the number of cell aggregates comprising at least a predetermined number of cells, e.g. three or more cells, in one example four or more cells. The method may particularly comprise determining the number of leukocyte aggregates and/or leukocyte-platelet aggregates comprising three or more cells. The presence of cell aggregates of a certain composition, e.g. having at least a given number of cells, may be associated with a certain medical condition or disease. For example, the presence of a leukocyte-platelet aggregate comprising two or more leukocytes, and in particular the presence of a leukocyte-platelet aggregate comprising three or more leukocytes, may be indicative of an infection. High concentrations of platelet-platelet aggregates may, for example, be indicative of complications in Covid-19 patients or patients suffering from cardiovascular disease.

In some embodiments, preparing the suspension comprises diluting the whole blood sample or the blood component sample, respectively, in a ratio of 1:10 to 1:1000, preferably 1:50 to 1:200, in one example 1:80 to 1:120. Diluting whole blood or blood fractions with ratios in these ranges may ensure that the cells and cell aggregates in suspension are sufficiently sparse that individual objects can be easily distinguished in the phase shift image while also providing objects of sufficiently high density to allow analysis of a large number of objects.

In a preferred embodiment, the preparation of the suspension does not involve lysis of erythrocytes, spheroidization of platelets and/or erythrocytes and/or labeling or staining of cells. For example, preparing the suspension may simply comprise adding the viscoelastic fluid to a blood sample, in particular a whole blood sample or a blood component sample, for example by diluting the blood sample in a ratio within the ranges described above. This may allow for a rapid processing of the sample, e.g. preventing spontaneous disintegration of cell aggregates from the sample. Furthermore, the sample preparation procedure described above may affect cell morphology and/or may result in disintegration of cell aggregates from the sample. In some examples, one or more coagulation inhibiting substances such as ethylenediamine tetraacetic acid (EDTA) may be contained in the viscoelastic fluid and/or may be added to the blood sample.

In some embodiments, the method may further comprise adding a platelet and/or leukocyte activating substance to induce platelet aggregation and/or leukocyte-platelet aggregation, e.g., to study the clotting process or leukocyte function, e.g., to study morphological changes and aggregate formation. An increase or decrease in the speed and/or extent of coagulation may be associated with certain pathological conditions, for example. Samples from patients with coronary artery disease may, for example, exhibit stronger clotting than samples from healthy individuals, see m.i. furman et al, j.am. col. Cardiol. Vol.31, no.3,292-296 (2009). The platelet activating substance may be selected, for example, from Adenosine Diphosphate (ADP), thrombin Receptor Activating Peptide (TRAP), epirphin, thrombin, von willebrand factor and C-reactive protein (CRP). Vice versa, inhibitors such as aspirin or clopidogrel may be added. Alternatively or in addition to platelet activation, a leukocyte-activating substance, such as a cytokine, may be added. In addition, drugs, such as checkpoint inhibitors, one or more antibody drug conjugates and/or one or more bispecific T-cell binding antibody constructs, may be added, for example, to study cell aggregation behavior and inhibition. In one example, the method can further comprise adding a substance that induces formation of aggregates comprising tumor cells in combination with white blood cells and/or platelets.

According to a first aspect, the invention further provides an apparatus for detecting cell aggregates of biological cells using a method according to any of the embodiments described herein. The apparatus includes a base configured to receive a microfluidic system including a measurement volume. The apparatus further includes a microscope configured to take phase-shifted images of biological cells in the measurement volume. The device further includes a microfluidic unit configured to receive a sample fluid including biological cells and a viscoelastic fluid from the sample, wherein the viscoelastic fluid includes a shear-thinning polymer having a molecular weight of 2MDa to 10MDa, and wherein the mass fraction of the shear-thinning polymer in the sample fluid is less than 0.2%. The microfluidic unit is configured to generate a sample fluid flow through the measurement volume to viscoelastically focus cell aggregates in the sample fluid flow in a focal plane of the microscope. The apparatus further includes a controller configured to identify cell aggregates in phase shifted images of the sample fluid flow obtained from the microscope.

The mount may be configured to hold the microfluidic system in a fixed reference position, for example, relative to a microscope. Preferably, the mount is configured to position the microfluidic system relative to the microscope, e.g., move the microfluidic system in one or more directions and/or tilt the microfluidic system about one or more axes. In some examples, the base may further include one or more fluid connectors for connecting ports of the microfluidic system, such as one or more input ports and output ports of the microfluidic system. The measurement volume of the microfluidic system may be, for example, a microfluidic channel or a part thereof. In some embodiments, the device may comprise a microfluidic system.

The microscope is a quantitative phase-contrast microscope, such as a digital holographic microscope or a stacked imaging device, configured to take phase-shifted images, i.e., images encoding the phase shift of light of one or more wavelengths as a function of position. Preferably, the microscope is configured to determine an absolute value of the phase shift. In other examples, the microscope may be configured only to determine the phase shift modulo 2pi. For example, the microscope may be configured to obtain a phase shift between the probe or imaging beam and the reference beam by interference of light, for example. In other examples, the microscope may be a stack imaging device configured to perform stack imaging without a reference beam, for example, by recording an interference pattern without a reference phase. The microscope may be configured to take one or more interference images and reconstruct a phase-shifted image from the one or more interference images.

In a preferred embodiment, the microscope is a digital holographic microscope configured to capture phase shift images as well as amplitude or intensity images, wherein the intensity images may encode light intensity as a function of position, such as light intensity reflected from or transmitted through an imaging sample (e.g., sample fluid stream), as a function of position in the imaging sample. For example, a digital holographic microscope may be configured to interfere with an image of an imaging sample (e.g., an imaging beam transmitted through the imaging sample) with a reference beam, which may or may not pass through the imaging sample. The digital holographic microscope may be configured to extract or reconstruct phase shift and intensity images from one or more interference images, for example by reconstructing a wavefront of light transmitted or reflected through the imaged sample. The digital holographic microscope may be an off-axis digital holographic microscope in which the imaging beam and the reference beam propagate along the same axis when interfering. Preferably, the digital holographic microscope is an off-axis digital holographic microscope in which the imaging beam and the reference beam interfere at an angle and may be configured to extract a phase-shifted image from a single interference image of the imaged sample. Such digital holographic microscopes are known, for example, from EP 152491A1 and EP 2357539 A1.

The microfluidic unit may for example comprise a reservoir for receiving a sample fluid and/or may comprise a well for receiving a reservoir containing a sample fluid, such as a test tube or a sample tube. The microfluidic unit may further comprise one or more fluidic connectors for connecting ports of the microfluidic system and/or ports of the base. The microfluidic unit may further comprise one or more pressure sources, e.g. pumps and/or one or more valves, for generating a sample fluid flow and/or further flows, e.g. one or more sheath fluid flows.

The controller may be implemented in hardware, software, or a combination thereof. The controller may, for example, include a processing device and a memory storing instructions for execution by the processing device to provide the functionality described herein. The controller may for example be configured to read out one or more phase shift images from the microscope and identify cell aggregates therein, e.g. as described above for the method according to the first aspect of the invention. The controller may further be configured to control some or all other components of the device, in particular the microfluidic unit and/or the sample preparation unit as described below. Preferably, the controller is configured to perform some or all of the steps of the method for detecting cell aggregates of biological cells according to one of the embodiments described herein.

In a preferred embodiment, the microfluidic system further comprises a hydrodynamic focus connection in fluid communication with the measurement volume. The hydrodynamic focusing connection may be configured to generate two or more sheath flows around the sample fluid flow to hydrodynamically focus the sample fluid flow in the measurement volume. The microfluidic unit may be configured to provide a sheath fluid to the hydrodynamic focus connection to hydrodynamic focus the sample fluid flow in the measurement volume such that cell aggregates in the sample fluid flow are focused in a focal plane of the microscope. At the hydrodynamic focus connection, a sample channel, which may be configured to direct a sample fluid flow to a measurement volume, for example, may intersect two or more sheath flow channels, each of which may be configured to direct a respective one of the sheath flows to the measurement volume, such that the respective sheath flow flows between the sample fluid flow and a respective wall of the measurement volume, for example.

In a preferred embodiment, the device further comprises a sample preparation unit configured to provide a viscoelastic fluid comprising a shear-thinning polymer having a molecular weight of 2MDa to 10MDa to prepare a sample fluid comprising biological cells from the sample and the viscoelastic fluid, wherein the mass fraction of the shear-thinning polymer in the sample fluid is less than 0.2%. The sample preparation unit may for example comprise a reservoir for receiving a sample or a part thereof, such as a whole blood sample or a blood component sample, or a well for receiving a reservoir containing a sample. The sample preparation unit may further comprise a reservoir for receiving a viscoelastic fluid and may be configured to mix the viscoelastic fluid and the sample or a portion thereof, for example by adding the viscoelastic fluid to the sample or vice versa. In some examples, the sample preparation unit and the microfluidic unit may be integrated into a single unit.

Preferably, the sample preparation unit is configured to adjust the mass fraction of the shear-thinning polymer in the sample fluid, for example at least 0.03% to 0.12%, in some examples at least 0% to 0.2%. The sample fluid may, for example, be configured to adjust the amount of viscoelastic fluid added to the sample fluid, the concentration of the shear-thinning polymer in the viscoelastic fluid, and/or the amount of another fluid (e.g., water or an aqueous solution) added to the sample fluid in addition to the viscoelastic fluid.

Additionally or alternatively, the sample preparation unit is configured to dilute the sample fluid in a ratio of 1:10 to 1:1000, preferably 1:50 to 1:200. Diluting the sample fluid with a given ratio may, for example, refer to adding a viscoelastic fluid and/or other fluid to the sample or portion thereof in an amount such that the sample or portion thereof constitutes each portion of the sample fluid in mass or volume. In a preferred embodiment, the sample preparation unit is further configured to adjust the dilution ratio, e.g. in the ranges described above.

