Hyperspectral quantitative imaging cytometry system

Abstract

The present invention relates to the hyperspectral detection of luminescence, in particular to the luminescence detection of solid phase samples excited using a radiation source.

Description

[0001] This is a divisional application, and its parent application is an application with an application date of May 29, 2020, an application number of 202080040450.7, and an invention title of "Hyperspectral Quantitative Imaging Cytometry System". Technical Field

[0002] The present invention relates to the hyperspectral detection of luminescence, and more particularly, to the luminescence detection of solid-phase samples excited by a radiation source. Background Art

[0003] The function of biological tissue is the result of the coordinated action of its cellular components. Each of these cells exhibits a specific phenotype resulting from its interaction with the tissue environment, and any dysregulation of these mechanisms can lead to diseases such as cancer. Therefore, the ability to analyze single-cell characteristics in a spatial context is crucial for understanding how tissues function in normal and diseased conditions and for helping to develop effective treatment methods.

[0004] Through traditional histological and morphological assessments, the outcome of a disease affecting a particular tissue may not always be apparent. Some structures and components of the tissue may be morphologically similar but exhibit significant differences in molecular composition due to disease-related dysregulation. In such cases, multiple and more specific staining methods (such as those provided by immunological methods) need to be employed to identify these differences. An example in this regard is the identification of immune cells within the tissue under study. Their presence may be due to a disease condition that affects the tissue and causes these cells to aggregate in the damaged area, or, on the other hand, their presence may be the primary cause of the disease affecting the tissue.

[0005] In both cases, it is necessary to correctly identify the lineage and functional state of these cells, especially in the case of small lymphocytes, where even the morphological assessment by experts is insufficient to reveal the nature, origin, and heterogeneity of these cells. Only multi-parameter immunophenotyping methods can enable the correct assessment of the characteristics and heterogeneity of cell infiltration. In addition, as important as being able to obtain multi-parameter information on tissue components is the association of different phenotypic recognition features with possible anatomical changes observed in the tissue. These changes can be identified by direct observation by experts in the field.

[0006] When applied to cell suspensions, the use of flow cytometry for multi-parameter analysis of single cells has proven to be fundamental for revealing cell phenotypic heterogeneity in normal and disease conditions. Modern flow cytometers can analyze dozens of parameters simultaneously, and the development of multi-spectral systems and the emergence of mass cytometry promise to drive an increase in these numbers in the near future. However, these techniques are unable to act on tissue samples without disturbing the native structure of the tissue and cannot study the composition of the extracellular environment of the tissue. On the other hand, although the standard choice for cell morphology visualization and spatial localization, microscope instruments are unable to perform quantitative and objective cell component analysis on a statistically significant number of cells and lack the standardization capabilities present in other methods such as flow cytometry.

[0007] Others have attempted to address some of these problems by adapting the configuration of flow cytometers to use laser scanning of samples fixed on microscope slides in order to excite fluorescent molecules on the sample and build a representation of the molecules present on the tissue pixel by pixel (laser scanning cytometry). This concept was later adapted to use mass spectrometry instead of fluorescence detection to increase the number of molecules analyzed simultaneously (imaging mass cytometry). However, both of these methods are rather slow due to the need to study biological tissue one pixel at a time.

[0008] Therefore, there is a need for a system that provides quantitative data on the size and expression of markers of cells and / or tissues fixed in a solid-phase sample support.

[0009] The most commonly used method at present is to perform multiple single-parameter studies using a conventional microscope. This method preserves the overall structure of the sample (usually biological solid tissue), but is not quantitative and lacks the multi-parameter dimension required for complex studies.

[0010] Alternatively, multi-parameter flow cytometry can be employed to obtain multi-parameter information on biological tissue, but at the cost of loss of spatial information due to the tissue dissociation required to obtain a single-cell suspension.

[0011] By adapting the configuration of a flow cytometer to scan samples fixed on microscope slides, laser scanning cytometry (LSC) has been developed. It uses a laser to excite fluorescent molecules on the sample and builds a representation of the molecules present on the tissue pixel by pixel. Therefore, although a snapshot system where all "colors" are sampled simultaneously, this is a very slow method. In addition, the potential of this technique for multiplexing remains very limited, only allowing the simultaneous study of 3 to 4 parameters.

[0012] In a similar manner, others have adapted mass cytometry to perform studies on solid tissues (Imaging Mass Cytometry - IMC). Different from LSC, IMC uses metal conjugates instead of fluorescent or chromogenic conjugates to reveal tissue components and also has a higher multiplexing potential. However, like LSC, since the samples are "imaged" on a single-pixel basis, this method also has the same drawback of being a very slow technique.

[0013] Alternatively, multispectral and hyperspectral capabilities have been applied to microscopy-based systems to increase the number of markers that can be analyzed simultaneously. Using two-dimensional sensors to sample data, these systems can sample multiple spatial positions simultaneously; however, these systems are designed to provide visual information rather than reproducible and quantifiable data, and they are designed to primarily provide high-resolution information on small amounts of biological material rather than analyzing large areas of tissue. Summary of the Invention

[0014] The present invention provides a solution to the above problems through the hyperspectral quantitative imaging cytometry system according to claim 1 and the method according to claim 15. Preferred embodiments of the present invention are defined in the dependent claims.

[0015] The present invention provides a system and method to obtain quantitative data on the size and expression of markers from cells and tissues of a biological sample fixed on a solid-phase sample support in a fast manner by focusing on the overall tissue structure with cellular resolution rather than subcellular resolution.

[0016] In a first aspect of the invention, the present invention provides a hyperspectral quantitative imaging cytometry system, comprising:

[0017] An observation area, the observation area including a sample holding part configured to hold one or more solid-phase samples,

[0018] At least one radiation source, the at least one radiation source being configured to irradiate the observation area,

[0019] A collection element, the collection element being configured to collect the radiation emitted or reflected by the sample irradiated by the at least one radiation source,

[0020] A multi-channel filtering element, the multi-channel filtering element being configured to selectively filter the wavelengths of the radiation collected by the collection element, and

[0021] An image sensor, the image sensor being configured to receive the filtered radiation and generate an image as a two-dimensional map of the sample, and the image sensor including a two-dimensional array of radiation detection elements.

[0022] Solid-phase samples are typically disposed on a solid-phase sample support. In one embodiment, the sample holding portion is configured to hold at least one solid-phase sample support, each support being adapted to receive a fixed sample (preferably, a biological sample). The support can have different materials (preferably, crystalline and optically transparent materials (e.g., glass or plastic)), and have different shapes and sizes (preferably, a rectangular shape (e.g., a microscope slide)).

