System and method for 3D imaging flow cytometry

Abstract

There is provided a system for three-dimensional (3D) imaging flow cytometry, comprising microfluidic device having disposed therein one or more channels each extending along a central axis, the one or more channels configured to continuously flow therethrough a fluid comprising a sample containing at least one object, the at least one object caused to be displaced, as the fluid flows through the one or more channels, away from the central axis and towards a target internal surface of the one or more channels to cause the at least one object to rotate, an imaging system optically coupled to the microfluidic device, the imaging system configured to acquire a plurality of two-dimensional (2D) images of the at least one object as the at least one object rotates, and a processing unit coupled to the imaging system, the processing unit configured to obtain the plurality of 2D images and to generate a 3D representation of the at least one object based on the plurality of 2D images.

Description

SYSTEM AND METHOD FOR 3D IMAGING FLOW CYTOMETRYCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims the benefit of United States Provisional Patent Application No. 63 / 648,752 filed on May 17, 2024, the contents of which are hereby incorporated by reference.FIELD

[0001] The improvements generally relate to the field of imaging flow cytometry, and more particularly to three-dimensional (3D) imaging flow cytometry.BACKGROUND

[0002] Flow cytometry is of practical interest in a number of fields due to the fact that it provides information on many individual samples to characterize an entire population. Imaging flow cytometry expands the amount of information provided by traditional flow cytometry by providing subcellular spatial resolution. The analysis opportunities for imaging flow cytometry data have been further increased with the rise of machine learning tools which benefit from rich datasets. As a result, flow imaging cytometry has become a fundamental tool in bioscience, among other applications.

[0003] As real samples are intrinsically three-dimensional (3D) in nature, 3D imaging is of great practical interest, with applications in research and diagnostics. However, while numerous two-dimensional (2D) imaging flow cytometry devices are currently available, few 3D imaging flow cytometry devices exist. Some 3D imaging flow cytometry devices use a thin light sheet to optically section a sample containing one or more cells, where the movement of a cell through the light sheet provides individual planes which are reconstructed in a subsequent step to obtain 3D information. Other 3D imaging flow cytometry devices capture and semi-immobilize (or trap) the samples while 3D is acquired by rotating the trapped object or rotating the imaging system around the object. Yet other 3D imaging flow cytometry devices use a modified microscope optical setup in which one image captured by the microscope is divided into sub-images in the back focal plane. The sub-images contain several perspectives of the sample which are algorithmically recombined with low resolution.

[0004] Therefore, existing 3D imaging flow cytometry techniques generally prove complex and there is a need for improvement.SUMMARY

[0005] In accordance with one aspect, there is provided a system for three-dimensional (3D) imaging flow cytometry. The system comprises a microfluidic device having disposed therein one or more channels each extending along a central axis, the one or more channels configured to continuously flow therethrough a fluid comprising a sample containing at least one object, the at least one object caused to be displaced, as the fluid flows through the one or more channels, away from the central axis and towards a target internal surface of the one or more channels to cause the at least one object to rotate, an imaging system optically coupled to the microfluidic device, the imaging system configured to acquire a plurality of two-dimensional (2D) images of the at least one object as the at least one object rotates, and a processing unit coupled to the imaging system, the processing unit configured to obtain the plurality of 2D images and to generate a 3D representation of the at least one object based on the plurality of 2D images.

[0006] In at least one embodiment in accordance with any previous / other embodiment described herein, a flow speed of the fluid flowing through the one or more channels is set to a limited flow speed, further wherein the limited flow speed and a difference in density between the at least one object and the fluid cause the at least one object to be displaced towards the target internal surface and to rotate at a given velocity.

[0007] In at least one embodiment in accordance with any previous / other embodiment described herein, the at least one object is displaced away from the central axis and towards the target internal surface of the one or more channels by applying a centrifugal force on the at least one object.

[0008] In at least one embodiment in accordance with any previous / other embodiment described herein, the at least one object is displaced away from the central axis and towards the target interface within the one or more channels by using at least one optical tweezer to apply a force on the at least one object.

[0009] In at least one embodiment in accordance with any previous / other embodiment described herein, the at least one object has at least one magnetic nanoparticle attached thereto, further wherein the at least one object is displaced away from the central axis andtowards the target internal surface by applying a magnetic force on the at least one magnetic nanoparticle.

[0010] In at least one embodiment in accordance with any previous / other embodiment described herein, the one or more channels comprises the target internal surface and at least one additional internal surface, the target internal surface and the at least one additional internal surface having different surface properties, further wherein the different surface properties create a non-uniform flow through the one or more channels that brings the at least one object to the position away from the central axis and towards the target internal surface.

[0011] In at least one embodiment in accordance with any previous / other embodiment described herein, the target internal surface is one of at least part of an upper surface and at least part of a lower surface of the one or more channels.

[0012] In at least one embodiment in accordance with any previous / other embodiment described herein, the one or more channels comprise a first channel having a first fluid flowing therethrough and a second channel disposed below the first channel and having a second fluid flowing therethrough, the first fluid having a first viscosity and the second fluid having a second viscosity lower than the first viscosity, further wherein the target internal surface is an interface between the first fluid and the second fluid.

[0013] In at least one embodiment in accordance with any previous / other embodiment described herein, the imaging system comprises at least one light source and a stationary image acquisition device, the at least one light source configured to illuminate the sample with light as the fluid flows through the one or more channels, and the image acquisition device configured to capture light scattered by the sample in response to illumination thereof for acquiring the plurality of 2D images of the at least one object as the fluid flows through the one or more channels and as the at least one object rotates.

[0014] In at least one embodiment in accordance with any previous / other embodiment described herein, the imaging system uses at least one of a fluorescence, a transmission, a phase contrast, a bright-field, and a dark-field imaging modality.

[0015] In at least one embodiment in accordance with any previous / other embodiment described herein, the processing unit is configured to identify common features in at leasttwo of the plurality of 2D images of the at least one object, determine relative positions of the common features in the at least two of the plurality of 2D images, determine angles between the relative positions of the common features and a reference position associated with a common point in the at least two of the plurality of 2D images, and generate the 3D representation of the at least one object comprising generating a 3D mesh based on the angles as determined, the 3D mesh indicative of 3D locations of the common features in the at least one object.

[0016] In at least one embodiment in accordance with any previous / other embodiment described herein, the at least one object has at least one marker attached thereto, further wherein the processing unit is configured to identify the common features by detecting the at least one marker in the at least two of the plurality of 2D images.

[0017] In at least one embodiment in accordance with any previous / other embodiment described herein, the at least one object is one of a single cell and a cluster of cells, further wherein the microfluidic device is configured to cause each cell to be controllably rotated within the one or more channels as the fluid flows therethrough, and the imaging system is configured to acquire the plurality of 2D images representative of full rotations of each cell.

[0018] In at least one embodiment in accordance with any previous / other embodiment described herein, the at least one object is an object encapsulated in a transparent gel to modify an effective size and / or density thereof.

[0019] In at least one embodiment in accordance with any previous / other embodiment described herein, the at least one object has a size ranging from about 5 microns to about 100 microns.