In some embodiments, the sample preparation unit is further configured to add one or more platelet-activating substances and/or one or more leukocyte-activating substances to the sample fluid and/or the sheath fluid. The sample preparation unit may for example comprise a respective reservoir for each of the one or more platelet-activating substances and/or for each of the one or more leukocyte-activating substances, and may be configured to add a predetermined amount of one or more of these substances to the sample fluid and/or the sheath fluid.

Preferably, the microfluidic cell is configured to control the flow rate of the sample fluid flow in the measurement volume, wherein the flow rate of the sample fluid flow may be for example 1mm/s to 1.0m/s, preferably 1mm/s to 250mm/s, in some examples 5mm/s to 100mm/s, in one example 8mm/s to 64mm/s. The microfluidic unit may for example be configured to regulate the flow of sample fluid provided to the hydrodynamic focus joint. The microfluidic unit may further be configured to regulate the flow of sheath fluid provided to the hydrodynamic focus connection, preferably such that the flow rate or flow of each of the two or more sheath flows may be individually controlled, e.g. to move or adjust the focus area to which the sample fluid flow is limited by the sheath flow.

In a preferred embodiment, the controller is configured to perform some or all of the steps for analysing phase shifted images of the method according to the first aspect of the invention described above. In particular, the controller may be configured to determine the number of cells in the cell aggregate identified in the phase shift image and/or the cell type of some or all of the cells in the cell aggregate identified in the phase shift image, e.g. as described above. The controller may be configured to identify platelet aggregates and/or leukocyte-platelet aggregates in the phase shift image. Preferably, the controller is further configured to determine the number of leukocyte-platelet aggregates comprising two or more leukocytes and/or the number of cell aggregates comprising three or more cells in the phase-shift image, for example as described above.

Drawings

Hereinafter, a detailed description of the present invention and exemplary embodiments thereof will be given with reference to the accompanying drawings. The figures are schematically shown in the drawing and,

Fig. 1: an apparatus for detecting cell aggregates of biological cells according to an exemplary embodiment of the present invention;

fig. 2: a microscope of the apparatus of fig. 1 according to an exemplary embodiment of the present invention;

Fig. 3: a flow chart of a method for detecting cell aggregates of biological cells using a quantitative phase-contrast microscope according to the present invention;

fig. 4: examples of phase shift images of single cells and cell aggregates obtained using the method according to an exemplary embodiment of the invention;

fig. 5a, fig. 5 b: image segmentation for identifying cell aggregate constituents according to exemplary embodiments of the present invention;

fig. 6: identification of platelet aggregates of different sizes according to exemplary embodiments of the present invention;

Fig. 7a, fig. 7 b: analysis of aggregate formation induced by various concentrations of shear-thinning polymer;

fig. 8: a flowchart of a method of detecting platelet aggregates using a quantitative phase-contrast microscope according to an exemplary embodiment of the present invention;

fig. 9: covid-19 experimental data of viral load versus fraction of platelet aggregates in patient blood samples;

Fig. 10a, 10b: experimental data on the fraction of platelet aggregates in blood samples of Covid-19 infected patients with different severity compared to healthy individuals;

fig. 11a, 11b: experimental determination of the cell number distribution of platelet aggregates in a blood sample of Covid-19 patients compared to healthy individuals;

Fig. 12: experimental determination of the ratio of the median phase shift (average optical height) of all pixel values in segmented cells divided by the diameter of individual cells for patients infected with Covid-19 of different severity compared to healthy individuals;

fig. 13: the experimental determination of the platelet size distribution of individual platelets and aggregated platelets in a blood sample of Covid-19 patients compared to healthy individuals; and

Fig. 14: the fractional time change of platelets and aggregated platelets in a sample from Covid-19 patients was determined using the method according to the invention, which conventional flow cytometry method was unable to resolve platelet aggregates compared to platelet counts determined using conventional flow cytometry methods.

Detailed Description

Fig. 1 shows a schematic view (not to scale) of an apparatus 100 for detecting cell aggregates 102A, 102B (not shown) of biological cells 104A, 104B (not shown) according to an exemplary embodiment of the invention. A schematic illustration (not to scale) of a microscope 108 of the apparatus 100 is depicted in fig. 2. The apparatus 100 may be used to perform a method for detecting cell aggregates according to any of the embodiments described herein, such as the method 300 described below with reference to fig. 3.

The apparatus 100 includes a base 106 configured to receive a microfluidic system 200, wherein the microfluidic system 200 includes a measurement volume 202 and a hydrodynamic focus connection 204. The measurement volume 202 and the hydrodynamic focus joint 204 may for example be arranged in a substrate comprising one or more layers, each of which may for example comprise or consist of glass, plastic (in particular transparent thermoplastic such as polymethyl methacrylate, PMMA), metal or a combination thereof.

The measurement volume 202 may be, for example, a microfluidic channel or a part thereof, and may have, for example, a width of 50 μm to 1000 μm in the view direction of fig. 1 and 2, a height of 30 μm to 500 μm in the Z direction of fig. 1 and 2, and a length of 50 μm to 60mm in the X direction of fig. 1 and 2. In one example, the measurement volume 202 has a rectangular cross-section with a width of 300 μm to 700 μm, e.g., 500 μm, and a height of 30 μm to 100 μm, e.g., 50 μm. The distance between the center of the measurement volume 202 (which may be, for example, in focus with the microscope 108) and the hydrodynamic focus connection 204 may be, for example, 30mm to 60mm, in some examples 35mm to 50mm, for example 40mm. The measurement volume 202 comprises a detection window 202A, which may be, for example, a transparent side wall of the measurement volume 202 or a part thereof, or may be a transparent window arranged in a side wall of the measurement volume 202. The detection window 202A is optimized for phase shift measurement. For example, the transmitted wavefront error of the detection window 202A may be less than λ/2, preferably less than λ/4, and in one example less than λ/8. The detection window 202A may comprise or consist of, for example, a transparent thermoplastic, borosilicate glass, and/or fused quartz. The microfluidic system 200 further comprises an illumination window 202B for illuminating the measurement volume 202, wherein the illumination window 202B may for example be arranged on the opposite side of the measurement volume 202 from the detection window 202A shown in fig. 2 and preferably also be optimized for phase shift measurement.

At hydrodynamic focus connection 204, sample channel 206A intersects with a plurality of sheath flow channels 206B such that sample fluid flow 208A from sample channel 206A into measurement volume 202 may be surrounded by two or more sheath flows 208B, which flow between sample fluid flow 208A and the respective walls of measurement volume 202. In the example of fig. 1, the microfluidic system 200 includes two perpendicular sheath flow channels 206B configured to generate a pair of perpendicular sheath flows 208B that sandwich the sample fluid flow 208A in the Z-direction of fig. 1 and 2 to hydrodynamically focus the sample fluid flow 208A in the Z-direction. The Z-direction may be aligned with the optical axis of the microscope 108, for example, i.e. may correspond to a direction perpendicular to the focal plane of the microscope 108. Additionally, the microfluidic system 200 may further include two horizontal or lateral sheath flow channels (not shown) configured to generate a pair of horizontal or lateral sheath flows sandwiching the sample fluid flow 208A in the middle of the view directions of fig. 1 and 2 so as to hydrodynamically focus the sample fluid flow 208A along the view directions of fig. 1 and 2.

In some embodiments, the microfluidic system 200 may not include the hydrodynamic focus joint 204, such as in the case where the cell aggregates 102A, 102B and individual cells 104A, 104B in the sample fluid stream 208A are focused by viscoelastic focus alone. In such an example, there may be no sheath flow in the measurement volume 204, and the sample fluid flow 208A may extend over the entire height of the measurement volume 202, e.g., from a bottom wall including the illumination window 202B to a top wall including the detection window 202A. In order to provide a sufficient confinement of the cell aggregates 102A, 102B and the individual cells 104A, 104B in the sample fluid flow 208A, a measurement volume 202 with a smaller height may be used in these cases, for example. The height of the measurement volume 202 may be, for example, 30 μm to 70 μm, in some examples 40 μm to 60 μm, for example 50 μm.

The mount 106 is configured to hold the microfluidic system 200 in a fixed reference position relative to the microscope 108. The mount 106 may also be configured to position the microfluidic system 200 relative to the microscope 108, e.g., move the microfluidic system 200 in one or more directions and/or tilt the microfluidic system 200 about one or more axes, e.g., align a center plane or centerline of the measurement volume 202 with a focal plane of the microscope 108.

The microscope 108 of the device 100 is a quantitative phase-contrast microscope, in particular a digital holographic microscope, configured to capture phase-shift images and intensity images of the sample fluid flow 208A in the measurement volume 202 through the detection window 202A. To this end, the microscope 108 includes an imaging system having an objective lens 110, a holographic imaging system 112, and an imaging lens 114, wherein the imaging system is configured to image a focal plane of the microscope 108 onto a camera 116, which may be a CCD or CMOS camera, for example. The microscope 108 further includes an illumination source 118 configured to illuminate the measurement volume 202 through an illumination window 202B. Microscope 108 further includes a microscope controller 108A for controlling holographic imaging system 112, camera 116, and/or illumination source 118.

The objective lens 110 may be, for example, a high NA objective lens having a numerical aperture greater than 0.4, in some examples, a numerical aperture greater than 0.5. The depth of field of the objective lens 110 may be less than 10 μm, preferably less than 5 μm, in one example 2 μm to 3 μm, wherein the depth of field may be defined, for example, as the minimum rayleigh length of the laser beam focused by the objective lens 110, for example, a wavelength of 1064 nm. This may allow for accurate focusing of objects in the measurement volume 202, such as the cell aggregates 102A, 102B, and may provide sufficient spatial resolution to resolve morphological features of individual cells.