[0023] The term "constituent" will be used to denote any molecule that occurs naturally in a cell or tissue. The terms "label" and "molecular label" will be used to denote any substance that is added to a sample to indicate the presence of a specific constituent that occurs naturally in the sample. The term "marker" will be used to define any constituent on a sample that naturally emits radiation or emits radiation due to the presence of a molecular label added to the sample; the marker is used to define the nature of a cell or tissue. The term "spectral signature" is used to denote the unique emission spectrum of a structure or pixel that results from the unique binding of a marker present in that structure or pixel. The term "list mode file" is used to denote a data file structure in which information about different elements of interest (such as biological structures) is stored, and each of these elements is represented by a line on a multi-line list.

[0024] At least one radiation source is provided to irradiate an observation area. Thus, when a sample is present in the observation area, the radiation interacts with the constituents of the sample and, if there are molecular labels used in conjunction with the sample, with the molecular labels. Through this interaction, the sample emits radiation due to any process such as scattering, fluorescence, phosphorescence, chemiluminescence, or selective absorption / transmission. The radiation emitted by the sample is collected using a collection element, and the radiation passes through a multi-channel filtering element and reaches an image sensor. The multi-channel filtering element should be understood as a filter whose spectral characteristics vary along the filter, thereby providing position-dependent filtering of the incident radiation.

[0025] The image sensor includes a two-dimensional array of radiation detection elements. The radiation detection elements receive the radiation and provide an output related to the radiation received at each radiation detection element. Thus, an image is produced that is a two-dimensional map of the sample. In one embodiment, the image sensor is a charge-coupled device (CCD), a complementary metal oxide semiconductor (CMOS), or an electron multiplying charge-coupled device (EMCCD).

[0026] Preferably, the multi-channel filtering element is disposed between and spaced apart from the collection element and the image sensor. In a particular embodiment, the system further includes at least one lens configured to project an image captured or collected by the collection element onto a plane; the multi-channel filtering element is located at the plane of the image projection such that the image is projected onto the filtering element; the system includes at least one additional lens configured to capture the intermediate image filtered by the multi-channel filtering element and project it onto the image sensor. In other words, the filtering element is located at (or slightly offset from) the exact plane where the first lens projects the intermediate image formed between the collection element and the image sensor. Thus, two lenses or, similarly, two sets of lenses can be used to form and collect the intermediate image.

[0027] In one embodiment, at least one radiation source is configured to emit radiation having a wavelength in the ultraviolet (UV), visible (VIS), or near-infrared (NIR) range, preferably in the range of 200 nm to 1200 nm, more preferably in the range of 350 nm to 950 nm. The system may include one or more radiation sources. In one embodiment, the system includes a plurality of radiation sources, each configured to emit radiation at a different wavelength interval, e.g., wavelengths included in the range of 525 nm to 625 nm.

[0028] In one embodiment, the filtering element is arranged to be movable between at least two positions, wherein each position of the filtering element selectively filters the wavelength of the radiation reaching each radiation detection element of the image sensor. In a preferred embodiment, the filtering element is arranged to be movable to a plurality of positions. Preferably, the movement of the filtering element is parallel to one of the spatial dimensions of the field of view (FOV) of the collection element.

[0029] In one embodiment, the filtering element is a continuous linear variable filter or a semi-continuous linear variable filter. Preferably, the filtering element is configured to filter radiation wavelengths between 200 nm and 1200 nm, more preferably between 350 nm and 950 nm.

[0030] In one embodiment, the observation region is interposed between at least one radiation source and the collection element such that the radiation from the radiation source passes through the observation region before being collected by the collection element, i.e., according to a trans-illumination configuration.

[0031] In one embodiment, at least one radiation source is disposed on the same side of the observation region as the collection element such that the observation region reflects the radiation before it is collected by the collection element, i.e., according to an epi-illumination configuration.

[0032] In one embodiment, at least one radiation source and a collection element are arranged such that a radiation beam from the radiation source is oriented at a non-zero angle with respect to the optical axis of the collection element, i.e., according to a dark-field configuration.

[0033] In one embodiment, at least one radiation source and a collection element (105) are arranged such that a radiation beam from the radiation source is oriented along the optical axis of the collection element, i.e., according to a bright-field configuration.

[0034] In a preferred embodiment, the system includes a plurality of radiation sources, each radiation source providing radiation of a given wavelength (A), and arranged according to different radiation modes (α) selected from bright-field epi-illumination, dark-field epi-illumination, bright-field transmission, and dark-field transmission.

[0035] In one embodiment, the system includes a processor. In a particular embodiment, the processor is part of a computer (e.g., a personal computer).

[0036] In one embodiment, the processor is configured to perform the following steps:

[0037] - Receive a plurality of wavelength-encoded two-dimensional maps of a sample, the plurality of wavelength-encoded two-dimensional maps being associated with a plurality of positions of a filtering element, wherein the wavelength-encoded two-dimensional map is an image generated by an image sensor based on the radiation it receives for the filtering element positions;

[0038] - Generate a plurality of monochromatic two-dimensional maps of the sample by combining portions of the wavelength-encoded two-dimensional maps of the sample corresponding to a specific wavelength;

[0039] - Construct a spectral cube containing the plurality of monochromatic two-dimensional maps;

[0040] - Identify the sample structure on the spectral cube and obtain its spectral features;

[0041] - Compare the obtained spectral features with a database of spectral features of known structures and / or decompose the spectral features and obtain an estimate of the abundance of each marker in each identified sample structure, where a marker is any component that naturally emits radiation on the sample or emits radiation due to the presence of a molecular marker added to the sample, and thus the marker is used to define the nature of the cell or tissue.

[0042] In an alternative embodiment, the system includes a processor configured to:

[0043] - Receive a plurality of wavelength-encoded two-dimensional maps of a sample, the plurality of wavelength-encoded two-dimensional maps being associated with a plurality of positions of a filtering element, wherein the wavelength-encoded two-dimensional map is an image generated by an image sensor based on the radiation it receives for the filtering element positions;

[0044] - generating a plurality of monochromatic two-dimensional maps of the sample corresponding to a specific wavelength (λ) by estimating the measurement signal at a specific wavelength (λ) from all wavelength-encoded two-dimensional maps via a multivariate interpolation process for each pixel of the monochromatic two-dimensional map;

[0045] -Build a spectral cube containing multiple monochromatic 2D maps;

[0046] - Identify sample structures on the spectral cube and obtain their spectral signatures;

[0047] - Comparing the obtained spectral signatures with a database of spectral signatures of known structures and / or decomposing the spectral signatures and obtaining an estimate of the abundance of each marker in each identified sample structure, a marker being any component of the sample that emits radiation naturally or due to the presence of a molecular marker added to the sample, and thus a marker is used to define the nature of a cell or tissue.

