Method and system for high-precision cancer cell discrimination

Abstract

Aspects of the present disclosure include systems for detecting and classifying cancer cells in a whole blood sample in a flow stream. A system according to certain embodiments includes a light source configured to irradiate cells of a whole blood sample in a flow stream; a light detection system including a photodetector that captures an image of light emitted from the irradiated cells; and a processor having a memory operably coupled to the processor, wherein the memory includes instructions stored thereon that, when executed by the processor, cause the processor to apply a neural network to the captured image in order to detect the presence of cancer cells in the whole blood sample having a minimum concentration of cancer cells and determine a classification of the cancer cells in a plurality of classifications. In some cases, the neural network is trained using images of samples labeled with a fluorophore ground truth marker. Methods of using the subject systems to detect and classify cancer cells in a whole blood sample are also described. Non-transitory computer readable storage media are also provided.

Description

Cross-references to related applications

[0001] This application claims priority to U.S. Provisional Patent Application Serial No. 63 / 813,216, filed May 28, 2025, and U.S. Provisional Patent Application Serial No. 63 / 748,726, filed January 23, 2025, pursuant to 35 USC § 119(e); the disclosure of which is incorporated herein by reference. Background Technology

[0002] Flow-based particle sorting systems, such as sorting flow cytometers, are used to sort particles in a fluid sample based on at least one measured characteristic of the particles. In a flow-based particle sorting system, particles (e.g., analyte-bound beads or individual cells in a fluid suspension) flow through a detection zone, where a sensor detects particles to be sorted contained in this type of flow. Upon detection of particles to be sorted, the sensor triggers a sorting mechanism that selectively separates particles of interest.

[0003] Particle sensing is typically performed by flowing a fluid through a detection area, exposing particles to irradiation from one or more lasers, and measuring the fluorescence from the particles. Particles or their components can be labeled with fluorescent dyes for easy detection, and multiple different particles or components can be detected simultaneously by labeling different particles or components with fluorescent dyes of different spectra. Detection is performed using one or more photoelectric sensors to independently measure the fluorescence of each different fluorescent dye.

[0004] Using data generated from the detected light, the distribution of each component can be recorded, and the desired material can be sorted. To sort particles in the sample, a droplet charging mechanism charges droplets containing the type of particles to be sorted at the break point of the flow stream. The droplets pass through an electrostatic field and are deflected into one or more collection containers according to the polarity and size of their charge. Uncharged droplets are not deflected by the electrostatic field.

[0005] Circulating tumor cells (CTCs) are typically (but not always) epithelial cells derived from solid tumors, entering the bloodstream of patients with various cancers at extremely low concentrations. CTCs shed from existing tumors or metastases often lead to the formation of secondary tumors. Secondary tumors are often difficult to detect and account for 90% of cancer deaths. Circulating tumor cells serve as a bridge between primary and metastatic tumors. Therefore, the identification and characterization of circulating tumor cells hold promise for the early detection and treatment management of metastatic epithelial malignancies. Detecting CTCs in cancer patients is an effective tool for the early diagnosis of primary or secondary cancer growth and for determining the prognosis of cancer patients undergoing treatment, as the number and characteristics of CTCs present in the blood of such patients are correlated with overall prognosis and response to therapy. Therefore, CTCs can serve as an early indicator of tumor spread or metastasis before the onset of clinical symptoms.

[0006] While the detection of circulating tumor cells (CTCs) holds significant prognostic and potential therapeutic value in cancer management and treatment, these rare cells are difficult to detect due to their elusive presence in the bloodstream. CTCs were first described in the 19th century, but only recent technological advancements have made them reliably detectable. It is believed that extremely low concentrations of CTCs are present in the peripheral blood of cancer patients. For example, it is estimated that there is one CTC for every 10 million normal blood cells in cancer patients. Although current technologies can identify CTCs and link them to disease, there is currently no method with sufficient sensitivity to reliably measure statistically significant cell numbers at different stages of the disease to guide the development of the most effective treatment regimen. Summary of the Invention

[0007] The inventors recognized the need for improved CTC detection. They also recognized that imaging flow cytometry methods could provide improved CTC detection.

[0008] This disclosure includes a system for detecting and classifying cancer cells in a flowing stream of whole blood samples. According to certain embodiments, the system includes a light source configured to irradiate cells of a whole blood sample in a flowing stream; a light detection system comprising a photodetector capturing an image of light emitted from the irradiated cells; and a processor comprising a memory operatively coupled to the processor, wherein the memory contains instructions stored thereon that, when executed by the processor, cause the processor to apply a neural network to the captured image to detect the presence of cancer cells in the whole blood sample with a minimum concentration of cancer cells and to determine the classification of the cancer cells among multiple classifications. In some cases, the neural network is trained using images of samples labeled with a fluorophore benchmark ground truth marker.

[0009] In some embodiments, the minimum concentration of cancer cells present in the whole blood sample is 0.5% or more, for example, 1% or more. In some cases, the minimum concentration of cancer cells present in the whole blood sample is 0.5% to 0.5%.

[0010] In some implementations, the memory includes instructions for classifying cancer cells in a sample into multiple categories. In some cases, the multiple categories include two or more categories, such as three or more categories, and may include four or more categories. In some cases, the multiple categories include ovarian cancer, T-lymphocytic carcinoma, breast cancer, colon cancer, or epithelial carcinoma. In some cases, the multiple categories include OVCAR, Jurkat, MCF-7, and adenocarcinoma categories.

[0011] In some cases, the memory contains instructions for classifying the cells based on one or more parameters calculated from the generated image data. In some cases, the memory contains instructions for applying a classifier neural network to classify the cells. In some cases, the neural network includes an input layer consisting of one or more parameters calculated from the generated image data. In some cases, the neural network includes an output layer that is a probability estimate based on cell identity (e.g., the cell is a cancer cell, circulating tumor cell). In some cases, the neural network includes a learning algorithm configured to learn from classification errors. In some cases, the memory contains instructions for updating the classification parameters based on one or more of the parameters calculated from the generated image data. In some cases, the update of the classification parameters is based on a comparison between the one or more parameters calculated from the generated image data and one or more parameters calculated from benchmark ground truth image data. In some cases, the memory contains instructions for generating benchmark ground truth image data.

[0012] In some implementations, the neural network includes a backpropagation neural network. In some cases, the neural network includes two or more stages, such as four or more stages. In some cases, the neural network includes at least one dropout stage. In some cases, each stage includes at least eight hidden layers, such as 32 or more hidden layers. In some cases, each stage includes eight to 64 hidden layers. In some cases, the neural network applies a sigmoid activation function, an Adam optimizer, an adaptive learning rate, a dynamic threshold, binary classification, a binary cross-entropy loss function, or any combination thereof. In some cases, the memory contains instructions for normalizing the intensity of the captured image. In some cases, the neural network is applied to normalize the image. In some cases, the neural network includes an input layer composed of one or more parameters computed from the generated image data.

[0013] In some embodiments, the image data includes one or more waveforms generated in response to the measurement light. In some cases, the memory contains instructions for generating an image of the cell from the image data. In some cases, the memory contains instructions for calculating image parameters based on one or more of the image data and the cell image. In some cases, the memory contains instructions for evaluating the morphology of the cell. In some cases, the memory contains instructions for evaluating the cell morphology by determining one or more of the cell size, the cell shape, and the degree of punctiformity of the cell. In some cases, the memory contains instructions for calculating cell parameters based on the graphic parameters. In some cases, the image parameters are quantized image parameters. In some cases, the parameters are selected from centroid, delta center of mass, diffusivity, eccentricity, major axis moment, maximum intensity, radial moment, minor axis moment, size, total intensity, light loss from the cell nucleus, forward scattered light from the cell nucleus, side scattered light from the cell nucleus, and combinations thereof. In some cases, the memory includes instructions for calculating the image parameters based on an image generated from fluorescence measured from the cell, light loss measured for the cell, forward scattered light measured for the cell, side scattered light measured for the cell, and combinations thereof.