Holographic imaging system 112 is configured to create an interference image on camera 116, for example, by interfering an imaging beam with a reference beam on camera 116. The imaging beam may be, for example, a beam that passes through the measurement volume 202 and propagates along a first optical path through the holographic imaging system 112 from the focal plane of the microscope 108 to the camera 116. The reference beam may be, for example, a beam propagating along a second optical path through the holographic imaging system 112 to the camera 116. In some examples, the reference beam may be split from the imaging beam, e.g., using a beam splitter or a diffraction grating, i.e., the reference beam may also pass through the measurement volume 202 and may propagate along a second optical path from the focal plane of the microscope 108 to the camera 116. In other examples, the reference beam may not pass through the measurement volume 202 and may be split, for example, from an imaging beam in front of the measurement volume 202.

The digital holographic microscope 108 may be an on-axis digital holographic microscope in which the imaging beam and the reference beam propagate along the same axis when interfering, i.e., interfering at an angle of 0 °. The microscope controller 108A may be configured to extract or reconstruct phase shifted images as well as intensity images of the sample fluid stream 208A in the measurement volume 202 from the plurality of interference images, for example, by varying the phase offset between the reference beam and the imaging beam using the holographic imaging system 112. Preferably, microscope 108 is an off-axis digital holographic microscope in which the imaging beam and the reference beam interfere at an angle. In this case, the microscope controller 108A may be configured to extract or reconstruct a phase shift image and an intensity image of the sample fluid stream 208A from the single interference image. Alternatively, microscope 108 may be a stacked imaging device and analysis of stacked images may be performed to classify cell aggregates.

The illumination source 118 is configured to illuminate the measurement volume 202 with spatially and/or temporally coherent light, wherein the coherence length of the illumination light may be, for example, greater than the field of view of the microscope 108, and the coherence time of the illumination light may be, for example, greater than the time delay between the imaging and reference beams, i.e., such that an interference pattern may be observed on the camera 116. The illumination source 108 may, for example, comprise a laser or a light emitting diode, and may be configured to emit monochromatic light, for example, at wavelengths of 500nm to 1100nm.

The microscope controller 108A may be implemented in hardware, software, or a combination thereof. Microscope controller 108A may be configured to provide phase shift and intensity images to another device, particularly to controller 124 of device 100, and may be controlled by another device (e.g., controller 124). In some examples, the microscope controller 108A or a portion thereof may be integrated into the controller 124. In addition to reconstructing phase shift and intensity images, the microscope controller 108A may be configured to analyze phase shift and/or intensity images, for example, as detailed below with respect to the method 300.

The device 100 further includes a microfluidic unit 120 and a sample preparation unit 122, which in some embodiments may be integrated into a single unit. The sample preparation unit 122 is configured to receive a liquid sample comprising biological cells, for example, in a test tube. In particular, the liquid sample may be a whole blood sample comprising individual cells, such as platelets 104A, white blood cells (white blood cells) 104B, and red blood cells (not shown). The whole blood sample may further include aggregates of blood cells, such as platelet aggregate 102A composed of a plurality of platelets, leukocyte-platelet aggregate 102B composed of one or more platelets and one or more leukocytes, and/or leukocyte aggregate (not shown) composed of a plurality of leukocytes. In other examples, the liquid sample may also be a blood component sample, such as a sample comprising one or more components of a whole blood sample. The sample preparation unit 122 is configured to prepare a sample fluid comprising biological cells from a sample by adding a viscoelastic fluid to the sample, e.g., as described in method 300 below.

The microfluidic unit 120 is configured to receive a sample fluid from the sample preparation unit 122 and is configured to generate a sample fluid flow 208A through the measurement volume 202 by providing the sample fluid to an inlet of the sample channel 206A. The microfluidic unit 120 is further configured to generate a sheath fluid stream 208B for hydrodynamically focusing the sample fluid stream 208A by providing the sheath fluid to an inlet of the sheath fluid channel 206B. The microfluidic unit 120 may for example comprise separate reservoirs for the sample fluid and the sheath fluid and one or more pumps for providing the sample fluid and the sheath fluid to separate inlets of the microfluidic system 200.

The apparatus 100 comprises a controller 124 configured to control the microscope controller 108A, the microfluidic unit 120 and/or the sample preparation unit 122. The controller 124 is further configured to analyze the phase shift images obtained from the microscope 108, and in particular to identify cell aggregates, such as cell aggregates 102A, 102B therein, such as described below with respect to method 300. Preferably, the controller 124 is configured to at least partially perform the method 300. The controller 124 may be implemented in hardware, software, or a combination thereof. The controller 124 may include, for example, a processing device (not shown) and a memory (not shown) that stores instructions for execution by the processing device to provide the functionality described herein. The controller 124 may include, for example, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), and/or a microcontroller.

Fig. 3 shows a flow chart of a method 300 for detecting cell aggregates of biological cells using a quantitative phase-contrast microscope, according to an exemplary embodiment of the invention. The method 300 may be used, for example, to detect the presence of cell aggregates in a sample (e.g., a whole blood sample), which is used as a non-limiting example for purposes of illustration below. In other examples, the sample may be, for example, a blood component sample. The method 300 may be implemented, for example, with the apparatus 100 and microfluidic system 200 of fig. 1 and 2, which are used below for purposes of example illustration. However, this is not meant to be limiting in any way, and the method 300 may also be implemented using different devices with quantitative phase-contrast microscopy and/or using different microfluidic systems. Furthermore, the method 300 is not limited to the order of execution shown in the flowchart of FIG. 3. Method 300 may be performed in any order, and portions thereof may be performed at least partially concurrently, such as steps 304 through 308, as long as is technically feasible.

In step 302, a suspension is prepared that includes a viscoelastic fluid and biological cells from a sample, such as single cells 104A, 104B and cell aggregates 102A, 102B. The suspension, hereinafter also referred to as sample fluid, may be prepared, for example, using the sample preparation unit 122 of the device 100, for example by adding a viscoelastic fluid to a whole blood sample in a container of the sample preparation unit 122 or vice versa. The viscoelastic fluid comprises a linear water-soluble shear-thinning polymer, such as poly (ethylene oxide) (PEO) or poly (vinyl pyrrolidone) (PVP), wherein the shear-thinning polymer has a molecular weight of from 2MDa to 10MDa, preferably from 3.5MDa to 4.5MDa, such as 4.0MDa. The sample fluid is prepared such that the mass fraction of shear-thinning polymer in the sample fluid is less than 0.2%, preferably 0.04% to 0.06%, for example 0.05%. In the sample fluid, the whole blood sample may be diluted in a ratio of 1:50 to 1:200, e.g. 1:100, for example by adding an appropriate amount of a viscoelastic fluid and/or another fluid such as water or an aqueous solution.

In step 302, the preparation of the suspension preferably does not include any of lysis of erythrocytes, spheroidization of platelets and erythrocytes, and labeling or staining of cells. In some embodiments, a coagulation inhibiting substance such as ethylenediamine tetraacetic acid (EDTA) may be added to the whole blood sample to prevent coagulation, for example, prior to or at the time of preparing the sample fluid. In some examples, a platelet-activating substance, such as thrombin receptor-activating peptide (TRAP), may be added to the whole blood sample or sample fluid.

In step 304, a stream 208A of sample fluid comprising individual cells 104A, 104B and cell aggregates 102A, 102B is generated by the measurement volume 202 of the microfluidic system 200, for example using the microfluidic unit 120. The viscoelastic fluid may exert hydrodynamic forces on the individual cells 104A, 104B and the cell aggregates 102A, 102B in the sample fluid stream 208A due to shear thinning caused by the shear thinning polymer. This may cause movement perpendicular to the flow direction such that the individual cells 104A, 104B and cell aggregates 102A, 102B are viscoelastically focused at a central region of the sample fluid flow 208A, which may be aligned with the focal plane of the microscope 108.

Meanwhile, in step 306, two or more sheath flows 208B may be generated by the measurement volume 202 to further hydrodynamically focus the sample fluid flow 208A in addition to viscoelastic focusing, for example by providing the sheath flows to the inlet of the sheath flow channel 206B of the microfluidic system 200. A pair of vertical sheath flows 208B may clamp the sample fluid flow 208A in the Z-direction of fig. 1 and 2, and a pair of horizontal sheath flows may clamp the sample fluid flow 208A in the view direction of fig. 1 and 2, thereby restricting the sample fluid flow 208A in two orthogonal directions. By adjusting the flow rate of the sheath flow 208B, the position of the sample fluid flow 208A in the measurement volume 202 can be controlled, for example, such that the sample fluid flow 208A flows along a centerline of the measurement volume 202 and the individual cells 104A, 104B and cell aggregates 102A, 102B contained therein are focused in a focal plane of the microscope 108. In some embodiments, a shear-thinning polymer included in the viscoelastic fluid may also be added to the sheath fluid for sheath fluid flow, e.g., such that the sheath fluid also becomes a shear-thinning fluid. In some examples, the method 300 may not include generating the sheath flow 208B in step 306, but rather the individual cells 104A, 104B and cell aggregates 102A, 102B in the sample fluid flow 208A may be focused in the focal plane of the microscope 108, for example, by viscoelastic focusing alone. For example, sheath flow may not be provided to the sheath flow channel 206B or the microfluidic system 200 without the sheath flow channel 206B, and the hydrodynamic focusing junction 204 may be used. The height of the measurement volume 202 along the Z-direction of fig. 2 may be independently selected and may be, for example, 30 μm to 70 μm, in some examples 40 μm to 60 μm, for example 50 μm.

To avoid damaging the cell aggregates 102A, 102B, the flow rates of the sample fluid stream 208A and the sheath fluid stream 208B are selected such that the cell aggregates 102A, 102B are subjected to a shear stress in the sample fluid stream 208A of less than 50Pa, preferably less than 10Pa, for example by independently regulating the flow rates through the sample channel 206A and the sheath fluid channel 206B. For example, the flow rate may be selected such that the flow rate of the sample fluid stream 208A in the measurement volume 202 is 5mm/s to 100mm/s, in one example 8mm/s to 64mm/s.