[0048] In one embodiment, the processor is configured to obtain the size and / or shape of the sample structure.

[0049] In one embodiment, the processor is configured to perform any of the aforementioned steps for multiple radiation wavelengths (Λ) and / or multiple radiation patterns (α), wherein each monochromatic two-dimensional map corresponds to radiation emitted at a specific wavelength (λ) when the sample is irradiated with a given radiation wavelength (Λ) and a given radiation pattern (α), and wherein the step of constructing a spectral cube is performed by combining the multiple monochromatic two-dimensional maps.

[0050] In one embodiment, the processor is configured to control the sequential recording of a plurality of wavelength-encoded two-dimensional maps of the sample that match the sequential displacement of the filter element. In another embodiment, the system comprises a second processor configured to control the sequential recording of a plurality of wavelength-encoded two-dimensional maps of the sample that match the sequential displacement of the filter element. Preferably, the filter element is a continuous linear variable filter or a semi-continuous linear variable filter, more preferably, a continuous linear variable filter.

[0051] In one embodiment, the system includes a memory for data storage. In a specific embodiment, the memory for data storage is a non-volatile computer memory such as a hard drive, EEPROM memory, or an optical disk.

[0052] In one embodiment, at least one radiation source is a laser, a light emitting diode or a lamp.The radiation source may be configured to provide monochromatic or broadband radiation.

[0053] In one embodiment, the system comprises a band pass filter interposed between the at least one radiation source and the observation area. Advantageously, the band pass filter enables selection of specific wavelengths of radiation emitted by the broadband radiation source.

[0054] In one embodiment, the collection element comprises a lens or a combination of lenses configured to simultaneously capture radiation from a plurality of spatial positions in the observation region.

[0055] In one embodiment, the magnification factor value of the collection element is less than 20, preferably less than 10, more preferably less than 2.

[0056] In one embodiment, the numerical aperture value of the collection element is greater than 0.25, preferably equal to or greater than 0.5.

[0057] In a second aspect of the invention, the present invention provides a method for obtaining data from a solid-phase sample using a hyperspectral quantitative imaging cytometry system according to any embodiment of the first aspect of the invention, the method comprising the following steps:

[0058] a) providing a sample;

[0059] b) irradiating the sample with radiation that interacts with the sample such that the sample emits radiation;

[0060] c) capturing the emitted radiation using a collection element;

[0061] d) filtering the emitted radiation using a multi-channel filtering element;

[0062] e) sequentially recording a plurality of wavelength-encoded two-dimensional maps of the sample that match the sequential displacement of the filtering element, wherein each position of the filtering element selectively filters the wavelength of the radiation reaching each radiation detection element of the image sensor, and wherein the image sensor generates a wavelength-encoded two-dimensional map based on the radiation it receives for each position of the filtering element;

[0063] f) generating a plurality of monochromatic two-dimensional maps of the sample by combining portions of the wavelength-encoded two-dimensional maps of the sample corresponding to a specific wavelength;

[0064] g) constructing a spectral cube containing a plurality of monochromatic two-dimensional maps;

[0065] h) identifying sample structures on the spectral cube and obtaining their spectral characteristics;

[0066] i) comparing the obtained spectral characteristics with a database of spectral characteristics of known structures and / or decomposing the spectral characteristics and obtaining an estimate of the abundance of each marker in each identified sample structure of the sample.

[0067] According to the method of the present invention, a sample located in an observation area is irradiated with radiation emitted by one or more radiation sources. The components of the sample can selectively absorb radiation of specific wavelengths and generally emit radiation at different wavelengths. Usually, the internal components of the sample may lack sufficient contrast for direct study. In such cases, specific classes of markers can be added to the sample to provide contrast and enable the detection of specific components. These components can be DNA, proteins, lipids, carbohydrates, etc., and multiple markers can be used to study multiple components simultaneously. Some markers can selectively absorb radiation of specific wavelengths and emit radiation of wavelengths complementary to the absorbed radiation when irradiated with a broadband radiation source (chromogenic markers). Other markers can emit radiation with a wavelength spectrum higher than the absorbed radiation when irradiated with high-energy radiation of a specific wavelength (fluorescent markers). The fluorescent or chromogenic markers can be fluorescent dyes or chromogens that have a natural affinity for specific molecules, or can be a combination of an affinity molecule (e.g., an antibody, a DNA reporter molecule, or other affinity molecules known in the literature) and a reporter molecule (e.g., a chromogen or a fluorescent dye).

[0068] The radiation emitted by the sample is collected by a collection element. After being collected by the collection element and before reaching the image sensor, the emitted radiation is directed through a multi-channel filtering element. Instead of a filtering element, multiple two-dimensional wavelength-encoded maps of the sample are sequentially recorded, where each position of the filtering element selectively filters the wavelength of the radiation reaching each radiation detection element (or group of radiation elements) of the image sensor. Thus, for each position of the filtering element, the image sensor generates a two-dimensional wavelength-encoded map based on the radiation it receives. In one embodiment, the filtering element moves parallel to one of the spatial dimensions of the field of view of the collection element, and the number of steps required for the complete displacement of the filtering element through the entire field of view defines the number of two-dimensional wavelength-encoded maps taken.

[0069] Multiple monochromatic two-dimensional maps of the sample are constructed from the multiple two-dimensional wavelength-encoded maps. In one embodiment, a monochromatic two-dimensional map of the sample is constructed by combining the portions of the two-dimensional wavelength-encoded maps corresponding to a specific wavelength. Thus, multiple monochromatic two-dimensional maps of the sample are obtained, where each of these monochromatic two-dimensional maps has information about the radiation emitted at a given wavelength when the sample is irradiated with a given radiation source. The set of different monochromatic two-dimensional maps obtained constitutes a spectral cube.

[0070] The image obtained using the image sensor is formed by multiple pixels, each pixel corresponding to the output of a radiation detection element. For each pixel, the method provides a discontinuous emission spectrum, where the spectral points corresponding to specific wavelengths are present in the monochromatic two-dimensional map associated with that wavelength.

[0071] Each discontinuous emission spectrum is typically formed by several overlapping pure emission spectra, where each pure emission spectrum corresponds to a specific marker. The degree of overlap or mixing of the pure emission spectra depends on the spatial distribution of the markers in the sample.