[0014] In some embodiments, the memory includes instructions for classifying cells of the sample based on the calculated cell parameters. In some cases, the memory includes instructions for classifying the cells based on a waveform generated from the measuring light, a generated cell image, image parameters calculated from the generated cell image, or a combination thereof. In some cases, the memory includes instructions for classifying the cells by assigning them to one or more particle population clusters. In some cases, the memory includes instructions for determining one or more sorting gates for classifying the cells in the sample. In some cases, the memory includes instructions for calculating one or more sorting gates for capturing circulating tumor cell population clusters and excluding normal cell population clusters. In some cases, the memory includes instructions for calculating a sorting gate that maximizes the inclusion rate of the circulating tumor cell population clusters. In some cases, the memory includes instructions for calculating a sorting gate that maximizes the exclusion of normal cell population clusters.

[0015] In some embodiments, the light source is configured to irradiate the sample with a frequency-modulated beam. In some cases, the light source includes one or more lasers. In some cases, the optical detection system includes multiple photodetectors. In some embodiments, one or more of the photodetectors are photomultiplier tubes. In some embodiments, one or more of the photodetectors are photodiodes (e.g., avalanche photodiodes, APDs). In some embodiments, the optical detection system includes a photodetector array, such as a photodetector array having multiple photodiodes or charge-coupled devices (CCDs).

[0016] In some cases, the system includes a sorting mechanism for distributing cells from the sample into multiple sample containers. In some cases, the sorting mechanism is configured to sort cells based on the presence, classification, or both of the cells. In some cases, the memory contains instructions for sorting cells based on the generated cell image. In some cases, the memory contains instructions for sorting cells based on the calculated image parameters.

[0017] This disclosure also includes methods for classifying cancer cells. According to some embodiments, the method includes irradiating a whole blood sample containing cells in a flowing stream with a light source, capturing an image of the light emitted from the irradiated cells using a light detection system with a photodetector, and applying a neural network to the captured image to detect the presence of cancer cells in the whole blood sample with a minimum concentration of cancer cells and to determine the classification of the cancer cells among multiple categories. In some cases, the method includes training the neural network using images of samples labeled with a fluorophore benchmark truth marker.

[0018] In some embodiments, the minimum concentration of cancer cells present in the whole blood sample is 0.5% or more, for example, 1% or more. In some cases, the minimum concentration of cancer cells present in the whole blood sample is 0.5% to 0.5%.

[0019] In the implementation scheme, the cancer cells in the sample are classified. In some cases, the multiple classifications include two or more classifications, such as three or more classifications, and include four or more classifications. In some cases, the multiple classifications include ovarian cancer classification, T-lymphoblastic cancer classification, breast cancer classification, colon cancer classification, or epithelial cancer classification. In some cases, the multiple classifications include OVCAR classification, Jurkat classification, MCF-7 classification, and adenocarcinoma classification.

[0020] In some cases, the method includes classifying the cells based on one or more parameters calculated from the generated image data. In some cases, the cells are classified using a classifier neural network. In some cases, the neural network includes an input layer consisting of one or more parameters calculated from the generated image data. In some cases, the neural network includes an output layer that is a probability estimate based on cell identity (e.g., the cell is a cancer cell, circulating tumor cell). In some cases, the neural network includes a learning algorithm configured to learn from classification errors. In some cases, the method includes updating the classification parameters based on one or more of the parameters calculated from the generated image data. In some cases, the update of the classification parameters is based on a comparison between the one or more parameters calculated from the generated image data and one or more parameters calculated from ground truth image data. In some cases, the method includes generating ground truth image data.

[0021] In some implementations, the neural network includes a backpropagation neural network. In some cases, the neural network includes two or more stages, such as four or more stages. In some cases, the neural network includes at least one dropout stage. In some cases, each stage includes at least eight hidden layers, such as 32 or more hidden layers. In some cases, each stage includes eight to 64 hidden layers. In some cases, the neural network applies a sigmoid activation function, an Adam optimizer, an adaptive learning rate, a dynamic threshold, binary classification, a binary cross-entropy loss function, or any combination thereof. In some cases, the method includes normalizing the intensity of the captured image. In some cases, the method includes applying the neural network to the normalized image. In some cases, the neural network includes an input layer composed of one or more parameters computed from the generated image data.

[0022] In some embodiments, the image data includes one or more waveforms generated in response to the measured light. In some cases, an image of the cell is generated from the image data. In some cases, the method includes calculating image parameters based on one or more of the image data and the cell image. In some cases, the method includes evaluating the morphology of the cell. In some cases, the cell morphology is evaluated by determining one or more of the cell size, the cell shape, and the degree of punctiformity of the cell. In some cases, the cell parameters are calculated based on the image parameters. In some cases, the image parameters are quantized image parameters. In some cases, the parameters are selected from centroid, centroid change, diffusivity, eccentricity, major axis moment, maximum intensity, radial moment, minor axis moment, size, total intensity, light loss of the cell nucleus, forward scattered light from the cell nucleus, side scattered light from the cell nucleus, and combinations thereof. In some cases, the image parameters are calculated based on an image generated from fluorescence measured from the cell, light loss measured from the cell, forward scattered light measured from the cell, side scattered light measured from the cell, and combinations thereof.

[0023] In some implementations, the method includes classifying cells in the sample based on the calculated cell parameters. In some cases, the cells are classified based on a waveform generated from the measured light, a generated cell image, image parameters calculated from the generated cell image, or a combination thereof. In some cases, classifying the cells includes assigning the cells to one or more particle population clusters. In some cases, the method includes determining one or more sorting gates for the classified cells in the sample. In some cases, the one or more sorting gates capture circulating tumor cell population clusters and exclude normal cell population clusters. In some cases, the sorting gates maximize the inclusion rate of the circulating tumor cell population clusters. In some cases, the sorting gates maximize the exclusion of normal cell population clusters.

[0024] In some embodiments, the method includes irradiating the whole blood sample in a flowing stream with a frequency-modulated beam. In some cases, the light source includes one or more lasers. In some cases, the light source includes a beam generator assembly configured to generate at least a first frequency-shifted beam and a second frequency-shifted beam. In some cases, the beam generator includes an acousto-optic deflector. In some cases, the beam generator includes a direct digital synthesizer (DDS) radio frequency comb generator. In some cases, the beam generator assembly is configured to generate a local oscillator beam. In some cases, the beam generator assembly is configured to generate multiple frequency-shifted comb beams.

[0025] In some cases, the light detection system comprises multiple photodetectors. In some embodiments, one or more of the photodetectors are photomultiplier tubes. In some embodiments, one or more of the photodetectors are photodiodes (e.g., avalanche photodiodes, APDs). In some embodiments, the light detection system comprises a photodetector array, such as a photodetector array having multiple photodiodes or charge-coupled devices (CCDs).

[0026] In some cases, the method includes sorting cells from the sample into multiple sample containers. In some cases, the cells are sorted based on their presence, classification, or both. In some cases, the cells are sorted based on the generated cell images. In some cases, the cells are sorted based on the calculated image parameters.

[0027] A non-transitory computer-readable storage medium is also provided, having instructions with an algorithm for classifying cells in a sample (e.g., a whole blood sample). According to certain embodiments, the non-transitory computer-readable storage medium has an algorithm for irradiating a whole blood sample containing cells in a flowing stream with a light source, an algorithm for capturing an image of light emitted from the irradiated cells using a light detection system with a photodetector, and an algorithm for applying a neural network to the captured image to detect the presence of cancer cells in the whole blood sample with a minimum concentration of cancer cells and to determine the classification of the cancer cells among multiple classifications. In some cases, the neural network is trained using images of samples labeled with fluorophore benchmark truth markers.

[0028] In some embodiments, the minimum concentration of cancer cells present in the whole blood sample is 0.5% or more, for example, 1% or more. In some cases, the minimum concentration of cancer cells present in the whole blood sample is 0.5% to 0.5%.

[0029] In the implementation scheme, the cancer cells in the sample are classified. In some cases, the multiple classifications include two or more classifications, such as three or more classifications, and include four or more classifications. In some cases, the multiple classifications include ovarian cancer classification, T-lymphoblastic cancer classification, breast cancer classification, colon cancer classification, or epithelial cancer classification. In some cases, the multiple classifications include OVCAR classification, Jurkat classification, MCF-7 classification, and adenocarcinoma classification.