[0020] In accordance with another aspect, there is provided a method for three- dimensional (3D) imaging flow cytometry. The method comprises receiving, at a computing device, a plurality of two-dimensional (2D) images of at least one object contained in a sample, the plurality of 2D images acquired as the at least one object rotates within one or more channels of a microfluidic device while fluid comprising the sample continuously flows through the one or more channels, the at least one object caused to be displaced within the microfluidic device, as the fluid flows through the one or more channels, away from a central axis of the one or more channels and towards a targetinternal surface of the one or more channels to cause the at least one object to rotate, and generating, at the computing device, a 3D representation of the at least one object based on the plurality of 2D images.

[0021] In at least one embodiment in accordance with any previous / other embodiment described herein, generating the 3D representation of the at least one object comprises identifying common features in at least two of the plurality of 2D images of the at least one object, determining relative positions of the common features in the at least two of the plurality of 2D images, determining angles between the relative positions of the common features and a reference position associated with a common point in the at least two of the plurality of 2D images, and generating a 3D mesh based on the angles as determined, the 3D mesh indicative of 3D locations of the common features in the at least one object.

[0022] In at least one embodiment in accordance with any previous / other embodiment described herein, identifying the common features comprises detecting, in the at least two of the plurality of 2D images, at least one marker attached to the at least one object.

[0023] In accordance with another aspect, there is provided a system for three- dimensional (3D) imaging flow cytometry. The system comprises a processing unit and a non-transitory memory communicatively coupled to the processing unit and comprising computer-readable program instructions executable by the processing unit for receiving a plurality of two-dimensional (2D) images of at least one object contained in a sample, the plurality of 2D images acquired as the at least one object rotates within one or more channels of a microfluidic device while fluid comprising the sample continuously flows through the one or more channels, the at least one object caused to be displaced within the microfluidic device, as the fluid flows through the one or more channels, away from a central axis of the one or more channels and towards a target internal surface of the one or more channels to cause the at least one object to rotate, and generating a 3D representation of the at least one object based on the plurality of 2D images.

[0024] Many further features and combinations thereof concerning embodiments described herein will appear to those skilled in the art following a reading of the instant disclosure.DESCRIPTION OF THE FIGURES

[0025] In the figures,

[0026] Fig. 1 A is a schematic diagram of an example of a system for 3D imaging flow cytometry, in accordance with one embodiment;

[0027] Fig. 1 B is a cross-sectional view of the microfluidic channel of Fig. 1A, in accordance with one embodiment;

[0028] Fig. 1 C is a side view of the microfluidic channel of the microfluidic device of Fig. 1A having a fluid with a low flow speed flowing therethrough, in accordance with one embodiment;

[0029] Fig. 1 D is a schematic diagram of an example multi flow layer configuration for the microfluidic device of Fig. 1A, in accordance with one embodiment;

[0030] Fig. 2A is a side view of the microfluidic channel of the microfluidic device of Fig. 1A, in accordance with one embodiment;

[0031] Fig. 2B is a top view of the microfluidic channel of the microfluidic device of Fig. 1A, in accordance with one embodiment

[0032] Fig. 3 is a block diagram of the processing unit of Fig. 1A, in accordance with one embodiment;

[0033] Fig. 4A illustrates dark-field imaging results obtained by the imaging system of Fig. 1A, in accordance with one embodiment;

[0034] Fig. 4B illustrates bright-field imaging results obtained by the imaging system of Fig. 1A, in accordance with one embodiment;

[0035] Fig. 4C illustrates simulations and 3D reconstructions obtained by the imaging system of Fig. 1A, in accordance with one embodiment;

[0036] Fig. 4D and Fig. 4E illustrate point reconstruction results obtained by the imaging system of Fig. 1A, in accordance with one embodiment;

[0037] Fig. 4F illustrates a 2D image of a cell captured by the imaging system of Fig. 1A, a mesh of the cell reconstructed by the processing unit of Fig. 1 A, and a cell surface generated by the processing unit of Fig. 1 A, in accordance with one embodiment;

[0038] Fig. 4G illustrates a fluorescence reconstruction snapshot obtained by the imaging system of Fig. 1A, in accordance with one embodiment;

[0039] Fig. 5 is a flowchart of an example method for performing 3D imaging flow cytometry, in accordance with one embodiment; and

[0040] Fig. 6 is a block diagram of an example computing device, in accordance with one embodiment.

[0041] It will be noted that throughout the appended drawings that like features are identified by like reference numerals.DETAILED DESCRIPTION

[0042] Described herein are systems and methods for 3D imaging flow cytometry, where one or more objects are imaged as they traverse a microfluidic device. In one embodiment, the one or more objects being imaged are cells or clusters of cells. It should however be understood that any other suitable objects may be imaged using the systems and methods described herein. As used herein, the term “imaging flow cytometry” refers to a group of techniques used to capture image data from one or more objects (e.g., cells) as they pass through a microfluidic device. As will be described further below, the one or more objects are brought into a flow channel of the microfluidic device, with different forces being applied at the top and the bottom of the one or more objects to cause the one or more objects to rotate as they flow through the microfluidic device and change their direction of rotation (for example, between a forward direction of rotation and a backward direction of rotation). As the one or more objects rotate, multiple 2D images are acquired and collected to produce a dataset of images from different angles. The acquired 2D image data is then combined to produce a 3D representation (also referred to herein as a 3D reconstruction) of the one or more objects.

[0043] Fig. 1A shows an example of a system 100 for 3D imaging flow cytometry. The system 100 comprises a microfluidic device 102 optically coupled to an imaging system 104, to which a processing unit 106 is coupled. As used herein, the term “microfluidic” refers to a device or system that comprises at least one channel (referred to herein as a “microfluidic channel” or “microchannel”) having microscale dimensions and which is configured to handle, process, and / or analyze a sample. The microfluidic device 102 comprises a body 108 bonded to a substrate 110. The body 108 and the substrate 1 10may be bonded to one another using any suitable technique including, but not limited to, plasma treatment or the use of a double-sided adhesive. In some embodiments, the microfluidic device 102 may be built inside a transparent substrate using local opto / chemical etching. 3D printing may also be used. The body 108 and the substrate 110 may be made of any suitable material. In one embodiment, the substrate 110 is made of glass and the body 108 is made of a mineral organic polymer such as polydimethylsiloxane (PDMS) (e.g., PDMS RTV-615 or PDMS Sylgard 184).

[0044] The body 108 has one or more microfluidic channels 1 12 disposed therein for transporting fluid(s) through the microfluidic device 102. Although a single microfluidic channel 112 is shown in Fig. 1A, it should be understood that this is for illustrative purposes only and that any suitable number of microfluidic channels 1 12 may apply. As used herein, the term “channel” refers to a pathway formed in or through a medium to allow movement of fluids, including, but not limited to, liquids. As used herein, the term “microfluidic channel” (or “microchannel”) refers to a channel of a microfluidic device or system having cross-sectional dimensions between about 4 pm and about 5000 pm. The microfluidic channel(s) 112 may have any suitable length and volume, any suitable shape (e.g., a U-shape), and any suitable configuration (e.g., linear or non-linear). Referring to Fig. 1 B in addition to Fig. 1A, the body 108 comprises a microfluidic channel 112 having a linear configuration with a central axis A (substantially parallel to the Z axis in the (X, Y, Z) coordinate system illustrated in Fig. 1 B), and at least one internal surface defining a pathway through which a fluid flows (e.g., in a typically laminar flow) along the direction indicated by straight arrow B (i.e. along a direction substantially parallel to the central axis A and to the Z axis). In the illustrated embodiment, the microfluidic channel 112 is straight and has a rectangular cross-section (with the cross-section being constant, i.e. having a constant width and a constant height) and comprises a substantially planar upper internal surface 1 13a, a substantially planar lower internal surface 113b opposite the upper internal surface 1 13a, and substantially planar opposite side surfaces 113c, 113d that interconnect the upper internal surface 1 13a to the lower internal surface 113b. In some embodiments, the microfluidic channel 1 12 may be substantially planar with a larger dimension along the Z axis (i.e. with an imaging chip or field of view having a large size), i.e. a dimension along the Z axis that is several orders of magnitude (e.g., two to three times) largerthan its dimensions along the X and Y axes. It should however be understoodthat the microfluidic channel 112 may have any suitable dimensions, and a cross-section of any suitable shape other than a rectangular cross-section.