[0072] After obtaining the spectral cube, the sample structures present in the sample are identified based on the acquired spectral and spatial information, and specific spectral features are obtained for each identified structure. Thus, a "list mode" file is obtained, in which information about the spectral features and spatial localization is stored for each identified structure.

[0073] Finally, the spectral features of each structure are compared with a set of known spectral features and / or the spectral features of each structure are decomposed (unmixed), and an estimate of the abundance of each marker for that structure is obtained.

[0074] That is, in the present embodiment, a monochromatic two-dimensional map is generated by combining the portions corresponding to specific wavelengths of the wavelength-encoded two-dimensional map. In an alternative embodiment of step f) of the method, a monochromatic two-dimensional map is generated by employing a multivariate interpolation process to obtain an estimate of the radiation received at a specific wavelength (λ) for each pixel based on the records from the wavelength-encoded two-dimensional map.

[0075] The multivariate interpolation calculation can be performed by various methods known in the literature such as polynomial interpolation, nearest neighbor interpolation, Kriging, inverse distance weighting, natural neighbor interpolation, radial basis function interpolation, trilinear interpolation, tricubic interpolation, spline interpolation, etc.

[0076] In a preferred embodiment, the multivariate interpolation process is a tricubic spline interpolation method.

[0077] In a specific embodiment, step h) uses spatial segmentation (also referred to as spatial clustering) to identify the structures in the spectral cube and obtains their specific spectral features by:

[0078] Determining the number of pixels corresponding to the sample structure,

[0079] Identifying the regions considered to represent the background,

[0080] Determining the regions in the background, and

[0081] Determining the background-corrected signal for each wavelength of the spectrum as:

[0082]

[0083] where N st and N bck are the number of pixels in the selected sample structure and the background, respectively; is the signal measured at pixel i of the sample structure; is the signal measured at pixel j of the background; S′ st is the background-corrected signal of the sample structure.

[0084] Advantageously, background correction enables the revelation of subtle variations in signal intensity.

[0085] In one embodiment, step h) includes obtaining the size and / or shape of the sample structure.

[0086] In a particular embodiment, step h) uses spatial segmentation (also referred to as spatial clustering) to identify the structures in the spectral cube and obtains its size by:

[0087] determining the number of pixels corresponding to the sample structure, and

[0088] determining the size of the structure as:

[0089]

[0090] where, N st is the number of pixels in the selected sample structure; P is the size of the pixel, and M is the total optical magnification of the system.

[0091] The purpose of step i) is to obtain information about the biological nature of the structures identified in the spectral cube. This can be achieved by comparing the spectral characteristics of each structure with those of a set of known biological structures. Additionally / or, the spectral characteristics of the structure can be decomposed, and the relative contribution of each biomarker studied can be obtained; the relative contributions of multiple biomarkers can help identify the nature of the structure based on the knowledge of experts in the field or a reference database based on pre-established ratios by experts.

[0092] In step i), the determination of the relative contribution of each biomarker of the sample to each structure of the spectral cube is performed using any spectral decomposition method known in the literature. In a preferred embodiment, assuming a linear mixing model of the emitter mixture, different methods (such as ordinary least squares (OLS), weighted least squares (WLS), generalized linear model (GLM), non-negative least squares (NNLS), etc.) can be used to solve the LMM problem. As a result of the decomposition process, multiple images are obtained for each biomarker of the sample. These monochromatic images will be referred to herein as "biomarker images" and represent the abundance of each biomarker in the sample.

[0093] In one embodiment, steps b) to f) are performed for a plurality of radiation wavelengths (Λ) and / or a plurality of radiation modes (α). In this embodiment, in step f), when the sample is irradiated with a given radiation mode (α) at a given radiation wavelength (Λ), each monochromatic two-dimensional map corresponds to the radiation emitted at a specific wavelength (λ), and step g) is performed by combining a plurality of monochromatic two-dimensional maps. A radiation source with a tunable wavelength and / or a selectable position or a plurality of radiation sources can be used to provide a plurality of radiation wavelengths and / or radiation modes.

[0094] Furthermore, in a specific embodiment, a spectral cube is constructed by sorting a plurality of monochromatic two-dimensional maps according to a specific wavelength (λ) and a given radiation wavelength (Λ).

[0095] In one embodiment, the solid-phase sample is a biological sample such as a sample from a human, an animal, a fungus, or a plant.

[0096] In one embodiment, the solid-phase sample is a biopsy of human tissue.

[0097] In one embodiment, the method includes a step of staining the sample with at least one molecular marker.

[0098] In one embodiment, during the staining step, at least one molecular marker is chromogenic or fluorescent.

[0099] In one embodiment, during the staining step, at least one marker is a combination of an affinity molecule and a reporter molecule, the affinity molecule has a natural affinity for at least one component of the sample, and the reporter molecule is a chromogen or a fluorescent dye.

[0100] In one embodiment, at least one of the affinity molecules that binds to the reporter molecule is an antibody.

[0101] In one embodiment, during the staining step, at least one marker is a single molecule that has a natural affinity for at least one component of the sample and is simultaneously a chromogen or a fluorescent dye.

[0102] All features described in this specification (including the claims, the description, and the drawings) and / or all steps of the described methods can be combined in any combination, except for combinations of mutually exclusive features and / or steps. BRIEF DESCRIPTION OF THE DRAWINGS

[0103] With reference to the accompanying drawings, these and other features and advantages of the present invention will be clearly understood in view of the detailed description of the present invention. The detailed description of the present invention becomes apparent through the preferred embodiments of the present invention, which are given only as examples and are not limited thereto.

[0104] Figures 1-A to 1-EFive embodiments of a hyperspectral quantitative imaging cytometry system according to the present invention are schematically shown.

[0105] Figure 2 A flowchart of a method according to an embodiment of the present invention is shown.

[0106] Figure 3 The process of acquiring a wavelength-encoded spatial map of a sample is schematically shown. Detailed Description

[0107] Figures 1-A to 1-E A hyperspectral quantitative imaging cytometry system 100 according to the present invention is schematically shown.

[0108] System 100 includes an observation area 104, and the observation area 104 includes a sample holding part 112 which is configured to hold one or more solid-phase samples. In one embodiment, the sample holding part is configured to hold at least one solid-phase sample support, and each support is adapted to accommodate a fixed sample, preferably a biological sample. These solid-phase sample supports can have different materials (preferably, crystalline and optically transparent materials (such as glass or plastic)), and can have different shapes and sizes (such as a rectangular shape (such as a microscope slide)).

[0109] Biological samples can be from humans, animals, fungi or plants, and can be used alone or in combination with molecular markers.