As the sample fluid stream 208A flows through the measurement volume 202, one or more phase shift images of the individual cells 104A, 104B and cell aggregates 102A, 102B in the sample fluid stream 208A are captured with the microscope 108. For example, the dilution ratio of the sample in the sample fluid may be selected such that each phase shift image contains 5 to 50 individual cells or cell aggregates in order to distinguish the individual cells and cell aggregates from each other. Preferably, a sequence of phase shifted images is employed, for example with a frame rate of 10 frames per second to 200 frames per second. This may allow a large number of individual cells and cell aggregates to be analyzed in a short time, thereby facilitating detection of the cell aggregate type that is rarely present in the sample.

In steps 310 and 312, the phase shift image captured in step 308 is analyzed, for example, using the microscope controller 108A and/or the controller 124 of the device 108. This includes identifying cell aggregates 102A, 102B and individual cells 104A, 104B in respective phase images in step 310. Cell aggregates and individual cells may be distinguished, for example, based on a combination of morphological parameters (e.g., average diameter and maximum phase shift), for example, by defining individual regions in a parameter space spanned by the morphological parameters, e.g., using one or more thresholds for the individual parameters. Additionally or alternatively, cell aggregates may also be identified using computer vision techniques such as neural network-based classifiers. Step 310 may also include determining the total number of individual cells, the total number of cell aggregates, and/or the fraction of cell aggregates, i.e., the ratio of the total number of cell aggregates to the total number of individual cells and cell aggregates.

In step 312, the individual cells 104A, 104B and cell aggregates 102A, 102B identified in step 310 may be further analyzed. This may include, inter alia, determining the number of cells in the cell aggregates 102A, 102B, determining the cell type of the individual cells 104A, 104B and the cell type of the cells in the cell aggregates 102A, 102B. To determine the number of cells in the cell aggregates 102A, 102B, the composition of the cell aggregates 102A, 102B may be identified by performing image segmentation on a portion of the phase shift image associated with the cell aggregates 102A, 102B, for example using a watershed algorithm as described in detail below with reference to fig. 5. One or more morphological parameters may then be determined for the composition of the cell aggregates 102A, 102B and the individual cells 104A, 104B to determine the cell type of the composition of the cell aggregates 102A, 102B and the composition of the individual cells 104A, 104B. For example, cell types may be determined by defining individual regions in a parameter space spanned by morphological parameters. In other examples, the cell type may be determined, for example, using regression analysis, linear discriminant analysis, decision tree classification, random forest classification, and/or neural network-based classifiers.

Step 312 is particularly useful for identifying platelet aggregates, leukocyte-platelet aggregates, and/or leukocyte aggregates in the phase-shift image, for example, for determining the total number or fraction of individual aggregates in the phase-shift image. Step 312 may also include determining the number or fraction of leukocyte-platelet aggregates comprising two or more leukocytes and/or the number or fraction of leukocyte-platelet aggregates and/or platelet aggregates comprising three or more cells, e.g., as an indicator of bacterial infection.

Fig. 4 shows four examples of phase shift images of individual cells and cell aggregates obtained using a method (e.g., method 300) according to an exemplary embodiment of the invention. Images are obtained from diluted and stabilized whole blood samples of patients in surgical intensive care. As a pre-analytical step, blood samples were diluted in viscoelastic polymer solution in a ratio of 1:100. The polymer solution consisted of 99.95% Phosphate Buffered Saline (PBS) and 0.05% PEO (4 MDa). During the measurement, a total flow of 1.6. Mu.l/s and a sample flow of 0.2. Mu.l/s were used. The left image contains a single platelet, the left image contains a platelet aggregate composed of three platelets, the right image contains a leukocyte-platelet aggregate composed of three leukocytes and a plurality of platelets, and the right image contains a leukocyte-platelet aggregate composed of a single leukocyte and a plurality of platelets.

Fig. 5a, 5b show examples of image segmentation for identifying components of cell aggregates according to an exemplary embodiment of the present invention, wherein fig. 5a shows examples of a leukocyte-platelet aggregate consisting of two leukocytes and a single platelet, and fig. 5b shows examples of a leukocyte-platelet aggregate consisting of two leukocytes and two platelets. The left graph represents the respective phase shift images, and the right graph represents the result after segmentation of the phase shift images. Segmentation is performed via watershed segmentation using an inverse distance transform. The inverse form of the standard transformation results in high intensity at the cell boundary and low intensity in the middle of the cell. In this case, the local minima in the inverse distance map ideally correspond to the centroid of the cell and are robust to high gradients inside the cell.

FIG. 6 illustrates the identification of individual cells and platelet aggregates of different sizes according to an exemplary embodiment of the present invention. For this purpose, two morphological parameters of the individual cells and of the cell aggregates are extracted from the phase-shift image, namely the average diameter (equivalent diameter) and the maximum phase shift (optical height maximum) plotted on the X and Y axes of the upper graph, respectively. In this parameter space, a plurality of regions are defined as indicated by solid black lines in the above figures, each region corresponding to an aggregate of a specific size, i.e., a single platelet (leftmost region), "small aggregate" (second region from left), an "aggregate 1" (third region from left), an "aggregate 2" (fourth region from left), an "aggregate 3" (third region from right), an "aggregate 4" (second region from right), and an "aggregate 5" (rightmost region). Fig. 6 provides an example of a simple and easily implemented method of analyzing phase contrast images of individual cells and cell aggregates. This analysis relies on only two morphological parameters, which can be easily obtained from the phase difference image and are robust to variations in image quality. Even though the information conveyed by these two morphological parameters is insufficient to allow the determination of the number of platelets in the aggregate, it still provides a fast and efficient method of classifying platelet aggregates by size, such as determining a histogram of the aggregate size distribution. For example, additional information, such as the number of platelets per set, may be obtained by determining additional morphological parameters and/or using a neural network-based classifier (e.g., mask R-CNN). See K.He, G.Gkioxari, P.Doll' r, and R.Girshick,"Mask RCNN",In:Proceedings of the IEEE International confer-ence on computer vision,2017,pp.2961–2969.

Fig. 7a, 7b show experimental results of studying the effect of the shear-thinning polymer on the formation of platelet aggregates, wherein fig. 7b is an enlarged view of the bottom a of fig. 7. To this end, a suspension comprising platelet concentrate and a viscoelastic fluid containing poly (ethylene oxide) (PEO) having a molecular weight of 4MDa and Phosphate Buffered Saline (PBS) was prepared, and the fraction of platelet aggregates of PEO at various concentrations in the suspension was determined as a function of time. For reference, the same experiment was performed with a suspension comprising only whole blood samples and Phosphate Buffered Saline (PBS), i.e. without PEO addition. At PEO mass fractions of 0.05% and 0.1%, no significant additional platelet aggregate formation was observed on the PBS reference. At a PEO mass fraction of 0.15%, a slight increase in the fraction of platelet aggregates was observed, whereas a PEO mass fraction of 0.2% had resulted in a significant increase in the fraction of platelet aggregates. This underscores the importance of selecting an appropriate mass fraction of the shear-thinning polymer in suspension to avoid affecting the measurement of polymer-induced cell aggregate formation.

Furthermore, it is an object of the present invention to provide a means for reliable clinical indicators for determining inflammation, thrombosis and infections, in particular bacterial and viral infections.

This object is achieved by a method for detecting cell aggregates of biological cells using a quantitative phase-contrast microscope according to claim 1 and by an apparatus for detecting cell aggregates of biological cells according to claim 18. Examples of which are described in detail in the dependent claims.

According to a second aspect, the present invention provides a method for detecting cell aggregates of biological cells using quantitative phase-contrast microscopy, wherein the method comprises (1) preparing a suspension comprising biological cells from a sample; (2) Generating a flow of the suspension along the microfluidic channel to viscoelastically and/or hydrodynamically focus the cell aggregates in the suspension in a focal plane of a quantitative phase-contrast microscope; (3) Capturing one or more phase-shifted images of biological cells in suspension using a quantitative phase-contrast microscope; and (4) identifying the cell aggregates in the one or more phase shift images. The sample is a whole blood sample or a blood component sample and identifying cell aggregates in the one or more phase-shifted images includes identifying platelet aggregates in the one or more phase-shifted images. The above numbering is for clarity only and does not imply a certain order of execution of the method. The methods may be performed in any order, and parts thereof may be performed at least partially concurrently, as long as they are technically feasible.

Preparing the suspension may for example comprise providing a sample, which may for example be withdrawn from the patient before performing the method, and diluting the sample with another fluid in a predetermined ratio, for example as detailed above for the method and apparatus according to the first aspect of the invention. In a preferred example, the suspension further comprises a viscoelastic fluid as detailed below.

The cell aggregates, in particular platelet aggregates, in the suspension may be focused by one or both of viscoelastic focusing and hydrodynamic focusing, wherein focusing may be achieved e.g. as detailed above for the method and the device according to the first aspect of the invention. In some examples, hydrodynamic focusing of the cell aggregates themselves, such as by generating one or more sheath flows, may provide sufficiently tight focusing of the cell aggregates, particularly the platelet aggregates, to obtain one or more phase shift images to identify the platelet aggregates therein, and viscoelastic focusing may not be required.

Capturing one or more phase shift images and identifying cell aggregates therein may be performed as detailed above for the method and apparatus according to the first aspect of the invention. Platelets and platelet aggregates can be distinguished from other cells and/or cell aggregates, for example, based on one or more morphologically-related parameters related to their size, shape, and/or structure, such as average diameter (equivalent diameter) and/or phase shift (optical height).