[0030] In some cases, the non-transitory computer-readable storage medium includes an algorithm for classifying the cells based on one or more parameters calculated from the generated image data. In some cases, the non-transitory computer-readable storage medium includes an algorithm for applying a classifier neural network to classify the cells. In some cases, the neural network includes an input layer consisting of one or more parameters calculated from the generated image data. In some cases, the neural network includes an output layer that is a probability estimate based on cell identity (e.g., the cell is a cancer cell, circulating tumor cell). In some cases, the neural network includes a learning algorithm configured to learn from classification errors. In some cases, the non-transitory computer-readable storage medium includes an algorithm for updating the classification parameters based on one or more of the parameters calculated from the generated image data. In some cases, the update of the classification parameters is based on a comparison between the one or more parameters calculated from the generated image data and one or more parameters calculated from benchmark ground truth image data. In some cases, the non-transitory computer-readable storage medium includes an algorithm for generating benchmark ground truth image data.

[0031] In some embodiments, the neural network includes a backpropagation neural network. In some cases, the neural network includes two or more stages, such as four or more stages. In some cases, the neural network includes at least one dropout stage. In some cases, each stage includes at least eight hidden layers, such as 32 or more hidden layers. In some cases, each stage includes eight to 64 hidden layers. In some cases, the neural network applies a sigmoid activation function, an Adam optimizer, an adaptive learning rate, a dynamic threshold, binary classification, a binary cross-entropy loss function, or any combination thereof. In some cases, the non-transitory computer-readable storage medium contains an algorithm for normalizing the intensity of the captured image. In some cases, the neural network is applied to normalize the image. In some cases, the neural network includes an input layer composed of one or more parameters computed from the generated image data.

[0032] In some embodiments, the image data includes one or more waveforms generated in response to the measurement light. In some cases, the non-transitory computer-readable storage medium includes an algorithm for generating an image of the cell from the image data. In some cases, the non-transitory computer-readable storage medium includes an algorithm for calculating image parameters based on one or more of the image data and the cell image. In some cases, the non-transitory computer-readable storage medium includes an algorithm for evaluating the morphology of the cell. In some cases, the non-transitory computer-readable storage medium includes an algorithm for evaluating the cell morphology by determining one or more of the cell size, the cell shape, and the degree of pointillism of the cell. In some cases, the non-transitory computer-readable storage medium includes an algorithm for calculating cell parameters based on the graphic parameters. In some cases, the image parameters are quantized image parameters. In some cases, the parameters are selected from centroid, centroid change, diffusivity, eccentricity, major axis moment, maximum intensity, radial moment, minor axis moment, size, total intensity, light loss from the cell nucleus, forward scattered light from the cell nucleus, side scattered light from the cell nucleus, and combinations thereof. In some cases, the non-transitory computer-readable storage medium includes an algorithm for calculating the image parameters based on an image generated from fluorescence measured from the cell, light loss measured from the cell, forward scattered light measured from the cell, side scattered light measured from the cell, and combinations thereof.

[0033] In some embodiments, the non-transitory computer-readable storage medium includes an algorithm for classifying cells of the sample based on the calculated cell parameters. In some cases, the non-transitory computer-readable storage medium includes an algorithm for classifying the cells based on a waveform generated from the measurement light, a generated cell image, image parameters calculated from the generated cell image, or a combination thereof. In some cases, the non-transitory computer-readable storage medium includes an algorithm for classifying the cells by assigning them to one or more particle population clusters. In some cases, the non-transitory computer-readable storage medium includes an algorithm for determining one or more sorting gates for classifying the cells in the sample. In some cases, the non-transitory computer-readable storage medium includes an algorithm for calculating one or more sorting gates for capturing circulating tumor cell population clusters and excluding normal cell population clusters. In some cases, the non-transitory computer-readable storage medium includes an algorithm for calculating a sorting gate that maximizes the inclusion rate of circulating tumor cell population clusters. In some cases, the non-transitory computer-readable storage medium includes an algorithm for calculating a sorting gate that maximizes the exclusion of normal cell population clusters.

[0034] In some cases, the non-transitory computer-readable storage medium includes algorithms for sorting cells of the sample into multiple sample containers. In some cases, the non-transitory computer-readable storage medium includes algorithms for sorting cells based on the presence, classification, or both of the cells. In some cases, the non-transitory computer-readable storage medium includes algorithms for sorting cells based on the generated cell image. In some cases, the non-transitory computer-readable storage medium includes algorithms. Attached Figure Description

[0035] A better understanding of the invention can be achieved by reading the following detailed description in conjunction with the accompanying drawings. The drawings include the following figures:

[0036] Figure 1A A neural network according to certain implementations is described, which is applied to captured images to determine the classification of cancer cells in a sample. Figure 1B A flowchart illustrating the classification of cancer cells in a sample according to certain implementation schemes is depicted.

[0037] Figure 2 A flow cytometry system according to certain embodiments is shown.

[0038] Figure 3A An image-assisted particle sorter according to certain implementation schemes is described. Figure 3B Image-assisted particle sorting data processing according to certain implementation schemes is described.

[0039] Figure 4A A functional block diagram of a particle analysis system according to certain implementation schemes is depicted. Figure 4B A flow cytometer according to certain implementation schemes is described.

[0040] Figure 5 A functional block diagram of an example particle analyzer control system according to certain implementations is shown.

[0041] Figure 6A A schematic diagram of a particle sorting system according to certain implementation schemes is depicted.

[0042] Figure 6B A schematic diagram of a particle sorting system according to certain implementation schemes is depicted.

[0043] Figure 7 A block diagram of a computing system according to certain implementation schemes is depicted.

[0044] Figure 8A Captured images and accuracy evaluations are depicted when classifying cancer cells in samples according to certain implementation schemes. Figure 8BThe classification of T lymphocyte cancer cells using a neural network trained on samples labeled with fluorophore-based truth markers according to certain implementation schemes is described.

[0045] Figure 9A A diagram is depicted showing the gating of whole blood samples with cancer cell lines according to certain implementation schemes. Figure 9B-9E A diagram illustrating gating of circulating tumor cells according to certain implementation schemes is depicted. Figure 9B Cells derived from the ovarian adenocarcinoma cell line (OVCAR) were depicted. Figure 9C Cells derived from the colorectal adenocarcinoma cell line (HT-29) were depicted. Figure 9D Cells derived from the mammalian adenocarcinoma cell line (MCF7) were depicted. Figure 9E Cells derived from the T-cell leukemia cell line (Jurkat cells) were depicted.

[0046] Figures 10A-10E An image confusion matrix is ​​depicted for circulating tumor cell types detected by a neural network architecture according to certain implementation schemes. Figure 10A Images of whole blood cells and a confusion matrix guide are provided. Figure 10B Cell images and confusion matrices of ovarian adenocarcinoma cell lines were depicted. Figure 10C Cell images and confusion matrices of colorectal adenocarcinoma cell lines were depicted. Figure 10D Cell images and confusion matrices of mammalian adenocarcinoma cell lines were depicted. Figure 10E Cell images and confusion matrices of T-cell leukemia cell lines were depicted. Detailed Implementation

[0047] This disclosure includes a system for detecting and classifying cancer cells in a flowing stream of whole blood. According to some embodiments, the system includes a light source configured to irradiate cells of a whole blood sample in a flowing stream; a light detection system including a photodetector that captures an image of light emitted from the irradiated cells; and a processor having a memory operatively coupled to the processor, wherein the memory contains instructions stored thereon that, when executed by the processor, cause the processor to apply a neural network to the captured image to detect the presence of cancer cells in the whole blood sample with a minimum concentration of cancer cells and to determine a classification of the cancer cells among multiple classifications, wherein the neural network is trained using images of samples labeled with a fluorophore benchmark ground truth marker. Methods for detecting and classifying cancer cells in whole blood samples using a subject system are also described. A non-transitory computer-readable storage medium is also provided.

[0048] Before describing the invention in more detail, it should be understood that the invention is not limited to the specific embodiments described, and therefore variations are possible. It should also be understood that the terminology used herein is for the purpose of describing specific embodiments only and is not intended to be limiting, as the scope of the invention will be limited only by the appended claims.