[0110] Preferably, molecular markers are used to reveal the components of biological samples. The reporting part of the molecular marker can emit light or selectively absorb radiation when irradiated. Molecular markers can show characteristic radiation spectra due to their physical structure or when bound to biological samples.

[0111] System 100 includes one or more radiation sources 101, 102, 103 to irradiate and excite the markers of the sample located in the observation area 104. Through this interaction, the sample emits radiation due to any process such as scattering, fluorescence, phosphorescence, chemiluminescence or selective absorption / transmission.

[0112] In one embodiment, the biological sample to be placed in the observation area 104 is biological tissue, that is, a collection of interconnected cells and their extracellular matrix that perform similar functions in a living organism. The components of biological tissue can naturally absorb light of specific wavelengths and generally emit radiation at different wavelengths. If the components of biological tissue lack sufficient contrast for direct study, specific classes of markers can be added to the sample to provide contrast and enable the detection of specific components. These components can be DNA, proteins, lipids, carbohydrates, etc., and multiple markers can be used to study multiple components simultaneously.

[0113] In response to radiation, components of the sample and / or molecular markers used in conjunction with the sample emit a radiation spectrum, which is captured by the collection element 105. The collected radiation is directed to the multi-channel filtering element 108 and then redirected to the image sensor 109.

[0114] Figure 1-A The illustrated embodiment includes three radiation sources 101, 102, 103 that emit light in the visible spectrum. However, a different number of radiation sources may be used. Additionally, any known radiation source suitable for exciting the target sample material may be used. For example, the radiation sources 101, 102, 103 may be lasers, light-emitting diodes (LEDs), and / or lamps. The laser or LED may be configured to emit multiple excitation wavelengths or a single wavelength. If the radiation source 101 produces radiation of more than one wavelength, a bandpass filter 107 may be located in front of the radiation source 101 to filter out any unwanted wavelengths before the radiation reaches the sample in the observation area 104.

[0115] In Figure 1-A the illustrated embodiment, each of the radiation sources 101, 102, 103 irradiates the entire observation area 104 at once (i.e., wide-field radiation).

[0116] In Figure 1-B the illustrated embodiment, there is only one radiation source 101, and the radiation source 101 irradiates the observation area 104 using a light beam 106 that comes from the same side as the side where the collection element 105 is located and is directed along the optical axis of the collection element 105 (bright-field epi-illumination). In Figure 1-C In another illustrated embodiment, the radiation source 102 irradiates the observation area 104 using a light beam 106 that comes from the same side as the side where the collection element 105 is located and is directed at a non-zero angle to the optical axis of the collection element 105 (dark-field epi-illumination). In Figure 1-D In another illustrated embodiment, the radiation source 103 irradiates the observation area 104 using a light beam 106 that comes from the side of the observation area opposite to the side where the collection element 105 is located and is directed along the optical axis of the collection element 105 (bright-field trans-illumination). In Figure 1-E In another illustrated embodiment, the radiation source 103 irradiates the observation area 104 using a light beam 106 that comes from the side of the observation area opposite to the side where the collection element 105 is located and is directed at a non-zero angle to the optical axis of the collection element 105 (dark-field trans-illumination).

[0117] Thus, in Figure 1-B and Figure 1-CIn the embodiment, the radiation sources 101, 102 and the collection element 105 are arranged on the same side of the observation region 104, such that the sample in the observation region 104 reflects the radiation before the radiation from the radiation sources 101, 102 is collected by the collection element 105. In Figure 1-D and Figure 1-E In the embodiment, the observation region 104 is interposed between the radiation source 103 and the collection element 105, such that the radiation from the radiation source 103 passes through the sample in the observation region 104 before being collected by the collection element 105.

[0118] In a preferred embodiment as shown in Figure 1-A shown, Figures 1-B to 1-E Two or more of the above-described imaging geometries as shown can coexist in the system and can be used sequentially to obtain complementary data.

[0119] The radiation emitted by the radiation sample is collected by the collection element 105. In a preferred embodiment, the collection element is configured to simultaneously capture the radiation from multiple spatial positions in the observation region 104. In one embodiment, the collection element is a lens or a combination of lenses.

[0120] In a preferred embodiment, the collection element has a low magnification factor (M) to achieve a large field of view (FOV) and obtain information about a larger two-dimensional area of the observation region 104, thereby obtaining information about the sample. Preferably, the magnification factor value (M) is less than 20, more preferably less than 10, and most preferably less than 2. The total magnification factor of the collection element 105 in combination with any other element of the system having a magnification factor can be selected to ensure the correct sampling frequency of the FOV by the image sensor 109. The sampling frequency of the FOV can be selected to distinguish individual cells in a biological tissue sample, but not small subcellular details. The image sensor 109 includes a two-dimensional array of radiation detection elements. The final magnification factor of the system can depend on the characteristics of the image sensor, e.g., the size of the radiation detection elements.

[0121] After being collected by the collection element 105 and before reaching the image sensor 109, the radiation is directed through the multi-channel filtering element 108. The multi-channel filtering element 108 selectively filters the radiation reaching each radiation detection element or group of radiation detection elements of the image sensor 109. In one embodiment, the multi-channel filtering element 108 is a continuous or semi-continuous variable bandpass filter.

[0122] In the present embodiment, the multi-channel filtering element is arranged to be displaceable in one or two spatial dimensions (x, y), such that the image sensor 109 is capable of generating a plurality of two-dimensional outputs, each output representing a two-dimensional (x, y) map of different wavelength encodings (λ = λ(y)) of the sample. Each wavelength-encoded two-dimensional map of the sample generated by the image sensor 109 can be sent to the processor 110 and / or stored in the memory 111 for further processing.

[0123] The image sensor 109 is a two-dimensional array sensor, where each radiation detection element in the array receives radiation from different two-dimensional spatial positions in the sample, generating an image that is a spatial (x, y) map of the sample under study. The image sensor 109 can be a two-dimensional photodetector array sensor, for example, a charge-coupled device (CCD), a complementary metal oxide semiconductor (CMOS), an electron multiplying CCD (EMCCD), or any other similar system for obtaining two-dimensional spatial data. The two-dimensional image sensor can be triggered to cumulatively sample data over a specific amount of time.