The method according to the second aspect of the invention may further comprise any combination of features described above for the method according to the first aspect of the invention, such as the method 300 of fig. 3, or vice versa. The method according to the second aspect of the invention may in particular comprise some or all of the additional method steps described above for the method according to the first aspect of the invention, such as the method 300 of fig. 3, or vice versa. The method according to the second aspect may be performed, for example, with an apparatus for detecting cell aggregates of biological cells according to any of the embodiments described herein, such as the apparatus 100 of fig. 1 and 2.

The inventors of the present invention have found that quantitative phase-contrast microscopy constitutes a reliable, versatile and accurate tool for quantitative analysis of platelet aggregates in blood samples. The inventors have also found that platelet aggregates, in particular their number, fraction and/or size, are useful biomarkers for inflammation and thrombosis as well as for a variety of infections, including viral infections such as Covid-19. Viscoelastic and/or hydrodynamic focusing of suspensions comprising cells from a blood sample allows focusing of cell aggregates of different sizes and in some instances also single cells in the focal plane of the microscope, which greatly facilitates identification of platelet aggregates and extraction of morphological parameters from phase shifted images.

In a preferred embodiment, the method according to the second aspect of the invention comprises determining the total number or fraction of platelet aggregates in the one or more phase shifted images, e.g. as detailed above. The fraction of platelet aggregates may be, for example, the ratio of the number of platelet aggregates to the total number of individual platelets and platelet aggregates. In another example, the fraction of platelet aggregates may be a ratio of the number of platelets in the platelet aggregates (i.e., all platelets contained in the platelet aggregates) to the total number of individual platelets and platelets in the platelet aggregates.

More generally, the fraction of platelet aggregates can be the ratio of molecules that characterize the number of platelet aggregates and/or the number of platelets in the platelet aggregates to the denominator that characterizes the total number of platelets and/or platelet-containing objects (i.e., individual platelets and platelet-containing aggregates). The molecule may for example be selected from the group consisting of the number of platelet aggregates and the number of platelets in the platelet aggregates. The denominator may be selected from the group consisting of a total number of individual platelets and platelet aggregates, a total number of platelets in individual platelets and platelet aggregates, a total number of all platelets (i.e., a total number of individual platelets and platelets in any type of aggregate), a total number of platelet-containing objects (i.e., individual platelets and any type of platelet-containing aggregates, such as a total number of platelet aggregates and leukocyte-platelet aggregates), a total number of all cells (i.e., any type of cells), and a total number of all objects (i.e., any type of individual cells and cell aggregates). The method may include determining some or all of the above scores.

Additionally or alternatively, the total number or fraction of platelet aggregates may also be determined by determining an amount related to or proportional to the total number or fraction of platelet aggregates. For example, the total number or fraction of platelet aggregates may be determined by determining the total number and fraction of aggregated platelets contained in any type of cell aggregates (platelets in the aggregates), respectively, e.g., as described in detail below. In other words, the molecule used to determine the fraction of platelet aggregates may also be the number of platelets in the aggregate.

The total number or fraction of platelet aggregates may be a valuable indicator of inflammation, thrombosis and infection, especially viral infection, and may be related, for example, to viral load and/or severity of infection. Thus, the method may further comprise using the total number or fraction of platelet aggregates as an indicator of complications in an infected patient, in particular Covid-19 patients, for example as an indicator of viral load and/or severity of infection, in particular Covid-19 severity of infection. The total number or fraction of platelet aggregates may be used, for example, as a prognostic indicator for determining the prognosis (e.g., the expected severity) of a patient, as a diagnostic indicator for determining whether a patient has an infection or a particular type of infection (e.g., for distinguishing between viral and bacterial infections and/or for distinguishing between different types of viral infections), and/or as a therapeutic indicator, e.g., for assessing whether a therapy is effective.

Activated platelets may be associated with increased serum troponin concentrations, for example. Troponin is a marker of cardiac tissue damage because it occurs, for example, in Covid-19 related myocarditis. The inventors also observed that there was a significant in-subject correlation between the total number or fraction of platelet aggregates and D-dimer concentration. An increase or decrease in the number of platelet aggregates may be associated with a corresponding increase or decrease in the D-dimer concentration. D-dimer is a laboratory marker for fibrinolysis when it occurs in thromboembolic events and inflammatory disorders. The inventors further observed that there was a significant in-subject correlation between platelet aggregates and serum Procalcitonin (PCT) concentrations. An increase or decrease in the number of platelet aggregates may be associated with a corresponding increase or decrease in PCT serum concentration. PCT is a laboratory marker of inflammation in infection.

Identifying the cell aggregates in the one or more phase shift images may also include determining the number of cells in the individual cell aggregates, e.g., as detailed above. This may be performed on some or all types of cell aggregates (e.g., leukocyte-platelet aggregates and/or leukocyte-leukocyte aggregates). In particular, identifying the platelet aggregates in the one or more phase shift images may include determining the number of cells in the individual platelet aggregates, i.e., determining how many platelets the identified platelet aggregates contain or consist of. The method may further comprise determining a size distribution of a plurality of cell aggregates, e.g. individual size distributions of one or more types of cell aggregates, in particular platelet aggregates, from the one or more phase shift images. The size distribution may be, for example, a cell number distribution, wherein determining the cell number distribution may, for example, comprise determining a number or fraction of cell aggregates as a function of the number of cells in the aggregates. In addition to or instead of the size distribution, one or more relevant parameters related to the size distribution may be determined, for example, an average value of the cell number distribution, a median value of the cell number distribution, a variation or standard deviation of the cell number distribution, and/or a position of one or more peaks in the cell number distribution.

The method may further comprise determining the number or fraction of platelet aggregates comprising or consisting of at least a predetermined number of cells, in particular determining the number or fraction of platelet aggregates comprising three or more cells. An increase in the number or fraction of larger platelet aggregates (e.g., comprising three or more cells) may be associated with an infection and/or a higher severity of the infection.

In some embodiments, identifying cell aggregates in the one or more phase-shifted images may further include identifying other types of cell aggregates in the one or more phase-shifted images, particularly white blood cell-platelet aggregates and/or white blood cell-white blood cell aggregates, e.g., as described above. White blood cells, white blood cell-platelets, and white blood cell aggregates may be distinguished from other cells and/or cell aggregates, for example, based on one or more morphologically-related parameters related to their size, shape, and/or structure, such as average diameter (equivalent diameter) and/or phase shift (optical height). Leukocyte-platelet aggregates, in particular their number, fraction and/or size, can also be useful biomarkers for inflammation and/or for a variety of infections, including bacterial infections and viral infections such as Covid-19.

The method may further comprise determining a total number or fraction of leukocyte-platelet aggregates in the one or more phase-shifted images. The fraction of leukocyte-platelet aggregates may be defined, for example, as similar to the fraction of platelet aggregates described above, wherein one or both of platelets and leukocytes may be used as references for defining the fraction, e.g., for defining molecules (e.g., the number of platelets in the leukocyte-platelet aggregates) and/or denominators (e.g., the total number of platelets and/or the total number of leukocytes).

Additionally or alternatively, the method may comprise determining the number or fraction of leukocyte-platelet aggregates comprising or consisting of at least a predetermined number of cells, in particular determining the number of leukocyte-platelet aggregates comprising three or more cells (i.e. independent of cell type). The method may further comprise determining a size distribution of the leukocyte-platelet aggregates and/or one or more relevant parameters related to the size distribution, for example as described above.

In some embodiments, the method may include determining a number or fraction of leukocyte-platelet aggregates including at least a predetermined number of leukocytes. The method may particularly comprise determining the number or fraction of leukocyte-platelet aggregates comprising two or more leukocytes and/or determining the number or fraction of leukocyte-platelet aggregates comprising three or more leukocytes. Additionally or alternatively, the method may comprise determining a white blood cell number distribution of white blood cell-platelet aggregates, for example by determining the number or fraction of white blood cell-platelet aggregates as a function of the white blood cell number in the aggregates, and/or determining one or more relevant parameters related to the white blood cell number distribution, such as an average, median and/or change in white blood cell number. The presence of larger leukocyte-platelet aggregates, in particular the presence of leukocyte-platelet aggregates comprising two or more leukocytes and/or the presence of leukocyte-platelet aggregates comprising three or more leukocytes, can be used as an indicator of inflammation and/or infection, in particular of a viral infection such as Covid-19 and/or a bacterial infection.

The method may further comprise determining a leukocyte type (e.g., neutrophil, eosinophil, basophil, lymphocyte or monocyte) of one or more leukocytes in the one or more phase-shifted images, particularly for some or all of the identified cell aggregates, e.g., for some or all of the leukocyte-platelet aggregates and/or the leukocyte-leukocyte aggregates. In one example, the method includes determining the number and type of leukocytes in a leukocyte-platelet aggregate comprising at least a predetermined number of cells and/or a leukocyte-platelet aggregate comprising at least a predetermined number of leukocytes (e.g., two or more leukocytes).

In some embodiments, the method includes determining the total number or fraction of aggregated platelets contained in any type of cell aggregates (e.g., platelets in platelet aggregates, as well as other types of cell aggregates such as platelets in leukocyte-platelet aggregates) in one or more phase shift images. The number or fraction of aggregated platelets may be related to the number or fraction of platelet aggregates and may also be used as an indicator of inflammation and/or infection.