[0045] Referring back to Fig. 1 A, the body 108 comprises an inlet 1 14a and an outlet 114b coupled to one or more microfluidic reservoirs and pumps (not shown) via microfluidic tubing (not shown). A sample 1 16 containing at least one cell (not shown) is comprised in a fluid 118 and injected into the microfluidic device 102 through the inlet 114a. In one embodiment, the sample 116 is suspended in the fluid 1 18 outside of the microfluidic device 102, and not necessarily while in the microfluidic device 102. The microfluidic channels 112 fluidly connect the inlet 114a to the outlet 114b, such that the fluid 118 is directed from the inlet 1 14a to the channel(s) 1 12, flows through the channel(s) 112, and is expelled via the outlet 114b.

[0046] Any suitable sample 1 16 may apply. In some embodiments, the sample 116 is a biological sample, such as blood, saliva, fine needle biopsy samples, dilutions of bodily fluids or extracted samples. The at least one cell contained in the sample 1 16 may be alive, dead, fixed, or possibly fixed and treated to physically expand its size (e.g., using a technique referred to as “expansion microscopy”) to more easily resolve its features, change its density, or modify its transparency. In addition, the at least one cell contained in the sample 116 may comprise a single cell or a combination (or group, also referred to as a “cluster”) of cells. In other words, the sample 116 may comprise one or more cells. For example, the sample 1 16 may comprise a single biological cell that is imaged using the system 100. In other examples, the sample 116 may contain an object (e.g., multiple cells) which are embedded (or encapsulated) into a medium (e.g., a transparent gel such as agarose) in order to modify (e.g., increase) the overall effective size, density, and / or optical properties of the object imaged using the system 100 to improve control and imaging. The object being imaged (i.e. the at least one cell) may thus have any suitable size. In some embodiments, the object’s size ranges from about 1 micron to about 500 microns, depending on the intended throughput and image resolution. In some embodiments, the object has a size ranging from about 5 microns to about 100 microns.

[0047] The microfluidic device 102 is configured to cause the fluid 118 (comprising the sample 116) to continuously flow therethrough, with the cell(s) in the sample 116 rotating within the microfluidic channel 112 and reaching different positions at different points in time as the fluid 1 18 flows through the microfluidic channel 112. For this purpose, themicrofluidic device 102 and operating conditions of the system 100 are designed such that, as the fluid 118 moves through the microfluidic device 102, each cell in the sample 116 is brought to a position that is away from the central axis A of the microfluidic channel 112 and towards one of the internal surfaces (referred to as a “target internal surface”) of the microfluidic channel 1 12. This results in the cell experiencing a shear force (i.e. a tangential force acting on the cell surface) that causes the cell to move within the microfluidic channel 112 by rotating along (or close to) the target internal surface. It should be understood that the target internal surface along (or close to) which the cell rotates may vary depending on the embodiment. In some embodiments, the imaging system 104 is positioned below the microfluidic device 102 (as illustrated in Fig. 1A) such that the cell is moved (e.g., using gravity) towards a lower target internal surface of the microfluidic channel 112. It should however be understood that, in other embodiments, the imaging system 104 may be positioned above the microfluidic device 102 and the cell moved towards an upper target internal surface of the microfluidic channel 1 12. In other embodiments, the target surface may be a side surface of the microfluidic channel 112. In addition, patterning may be used on a surface of the microfluidic channel 1 12 to include some (i.e. part of) or the entirety of the patterned surface as the target surface. In other words, the target surface may comprise at least part of a surface of the microfluidic channel 1 12. It should also be understood that, while reference is made herein to the cell rotating along (or close to) an internal surface of the microfluidic channel 112, such a surface may not refer to the interface of a static surface (as illustrated in Fig. 1 A) but may instead be the interface of two (2) liquids, with one liquid traveling next to (or below) the other liquid.

[0048] Each cell may be controllably rotated in multiple stages to produce multiple directions of rotation. More than one rotation may then be used to capture the system’s temporal dynamics. In particular, the cell may be observed by the imaging system 104 as it rotates in such a way that multiple full rotations are captured in the same recording. The imaging system 104 may, in some embodiments, be configured to capture at least ten (10) images during a full rotation of the cell. Any suitable number of half rotations (i.e. any suitable degree of rotation) may apply. For instance, in one embodiment, instead of a 360 degree rotation (i.e. one full rotation) of the cell being captured in a given recording, 180 (i.e. one half of a full rotations), a 540 degree rotation (i.e. one and a half full rotations), or a 720 degree rotation (i.e. two full rotations) of the cell may be captured. In oneembodiment, at least a 175 degree rotation and up to about a 720 degree rotation of the cell can be achieved. It should however be understood that the cell may continue to rotate to values in between the above-mentioned increments (e.g. to any suitable value between 175 degrees and 720 degrees), and the data acquired by the system 100 may be used even if it does not fall on a specific full number of integer rotations. For example, a 425 degree rotation may be suitable in some embodiments. It should however be understood that, in cases where the rotation is incomplete such that a full rotation of the cell is not captured (e.g., a rotation of less than 180 degrees is captured by the imaging system 104), information about the sample may still be partially reconstructed. Thus, multiple rotations may be captured by the imaging system 104 in a given recording and used for reconstruction purposes, although any amount of rotation may apply provided the rotation is sufficiently large to reconstruct information about the cell.

[0049] In addition, the microfluidic device 102 may be operated in multiple discrete stages that use the control of flow to change the direction of rotation. For example, when imaging an L or T junction, the cell changes directions as it moves forward, causing the cell to rotate in a different direction. The direction of rotation may depend on the applied fluid pressure and may be changed in one imaging session.

[0050] The rotation of the cell(s) contained in the sample 16 may be achieved using any suitable technique. In one embodiment, rotation of the cell(s) is achieved by controlling the flow speed of the fluid 1 18 (where the flow is driven by a syringe pump or a pressure difference) and using gravity to bring the cell(s) downward. In particular, any suitable control means (e.g., a controller coupled to the syringe pump) is used to create a flow speed (referred to herein as a “low flow speed” or a “limited flow speed”) that is slower compared to the flow speed used in existing microfluidic devices. The flow speed may be controlled dynamically (e.g., using the controller) to cause the cell(s) to rotate at a given velocity. Any suitable velocity may apply. In one embodiment, the value of the flow speed is set such that the cell(s) rotate once very few (e.g., three (3)) seconds. The flow speed of existing microfluidic devices typically causes cells to rotate at a velocity that exceeds about 50 centimeters per second and, in one embodiment, it is proposed herein to use a flow speed that causes the cells to rotate at a velocity of about 50 microns per second in the microfluidic device 102. This low flow speed and the difference in density between the cell(s) and the surrounding media of the microfluidic device 102 (e.g., a density difference between the cell(s) and the fluid 118) causes the cell(s) to move systematically to aninterface (referred to herein as a target internal surface). It should be understood that, in some embodiments, the cell(s) may first be brought towards the target internal surface (due to the difference in density) prior to the flow speed being controlled and set to the limited flow speed in order to cause the cell(s) to move at a desired velocity and rotate. In one embodiment, a density difference between the more dense to less dense of the two between about 1 .05 and about 2 may be achieved, although it should be understood that it is desirable for the density difference to be as large as possible in the context of not perturbing the sample. In some embodiments, the cell(s) may be more dense than the fluid. It should however be understood that, in other embodiments, the fluid may be more dense than the cell(s).