[0124] Figure 2 A flowchart of a method according to an embodiment of the present invention is shown. A sample (which may already be labeled and stained) is placed in the observation area 104 (201). The first radiation sources 101, 102, 103 are activated (202), the filtering element 108 is in its initial position (203), and the image sensor 109 records a first image representing a first wavelength-encoded two-dimensional map of the biological sample (204). Then the filtering element 108 is moved to a second position (203), and a second wavelength-encoded two-dimensional map is obtained. This process continues until the last position of the filtering element 108 is reached. If multiple radiation modes (α) or radiation wavelengths (Λ) are used, the second radiation sources 101, 102, 103 are activated (202), the filtering element 108 is moved to its initial position (203), and a new image corresponding to the new wavelength-encoded two-dimensional map is taken. This process continues until the last position of the filtering element 108 is reached, and the process is repeated for all radiation sources.

[0125] The processor 110 takes a plurality of wavelength-encoded two-dimensional maps 301 of the biological sample and constructs a plurality of monochromatic two-dimensional maps 304 of the sample (205) by combining the portions of the wavelength-encoded two-dimensional maps 301 corresponding to a specific wavelength. Each of these monochromatic two-dimensional maps has spatial correlation (x, y) information regarding the radiation emitted at a given wavelength (A) when the sample is irradiated with a given radiation mode (α) at a given radiation wavelength (A). The combination of these plurality of monochromatic two-dimensional maps 304 constitutes a five-dimensional (x, y, λ, Λ, α) data set 305, which can be reduced to a spectral cube 306 having two spatial dimensions (x, y) and one spectral dimension (α, Λ, λ).

[0126] Although in this exemplary embodiment, the monochromatic two-dimensional map is generated by combining portions of the wavelength-encoded two-dimensional map corresponding to specific wavelengths, in alternative embodiments, the monochromatic two-dimensional map is generated by employing a multivariate interpolation process to obtain an estimate of the radiation received at that specific wavelength (λ) for each pixel based on the recordings from the wavelength-encoded two-dimensional map.

[0127] Processor 110 captures the spectral cube 306 generated by the biological sample and performs a spatial segmentation (206) of the data locations in order to identify meaningful biological sample structures (e.g., cells) and obtain the spectral characteristics (207) of each of these structures. The processor 110 then compares the spectral characteristics of each structure with a database of spectra of known structures stored in the memory 111 (208). Alternatively or concurrently with step 208, the processor 110 performs an estimation of the abundance of each marker (i.e., spectral decomposition) in each of the identified sample structures by determining the relative contribution of each marker to the spectral characteristics of each structure identified in the spectral cube (209). The information obtained regarding the nature of the identified structures and / or regarding the abundance of each marker can be correlated by an expert in the field with other relevant information about the sample (210).

[0128] A commonly accepted model for the mixture of emitters required to perform spectral decomposition is the linear mixing model (LMM), which assumes a linear combination of the abundances of the emitters. Multiple monochromatic two-dimensional maps provide a discontinuous emission spectrum for each pixel, where the spectral points corresponding to a specific wavelength are present in the monochromatic two-dimensional map corresponding to that wavelength.

[0129] According to the LMM, it is assumed that the discontinuous emission spectrum is a linear combination of the spectra of individual markers. The emissions of M excitation markers from N pixels or structures are captured at L excitation wavelengths (Λ) and produce L individual signals (channels). Thus, each pixel or structure is represented by a vector of L channels, which contains the sum of the contributions of the M markers for each channel. The LMM can be written in matrix form as:

[0130] Y = AH

[0131] where Y is an L×N matrix of the detected intensities, A is an L×M matrix containing the mixture of the expected emissions of each of the M markers in each of the L spectral channels, and H is an M×N matrix of the true marker concentrations for each structure.

[0132] The uncertainty in the measurements can also be taken into account by incorporating noise into the model. Two noise models are commonly employed: the first model is the additive Gaussian noise (white noise) model, where the above equation is modified to:

[0133] Y = AH + R

[0134] Wherein, R is a matrix formed by independent and identically distributed Gaussian variables with zero mean. The second model uses a Poisson process to construct a model of photon emission.

[0135] Different methods such as ordinary least squares (OLS), weighted least squares (WLS), generalized linear model (GLM), non - negative least squares (NNLS), etc. can be used to solve the LMM problem and obtain a spatial map of the abundance of each biomarker, which can be sent to a processor and / or stored in a memory for further processing.

[0136] Figure 3 An embodiment representing the steps of obtaining a wavelength - encoded two - dimensional map of a biological sample (A) and its conversion to a monochromatic two - dimensional map of biological samples (B & C), where the monochromatic two - dimensional maps form a spectral cube of sample (D) when combined. This process uses all consecutive images taken at different positions of the filter element 108 for a given radiation wavelength (Λ) and a given radiation mode (α), and then proceeds in the (x, y, λ) dimensional space. If multiple radiation wavelengths (Λ) and / or multiple radiation modes (α) are used, each set of images is processed independently, where each set of images corresponds to a given radiation wavelength (Λ) and a given radiation mode (α).

[0137] Each wavelength - encoded two - dimensional map 301(A) is obtained by moving the filter element 108 parallel to one of the spatial dimensions (y) of the FOV of the collection element in a number of consecutive discrete steps. The number of steps required for the complete displacement of the filter element through the entire FOV 303 defines the number of images taken, and thus the number of wavelength - encoded two - dimensional maps.

[0138] Each wavelength - encoded two - dimensional map 301 is divided (sliced) into "n" bands (B) along the "y" spatial dimension. The number of bands "n" corresponds to the number of steps required for the complete displacement of the filter through one band of the entire FOV. Each band is combined with bands from other wavelength - encoded two - dimensional maps 301 corresponding to the same wavelength to reproduce a monochromatic two - dimensional map 304 of the biological sample, where each monochromatic two - dimensional map 304 corresponds to a specific wavelength. The entire set of monochromatic two - dimensional maps 304 constitutes a spectral cube 306. In Figure 3 which, the bands corresponding to the same wavelength are represented by the same pattern.

[0139] In an alternative embodiment, a monochromatic two - dimensional map is generated from the wavelength - encoded two - dimensional map by employing a multivariate interpolation process to obtain an estimate of the radiation received at a specific wavelength (λ) for each pixel based on the records from the wavelength - encoded two - dimensional map.

[0140] According to the present invention, the following clauses are provided herein:

[0141] Clause 1. A hyperspectral quantitative imaging cytometry system (100) comprising:

[0142] An observation area (104), the observation area (104) comprising a sample holding part configured to hold one or more solid-phase samples,

[0143] At least one radiation source (101, 102, 103), the at least one radiation source (101, 102, 103) being configured to irradiate the observation area (104),

[0144] A collection element (105), the collection element (105) being configured to collect radiation emitted or reflected by the sample irradiated by the at least one radiation source (101, 102, 103),

[0145] A multi-channel filtering element (108), the multi-channel filtering element (108) being configured to selectively filter the wavelengths of the radiation collected by the collection element (105), and

[0146] An image sensor (109), the image sensor (109) being configured to receive the filtered radiation and generate an image as a two-dimensional map of the sample, and the image sensor (109) comprising a two-dimensional array of radiation detection elements.