The method may further include determining a granularity metric from the one or more phase-shifted images, wherein the granularity metric characterizes (e.g., quantifies) the granularity of one or more cells in the one or more phase-shifted images. The granularity measure may be, for example, a quantity or parameter related to the granularity of the individual cells. The one or more cells may include a single cell and/or a cell in an aggregate of cells (i.e., a single cell in one or more aggregates of cells rather than an entire aggregate of cells). The particle size of the cells may, for example, be indicative of the internal structure of the cells, and in particular the degree of irregularity of the internal structure (e.g. the presence of voluminous and/or voluminous young platelets in the aggregate). In conventional flow cytometry, side scattering of an incident laser beam may be used, for example, to detect the granularity of cells. In the method according to the invention, the granularity measure may for example be determined corresponding to or based on spatial variations (e.g. standard deviation and/or peak to peak variations) of the phase shift (optical height) within the cell, spatial correlation (e.g. correlation length) of the phase shift, aspect ratio of the cell and/or ratio of phase shift (e.g. average or median of optical height) to size of the cell (e.g. diameter or equivalent diameter in the phase shift image). In one example, the method includes determining a first particle size metric value for one or more (individual) cells in an aggregate of cells and a second particle size metric value for one or more single cells.

The method may further include determining a size distribution of the plurality of cells and/or one or more relevant parameters (e.g., average, median, variance, and/or location of peaks or maxima) related to the size distribution from the one or more phase shift images. The plurality of cells may include individual cells and/or cells in an aggregate of cells (i.e., individual cells in one or more aggregates of cells rather than an entire aggregate of cells). The size distribution may be, for example, a size distribution of individual cells (individual cells and/or cells in an aggregate), wherein the size may be, for example, a physical size of the individual cells, such as a diameter or longitudinal extension, a cross-sectional area of the individual cells (e.g., equivalent diameter), an average, median or maximum phase shift of the individual cells, or a combination thereof. In some embodiments, the method comprises determining a distribution of diameters of a plurality of cells and a distribution of average or median phase shifts of the plurality of cells. In one example, the method includes determining a first size distribution (e.g., distribution of diameters and/or distribution of average or median phase shifts) of a plurality of (individual) cells in an aggregate of cells and a second size distribution (e.g., distribution of diameters and/or distribution of average or median phase shifts) of the plurality of individual cells. The method may further comprise analyzing the composition of the aggregate and the size distribution and/or phase shift of individual platelets in the aggregate, for example to obtain a size distribution of cells in different types of cell aggregates (e.g. platelets in platelet-platelet aggregates and platelets in leukocyte-platelet aggregates or platelets in platelet-platelet aggregates comprising different numbers of platelets).

The granularity metric, the size distribution and/or one or more related parameters related to the size distribution may be determined, for example, for a particular type of individual and/or aggregated cells. In some embodiments, the determination may be made separately for a plurality of cell types and/or a plurality of cell aggregate types (i.e., separately for each type), such as for one or more of individual platelets, platelets in platelet aggregates, platelets in leukocyte-platelet aggregates, aggregated platelets in any type of cell aggregates, individual leukocytes, leukocytes in leukocyte-platelet aggregates, and combinations thereof.

In a preferred embodiment, the suspension further comprises a viscoelastic fluid. In particular, the viscoelastic fluid may comprise a shear-thinning polymer having a molecular weight of 2MDa to 10MDa, in one example 3.5MDa to 4.5MDa. The mass fraction of the shear-thinning polymer in the suspension may be less than 0.2%, in some examples from 0.03% to 0.12%, preferably from 0.04% to 0.06%. This may allow for reliable viscoelastic focusing of the cell aggregates in the focal plane of the microscope.

The viscoelastic and/or hydrodynamic focusing may be adjusted such that single cells in suspension, in particular single platelets, are also focused in the focal plane of the quantitative phase contrast microscope, for example by a suitable choice of the composition of the viscoelastic fluid, the flow rate of the suspension, the physical size of the microfluidic channels, the arrangement and/or number of microfluidic channels relative to the focal point of the microscope, the arrangement and flow rate of sheath flows for hydrodynamic focusing as detailed above.

Preferably, the preparation of the suspension does not include lysis of erythrocytes, spheroidization of platelets and/or erythrocytes, and/or labeling or staining of cells, e.g., as detailed above. For example, preparing a suspension may simply involve diluting a blood sample drawn from a patient and optionally adding a viscoelastic fluid or substance and/or a coagulation inhibiting substance thereto.

In some embodiments of the method according to the first aspect of the invention and in some embodiments of the method according to the second aspect of the invention, identifying the cell aggregates in the one or more phase shifted images may further comprise identifying cell aggregates comprising one or more bacteria, in particular cell aggregates comprising one or more bacteria and blood cells, such as cell aggregates consisting of one or more bacteria and one or more platelets and/or one or more white blood cells. In one example, identifying the cell aggregates in the one or more phase shift images may further include identifying the cell aggregates comprising one or more bacteria and/or one or more tumor cells.

In some embodiments of the method according to the first aspect of the invention and in some embodiments of the method according to the second aspect of the invention, the method may comprise taking phase shift images of biological cells in suspension at different flow rates or flows of the suspension flow (e.g. at two or more flow rates/amounts) and/or different shear rates of the suspension flow (e.g. at two or more shear rates), i.e. one or more phase shift images may be taken at different flow rates and/or different shear rates. Capturing the one or more phase-shifted images may, for example, include capturing a first phase-shifted image at a first flow rate or amount and/or at a first shear rate and capturing a second phase-shifted image at a second flow rate or amount and/or at a second shear rate different from the first flow rate/amount and the first shear rate, respectively.

According to a second aspect, the present invention also provides a device for detecting cell aggregates of biological cells using a method according to any of the embodiments of the second aspect of the invention described herein. The apparatus includes a base configured to receive a microfluidic system including a measurement volume. The apparatus further comprises a microscope configured to take phase-shifted images of biological cells in the measurement volume. The device further comprises a microfluidic unit configured to receive a sample fluid comprising biological cells from the blood sample, wherein the microfluidic unit is configured to generate a sample fluid flow through the measurement volume to viscoelastically and/or hydrodynamically focus cell aggregates in the sample fluid flow in a focal plane of the microscope. The apparatus further includes a controller configured to identify platelet aggregates in phase-shifted images of the sample fluid flow obtained from the microscope.

Some or all of the components of the apparatus according to the second aspect of the invention may be formed or implemented as described above for the apparatus according to the first aspect of the invention, and vice versa. The apparatus according to the second aspect of the present invention may comprise some or all of the features and/or components described above for the apparatus according to the first aspect of the present invention and vice versa. In some embodiments, the apparatus according to the second aspect of the invention may be implemented as described above for the apparatus 100 of fig. 1 and 2.

The microfluidic unit is configured to receive a sample fluid, such as a suspension as described above, comprising biological cells from a blood sample, such as a whole blood sample or a blood component sample. In some embodiments, the sample fluid may further comprise a viscoelastic fluid as detailed below. The microfluidic unit may be configured to focus the cell aggregates in the sample fluid by one or both of viscoelastic focusing and hydrodynamic focusing. The microfluidic system may for example comprise a hydrodynamic focus connection in fluid communication with the measurement volume, and the microfluidic unit may be configured to provide a sheath fluid to the hydrodynamic focus connection, e.g. as detailed above.

The microfluidic unit may be configured to receive and/or generate a sample fluid flow having a varying flow rate or flow rate, e.g., increasing shear rate or shear stress to detect the strength of interaction of cell aggregates and their compositions. Differences in aggregate concentration or composition at different shear stress levels may be indicative of inflammatory, thrombotic, and infectious states. Such assay conditions may be combined with an activator or inhibitor for cell aggregate formation to detect the presence or absence of molecular interaction mechanisms between cells and their respective intensities. Generating a sample fluid flow with varying flow rates or flows may, for example, allow phase shift images to be taken at different flow rates or flows as described above.

The controller is configured to identify a particular type of cell aggregate, i.e., platelet aggregate. In addition, the controller may be configured to identify one or more other types of cell aggregates, such as white blood cell-platelet aggregates, and/or one or more types of single cells, such as single platelets and/or single white blood cells. The controller may be configured to identify individual cell aggregates and/or individual cells, as detailed above. Preferably, the controller is configured to perform some or all of the steps of the method according to any of the embodiments of the first and/or second aspects of the invention described herein.

The controller may be particularly configured to determine a total number or fraction of platelet aggregates in the phase shift image and/or to determine a number or fraction of platelet aggregates comprising at least a predetermined number of cells (e.g. a number or fraction of platelet aggregates comprising three or more cells), e.g. as described above.

In some embodiments, the controller may be configured to identify white blood cell-platelet aggregates in the phase-shifted image, e.g., as described above. Preferably, the controller is configured to determine a total number or fraction of white blood cell-platelet aggregates in the phase shift image, the number or fraction of white blood cell-platelet aggregates in the phase shift image comprising at least a predetermined number of cells (e.g. three or more cells) and/or the number or fraction of white blood cell-platelet aggregates in the phase shift image comprising at least a predetermined number of white blood cells (e.g. two or more white blood cells and/or three or more white blood cells). The controller may further be configured to determine a white blood cell type of the one or more white blood cells in the phase shift image, e.g. as described above.

The controller may be configured to determine a size distribution of the plurality of cells, one or more relevant parameters related to the size distribution and/or a granularity metric from the phase shift image, wherein the granularity metric characterizes granularity of one or more cells in the phase shift image, e.g. as described above.

In a preferred embodiment, the sample fluid further comprises a viscoelastic fluid. The viscoelastic fluid may in particular comprise a shear-thinning polymer having a molecular weight of 2MDa to 10MDa, in one example 3.5MDa to 4.5MDa. The mass fraction of the shear-thinning polymer in the sample fluid may be less than 0.2%, in some examples from 0.03% to 0.12%, preferably from 0.04% to 0.06%. In some embodiments, the device may include a sample preparation unit configured to provide a viscoelastic fluid comprising a shear-thinning polymer having a molecular weight of 2MDa to 10MDa, in one example 3.5MDa to 4.5MDa, to prepare a sample fluid comprising biological cells from the sample and the viscoelastic fluid, wherein the mass fraction of the shear-thinning polymer in the sample fluid is less than 0.2%. The sample preparation unit may be configured to adjust the mass fraction of the shear-thinning polymer in the sample fluid and/or to dilute the sample fluid, for example as detailed above.