[0051] Referring to Fig. 1 C in addition to Fig. 1A, the low flow speed and density difference in turn causes the cell(s) (e.g., cell 1 19) contained in the sample 1 16 to leave the center of the laminar flow through the microfluidic device 102 (i.e. to sink away from the central axis A of the microfluidic channel(s) 1 12 towards the bottom of the channel(s) 112) and to rotate along (i.e. adjacent to or on) the target internal surface (e.g., the lower internal surface 1 13b) of the microfluidic channel 112. The low flow speed may be created using any suitable technique or device(s) including, but not limited to, using syringe pumps to deliver a constant (i.e. low) flow speed, using a constant pressure source to define a fixed pressure drop across the microfluidic channel 112 in order to achieve the low flow speed, or positioning the inlet 114a and the outlet 114b of the microfluidic device 102 at different heights to achieve the low flow speed. In some embodiments, the dimensions of the microfluidic device 102 may be modified such that the flow speed is lower (i.e. the flow velocity is slower) in a wider part of the microfluidic device 102, namely in front of the imaging region. Other embodiments may apply.

[0052] In another embodiment, multiple densities of liquids may be used to create an interface (i.e. the target internal surface) towards which the cell(s) are brought to cause them to rotate. Depending on the implementation, the interface may be a liquid-to-liquid interface or a liquid-to-solid interface that can be chemically modified. The liquid-to-liquid interface may be created by premixing two fluids and injecting them into the microfluidic channel 112 prior to imaging. The liquid-to-liquid interface is then created between the two fluids as the premixed fluids flow through the microfluidic device 102. The liquid-to-solid interface may be created by connecting several microfluidic channels as in 102. The microfluidic device 102 may indeed be configured to comprise multiple microfluidicchannels as in 1 12 arranged in any suitable manner (e.g., at an angle such as in a perpendicular arrangement, side-by-side, or in a vertical stack as shown in Fig. 1 D). In particular, the microfluidic channels 1 12 may be positioned one above the other and be composed of medias of different physical properties, such as viscosities. The microfluidic channels 112 may also be placed at an angle (i.e. in the direction of the Z axis) in order to provide multifocal image planes. Imaging may be performed, using the imaging system 104, at a given one of the microfluidic channels 112 (referred to herein as a “main channel”). The main channel may comprise the cell(s) to be imaged and two additional microfluidic channels as in 112 or more microfluidic channels as in 112 (e.g., for forming a Y, T or cross junction device) may be connected to the main channel. The additional (two or more) microfluidic channels (or abrupt changes in channel geometry such as widening or shrinking) may be used to manipulate the movement of the cell(s) contained in the main channel, by creating the liquid-to-solid interface between the main channel and the additional channels.

[0053] Fig. 1 D illustrates an example stacked-flow configuration 150 forthe microfluidic device 102, in which two (2) microfluidic channels 152a, 152b are stacked vertically along the Y axis, with the first microfluidic channel 152a being disposed above the second microfluidic channel 152b. A first fluid 154a having a first viscosity (referred to herein as a higher-viscosity fluid) is injected into the first microfluidic channel 152a (along a direction illustrated by arrow 156a), flows through the first microfluidic channel 152a, and exits therefrom (along the direction illustrated by arrow 158a). A second fluid 154b having a second viscosity lower than the first viscosity (referred to herein as a lower-viscosity fluid) is injected into the second microfluidic channel 152b (along a direction illustrated by arrow 156b), flows through the second microfluidic channel 152b, and exits therefrom (along the direction illustrated by arrow 158b). The higher-viscosity fluid 154a and the lower-viscosity fluid 154b do not readily mix, which creates within the microfluidic device 102 different flow profiles with discontinuities in between that can be used to rotate cell(s) contained in the fluids 154a, 154b at the interface between the fluids 154a, 154b, with the cell(s) being trapped at this interface. In one embodiment, each fluid 154a, 154b, and accordingly each microfluidic channel 152a, 152b contains one or more cells therein. In another embodiment, one of the fluids 154a, 154b (and accordingly a corresponding microfluidic channel 152a, 152b) contains one or more cell(s) therein.

[0054] In another embodiment, rotation of the sample 116 is achieved by creating an attractive interaction between the cell surface and the target internal surface (e.g., the upper internal surface 113a, the lower internal surface 113b , or one of the side surfaces 113c, 113d) of the microfluidic channel 1 12, such that the cell rolls once in contact with the target internal surface. For example, the cell(s) contained in the sample 1 16 may be brought towards the target internal surface of the microfluidic device 102 through the use of magnetic nanoparticles. As understood by those skilled in the art, the term “magnetic nanoparticles” refers to a class of ultrafine particles (referred to as “nanoparticles”) which have a diameter between about 1 and 100 nanometers and magnetic properties that allow the particles to be manipulated using magnetic fields. The magnetic nanoparticles may be generated or synthesized using any suitable technique including, but not limited to, coprecipitation, sol-gel, ultrasonification, sonochemical processing, thermal deposition, gasphase synthesis, plasma, microwave irradiation, spray pyrolysis, laser pyrolysis, mechanical milling, and arc discharge. Each cell in the sample 1 16 may be labelled with (i.e. have attached thereto using any suitable technique) one or more such magnetic nanoparticles. In order to bring the cell(s) towards the target internal surface, the magnetic nanoparticles which are attached to the cells may be manipulated (i.e., transported through the microfluidic channel 112 and attracted towards the target internal surface) by applying magnetic forces thereto. The magnetic forces may be generated and applied on the magnetic nanoparticles using any suitable device or technique including, but not limited to, permanent magnet(s) or electromagnet(s) which may be external to the microfluidic device 102, magnetic micro-wires (also referred to as micro-coils) which may be embedded in the microfluidic device 102, and magnetic thin films (also referred to as micro-strips) which may be embedded in the microfluidic device 102.

[0055] The cell(s) contained in the sample 116 may also be brought towards the target internal surface of the microfluidic device 102 through the use of optical tweezers (e.g., without a need for a density difference between the cell(s) and the fluid 1 18). As understood by those skilled in the art, the term “optical tweezers” refers to techniques in which an incident light (e.g., laser) beam is focused through a lens in order to exert forces on microscopic objects and yield changes in the objects’ displacement that accompany the application of the forces. Using such optical tweezers, forces can thus be applied to the cell(s) contained in the sample 116 in order to move the cell(s) towards the target internal surface of the microfluidic channel 112 and thus lead to a rotation in the cell(s).