[0049] When a numerical range is provided, it should be understood that every intermediate value between the upper and lower limits of the range (unless otherwise explicitly stated by the context, up to one-tenth of the lower limit unit) and any other stated value or intermediate value within the stated range are covered by this invention. The upper and lower limits of these smaller ranges may be independently included within that smaller range and also covered by this invention, but are subject to any specific exclusions within the emphasized range. When the emphasized range includes one or two limits, the range excluding one or both of the included limits is also included in this invention.

[0050] The numerical ranges given in this article are preceded by the term "approximately". The term "approximately" is used to provide literal support for the exact number that follows, as well as numbers that are close to or approximate to the number following the term. In determining whether a number is close to or approximates a specifically listed number, an unlisted number that is close to or approximates can be a number that provides a basic equivalent to the specifically listed number in the presented context.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Representative illustrative methods and materials are now described, although any methods and materials similar to or equivalent to those described herein may also be used in the practice or testing of this invention.

[0052] All publications and patents referenced in this specification are incorporated herein by reference as if each individual publication or patent were specifically and individually indicated to be incorporated by reference, and are incorporated herein by reference to disclose and describe the methods and / or materials relating to the publication reference. References to any publication refer to its disclosure prior to the filing date and should not be construed as an admission that the invention is not entitled to a prior invention prior to such publication. Furthermore, the publication dates provided may differ from the actual publication dates, which may require independent verification.

[0053] It should be noted that, as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural indicators unless the context clearly indicates otherwise. It should also be noted that claims may be drafted to exclude any optional elements. Therefore, this statement is intended to serve as a precondition for combining elements of the claim using exclusive terms such as “unique,” ​​“only,” or using a negative limiting term.

[0054] As will be apparent to those skilled in the art upon reading this disclosure, each of the various embodiments described and illustrated herein has discrete components and features that can be readily separated from or combined with features of any of the other several embodiments without departing from the scope or spirit of the invention. Any described method may be performed in the order of the described events or in any other logically possible order.

[0055] For the sake of grammatical fluency and functional interpretation, the apparatus and method have been or will be described. However, it should be clearly understood that, unless expressly stated in accordance with 35 USC §112, the claims should not be construed as necessarily being limited in any way by the interpretation of “apparatus” or “step”, but should be given the full meaning and scope of the definitions provided by the claims under the doctrine of equivalents, and where the claims are expressly stated in accordance with 35 USC §112, they should be given all legal equivalents under 35 U.S.SC §112.

[0056] A system for classifying cancer cells in whole blood samples from a flowing stream.

[0057] As summarized above, aspects of this disclosure include systems for detecting and classifying cancer cells in a flowing stream of whole blood samples. According to certain embodiments, the system includes a light source configured to irradiate cells of a whole blood sample in a flowing stream; a light detection system comprising a photodetector capturing an image of light emitted from the irradiated cells; and a processor having a memory operatively coupled to the processor, wherein the memory contains instructions stored thereon that, when executed by the processor, cause the processor to apply a neural network to the captured image to detect the presence of cancer cells in the whole blood sample with a minimum concentration of cancer cells and to determine the classification of the cancer cells among multiple classifications. In some cases, the neural network is trained using images of samples labeled with a fluorophore benchmark ground truth marker.

[0058] In some cases, the system is configured to classify cancer cells using only high-order imaging features of the computed image parameters. In some implementations, the subject system provides a high-throughput and robust method for generating parameters from image data that allow assessment of the type and morphological characteristics of cancer cells in a sample. In some cases, the subject system is configured to classify rare cells in a sample (and sort them as described below). In some cases, the system allows identification of cancer cells, such as circulating tumor cells, in a sample. Therefore, the subject system can be used for diagnosis, and in some cases, for cancer staging in subjects from whom samples are obtained.

[0059] In some implementations, the system described herein can improve the accuracy of classifying cancer cells in a sample by 5% or more, such as 10% or more, 15% or more, 25% or more, 50% or more, 75% or more, and including 99% or more. In some cases, the subject-matter approach allows for monitoring disease progression, such as determining cancer staging or identifying the extent of cancer metastasis.

[0060] In some implementations, the subject method can provide high-throughput cell sampling, for example, wherein cells in a sample can be sorted multiple times within a timeframe of 10 minutes or less, such as 9 minutes or less, such as 8 minutes or less, such as 7 minutes or less, such as 6 minutes or less, such as 5 minutes or less, such as 4 minutes or less, such as 3 minutes or less, such as 2 minutes or less, and including 1 minute or less. As described in more detail below, in some cases, cancer cells sorted in the sample can be further processed or prepared, for example, wherein 50% or more, such as 60% or more, 70% or more, 80% or more, 90% or more, 95% or more, 97% or more, and including 99% or more cancer cells in the sample are suitable for downstream biological assays.

[0061] In the implementation scheme, the sample irradiated in the flow stream can be a biological sample. A "biological sample" can refer to a whole organism, plant, fungus, or subset of animal tissue, cells, or components, and in some cases may be present in blood, mucus, lymph, synovial fluid, cerebrospinal fluid, saliva, bronchoalveolar lavage fluid, amniotic fluid, amniotic cord blood, urine, vaginal secretions, and semen. Therefore, a "biological sample" refers both to a natural organism or a subset of its tissues and to homogenates, lysates, or extracts prepared from an organism or a subset of its tissues, including but not limited to, plasma, serum, cerebrospinal fluid, lymph, sections (skin, respiratory, gastrointestinal, cardiovascular, and genitourinary tracts), tears, saliva, breast milk, blood cells, tumors, and organs. Biological samples can be any type of biological tissue, including healthy tissue and diseased tissue (e.g., cancerous, malignant, necrotic, etc.). In some implementations, the biological sample is a liquid sample, such as blood or its derivatives, such as plasma, tears, urine, semen, etc. In some cases, the sample is a blood sample, including whole blood, such as blood obtained from venipuncture or finger prick (where the blood may or may not be mixed with any reagents such as preservatives, anticoagulants, etc., before testing). In some cases, the sample is a whole blood sample.

[0062] In some embodiments, the source of the sample is a “mammal” or “milk animal,” terms widely used to describe organisms in the class Mammalia, including carnivores (e.g., dogs and cats), rodents (e.g., mice, guinea pigs, and rats), and primates (e.g., humans, chimpanzees, and monkeys). In some cases, the subject is a human. The method can be applied to samples obtained from human subjects of both sexes at any developmental stage (i.e., newborns, infants, adolescents, teenagers, and adults), wherein in some embodiments, the human subject is an adolescent, teenager, or adult. While the invention is applicable to samples from human subjects, it should be understood that the method can also be performed on samples from other animal subjects (i.e., “non-human subjects”), such as, but not limited to, birds, mice, rats, dogs, cats, livestock, and horses.

[0063] In some cases, cancer cells in a sample constitute 1% or less of the cells in the sample (e.g., a whole blood sample), for example, 0.5% or less, 0.1% or less, 0.05% or less, 0.01% or less, 0.005% or less, 0.001%, and include cases where cancer cells constitute 0.0001% of the cells in the sample. In some cases, the sample contains a minimum concentration of cancer cells, for example, at least about 0.01%, at least about 0.05%, at least about 0.1%, at least about 0.5%, at least about 1%, at least about 2%, at least about 3%, at least about 4%, at least about 5%, at least about 6%, at least about 7%, at least about 8%, at least about 9%, and includes a minimum concentration of at least about 10%. In some cases, the minimum concentration of cancer cells in the sample is from 0.1% to 10%, for example, from 0.5% to 5%.

[0064] In some implementations, the memory contains instructions for identifying rare cells in the sample. The term "rare cell" as used herein is used in its conventional sense, referring to a cell that constitutes 3% or less, for example 2.5% or less, for example 2% or less, for example 1.5% or less, for example 1% or less, for example 0.5% or less, for example 0.1% or less, for example 0.05% or less, for example 0.01% or less, for example 0.005% or less, for example 0.001%, and includes cases where rare cells constitute 0.0001% or less of the cells in the sample. In some cases, the cancer cells in the sample are circulating tumor cells. In some cases, circulating tumor cells refer to tumor cells present in biological fluid samples that have leaked from a primary tumor within the subject. In some cases, circulating tumor cells are present in biological samples from the subject's circulatory system (e.g., blood samples such as whole blood samples, plasma samples, etc.). In some cases, circulating tumor cells are present in biological samples from the subject's lymphatic system (e.g., lymph). In some cases, the cells are ovarian cancer cells, T-lymphocytic cancer cells, breast cancer cells, colon cancer cells, or epithelial cancer cells. In some cases, circulating tumor cells include one or more OVCAR cells, Jurkat cells, MCF-7 cells, and colon adenocarcinoma cells. In some cases, circulating tumor cells are homogeneous clusters. In some cases, circulating tumor cells are heterogeneous clusters.