[0147] Clause 2. The hyperspectral quantitative imaging cytometry system (100) according to the preceding clause, wherein the filtering element (108) is arranged to be movable between at least two positions, and wherein each position of the filtering element (108) selectively filters the wavelengths of the radiation reaching each radiation detection element of the image sensor (109).

[0148] Clause 3. The hyperspectral quantitative imaging cytometry system (100) according to clause 2, wherein the system further comprises a processor (110), the processor (110) being configured to:

[0149] - Receive a plurality of wavelength-encoded two-dimensional maps (301) of the sample, the plurality of wavelength-encoded two-dimensional maps (301) being associated with a plurality of positions of the filtering element (108), wherein the wavelength-encoded two-dimensional map is an image generated by the image sensor (109) based on the radiation it receives for the positions of the filtering element (108);

[0150] - Generate (205) a plurality of monochromatic two-dimensional maps (304) of the sample by combining portions of the wavelength-encoded two-dimensional maps (301) of the sample corresponding to a specific wavelength (λ);

[0151] - Construct a spectral cube (306) comprising the plurality of monochromatic two-dimensional maps (304);

[0152] - Identify the sample structure (206) on the spectral cube (306) and obtain its spectral characteristics (207);

[0153] - Compare the obtained spectral characteristics with a database of spectral characteristics of known structures (208), and / or decompose the spectral characteristics and obtain (209) an estimate of the abundance of each marker in each identified sample structure.

[0154] Clause 4. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses, wherein, according to a transmission configuration, the observation region (104) is interposed between at least one radiation source (103) and the collection element (105) such that the radiation from the radiation source (103) passes through the observation region (104) before being collected by the collection element (105).

[0155] Clause 5. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses, wherein at least one radiation source (102), the observation region (104), and the collection element (105) are arranged according to a dark-field configuration.

[0156] Clause 6. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses, wherein at least one radiation source (101, 103), the observation region (104), and the collection element (105) are arranged according to a bright-field configuration.

[0157] Clause 7. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses, wherein, according to an epi-illumination configuration, the direction of at least one radiation source (101, 102) is towards the observation region (104) such that the observation region (104) reflects the radiation from the radiation source (101, 102) before the radiation from the radiation source (101, 102) is collected by the collection element (105).

[0158] Clause 8. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses, wherein the system (100) further includes a memory (111).

[0159] Clause 9. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses, wherein the system (100) further includes at least one band-pass filter (107) interposed between at least one radiation source (101, 102, 103) and the observation region (104).

[0160] Clause 10. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses, wherein at least one radiation source (101, 102, 103) is a laser, a light-emitting diode, or a lamp.

[0161] Clause 11. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses, wherein the collection element (105) comprises a lens or a combination of lenses configured to simultaneously capture radiation from a plurality of spatial positions in the observation region.

[0162] Clause 12. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses,

[0163] wherein the magnification factor value (M) of the collection element (105) is less than 20, preferably less than 10, and most preferably less than 2; and / or

[0164] wherein the numerical aperture value of the collection element (105) is greater than 0.25, preferably equal to or greater than 0.5.

[0165] Clause 13. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding clauses, wherein the filtering element (108) is a continuous linear variable filter or a semi - continuous linear variable filter between 200 nm and 1200 nm, preferably between 350 nm and 950 nm.

[0166] Clause 14. A method of obtaining data from a solid - phase sample using a hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, comprising the steps of:

[0167] a) Providing a sample (201);

[0168] b) Radiating the sample (202) with radiation that interacts with the sample such that the sample emits radiation;

[0169] c) Capturing the emitted radiation using the collection element (105);

[0170] d) Filtering the emitted radiation using a multi - channel filtering element (108);

[0171] e) Sequentially recording (204) a plurality of wavelength - encoded two - dimensional maps (301) of the sample that match the sequential displacement (203) of the filtering element (108), wherein each position of the filtering element (108) selectively filters the wavelength (λ) of the radiation reaching each radiation - detecting element of the image sensor (109), and wherein the image sensor (109) generates a wavelength - encoded two - dimensional map based on the radiation it receives for each position of the filtering element (108);

[0172] f) Generating (205) a plurality of monochromatic two - dimensional maps (304) of the sample by combining portions of the wavelength - encoded two - dimensional maps (301) of the sample corresponding to a specific wavelength (λ) of the emitted radiation;

[0173] g) Construct a spectral cube (306) comprising a plurality of monochromatic two-dimensional images (304);

[0174] h) Identify a sample structure (206) on the spectral cube (306) and obtain its spectral characteristics (207);

[0175] i) Compare the obtained spectral characteristics with a database of spectral characteristics of known structures (208), and / or decompose the spectral characteristics and obtain (209) an estimate of the abundance of each marker in each identified sample structure.

[0176] Clause 15. The method according to Clause 14, wherein steps b) to f) are performed for a plurality of radiation wavelengths (A) and / or a plurality of radiation modes (α), wherein, in step f), when the sample is irradiated with a given radiation wavelength (A) and a given radiation mode (α), each monochromatic two-dimensional image (304) corresponds to radiation emitted at a specific wavelength (λ), and wherein step g) is performed by combining the plurality of monochromatic two-dimensional images (304).

Claims

1. A hyperspectral quantitative imaging cytometry system (100), comprising: an observation region (104), the observation region (104) including a sample holding portion configured to hold one or more solid-phase samples, at least one radiation source (101, 102, 103), the at least one radiation source (101, 102, 103) being configured to irradiate the observation region (104), a collection element (105), the collection element (105) being configured to collect radiation emitted or reflected by a sample irradiated by the at least one radiation source (101, 102, 103), wherein the collection element (105) has a magnification factor value (M) equal to or less than 20 and a numerical aperture value equal to or greater than 0.25, a multi-channel filtering element (108), the multi-channel filtering element (108) being configured to selectively filter the wavelengths of the radiation collected by the collection element (105), and an image sensor (109), the image sensor (109) being configured to receive the filtered radiation and generate an image as a two-dimensional map of the sample, and the image sensor (109) includes a two-dimensional array of radiation detection elements.

2. The hyperspectral quantitative imaging cytometry system (100) according to the preceding claim, wherein, the filtering element (108) is arranged to be movable between at least two positions, wherein each position of the filtering element (108) selectively filters the wavelengths of the radiation reaching each radiation detection element of the image sensor (109).