Fig. 8 shows a flowchart of a method 800 for detecting cell aggregates of biological cells using a quantitative phase-contrast microscope, according to an exemplary embodiment of the invention. The method 800 may be used, for example, to detect and analyze platelet aggregates and leukocyte-platelet aggregates in a patient's blood sample as an indicator of infection (e.g., viral infection, such as Covid-19), which is used below as a non-limiting example for illustration purposes. The method 800 may be implemented, for example, with the apparatus 100 and microfluidic system 200 of fig. 1 and 2, which are used below for illustrative purposes. However, this is not meant to be limiting in any way, and the method 800 may also be implemented using different devices with quantitative phase-contrast microscopy and/or using different microfluidic systems. In some embodiments, the controller 124 of the apparatus 100 may be configured to perform some or all of the steps of the method 800. The method 800 is not limited to the order of execution shown in the flowchart of fig. 8. Method 800 may be performed in any order, and portions thereof may be performed at least partially concurrently, such as step 810 and step 812, as long as is technically feasible.

In step 802, method 800 includes preparing a suspension including biological cells from a whole blood sample or from a blood component sample, for example, using sample preparation unit 122. For example, a sample may be drawn from a patient having a suspected or confirmed infection and may, for example, contain arterial and/or venous blood prior to performing method 800. In some examples, one or more coagulation inhibiting substances, such as ethylenediamine tetraacetic acid (EDTA), heparin, or citrate, may be added to the sample or suspension to prevent coagulation. To prepare the suspension, the sample may be diluted, for example in a ratio of 1:50 to 1:200, in one example in a ratio of 1:100. In some embodiments, a viscoelastic fluid or viscoelastic substance, such as a shear-thinning polymer, may be added to prepare a viscoelastic suspension, for example as detailed above for step 302 of method 300.

In step 804, a flow 208A of suspension is generated along the measurement volume 202 of the microfluidic system 200, for example using the microfluidic unit 120. The generated flow 208A causes the cell aggregates 102A, 102B, in particular the platelet aggregates 102A, in suspension to be viscoelastically and/or hydrodynamically focused in the focal plane of the microscope 108, for example as described above for steps 304 and 306 of the method 300. In some embodiments, the suspension may not be viscoelastic and may employ hydrodynamic focusing alone. Preferably, a combination of viscoelastic and hydrodynamic focusing is employed.

The method 800 further includes, in step 808, capturing one or more phase shift images of biological cells (e.g., individual platelets 104A, platelet aggregates 102A, individual white blood cells 104B, and/or white blood cell-platelet aggregates 102B) in suspension using the microscope 108, e.g., step 308 of the method 300 as described above.

In step 810, platelet aggregates 102A in one or more phase shift images are identified and analyzed, for example, as a diagnostic, prognostic, and/or therapeutic indicator of infection, particularly as an indicator of an already occurring and/or impending patient complication. The identification and analysis of the platelet aggregates 102A may be similar to steps 310 and 312 of the method 300 described above.

Step 810 may include, inter alia, determining a total number or fraction of platelet aggregates in the one or more phase shift images, such as a percentage of platelets contained in the platelet aggregates and/or a ratio of the number of platelet aggregates to the total number of individual platelets and platelet aggregates. Additionally or alternatively, step 810 may also include determining the fraction of aggregated platelets contained in any type of cell aggregate (platelets in the aggregate). As detailed below with reference to fig. 9,10 a and 10b, the portion of the platelet aggregate and the portion of the platelets in the aggregate (aggregated platelets) may be indicators of complications in infected patients, particularly Covid-19 patients. Additionally or alternatively, step 810 may also include determining a cell number distribution of the platelet aggregates, e.g., as described below with reference to fig. 11a, 11b, and/or determining a number or fraction of the platelet aggregates comprising at least a predetermined number of cells (e.g., a weight or range of one wing of the cell number distribution).

Step 810 may also include determining a granularity measure of individual platelets, platelets in platelet aggregates, and/or aggregated platelets in any type of cell aggregates from the one or more phase shift images, for example as described in detail below with reference to fig. 12. Additionally or alternatively, step 810 may also include determining a size distribution of individual platelets, platelets in platelet aggregates, and/or aggregated platelets in any type of cell aggregates, and/or determining one or more relevant parameters related to the size distribution from one or more phase shift images, e.g., as described below with reference to fig. 13.

In some embodiments, the method 800 may further include, in step 812, identifying the leukocyte-platelet aggregate 102B in the one or more phase shift images and analyzing the leukocyte-platelet aggregate 102B. The leukocyte-platelet aggregate 102B can also be used as a diagnostic, prognostic, and/or therapeutic indicator of infection, particularly in combination with information obtained from the platelet aggregate 102A. The identification and analysis of the leukocyte-platelet aggregates 102B may be similar to steps 310 and 312 of the method 300 described above.

Step 812 may include, inter alia, determining a total number or fraction of white blood cell-platelet aggregates in the one or more phase shift images, such as a percentage of platelets contained in the white blood cell-platelet aggregates and/or a ratio of a number of white blood cell-platelet aggregates to a total number of individual cells and cell aggregates of any type. Additionally or alternatively, step 812 may further comprise determining one or more of a cell number distribution of the leukocyte-platelet aggregates, a number or fraction of the leukocyte-platelet aggregates comprising at least a predetermined number of cells, and a number or fraction of the leukocyte-platelet aggregates comprising at least a predetermined number of leukocytes. Additionally or alternatively, step 812 may further include determining one or more of a granularity metric, a size distribution, and/or one or more related parameters related to the size distribution, the one or more parameters being for individual leukocytes in a leukocyte-platelet aggregate, platelets, and/or aggregated leukocytes in a leukocyte-platelet aggregate, and any type of cell aggregate.

Fig. 9 to 13 show the results of the study performed by the inventors of the present invention on a control group of patients and healthy individuals suffering from acute Covid-19 infection (no acute Covid-19 infection and no chronic cardiovascular disease or inflammatory disease). By using the method for detecting cell aggregates of biological cells according to the present invention, a blood sample of a subject is analyzed with a quantitative phase contrast microscope, and the result is obtained.

FIG. 9 depicts experimental data relating viral load to the fraction of platelet aggregates in patient blood samples for Covid-19 patients. The fraction of platelet aggregates is determined with a quantitative phase-contrast microscope using the method for detecting cell aggregates of biological cells according to the invention. The figure shows a clear correlation between the viral load of the patient and the fraction of platelet aggregates observed, indicating that platelet aggregates constitute a promising biomarker, in particular for Covid-19, and in general for viral infection. Healthy individuals were found to have a platelet aggregate score of less than 3%, while scores above this threshold indicated Covid-19 infection.

Fig. 10a, 10b depict the fraction of platelets in aggregates (ratio of the number of platelets in any type of aggregate to the total number of platelets) for different groups of patients treated in Intensive Care (ICU). In each sub-graph, the left dataset is a healthy individual, the central dataset Covid-19 patient has a World Health Organization (WHO) Covid-19 severity score of 3-5 ("mild"), and the right dataset Covid-19 patient has a WHO Covid-19 severity score of greater than or equal to 8 ("severe"). For detailed information about the severity score of WHO Covid-19, see "Lancet infectious disease" (LANCET INFECT DIS) 2020,20:e192-197. Fig. 10a shows the results of samples taken from patients on the first day of a hospital or on the first day of an Intensive Care Unit (ICU), while fig. 10b shows the results averaged over the whole residence time of each patient in the ICU, with new samples taken daily where each patient stays in the ICU. In both cases, a significant increase in the fraction of platelets in the aggregate was observed for patients with severe Covid-19, which demonstrates the potential of platelets in the aggregate as a prognostic indicator of Covid-19 in the case of fig. 10 a. Similar results were also observed for the fraction of platelet aggregates in the samples.

FIGS. 11a and 11b depict experimentally determined cell number distribution of platelet aggregates in blood samples of Covid-19 patients. FIG. 11a shows a histogram of the number of platelets in the identified platelet aggregates of the severe Covid-19 patient group compared to healthy individuals. Figure 11b shows the same data but including data for the mild Covid-19 patient group. Although Covid-19 patients exhibited lower levels of platelet aggregates consisting of two platelets, the number of platelet aggregates consisting of three or more platelets increased. This suggests that the number or fraction of platelet aggregates comprising three or more cells and/or the average number of platelets per platelet aggregate may be used as an indicator of Covid-19 infection and its severity.

FIG. 12 depicts an example of experimentally determined granularity metrics for Covid-19 infected patients with different severity as compared to healthy individuals. The upper plot shows the ratio of the median optical height value (median phase shift for all pixel values in the segmented cells, "average optical height") to diameter (arbitrary units) for individual platelets, while the lower plot shows the same amount of aggregated platelets contained in any type of aggregate. This ratio is known to be related to the particle size of the individual cells and thus constitutes an example of a particle size measure. Platelets in samples from severely infected patients showed lower particle sizes than in samples from healthy individuals.

Figure 13 depicts the size distribution of individual platelets (upper column) and aggregated platelets in any type of aggregate (lower column) of healthy individuals (left column), mild Covid-19 patients (middle column) and severe Covid-19 patients (right column) determined experimentally. The figure shows a histogram of equivalent diameters of individual platelets determined from a phase shift image obtained by the method according to the invention. For Covid-19 patients, the distribution of platelet sizes was shifted slightly to the right to larger diameters. For patients with severe infections, a significant second peak can be observed at a diameter of about 5 μm, which may be related to the occurrence of reticulocytes (i.e., immature platelets, which are highly active and prone to aggregation).