[0065] In implementations, the light source can be any suitable broadband or narrowband light source. Depending on the components in the sample (e.g., cells exhibiting intercellular interactions, cells demonstrating intracellular transport, etc.), the light source can be configured to emit light of multiple wavelengths, ranging from 200 nm to 1500 nm, for example 250 nm to 1250 nm, for example 300 nm to 1000 nm, for example 350 nm to 900 nm, and including 400 nm to 800 nm. For example, the light source can include a broadband light source emitting light with wavelengths from 200 nm to 900 nm. In other cases, the light source includes a narrowband light source emitting light with wavelengths ranging from 200 nm to 900 nm. For example, the light source can be a narrowband LED (1 nm - 25 nm) emitting light with wavelengths ranging from 200 nm to 900 nm. In some implementations, the light source is a laser. In some cases, the subject system is a gas laser, such as a helium-neon laser, an argon laser, a krypton laser, a xenon laser, a nitrogen laser, a CO2 laser, a CO laser, an argon-fluorine (ArF) excimer laser, a krypton-fluorine (KrF) excimer laser, a xenon-chlorine (XeCl) excimer laser, or a xenon-fluorine (XeF) excimer laser, or combinations thereof. In other cases, the subject system includes dye lasers, such as stilbene, coumarin, or rhodamine lasers. In still other cases, lasers of interest include metal vapor lasers, such as helium-cadmium (HeCd) lasers, helium-mercury (HeHg) lasers, helium-selenium (HeSe) lasers, helium-silver (HeAg) lasers, strontium lasers, neon-copper (NeCu) lasers, copper lasers, or gold lasers, or combinations thereof. In other cases, the subject system includes solid-state lasers, such as ruby ​​lasers, Nd:YAG lasers, NdCrYAG lasers, Er:YAG lasers, Nd:YLF lasers, Nd:YVO4 lasers, Nd:yCa4O(BO3)3 lasers, Nd:YCOB lasers, Ti:sapphire lasers, thulium YAG lasers, ytterbium YAG lasers, ytterbium₂O₃ lasers, or cerium-doped lasers and combinations thereof.

[0066] In other embodiments, the light source is a non-laser light source, such as a lamp, including but not limited to halogen lamps, deuterium arc lamps, xenon arc lamps, light-emitting diodes (LEDs), such as broadband LEDs with a continuous spectrum, superluminescent LEDs, semiconductor LEDs, broadband LED white light sources, and multi-LED integration. In some cases, the non-laser light source is a stable fiber-coupled broadband light source, a white light source, and other light sources or any combination thereof.

[0067] The light source can be positioned at any suitable distance from the sample (e.g., the flow stream in a flow cytometer), for example, at a distance of 0.001 mm or greater, such as 0.005 mm or greater, such as 0.01 mm or greater, such as 0.05 mm or greater, such as 0.1 mm or greater, such as 0.5 mm or greater, such as 1 mm or greater, such as 5 mm or greater, such as 10 mm or greater, such as 25 mm or greater, and including at a distance of 100 mm or greater. Furthermore, the light source irradiates the sample at any suitable angle (e.g., relative to the vertical axis of the flow stream), for example, in the range of 10° to 90°, such as 15° to 85°, such as 20° to 80°, such as 25° to 75°, and including angles of 30° to 60°, such as 90°.

[0068] The light source can be configured to irradiate the sample continuously or at discontinuous intervals. In some cases, the system includes a light source configured to continuously irradiate the sample, such as a continuous-wave laser that continuously irradiates the flow at the probe point in a flow cytometer. In other cases, the system of interest includes a light source configured to irradiate the sample at discontinuous intervals, such as every 0.001 ms, every 0.01 ms, every 0.1 ms, every 1 ms, every 10 ms, every 100 ms, and including every 1000 ms, or some other interval. When the light source is configured to irradiate the sample at discontinuous intervals, the system may include one or more additional components to intermittently irradiate the sample with the light source. For example, the subject system in these embodiments may include one or more laser beam choppers, manually or computer-controlled beam stoppers, for shielding the sample and exposing the sample to the light source.

[0069] In some implementations, the light source is a laser. Lasers of interest may include pulsed lasers or continuous-wave lasers. For example, the laser may be a gas laser, such as a helium-neon laser, an argon laser, a krypton laser, a xenon laser, a nitrogen laser, a CO2 laser, a CO laser, an argon-fluorine (ArF) excimer laser, a krypton-fluorine (KrF) excimer laser, a xenon-chlorine (XeCl) excimer laser, or a xenon-fluorine (XeF) excimer laser, or a combination thereof; a dye laser, such as a stilbene, coumarin, or rhodamine laser; or a metal vapor laser, such as a helium-cadmium (HeCd) laser, a helium-mercury (HeHg) laser, a helium-selenium (HeSe) laser, a helium-silver (HeAg) laser, a strontium laser, or a neon-copper (Neon) laser. NeCu lasers, copper lasers, or gold lasers and combinations thereof; solid-state lasers, such as ruby ​​lasers, Nd:YAG lasers, NdCrYAG lasers, Er:YAG lasers, Nd:YLF lasers, Nd:YVO4 lasers, Nd:yCa4O(BO3)3 lasers, Nd:YCOB lasers, Ti:sapphire lasers, thallium YAG lasers, ytterbium YAG lasers, ytterbium₂O₃ lasers, or cerium-doped lasers and combinations thereof; semiconductor diode lasers, optically pumped semiconductor lasers (OPSL), or frequency-doubled or third-doubled embodiments of any of the above lasers.

[0070] In some embodiments, the light source is a beam generator configured to generate two or more frequency-shifted beams. In some cases, the beam generator includes a laser, a radio frequency (RF) generator configured to apply an RF drive signal to the acousto-optic device to generate laser beams with two or more angle deflections. In these embodiments, the laser can be a pulsed laser or a continuous-wave laser. For example, the laser in the beam generator of interest can be a gas laser, such as a helium-neon laser, an argon laser, a krypton laser, a xenon laser, a nitrogen laser, a CO2 laser, a CO laser, an argon-fluorine (ArF) excimer laser, a krypton-fluorine (KrF) excimer laser, a xenon-chlorine (XeCl) excimer laser, or a xenon-fluorine (XeF) excimer laser or a combination thereof; a dye laser, such as a stilbene, coumarin, or rhodamine laser; or a metal vapor laser, such as a helium-cadmium (HeCd) laser, a helium-mercury (HeHg) laser, or a helium-selenium (H) laser. eSe) lasers, helium-silver (HeAg) lasers, strontium lasers, neon-copper (NeCu) lasers, copper lasers, or gold lasers and combinations thereof; solid-state lasers, such as ruby ​​lasers, Nd:YAG lasers, NdCrYAG lasers, Er:YAG lasers, Nd:YLF lasers, Nd:YVO4 lasers, Nd:yCa4O(BO3)3 lasers, Nd:YCOB lasers, Ti:sapphire lasers, thallium YAG lasers, ytterbium YAG lasers, ytterbium₂O₃ lasers, or cerium-doped lasers and combinations thereof.

[0071] The acousto-optic device can be any convenient acousto-optic scheme configured to frequency-shift a laser using applied acoustic waves. In some embodiments, the acousto-optic device is an acousto-optic deflector. The acousto-optic device in the subject system is configured to generate an angle-deflected laser beam from light from a laser and an applied radio frequency (RF) drive signal. The RF drive signal can be applied to the acousto-optic device using any suitable RF drive signal source, such as a direct digital synthesizer (DDS), arbitrary waveform generator (AWG), or electrical pulse generator.