[0056] In addition, the internal surfaces 113a, 113b, 1 13c, and 1 13d of the microfluidic channel 112 may be designed to have different surface properties (e.g., made of different materials or of materials with different surface coatings) such that the flow through the microfluidic channel 112 is not uniform across the microfluidic device 102, causing the cell(s) to be brought towards the target internal surface (e.g., the upper internal surface 113a or the lower internal surface 1 13b), thus resulting rotation of the cell(s) contained in the sample 116. The internal surfaces 113a, 113b, 113c, and 113d of the microfluidic channel 112 may comprise any suitable surface and any suitable material including, but not limited to, optically transparent materials, polyethylene glycol (PEG) and its derivatives, biotinylated surfaces, antibody-surfaces, fluorinated surfaces, silanized surfaces and their derivatives, membrane-intercalating derivatives, Poly-L-Lysine, plasma-cleaned surfaces, ozone-cleaned surfaces, and solutions prepared with a piranha- etched solution.

[0057] In yet another embodiment, rotation of the cell(s) contained in the sample 116 may be achieved by centrifugation. In particular, a centrifugal force may be applied on the cell(s) to cause the latter to rotate. This may be achieved using a curved or spiraling microfluidic channel 112. Other embodiments may apply. For instance, before operating the microfluidic device 102 and attaching the latter to reservoirs and pumps via microfluidic tubing, cell(s) may be prepared on a surface by centrifuging the microfluidic device 102 with cell(s) inside to bring the cell(s) onto the surface. The microfluidic device 103 may then be attached to the reservoirs and pumps and coupled to the imaging system 104 (e.g., placed under a microscope).

[0058] Fig. 2A and Fig. 2B respectively illustrate a side view and a top view of the microfluidic channel 1 12, with a cell 202 of the sample (reference 116 in Fig. 1A) rotating as it flows through the microfluidic channel 112. In Fig. 2A, only the lower portion of the microfluidic channel 112, i.e. the portion delimited by the channel’s lower internal surface 113b and the central axis A. The upper portion of the microfluidic channel 112, i.e. the portion delimited by the channel’s upper internal surface (reference 1 13a in Fig. 1 B) and the central axis A, is not shown for clarity purposes. As can be seen from Fig. 2A, the cell 202 is not positioned along the central axis A of the microfluidic channel 112 but has descended towards a bottom portion of the microfluidic channel 112, adjacent the lower inner surface 113b of the microfluidic channel 112. As can be further seen from Fig. 2A, the cell 202 rotates along a direction indicated by curved arrow C as the fluid (reference118 in Fig. 1A) flows through the microfluidic device (reference 102 in Fig. 1A) along the direction indicated by straight arrow B. In the illustrated embodiment, the cell 202 rotates along the direction C about an axis of rotation (not shown) substantially parallel to the X axis (shown in Fig. 2B) and flows along the direction B which is substantially parallel to the central axis A and to the Z axis (shown in Fig. 1 B). Fig. 2A illustrates five (5) different positions 204i, 2042, 204s, 2044, and 204s of the cell 202 as it is rotated, the different positions 204i, 2042, 204s, 204 , and 204s being captured by the imaging system 104, and more particularly by the image acquisition device 124 as the cell 202 flows through the microfluidic device 102.

[0059] Although reference is made herein to the particular orientation of the microfluidic device 102 and related components relative to the X, Y, and Z axes (e.g., as shown in Fig. 1 B), it should be understood that the X and Z axes may be rotated (and thus oriented differently from what is shown in Fig. 1 B) by rotating the microfluidic device 102 around the Y axis or rotating the image acquisition device 124. The images collected by the image acquisition device 124 may also be digitally rotated. In addition, the axes may be rotated by physically orientating the microfluidic device 102 differently with respect to the imaging system 104. As such, the orientations of the X, Y, and Z axes described and illustrated herein should be understood as being exemplary only and are not intended to be limiting.

[0060] Referring back to Fig. 1A, the imaging system 104 is used to image the sample 116 as it flows through the microfluidic device 102. In particular, the imaging system 104 acquires multiple images of each cell as the sample 116 flows through the microfluidic channel 1 12 and the cells(s) contained therein rotate. Any suitable imaging system 104 including, but not limited to, a fluorescence, transmission, phase contrast, bright-field, and dark-field imaging system, may be used. It should be understood that the imaging system 104 may combine multiple imaging modalities or techniques. In particular, several imaging modalities or techniques may be used simultaneously.

[0061] In the illustrated embodiment, the imaging system 104 comprises a light source 120 which is configured to generate a light beam 122 that illuminates the sample 116 (see Fig. 1 A). For example, the light source 120 may be a laser configured to generate a laser beam. In some applications, the light source 120 may be used to produce fluorescence. The light beam 122 is output by the light source 120 towards an exposed surface (e.g., a bottom surface 123) of the substrate 1 10, at an angle (not shown) that is substantiallyorthogonal (i.e. at about ninety (90) degrees) to the surface 123. The light source 120 scatters light off the sample 116, with the light scattered by the sample 116 being characteristic to the cell(s) contained in the sample 116 and to the cell components. Although a single light source 120 is shown in Fig. 1 A, it should be understood that this is for illustration purposes only and that the imaging system 104 may comprise multiple light sources as in 120. In addition, the light source(s) 120 may be actuated (i.e. turned on and off) to switch between modes.

[0062] In some embodiments, the cell(s) contained in the sample 116 are labelled with at least one marker, such that light is absorbed and emitted in a given band of wavelengths due to the presence of the at least one marker. This may in turn allow the imaging system 104 to determine the location of at least one marker relative to the cell as well as determine the number of markers associated with the sample 116. At least one marker may be attached at any suitable location of the cell(s), such as on the surface thereof. At least one marker may be attached to the cell(s) prior to imaging of the sample 116 or may be attached to and detached from the cell(s) dynamically during imaging. Any suitable markers including, but not limited to, one or more fluorescent markers, one or more antibody-labelled nanoparticles (e.g., having a diameter of about tens to hundreds of nanometers), and / or one or more immunoplasmonic biomarkers (attached using immunochemistry techniques), may be used. In some embodiments, the antibody-labelled nanoparticles used as markers have a spherical or oblong shape and are made of any suitable material including, but not limited to, polymer, gold, a gold / silver allow, or silver. The nanoparticle shape can contribute to spectral properties including color and polarization. The nanoparticles may contain one or more fluorophores or be intrinsically fluorescent. The number of nanoparticles bound to the cell(s) may be subsequently used for identification (e.g., barcoding).

[0063] The imaging system 104 further comprises an image acquisition device 124 which is configured to capture a sequence of images of the illuminated sample 116 as the latter traverses the microfluidic device 102. In particular, the image acquisition device 124 may be configured to acquire multiple 2D images of each cell in the sample 1 16 as the cell rotates, with the image acquisition device 124 remaining stationary. The sequence of 2D images thus describes the rotation of the cell and provides a 360 degree view of the cell’s structure. The image acquisition device 124 may comprise any suitable device including, but not limited to, a 2D camera or a microscope. The image acquisition device124 may be triggered to capture the images by the entry of a cell into the device’s frame. The image acquisition device 124 may be configured to capture one or more focal planes. The images may be stored in any suitable storage media including, but not limited to, a memory and / or a database (not shown) associated with the imaging system 104.

[0064] The processing unit 106 is configured to process the images acquired by the image acquisition device 124. In particular, the processing unit 106 is configured to algorithmically recombine the images to generate a 3D reconstruction of the sample 116, and to perform an analysis on the 3D reconstruction. Fig. 3 illustrates an example of the processing unit 106, in accordance with one embodiment, where the example processing unit 106 comprises an input unit 302, a tracking unit 304, a 3D reconstruction unit 306, an analysis unit 308, and an output unit 310.