3. The hyperspectral quantitative imaging cytometry system (100) according to claim 2, wherein, the system further includes a processor (110), the processor (110) being configured to: - receive a plurality of wavelength-encoded two-dimensional maps (301) of the sample, the plurality of wavelength-encoded two-dimensional maps (301) being associated with a plurality of positions of the filtering element (108), wherein the wavelength-encoded two-dimensional map is an image generated by the image sensor (109) based on the radiation received by it for the positions of the filtering element (108); - generate (205) a plurality of monochromatic two-dimensional maps (304) of the sample by combining portions of the wavelength-encoded two-dimensional maps (301) of the sample corresponding to a specific wavelength (λ); - construct a spectral cube (306) containing the plurality of monochromatic two-dimensional maps (304); - identify a sample structure (206) on the spectral cube (306) and obtain its spectral characteristics (207); - compare the obtained spectral characteristics with a database of spectral characteristics of known structures (208), and / or decompose the spectral characteristics and obtain (209) an estimate of the abundance of each marker in each identified sample structure, the marker being any component that naturally emits radiation on the sample or emits radiation due to the presence of a molecular marker added to the sample.

4. The hyperspectral quantitative imaging cytometry system (100) according to claim 2, Wherein, the system further includes a processor (110), and the processor (110) is configured to: - receive a plurality of wavelength-encoded two-dimensional maps (301) of the sample, the plurality of wavelength-encoded two-dimensional maps (301) being associated with a plurality of positions of the filtering element (108), wherein the wavelength-encoded two-dimensional map is an image generated by the image sensor (109) based on the radiation received by it for the positions of the filtering element (108); - generate (205) a plurality of monochromatic two-dimensional maps (304) of the sample corresponding to a specific wavelength (λ) by adopting a multivariate interpolation process to obtain an estimate of the radiation received at the specific wavelength (λ) for each pixel based on the records from the wavelength-encoded two-dimensional maps; - construct a spectral cube (306) containing the plurality of monochromatic two-dimensional maps (304); - identify a sample structure (206) on the spectral cube (306) and obtain its spectral characteristics (207); - compare the obtained spectral characteristics with a database of spectral characteristics of known structures, and / or decompose the spectral characteristics and obtain (209) an estimate of the abundance of each marker in each identified sample structure, where the marker is any component that naturally emits radiation on the sample or emits radiation due to the presence of a molecular marker added to the sample.

5. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, According to the transmission configuration, the observation region (104) is interposed between at least one radiation source (103) and the collection element (105), such that the radiation of the radiation source (103) passes through the observation region (104) before being collected by the collection element (105).

6. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, At least one radiation source (102), the observation region (104) and the collection element (105) are arranged according to the dark field configuration.

7. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, At least one radiation source (101, 103), the observation region (104) and the collection element (105) are arranged according to the bright field configuration.

8. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, According to the epi-illumination configuration, the direction of at least one radiation source (101, 102) is towards the observation region (104), such that the observation region (104) reflects the radiation of the radiation source (101, 102) before the radiation of the radiation source (101, 102) is collected by the collection element (105).

9. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, The system (100) further includes a memory (111).

10. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, the system (100) further comprises at least one band - pass filter (107) interposed between at least one radiation source (101, 102, 103) and the observation area (104).

11. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, at least one radiation source (101, 102, 103) is a laser, a light - emitting diode or a lamp.

12. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, the collection element (105) comprises a lens or a combination of lenses configured to simultaneously capture radiation from multiple spatial positions in the observation area.

13. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, the magnification factor value (M) of the collection element (105) is equal to or less than 10, preferably equal to or less than 2; and / or wherein, the numerical aperture value of the collection element (105) is equal to or greater than 0.

5.

14. The hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, wherein, the filtering element (108) is a continuous linear variable filter or a semi - continuous linear variable filter, preferably between 200 nm and 1200 nm, more preferably between 350 nm and 950 nm.

15. A method for obtaining data from a solid - phase sample using the hyperspectral quantitative imaging cytometry system (100) according to any one of the preceding claims, comprising the following steps: Step a) providing the solid - phase sample (201); Step b) irradiating the sample (202) with radiation that interacts with the sample such that the sample emits radiation; Step c) capturing the emitted radiation using the collection element (105); Step d) filtering the emitted radiation using the multi - channel filtering element (108); Step e) sequentially recording (204) a plurality of wavelength - encoded two - dimensional maps (301) of the sample that match the sequential displacement (203) of the filtering element (108), wherein each position of the filtering element (108) selectively filters the wavelength (λ) of the radiation reaching each radiation - detecting element of the image sensor (109), and wherein the image sensor (109) generates a wavelength - encoded two - dimensional map based on the radiation it receives for each position of the filtering element (108); Step f) generating (205) a plurality of monochromatic two - dimensional maps (304) of the sample based on the wavelength - encoded two - dimensional maps (301) of the sample; Step g) constructing a spectral cube (306) containing the plurality of monochromatic two - dimensional maps (304); Step h) identifying the sample structure (206) on the spectral cube (306) and obtaining its spectral characteristics (207); Step i) Compare the spectral features obtained with a database of spectral features of known structures (208), and / or decompose the spectral features and obtain (209) an estimate of the abundance of each marker in each identified sample structure, where a marker is any component that emits radiation naturally on the sample or due to the presence of a molecular label added to the sample.

16. The method according to claim 15, wherein, Steps b) to f) are performed for a plurality of radiation wavelengths (Λ) and / or a plurality of radiation modes (α), where, in step f), when the sample is irradiated with a given radiation wavelength (Λ) and a given radiation mode (α), each monochromatic two-dimensional map (304) corresponds to radiation emitted at a specific wavelength (λ), and wherein step g) is performed by combining the plurality of monochromatic two-dimensional maps (304).

17. The method according to any one of claims 15 or 16, wherein, The monochromatic two-dimensional map (304) is generated by combining portions of the wavelength-encoded two-dimensional map corresponding to a specific wavelength (300).

18. The method according to any one of claims 15 or 16, wherein, The monochromatic two-dimensional map is generated by employing a multivariate interpolation process to obtain an estimate of the radiation received at the specific wavelength (λ) for each pixel based on the records from the wavelength-encoded two-dimensional map.

19. The method according to claim 18, wherein, The multivariate interpolation process is a tricubic spline interpolation method.

20. The method according to any one of claims 15 to 19, wherein, In step a), a specific class of label has been added to the solid-phase sample (201) to provide contrast and enable the detection of specific components.

Images