FIG. 14 depicts the change over time ("days in the ICU") in the fraction of individual platelets ("individual thrombocytes"), the fraction of aggregated platelets ("aggregated thrombocytes") and the total fraction of platelets ("thrombocytes") in a sample of Covid-19 patients determined using the method according to the invention, wherein the fractions shown in FIG. 14 are normalized to the total number of cells. For comparison, platelet counts from the same samples as determined using conventional flow cytometry methods ("thrombocytes sysmex") were used as a reference. The latter was determined with a Sysmex XN flow cytometer (Sysmex Eu-rope GmbH, norderstedt, germany) which did not resolve platelet aggregates.

The embodiments of the invention disclosed herein are merely comprised of specific examples for the purpose of illustration. The present invention may be implemented in various ways and many modifications without changing the basic characteristics. Accordingly, the invention is limited only by the claims that follow.

REFERENCE SIGNS LIST

100-Device for detecting cell aggregates

102A-platelet aggregates

102B-leukocyte-platelet aggregates

104A-platelets

104B-white blood cells

106-Base

108-Microscope

108A-microscope controller

110-Object

112-Holographic imaging system

14-Imaging lens

116-Camera

118-Illumination source

120-Microfluidic cell

122-Sample preparation Unit

124-Controller

200-Microfluidic system

202-Measurement volume

202A-detection window

202B-Lighting Window

204-Hydrodynamic focus joint

206A-sample channel

206B-sheath flow channel

208A-sample fluid flow

208B-sheath flow

300-Method for detecting cell aggregates

302-Step of preparing a suspension comprising biological cells from a sample and a viscoelastic fluid

304 Step of generating a suspension flow

306-Step of generating two or more sheath flows

308-A step of taking one or more phase shifted images

310-Step of identifying cell aggregates in one or more phase-shifted images

312-Step of analyzing cell aggregates in one or more phase-shifted images

800-Method for detecting cell aggregates

802-Step of preparing a suspension comprising biological cells from a blood sample

804 Step of generating a suspension flow

806-A step of taking one or more phase shifted images

808-Step of identifying and analyzing platelet aggregates

810-Step of identifying and analyzing leukocyte-platelet aggregates

Claims

1. A method (800) of detecting a cell aggregate (102A, 102B) of biological cells (104A, 104B) using a quantitative phase-contrast microscope (108), the method (800) comprising:

Preparing a suspension comprising biological cells (104A, 104B) from a sample;

Generating a flow (208A) of the suspension along the microfluidic channel (202) to viscoelastically and/or hydrodynamically focus the cell aggregates (102A, 102B) in the suspension in a focal plane of the quantitative phase-contrast microscope (108);

capturing one or more phase-shifted images of biological cells (104A, 104B) in suspension using a quantitative phase-contrast microscope (108); and

Identifying an aggregate of cells (102A, 102B) in the one or more phase shift images,

Wherein the sample is a whole blood sample or a blood component sample and identifying cell aggregates (102A, 102B) in the one or more phase shift images comprises identifying platelet aggregates in the one or more phase shift images.

2. The method (800) of claim 1, further comprising determining a total number or fraction of platelet aggregates (102A) in the one or more phase shifted images.

3. The method (800) of claim 2, further comprising using the total number or fraction of platelet aggregates (102) as an indicator of complications in infected patients, particularly Covid-19 patients.

4. The method (800) of any of the preceding claims, wherein identifying the cell aggregates (102A, 102B) in the one or more phase shift images comprises determining a number of cells in the individual cell aggregates (102A, 102B).

5. The method (800) according to any one of the preceding claims, further comprising determining the number or fraction of platelet aggregates comprising at least a predetermined number of cells, in particular determining the number or fraction of platelet aggregates comprising three or more cells.

6. The method (800) of any of the preceding claims, wherein identifying cell aggregates (102A, 102B) in the one or more phase-shifted images further comprises identifying white blood cell-platelet aggregates (102B) in the one or more phase-shifted images.

7. The method (800) of claim 6, further comprising:

Determining a total number or fraction of leukocyte-platelet aggregates (102B) in the one or more phase-shifted images; and/or

The number or fraction of leukocyte-platelet aggregates (102B) comprising at least a predetermined number of cells is determined, in particular the number of leukocyte-platelet aggregates (102B) comprising three or more cells is determined.

8. The method (800) of claim 6 or 7, further comprising determining a number or fraction of leukocyte-platelet aggregates (102B) comprising at least a predetermined number of leukocytes.

9. The method (800) of claim 8, wherein:

The method (800) includes determining a number or fraction of white blood cell-platelet aggregates (102B) comprising two or more white blood cells (104B) and/or determining a number or fraction of white blood cell-platelet aggregates (102B) comprising three or more white blood cells (104B); and

The presence of a leukocyte-platelet aggregate comprising two or more leukocytes and/or the presence of a leukocyte-platelet aggregate comprising three or more leukocytes is used as an indicator of infection, in particular of a viral infection and/or a bacterial infection.

10. The method (800) of any of the preceding claims, wherein the method (800) further comprises determining a total number or fraction of aggregated platelets contained in any type of cell aggregates (102A, 102B) in the one or more phase shifted images.

11. The method (800) of any of the preceding claims, wherein the method (800) further comprises determining a granularity metric from the one or more phase-shifted images, wherein the granularity metric characterizes granularity of one or more cells in the one or more phase-shifted images.

12. The method (800) of any of the preceding claims, wherein the method (800) further comprises determining a size distribution of a plurality of cells and/or one or more parameters related to the size distribution from the one or more phase shifted images.

13. The method (800) according to claim 11 or 12, wherein the granularity measure, the size distribution and/or one or more parameters related to the size distribution are determined for a specific type of individual cells and/or aggregated cells, preferably for individual platelets, platelets in platelet aggregates, platelets in leukocyte-platelet aggregates and/or aggregated platelets in any type of cell aggregates.

14. The method (800) of any of the preceding claims, wherein the suspension further comprises a viscoelastic fluid, in particular wherein the viscoelastic fluid comprises a shear-thinning polymer having a molecular weight of 2-10 MDa, and wherein the mass fraction of the shear-thinning polymer in the suspension is less than 0.2%.

15. The method (800) according to any one of the preceding claims, wherein the viscoelastic and/or hydrodynamic focusing is adapted such that individual cells (104A, 104B) in the suspension are also focused in the focal plane of the quantitative phase-contrast microscope.

16. The method (800) of any of the preceding claims, wherein preparing the suspension does not comprise lysis of erythrocytes, spheroidization of platelets and/or erythrocytes, and/or labeling or staining of cells.

17. The method (800) of any of the preceding claims, wherein capturing one or more phase-shifted images of the biological cells (104A, 104B) in the suspension using the quantitative phase-contrast microscope (108) comprises capturing a first phase-shifted image of the biological cells (104A, 104B) in the suspension at a first flow rate of the suspension and/or a first shear rate of the suspension, and capturing a second phase-shifted image of the biological cells (104A, 104B) in the suspension at a second flow rate of the suspension different from the first flow rate and/or a second shear rate of the suspension different from the first shear rate.

18. Device (100) for detecting cell aggregates (102A, 102B) of biological cells (104A, 104B) using the method (800) according to any one of the preceding claims, the device (100) comprising:

A base (106) configured to receive a microfluidic system (200) comprising a measurement volume (202);

A microscope (108) configured to take phase-shifted images of biological cells (104A, 104B) in a measurement volume (202);

A microfluidic unit (120) configured to receive a sample fluid comprising biological cells (104A, 104B) from a blood sample, wherein the microfluidic unit (120) is configured to generate the sample fluid flow (208A) through the measurement volume (202) to viscoelastically and/or hydrodynamically focus cell aggregates (102A, 102B) in the sample fluid flow (208A) in a focal plane of the microscope (108); and

A controller (124) configured to identify platelet aggregates (102A) in phase shifted images of a sample fluid stream (208A) obtained from the microscope (108).

19. The device (100) of claim 18, wherein the controller (124) is further configured to determine a total number or fraction of platelet aggregates (102) in the phase shift image, the total number or fraction of platelet aggregates (102) comprising a number or fraction of platelet aggregates of at least a predetermined number of cells in the phase shift image and/or a total number or fraction of aggregated platelets contained in any type of cell aggregates (102A, 102B) in the phase shift image.

20. The device (100) according to claim 18 or 19, wherein the controller (124) is further configured to identify white blood cell-platelet aggregates (102B) in the phase shift image, in particular wherein the controller (124) is configured to determine a total number or fraction of white blood cell-platelet aggregates (102B) in the phase shift image, the number or fraction of white blood cell-platelet aggregates (102B) comprising at least a predetermined number of cells in the phase shift image, and/or the number or fraction of white blood cell-platelet aggregates (102B) comprising at least a predetermined number of white blood cells in the phase shift image.

21. The apparatus (100) of any one of claims 18 to 20, wherein the controller (124) is further configured to determine a size distribution of a plurality of cells from the phase-shift image, one or more parameters related to the size distribution, and/or a granularity metric, wherein the granularity metric characterizes granularity of one or more cells in the phase-shift image.

22. The device (100) according to any one of claims 18 to 21, wherein the sample fluid further comprises a viscoelastic fluid, in particular wherein the viscoelastic fluid comprises a shear-thinning polymer having a molecular weight of 2 to 10MDa, and the mass fraction of the shear-thinning polymer in the sample fluid is less than 0.2%.

23. The device (100) of claim 22, further comprising a sample preparation unit (122) configured to provide the viscoelastic fluid comprising a shear-thinning polymer having a molecular weight of 2-10 MDa to prepare the sample fluid comprising biological cells (104A, 104B) from the sample and the viscoelastic fluid, wherein a mass fraction of the shear-thinning polymer in the sample fluid is less than 0.2%.

24. The apparatus (100) of any one of claims 18 to 23, wherein the controller (124) is configured to perform the method (800) of any one of claims 1 to 17.