[0065] The input unit 302 is configured to obtain the 2D images acquired by the image acquisition device 124. These images may be obtained directly from the image acquisition device 124 or retrieved from the storage media associated with the imaging system 104. The images are provided to the tracking unit 304, which is configured to apply a tracking algorithm to the 2D image data to identify points or extended shapes (also referred to herein as “features”) in each 2D image and associated features from one 2D image to the next for creating tracking segments for each cell. The features may include any optically distinct feature on the cell, i.e. any point (or extended shape) in the 2D images with optical properties that make the point (or extended shape) distinct from its surroundings (i.e. from other points or extended shapes in the 2D image), as illustrated for example in Fig. 4B described further below. Examples of features include, but are not limited to, cellular membranes, cellular organelles such as the nucleus, and non-symmetrical extensions of the cell. In particular, the tracking unit 304 is configured to identify, from the multiple 2D images, common features in at least two 2D images of a same cell. When the cell(s) contained in the sample 116 are labelled with markers (e.g. nanoparticles or fluorescent biomarkers), the tracking unit 304 may be configured to identify the common features by detecting (e.g., visualizing) the markers in the 2D images. The markers may be one or more fluorescent or optically distinct markers, or one or more antibody-labelled nanoparticles including, but not limited to, protein targets, structures such as the cellular cytoskeleton, DNA or specific regions of the DNA, RNA. The features identified by the tracking unit 304 may therefore include, but are not limited to, fluorescent labelling such as antibodies, specific dyes (e.g., attached to the outside of a cell or brought within thecell), DNA intercalating dyes, and fluorescent proteins expressed by the cell. The features may also include, but are not limited to, autofluorescence from the cell or patterned structures due to photobleaching. The identified features may be used to determine a number of characteristics of the cell(s) and / or the sample 116 including, but not limited to, the cell type, cell state, disease associated with the cell, and therapeutics added to the sample 116.

[0066] The tracking unit 304 is also configured to apply a position prediction algorithm to determine the relative positions of the common features identified in the 2D images. For this purpose, the tracking unit 304 may be configured to integrate the features into a model that computes the inverse of how 3D physical movement of the sample 1 16 (i.e. rotation and translation) manifests in the recorded image data. In doing so, the tracking unit 304 identifies the most likely 3D positions that would give rise to a recorded set of images. For each cell being imaged, the tracking unit 304 may also compute the angles between the relative positions of the common features and the position (referred to herein as a “reference position”) of a common point of the cell being imaged. The tracking unit 304 further determines, based on the tracking segments and on the determined relative positions (and angles), the cell rotation rate (or velocity) over time. The tracking unit 304 then assigns, based on the combined tracked features and cell velocity, a set of parameters to each of the 2D images acquired by the image acquisition device 124. The parameters include, but are not limited to, the relative angle and coordinates of each cell depicted in the 2D image.

[0067] The 3D reconstruction unit 306 is then configured to generate, based on data received from the tracking unit 304, a 3D reconstruction of each cell contained in the sample 1 16. The 3D reconstruction is representative of the 3D locations of the features present in each cell and may serve as an estimate of the cell surface. The 3D reconstruction unit 306 is configured to use the coordinates associated with each 2D image to map features of the image into the 3D space. The features of the 2D image which are mapped into 3D space comprise the features tracked by the tracking unit 304 (as described herein above) and may also comprise one or more features which were not tracked by the tracking unit 304 (e.g., inhomogeneities in the optical properties of the cell) but nonetheless observable within the 2D image. The 3D reconstruction unit 306 may use any suitable technique to generate the 3D reconstruction. In one embodiment, Poisson’s surface reconstruction may be used to generate a surface mesh of the cell. The 3Dreconstruction unit 306 may also use any suitable technique to generate a 3D reconstruction in which the cell surface is color-coded according to the cell’s curvature. In another embodiment, the 3D reconstruction may be performed using a Radon transformation or projection-slice methods, whereby the 2D projected images are reconstructed. Individual points within the cell (e.g., structures of DNA markers or the distribution of fluorescent markers) may also be counted or reconstructed.

[0068] In some embodiments, the relative 3D positions of individual points may be calculated by extracting motion speeds from the 2D image. In this case, the 3D reconstruction is not made from the 2D images but rather composed of the extracted positions. In particular, the reconstruction may be performed by using the physical property of a rotating object, i.e. that all parts of that object must have the same angular velocity and move in the 2D image as a function of their coordinates within the object. In other words, an object of a larger radius will move faster in 3D space but still have the same angular velocity. By tracking objects (i.e. cells) over time in two (2) or three (3) dimensions, detailed 3D spatial information can be extracted from the measured velocities. Using the Radon transformation, the computed angular rotation may also be used to directly reconstruct a 3D visualization even from objects that have not been tracked but which appear in the captured 2D images. An example could be a low-contrast part of a 2D image. Other embodiments may apply, for example, modifying the pointspread function of the microscope to encode extractable 3D information into 2D images.

[0069] The 3D reconstruction is then sent by the 3D reconstruction unit 306 to the analysis unit 308, which is configured to analyze and quantify the 3D data associated with the 3D reconstruction for various applications. The applications include, but are not limited to, determination of cell characteristics and function, diagnosis of health disorders such as cancers, biomarker detection, microorganism detection, cell counting, cell sorting or classification, virus detection, blood cell counts, fine needly biopsy diagnostics, therapy testing, toxicity testing, quantification of molecular densities and distribution of molecular densities inside or outside a cell, and response to stimuli (e.g., chemical or physical such as electroporation). For example, the analysis unit 308 may be configured to quantify the number and distribution of bound objects (e.g. antibody-labelled nanoparticles or other markers attached to the cell(s)), measure the 3D shape of the sample represented by the 3D reconstruction, and measure deviations from the canonical spherical shape. The analysis unit 308 may also be configured to use the 3D positions of features in the 3Dreconstruction to calculate the relative distances of the features, and to determine whether the features are internal or external relative to the imaged cell or to imaged organelles. This may be of particular interest in measuring cellular responses to stimuli. In some embodiments, the analysis unit 308 may also be configured to cause the 3D images and data recorded by the image acquisition device 124 to train a neural network algorithm to infer high-resolution data from individual images or other lower-resolution reconstructions. The outcome of the analysis performed by the analysis unit 308 may be provided to the output unit 310 for transmission to a user in any suitable manner. For example, the output unit 310 may generate an output in any suitable format (e.g., a report in tabular format) for presentation to the user via a suitable output means (e.g., a display on a client device).

[0070] Reference will now be made to Fig. 4A, Fig. 4B, Fig. 4C, 4D, and Fig. 4E, which show results of an optical simulation that illustrates the difference between frames as a cell rotates within the microfluidic device (reference 102 of Fig. 1A).

[0071] Fig. 4A illustrates results 400 obtained by the imaging system (reference 104 of Fig. 1A) when implementing dark-field imaging. In this example, the imaging system 104 captured 200 images (or frames) of a cell (frames 1 , 50, and 150 being shown as 402i, 40250, and 402isofor illustration purposes) at different orientations and the processing unit (reference 106 in Fig. 1 A) combined the images to produce a 3D reconstruction 404 of the cell.