[0072] In an embodiment, the controller is configured to apply radio frequency drive signals to the acousto-optic device to generate a desired number of angle-deflected laser beams in the output laser beam, for example, to apply 3 or more radio frequency drive signals, such as 4 or more radio frequency drive signals, such as 5 or more radio frequency drive signals, such as 6 or more radio frequency drive signals, such as 7 or more radio frequency drive signals, such as 8 or more radio frequency drive signals, such as 9 or more radio frequency drive signals, such as 10 or more radio frequency drive signals, such as 15 or more radio frequency drive signals, such as 25 or more radio frequency drive signals, such as 50 or more radio frequency drive signals, and includes being configured to apply 100 or more radio frequency drive signals.

[0073] In some cases, in order to generate an intensity distribution of an angle-deflected laser beam in the output laser beam, the controller is configured to apply an RF drive signal having an amplitude that varies from about 0.001 V to about 500 V (e.g., about 0.005 V to about 400 V, about 0.01 V to about 300 V, about 0.05 V to about 200 V, about 0.1 V to about 100 V, about 0.5 V to about 75 V, about 1 V to 50 V, about 2 V to 40 V, about 3 V to about 30 V, and including about 5 V to about 25 V). In some implementations, the radio frequency drive signal for each application has a frequency of about 0.001 MHz to about 500 MHz, for example about 0.005 MHz to about 400 MHz, for example about 0.01 MHz to about 300 MHz, for example about 0.05 MHz to about 200 MHz, for example about 0.1 MHz to about 100 MHz, for example about 0.5 MHz to about 90 MHz, for example about 1 MHz to about 75 MHz, for example about 2 MHz to about 70 MHz, for example about 3 MHz to about 65 MHz, for example about 4 MHz to about 60 MHz, and includes about 5 MHz to about 50 MHz.

[0074] In some embodiments, the controller has a processor with a memory operatively coupled to the processor, such that the memory contains instructions stored thereon that, when executed by the processor, cause the processor to generate an output laser beam having angle-deflected laser beams with a desired intensity distribution. For example, the memory may include instructions to generate two or more angle-deflected laser beams of equal intensity (e.g., three or more, four or more, five or more, ten or more, 25 or more, or 50 or more), and the memory may include instructions to generate 100 or more angle-deflected laser beams of equal intensity. In other embodiments, the memory may include instructions to generate two or more angle-deflected laser beams of different intensities (e.g., three or more, four or more, five or more, ten or more, 25 or more, or 50 or more), and the memory may include instructions to generate 100 or more angle-deflected laser beams of different intensities.

[0075] In some embodiments, the controller has a processor with a memory operatively coupled to the processor, such that the memory contains instructions stored thereon that, when executed by the processor, cause the processor to generate an output laser beam with increasing intensity along a horizontal axis from its edge to its center. In these cases, the intensity of the angularly deflected laser beam at the center of the output beam can be from 0.1% to about 99%, for example from 0.5% to about 95%, for example from 1% to about 90%, for example from about 2% to about 85%, for example from about 3% to about 80%, for example from about 4% to about 75%, for example from about 5% to about 70%, for example from about 6% to about 65%, for example from about 7% to about 60%, for example from about 8% to about 55%, and from about 10% to about 50% of the intensity of the angularly deflected laser beam at the edge of the output laser beam along the horizontal axis. In other embodiments, the controller has a processor with a memory operatively coupled to the processor, such that the memory contains instructions stored thereon that, when executed by the processor, cause the processor to generate an output laser beam with increasing intensity along a horizontal axis from its edge to its center. In these cases, the intensity of the angularly deflected laser beam at the edge of the output beam can be from 0.1% to about 99%, for example from 0.5% to about 95%, for example from 1% to about 90%, for example from about 2% to about 85%, for example from about 3% to about 80%, for example from about 4% to about 75%, for example from about 5% to about 70%, for example from about 6% to about 65%, for example from about 7% to about 60%, for example from about 8% to about 55%, and from about 10% to about 50% of the intensity of the angularly deflected laser beam at the center of the output laser beam along the horizontal axis. In some other embodiments, the controller has a processor with a memory operatively coupled to the processor, such that the memory contains instructions stored thereon, which, when executed by the processor, cause the processor to generate an output laser beam with a Gaussian intensity distribution along a horizontal axis. In still other embodiments, the controller has a processor with a memory operatively coupled to the processor, such that the memory contains instructions stored thereon, which, when executed by the processor, cause the processor to generate an output laser beam with a top-hat intensity distribution along a horizontal axis.

[0076] In implementations, the beam generator of interest can be configured to generate angle-deflected laser beams within spatially separated output laser beams. Depending on the applied RF drive signal and the desired irradiance distribution of the output laser beams, the angle-deflected laser beams can be spaced apart by 0.001 μm or more, for example, 0.005 μm or more, for example, 0.01 μm or more, for example, 0.05 μm or more, for example, 0.1 μm or more, for example, 0.5 μm or more, for example, 1 μm or more, for example, 5 μm or more, for example, 10 μm or more, for example, 100 μm or more, for example, 500 μm or more, for example, 1000 μm or more, and including 5000 μm or more. In some implementations, the system is configured to generate angle-deflected laser beams within the output laser beams, which overlap, for example, with adjacent angle-deflected laser beams along the horizontal axis of the output laser beams. The overlap between adjacent angle-deflected laser beams (e.g., overlap of beam points) can be 0.001 μm or greater, such as 0.005 μm or greater, such as 0.01 μm or greater, such as 0.05 μm or greater, such as 0.1 μm or greater, such as 0.5 μm or greater, such as 1 μm or greater, such as 5 μm or greater, such as 10 μm or greater, and includes 100 μm or greater overlap.

[0077] In some cases, the beam generator configured to generate two or more frequency-shifted beams includes a laser excitation module, as described in Diebold et al., Nature Photonics Vol. 7(10); 806-810. The disclosures of (2013) and those set forth in U.S. Patent Nos. 9,423,353, 9,784,661, 9,983,132, 10,006,852, 10,078,045, 10,036,699, 10,222,316, 10,288,546, 10,324,019, 10,408,758, 10,451,538, 10,620,111 and U.S. Patent Publications Nos. 2017 / 0133857, 2017 / 0328826, 2017 / 0350803, 2018 / 0275042, 2019 / 0376895 and 2019 / 0376894 are incorporated herein by reference.

[0078] In the implementation, the system includes a light detection system having multiple photodetectors for measuring light from cells in a sample. In some cases, one or more photodetectors of the light detection system are light-loss photodetectors. In some cases, one or more photodetectors of the light detection system are configured to measure scattered light. In some cases, one or more photodetectors of the light detection system are configured to measure side-scattered light. In some cases, one or more photodetectors of the light detection system are configured to measure forward-scattered light. In some cases, one or more photodetectors of the light detection system are configured to measure back-scattered light. Photodetectors of interest may include, but are not limited to, optical sensors such as active pixel sensors (APS), avalanche photodiodes (APDs), image sensors, charge-coupled devices (CCDs), enhancement-mode charge-coupled devices (ICCDs), light-emitting diodes, photon counters, calorimeters, thermoelectric detectors, photoresistors, photovoltaic cells, photodiodes, photomultiplier tubes, phototransistors, quantum dot photoconductors, or combinations thereof, as well as other photodetectors. In some implementations, a charge-coupled device (CCD), a semiconductor charge-coupled device (CCD), an active pixel sensor (APS), a complementary metal-oxide-semiconductor (CMOS) image sensor, or an N-type metal-oxide-semiconductor (NMOS) image sensor is used to measure light from the sample.

[0079] In some embodiments, the optical detection system of interest comprises multiple photodetectors. In some cases, the optical detection system comprises multiple solid-state detectors, such as photodiodes. In some embodiments, the optical detection system comprises a photodetector array, such as a photodiode array. In these embodiments, the photodetector array may include four or more photodetectors, such as 10 or more, 25 or more, 50 or more, 100 or more, 250 or more, 500 or more, 750 or more, and may include 1000 or more. For example, the detector may be a photodiode array having four or more photodiodes, such as 10 or more, 25 or more, 50 or more, 100 or more, 250 or more, 500 or more, 750 or more, and may include 1000 or more.