[0072] Fig. 4B illustrates results 410 obtained by the imaging system (reference 104 of Fig. 1A) when implementing bright-field imaging where the illumination source (e.g., the light source 120 in Fig. 1) faces the light collection system (e.g., the image acquisition device 124 of Fig. 1) and contrast is produced through differential absorption and differential refraction through the sample and surrounding environment. In this example, the imaging system 104 captured a given number of images (or frames) of a cell (with frames 412i, 4122, and 412s being shown for illustration purposes) at three (3) different orientations (as shown by the arrows indicating illustrative orientation markers that represent rotation of the cell).

[0073] Fig. 4C illustrates further results 420 of simulations and reconstructions obtained using the system 100, in accordance with another embodiment. A plurality of images (or frames) of a rotating sample (with frames 422i, 4222, 422s, 4224, and 422s being shownfor illustration purposes) are captured at different orientations using the imaging system 104. The processing unit 106 (and more particularly the 3D reconstructions unit 306) is configured to combine the images to produce a 3D reconstruction 424 of the cell.

[0074] Fig. 4D illustrates further results of simulations and reconstructions obtained using the system 100, in accordance with yet another embodiment. Image 430a of Fig. 4D show a single frame with localized positions, and image 430b of Fig. 4D show a single frame with all estimated positions. Image 430c of Fig. 4D shows the estimated X positions over all time points, and image 430d of Fig. 4D illustrates the calculated Z and X positioned based on the measured movement vectors. It should be noted that the axes illustrated in Fig. 4D are not to scale.

[0075] Fig. 4E illustrates results obtained by the system 100 when reconstructing individual points within a cell in 3D. The left image 431 of Fig. 4E shows a single frame snapshot acquired using the imaging system 100, and the left image 436 shows the reconstruction. As noted above, examples of such individual points include, but are not limited to, structures of DNA markers or the distribution of fluorescent markers. These individual points may be reconstructed or counted using the 3D reconstruction unit 306. In the embodiment illustrated in Fig. 4E, multiple points as in 432 provided within the cell 434 are reconstructed (see reconstruction 436) using global sine-wave fitting, such that the position of the points 432 (e.g., along the X and Z axes) can be determined and tracked in real-time as the cell 434 rotates through the microfluidic channel (reference 112 of Fig. 1A) of the microfluidic device (reference 102 in Fig. 1A).

[0076] Fig. 4F illustrates results 440 obtained when reconstructing a 3D shape using nanoparticles attached to cell(s) in the sample being imaged. In this example, a 2D image 442 of a cell moving through the microfluidic channel (reference 1 12 of Fig. 1A) of the microfluidic device (reference 102 in Fig. 1A) is captured by the imaging system (reference 104 in Fig. 1A). Image 444 illustrates the mesh 446 of the cell depicted in the 2D image 442, as reconstructed by the processing unit 106, and more particularly using the 3D reconstruction unit (reference 306 in Fig. 3). Image 448 further shows the cell surface 450 colored based on a characterization of the cell’s curvature, as generated by the 3D reconstruction unit 306.

[0077] Fig. 4G illustrates a fluorescence reconstruction snapshot 450 obtained for yeast flowing through the microfluidic channel (reference 1 12 of Fig. 1A) of the microfluidic device (reference 102 in Fig. 1A). The results illustrated in Fig. 4G are based on fluorescence measurements acquired by the imaging system (reference 104 in Fig. 1A) over time, as two yeast cells 452a, 452b move through and rotate within the microfluidic channel 1 12. Fig. 4G shows the position of two fluorescence markers 454a, 454b respectively attached to the yeast cells 452a, 452b. In particular, the snapshot 450 shows the respective positions of the markers 454a, 454b at a time equal to 0.42 seconds.

[0078] Fig. 5 is a flowchart of an example method 500 for performing 3D imaging flow cytometry. The method 500 is illustratively performed by the processing unit (reference 106) of Fig. 1A. The method 500 comprises, at step 502, receiving a plurality of two- dimensional (2D) images of at least one object (e.g., a cell or a cluster of cells). The plurality of 2D images is acquired (e.g., by the imaging system 104, in the manner described herein above with reference to Fig. 1 A) as the at least one object rotates within at least one channel of a microfluidic device while fluid continuously flows through the at least one channel, the fluid comprising a sample containing the at least one object. The at least one object is caused to rotate as fluid flows through the microfluidic device, in the manner described herein above with reference to Fig. 1A, Fig. 1 B, Fig. 2A, and Fig. 2B. The method 500 further comprises, at step 504, generating a 3D representation of the at least one object based on the plurality of 2D images. The 3D representation may be generated by the 3D reconstruction unit (reference 206) of Fig. 3, in the manner described herein above.

[0079] As described herein above, in one embodiment, generating the 3D representation of the at least one object at step 504 comprises identifying common features in at least two of the plurality of 2D images of the at least one object, determining relative positions of the common features in the at least two of the plurality of 2D images, determining angles between the relative positions ofthe common features and a reference position associated with a common point in the at least two of the plurality of 2D images, and generating a 3D mesh based on the angles as determined, the 3D mesh indicative of 3D locations of the common features in the at least one object.

[0080] As described herein above, in one embodiment, identifying the common features comprises visualizing, on the at least two ofthe plurality of 2D images, at least one marker(e.g., at least one nanoparticle or at least one fluorescent biomarker) attached to the at least one object prior to imaging being performed by the imaging system.

[0081] Referring now to Fig. 6, the system 100 of Fig. 1A and / or the method 500 of Fig.5, respectively, may be implemented using a computing device 600. For simplicity only one computing device 600 is shown but the system 100 and / or the method 500 may involve more computing devices 600 which may be the same or different types of devices. The computing device 600 comprises a processing unit 602 and a memory 604 which has stored therein computer-executable instructions 606. The processing unit 602 may comprise any suitable devices configured to implement the system 100 and / or the method 500 such that instructions 606, when executed by the computing device 600 or other programmable apparatus, may cause the functions / acts / steps of the system 100 and / or the method 500 described herein to be executed. The processing unit 602 may comprise, for example, any type of general-purpose microprocessor or microcontroller, a digital signal processing (DSP) processor, a central processing unit (CPU), an integrated circuit, a field programmable gate array (FPGA), a reconfigurable processor, other suitably programmed or programmable logic circuits, or any combination thereof.

[0082] The memory 604 may comprise any suitable known or other machine-readable storage medium. The memory 604 may comprise non-transitory computer readable storage medium, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. The memory 604 may include a suitable combination of any type of computer memory that is located either internally or externally to device, for example random-access memory (RAM), read-only memory (ROM), compact disc readonly memory (CDROM), electro-optical memory, magneto-optical memory, erasable programmable read-only memory (EPROM), and electrically-erasable programmable read-only memory (EEPROM), Ferroelectric RAM (FRAM) or the like. Memory 604 may comprise any storage means (e.g., devices) suitable for retrievably storing machine- readable instructions 606 executable by processing unit 602.

[0083] In some embodiments, the systems and methods described herein may be used to quantify the number of nanoparticles bound to a cell and to reconstruct features from cell images in 3D.