[0080] The photodetectors can be arranged in any geometric configuration as needed, with arrangements of interest including, but not limited to, square, rectangular, trapezoidal, triangular, hexagonal, heptagonal, octagonal, nonagonal, decagonal, dodecagonal, circular, elliptical, and irregular pattern configurations. The photodetectors in the photodetector array can be oriented relative to each other (with reference to the XZ plane) at an angular range of 10° to 180° (e.g., 15° to 170°, 20° to 160°, 25° to 150°, 30° to 120°, and including 45° to 90°). The photodetector array can be any suitable shape and can be linear, such as square, rectangular, trapezoidal, triangular, hexagonal, etc.; curved, such as circular, elliptical; and irregular, such as a parabolic base coupled to the top of a plane. In some embodiments, the photodetector array has a rectangular effective surface.

[0081] Each photodetector (e.g., photodiode) in the array may have an effective surface with a width ranging from 5 μm to 250 μm, for example, 10 μm to 225 μm, for example, 15 μm to 200 μm, for example, 20 μm to 175 μm, for example, 25 μm to 150 μm, for example, 30 μm to 125 μm, and including 50 μm to 100 μm, and a length ranging from 5 μm to 250 μm, for example, 10 μm to 225 μm, for example, 15 μm to 200 μm, for example, 20 μm to 175 μm, for example, 25 μm to 150 μm, for example, 30 μm to 125 μm, and including 50 μm to 100 μm, wherein the surface area of ​​each photodetector (e.g., photodiode) in the array ranges from 25 μm to 100 μm. 2 Up to 10000 μm 2 For example, 50 to μm 2 Up to 9000 μm 2 For example, 75 to μm 2 Up to 8000 μm 2 For example, 100 to μm 2 Up to 7000 μm 2 For example, 150 to μm 2 Up to 6000 μm 2 And including 200 to μm 2 Up to 5000 μm 2 .

[0082] The size of the photodetector array can vary depending on the amount and intensity of light, the number of photodetectors, and the required sensitivity, and its length can range from 0.01 mm to 100 mm, for example, 0.05 mm to 90 mm, 0.1 mm to 80 mm, 0.5 mm to 70 mm, 1 mm to 60 mm, 2 mm to 50 mm, 3 mm to 40 mm, 4 mm to 30 mm, and including 5 mm to 25 mm. The width of the photodetector array can also vary, ranging from 0.01 mm to 100 mm, for example, 0.05 mm to 90 mm, 0.1 mm to 80 mm, 0.5 mm to 70 mm, 1 mm to 60 mm, 2 mm to 50 mm, 3 mm to 40 mm, 4 mm to 30 mm, and including 5 mm to 25 mm. Therefore, the effective surface area of ​​the photodetector can range from 0.1 mm. 2 Up to 10000 mm 2 For example, 0.5 mm 2 Up to 5000 mm 2 For example, 1 mm 2 Up to 1000 mm 2 For example, 5 mm 2 Up to 500 mm 2 And includes 10 mm 2 Up to 100 mm 2 .

[0083] The photodetector of interest is configured to measure collected light at one or more wavelengths (e.g., two or more wavelengths, five or more different wavelengths, ten or more different wavelengths, 25 or more different wavelengths, 50 or more different wavelengths, 100 or more different wavelengths, 200 or more different wavelengths, 300 or more different wavelengths), and includes measuring light emitted by a sample in a flowing stream at 400 or more different wavelengths.

[0084] In some embodiments, the photodetector is configured to measure light collected within a wavelength range (e.g., 200 nm–1000 nm). In some embodiments, the photodetector of interest is configured to collect the spectrum of light within a wavelength range. For example, the system may include one or more detectors configured to collect the spectrum of light within one or more wavelength ranges of 200 nm–1000 nm. In other embodiments, the detector of interest is configured to measure light from a sample in a flowing stream at one or more specific wavelengths. For example, the system may include one or more detectors configured to measure light at one or more of the following wavelengths: 450 nm, 518 nm, 519 nm, 561 nm, 578 nm, 605 nm, 607 nm, 625 nm, 650 nm, 660 nm, 667 nm, 670 nm, 668 nm, 695 nm, 710 nm, 723 nm, 780 nm, 785 nm, 647 nm, 617 nm, and any combination thereof.

[0085] The optical detection system is configured to measure light continuously or at discontinuous intervals. In some cases, the photodetector of interest is configured to measure the collected light continuously. In other cases, the optical detection system is configured to perform measurements at discontinuous intervals, such as every 0.001 milliseconds, every 0.01 milliseconds, every 0.1 milliseconds, every 1 millisecond, every 10 milliseconds, every 100 milliseconds, and including every 1000 milliseconds or some other interval.

[0086] In one embodiment, the memory includes instructions for detecting the presence of cancer cells in a sample (e.g., a whole blood sample) at a minimum concentration, such as at least about 0.01%, at least about 0.05%, at least about 0.1%, at least about 0.5%, at least about 1%, at least about 2%, at least about 3%, at least about 4%, at least about 5%, at least about 6%, at least about 7%, at least about 8%, at least about 9%, and including a minimum concentration of at least about 10%. In some cases, the memory includes instructions for detecting the presence of cancer cells having a minimum concentration in the sample of 0.1% to 10%, for example, 0.5% to 5%.

[0087] In some embodiments, the system includes a processor containing memory operatively coupled to the processor, wherein the memory contains instructions stored thereon that, when executed by the processor, cause the processor to apply a neural network to the captured image to detect the presence of cancer cells in a whole blood sample with a minimum concentration of cancer cells. In some embodiments, the memory also contains instructions for determining cancer cell classification. In some embodiments, the memory contains instructions for classifying cancer cells in the sample based on image data (e.g., waveforms generated in each photodetector channel), generated cell images, calculated image parameters, or combinations thereof. In some cases, the multiple classifications include two or more classifications, such as three or more classifications, such as four or more classifications, such as five or more classifications, such as six or more classifications, such as seven or more classifications, such as eight or more classifications, such as nine or more classifications, and include ten or more classifications. In some cases, the multiple classifications include ovarian cancer classification, T-lymphocytic carcinoma classification, breast cancer classification, colon cancer classification, or epithelial carcinoma classification. In some cases, the multiple classifications include OVCAR classification, Jurkat classification, MCF-7 classification, and adenocarcinoma classification.

[0088] In some implementations, the memory contains instructions for applying machine learning algorithms to captured images to determine cancer cell classification. Any convenient machine learning algorithm can be implemented, including but not limited to linear regression, logistic regression, Naive Bayes, k-nearest neighbors (kNN), random forest, decision tree, support vector machine, gradient boosting, and clustering algorithms. In some implementations, the system is configured to apply neural networks, such as artificial neural networks, convolutional neural networks, or recurrent neural networks. In some implementations, the system is configured to implement Python scripts. In some implementations, the system is configured to apply artificial neural networks (e.g., convolutional neural networks, etc.), Bayesian statistics, learning automata, hidden Markov models, linear classifiers, quadratic classifiers, association rule learning, and / or similar techniques.

[0089] In some implementations, a neural network comprises a network of nodes. Nodes can be organized into layers, where the first layer is the input layer, from which data flows. The neural network can also include output layers from which transformed data flows out. Each node can have multiple inputs and a single output (e.g., input layer nodes have only a single input). The output of a node represents a linear combination of its inputs. In other words, the input can be multiplied by an associated constant. The product can be accumulated along a node path with a constant offset. This constant offset, or “bias,” can represent another degree of freedom that can be adjusted during training. For example, the constant offset could be a threshold, as it can reduce node values ​​below zero, causing the activation function to output zero.

[0090] The obtained value is evaluated using an activation function and used as the node's output. Within a neural network, nodes in a given layer are connected to each node in adjacent layers. A neural network can be trained by using gradient descent and an error function to compare the network's expected output with its actual output, thereby minimizing the network error. The weights of one or more nodes can be adjusted to simulate the desired result produced by the network.