[0084] In some embodiments, the systems and methods described herein may allow to continuously process and image cells in a flow of a microfluidic device, without the need for a cell capture or immobilization step to be performed. This may in turn lead to a significant reduction in the number of components, thus simplifying the resulting imaging device. The systems and methods described herein may also enable the replacement of most of the single-use plastic used in current devices. The systems and methods described herein may also allow to increase the signal-to-noise ratio, enhance image resolution, and achieve improved throughput. Furthermore, using the systems and methods described herein may allow to perform rapid cytopathological analysis in solution without destroying the original samples. Digital differentiation and sorting of cells may be provided for subsequent diagnostics. The potential for a low-cost and portable device using the setup proposed herein may also enable outpatient diagnostics. In addition, the systems and methods described herein may prove compatible with multiple imaging techniques and downstream processing (e.g., fluorescence-activated cell sorting, or DNA and RNA quantification and sequencing) devices. The systems and methods described herein may also prove compatible with upstream processing or processing from other instruments. The proposed setup is also live cell or fixed cell compatible.

[0085] The above description is meant to be exemplary only, and one skilled in the art will recognize that changes may be made to the embodiments described without departing from the scope of the invention disclosed. Still other modifications which fall within the scope of the present invention will be apparent to those skilled in the art, in light of a review of this disclosure.

[0086] Various aspects of the systems and methods described herein may be used alone, in combination, or in a variety of arrangements not specifically discussed in the embodiments described in the foregoing and is therefore not limited in its application to the details and arrangement of components set forth in the foregoing description or illustrated in the drawings. For example, aspects described in one embodiment may be combined in any manner with aspects described in other embodiments. Although particular embodiments have been shown and described, it will be apparent to those skilled in the art that changes and modifications may be made without departing from this invention in its broader aspects. The scope of the following claims should not be limited by the embodiments set forth in the examples, but should be given the broadest reasonable interpretation consistent with the description as a whole.

Claims

WHAT IS CLAIMED IS:1 . A system for three-dimensional (3D) imaging flow cytometry, the system comprising: a microfluidic device having disposed therein one or more channels each extending along a central axis, the one or more channels configured to continuously flow therethrough a fluid comprising a sample containing at least one object, the at least one object caused to be displaced, as the fluid flows through the one or more channels, away from the central axis and towards a target internal surface of the one or more channels to cause the at least one object to rotate ; an imaging system optically coupled to the microfluidic device, the imaging system configured to acquire a plurality of two-dimensional (2D) images of the at least one object as the at least one object rotates; and a processing unit coupled to the imaging system, the processing unit configured to obtain the plurality of 2D images and to generate a 3D representation of the at least one object based on the plurality of 2D images.

2. The system of claim 1 , wherein a flow speed of the fluid flowing through the one or more channels is set to a limited flow speed, further wherein the limited flow speed and a difference in density between the at least one object and the fluid cause the at least one object to be displaced towards the target internal surface and to rotate at a given velocity.

3. The system of claim 1 , wherein the at least one object is displaced away from the central axis and towards the target internal surface of the one or more channels by applying a centrifugal force on the at least one object.

4. The system of claim 1 , wherein the at least one object is displaced away from the central axis and towards the target interface within the one or more channels by using at least one optical tweezer to apply a force on the at least one object.

5. The system of claim 1 , wherein the at least one object has at least one magnetic nanoparticle attached thereto, further wherein the at least one object is displaced away from the central axis and towards the target internal surface by applying a magnetic force on the at least one magnetic nanoparticle.

6. The system of claim 1 , wherein the one or more channels comprises the target internal surface and at least one additional internal surface, the target internal surface and the at least one additional internal surface having different surface properties, further wherein the different surface properties create a non-uniform flow through the one or more channels that brings the at least one object to the position away from the central axis and towards the target internal surface.

7. The system of any one of claims 1 to 6, wherein the target internal surface is one of at least part of an upper surface and at least part of a lower surface of the one or more channels.

8. The system of any one of claims 1 to 6, wherein the one or more channels comprise a first channel having a first fluid flowing therethrough and a second channel disposed below the first channel and having a second fluid flowing therethrough, the first fluid having a first viscosity and the second fluid having a second viscosity lower than the first viscosity, further wherein the target internal surface is an interface between the first fluid and the second fluid.

9. The system of any one of claims 1 to 8, wherein the imaging system comprises at least one light source and a stationary image acquisition device, the at least one light source configured to illuminate the sample with light as the fluid flows through the one or more channels, and the image acquisition device configured to capture light scattered by the sample in response to illumination thereof for acquiring the plurality of 2D images of the at least one object as the fluid flows through the one or more channels and as the at least one object rotates.

10. The system of claim 9, wherein the imaging system uses at least one of a fluorescence, a transmission, a phase contrast, a bright-field, and a dark-field imaging modality.11 . The system of any one of claims 1 to 10, wherein the processing unit is configured to: identify common features in at least two of the plurality of 2D images of the at least one object; determine relative positions of the common features in the at least two of the plurality of 2D images;determine angles between the relative positions of the common features and a reference position associated with a common point in the at least two of the plurality of 2D images; and generate the 3D representation of the at least one object comprising generating a 3D mesh based on the angles as determined, the 3D mesh indicative of 3D locations of the common features in the at least one object.

12. The system of claim 11 , wherein the at least one object has at least one marker attached thereto, further wherein the processing unit is configured to identify the common features by detecting the at least one marker in the at least two of the plurality of 2D images.

13. The system of any one of claims 1 to 12, wherein the at least one object is one of a single cell and a cluster of cells, further wherein the microfluidic device is configured to cause each cell to be controllably rotated within the one or more channels as the fluid flows therethrough, and the imaging system is configured to acquire the plurality of 2D images representative of full rotations of each cell.

14. The system of any one of claims 1 to 13, wherein the at least one object is an object encapsulated in a transparent gel to modify an effective size and / or density thereof.

15. The system of any one of claims 1 to 14, wherein the at least one object has a size ranging from about 5 microns to about 100 microns.

16. A method for three-dimensional (3D) imaging flow cytometry, the method comprising: receiving, at a computing device, a plurality of two-dimensional (2D) images of at least one object contained in a sample, the plurality of 2D images acquired as the at least one object rotates within one or more channels of a microfluidic device while fluid comprising the sample continuously flows through the one or more channels, the at least one object caused to be displaced within the microfluidic device, as the fluid flows through the one or more channels, away from a central axis of the one or more channels and towards a target internal surface of the one or more channels to cause the at least one object to rotate; andgenerating, at the computing device, a 3D representation of the at least one object based on the plurality of 2D images.

17. The method of claim 16, wherein generating the 3D representation of the at least one object comprises: identifying common features in at least two of the plurality of 2D images of the at least one object; determining relative positions of the common features in the at least two of the plurality of 2D images; determining angles between the relative positions of the common features and a reference position associated with a common point in the at least two of the plurality of 2D images; and generating a 3D mesh based on the angles as determined, the 3D mesh indicative of 3D locations of the common features in the at least one object.

18. The method of claim 17, wherein identifying the common features comprises detecting, in the at least two of the plurality of 2D images, at least one marker attached to the at least one object.

19. A system for three-dimensional (3D) imaging flow cytometry, the system comprising: a processing unit; and a non-transitory memory communicatively coupled to the processing unit and comprising computer-readable program instructions executable by the processing unit for: receiving a plurality of two-dimensional (2D) images of at least one object contained in a sample, the plurality of 2D images acquired as the at least one object rotates within one or more channels of a microfluidic device while fluid comprising the sample continuously flows through the one or more channels, the at least one object caused to be displaced within the microfluidic device, as the fluid flows through the one or more channels, away from a central axis of the oneor more channels and towards a target internal surface of the one or more channels to cause the at least one object to rotate; and generating a 3D representation of the at least one object based on the plurality of 2D images.

Images