[0091] In some cases, the memory contains instructions for classifying cells based on one or more parameters computed from generated image data. In some cases, the memory contains instructions for applying a classifier neural network to classify cells. In some cases, the neural network includes an input layer consisting of one or more parameters computed from the generated image data. In some cases, the input layer contains only imaging parameters. In some cases, the neural network includes an output layer that provides a probability estimate based on cell identity (e.g., the cell is a cancer cell, circulating tumor cell). In some cases, the neural network includes a learning algorithm configured to learn from classification errors. In some cases, the memory contains instructions for updating classification parameters based on one or more parameters computed from the generated image data. In some cases, the update of classification parameters is based on a comparison between one or more parameters computed from the generated image data and one or more parameters computed from ground truth image data. In some cases, the neural network includes a learning algorithm configured to learn from classification errors. In some cases, the memory contains instructions for generating ground truth image data. In some cases, the memory contains instructions for updating classification parameters based on one or more parameters computed from captured images. In some cases, the update of classification parameters is based on a comparison between one or more parameters calculated from the captured image and one or more parameters calculated from the baseline ground truth image data.

[0092] In some cases, samples are labeled with fluorophore-based truth markers. In some implementations, specific fluorescent dyes are used as truth markers to identify different cancer cell types during training. In some cases, fluorescent dyes of interest may include, but are not limited to: bodipyrrole dyes, coumarin dyes, rhodamine dyes, acridine dyes, anthraquinone dyes, arylmethane dyes, diarylmethane dyes, chlorophyll-containing dyes, triarylmethane dyes, azo dyes, diazo dyes, nitro dyes, nitroso dyes, phthalocyanine dyes, anthocyanine dyes, asymmetric anthocyanine dyes, quinone imine dyes, azin dyes, eurhodin dyes, safranin dyes, indomethacin dyes, indophenol dyes, fluorine dyes, oxazine dyes, oxazolone dyes, thiazine dyes, thiazolium dyes, xanthine dyes, fluorene dyes, pyrrolidine dyes, fluorine dyes, rhodamine dyes, phenanthridine dyes, squaric acid cyanine, bodipyrroles, roxitanes, naphthalene, coumarin, oxadiazole, anthracene, pyrene, acridine, arylmethyne, or tetrapyrroles and combinations thereof. In some embodiments, the conjugate may comprise two or more dyes, such as those selected from boroquinone dipyrrole dyes, coumarin dyes, rhodamine dyes, acridine dyes, anthraquinone dyes, arylmethane dyes, diarylmethane dyes, chlorophyll-containing dyes, triarylmethane dyes, azo dyes, diazo dyes, nitro dyes, nitroso dyes, phthalocyanine dyes, cyanine dyes, asymmetric cyanine dyes, quinone imine dyes, azin dyes, eurhodin dyes, safranin dyes, indigo dyes, indophenol dyes, fluorine dyes, oxazine dyes, oxazolone dyes, thiazine dyes, thiazolium dyes, xanthine dyes, fluorene dyes, pyrrolin dyes, fluorine dyes, rhodamine dyes, phenanthridine dyes, squaric acid cyanine, boroquinone dipyrroles, roxitanes, naphthalene, coumarin, oxadiazole, anthracene, pyrene, acridine, arylmethyst, or tetrapyrroles and combinations thereof.

[0093] In some embodiments, the fluorescent dyes of interest may include, but are not limited to, fluorescein isothiocyanate (FITC), phycoerythrin (PE) dyes, polydinophytoxanthophyll-chlorophyll-cyanine dyes (e.g., PerCP-Cy5.5), phycoerythrin-cyanine dyes (PE-Cy7), allophycocyanin (APC) dyes (e.g., APC-R700), allophycocyanin-cyanine dyes (e.g., APC-Cy7), and coumarin dyes (e.g., V450 or V500). In some cases, fluorescent dyes may include one or more of the following: 1,4-bis-(o-methylstyryl)benzene (bis-MSB 1,4-bis[2-(2-methylphenyl)vinyl]benzene), C510 dye, C6 dye, Nile Red dye, T614 dye (e.g., N-[7-(methanesulfonamide)-4-oxo-6-phenoxychrome-3-yl]formamide), LDS 821 dye ((2-(6-(p-dimethylaminophenyl)-2,4-penten-1,3,5-hextrienyl)-3-ethylbenzothiazole perchlorate), and mFluor dye (e.g., mFluor Red dye, such as mFluor 780NS).

[0094] Fluorescent dyes of interest may include, but are not limited to, fluorescein, hydroxycoumarin, aminocoumarin, methoxycoumarin, Cascade Blue, Pacific Blue, Pacific Orange, Lucifer yellow, NBD, R-phycoerythrin (PE), PE-Cy5 conjugate, PE-Cy7 conjugate, Red 613, PerCP, TruRed, FluorX, BODIPY-FL, TRITC, X-rhodamine, rhodamine B, Texas Red, allophycocyanin (APC), APC-Cy7 conjugate, Cy2, Cy3, Cy3B, Cy3.5, Cy5, Cy5.5, Cy7, Hoechst 33342, DAPI, Hoechst 33258, SYTOX Blue, chromomycin A3, photomycin, YOYO-1, ethidium bromide, acridine orange, SYTOX Green, TOTO-1, TO-PRO-1, thiazole orange, propidium iodide (PI), and LDS. 751, 7-AAD, SYTOX Orange, TOTO-3, TO-PRO-3, DRAQ5, Indo-1, Fluo-3, DCFH, DHR, SNARF, Y66H, Y66F, EBFP, EBFP2, Azurite Blue, GFPuv, T-Sapphire, TagBFP, Cerulean, mCFP, ECFP, CyPet, Y66W, dKeima-Red, mKeima-Red, TagCFP, AmCyan1, mTFP1 (Teal), S65A, Midoriishi-Cyan, wild-type GFP, S65C, TurboGFP, TagGFP, TagGFP2, AcGFP1, S65L, Emerald, S65T, EGFP, Azami-Green, ZsGreen1, Dronpa-Green, TagYFP, EYFP, Topaz , Venus, mCitrine, YPet, TurboYFP, PhiYFP, PhiYFP-m, ZsYellow1, mBanana, Kusabira-Orange, mOrange, mOrange2, mKO, TurboRFP, tdTomato, DsRed-Express2, TagRFP, DsRed monomer, DsRed2 ("RFP"), mStrawberry, TurboFP602, AsRed2, mRFP1, J-Red, mCherry, HcRed1, mKate2, Katushka (TurboFP635), mKate(TagFP635) 、TurboFP635 、mPlum 、mRaspberry 、mNeptune 、E2-Crimson 、Monochlorobimane 、Fluid Barrier 350 Alexa Fluor 405 AlexaFluor 430, Alexa Fluor 488, Alexa Fluor 500, Alexa Fluor 514, Alexa Fluor 532, Alexa Fluor 546, Alexa Fluor 555, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor610, Alexa Fluor 633, Alexa Fluor 647, Alexa Fluor 660, Alexa Fluor 680, Alexa Fluor 700, Alexa Fluor 750, Alexa Fluor 790 is a HyPer compound with a range of active ingredients: 7-AAD, Alexa Fluor 488, Alexa Fluor 647, and Alexa Fluor 700, AmCyan, APC, APC-Cy7, APC-H7, APC-R700, BB660-P2, BB790-P, BUV395, B UV615, BUV661, BV570, BV605, BV650, BV711, BV750, BV786, BYG584-P, Calcein AM, Calcein Blue AM, CFSE, DAPI, DRAQ5, DRAQ7, FITC, Fluo-4 AM, FVS440UV, FVS450, FVS510, FVS520, FVS570, FVS575V, FVS620, FVS660, FVS700, FVS780, Indo-1 Hi, Indo-1 Lo、JC-1、MythStatus Network、MythStatus TMRE, PacificBlue, PE, PE-CF594, PE-Cy5, PE-Cy7, PerCP, PerCP-Cy5.5, P.S I, R718, RB545, RB613, RB744, RB780, RY586, RY610, V450, V500 Via-Probe Green、Via-Probe Network、VPD450、Alexa Fluor532、Alexa Fluor 561、Alexa Fluor 660、APC-eFluor 780、APC / Fire750, APC / Fire 810, BV785, eBFP, eCFP, eFluor 450, eFluor 506, eFluor 660, eGFP, eYFP, Hoechst 33258, KIRAVIA Blue 520, mCherry, NFB510, NFB530, NFB555, NFB585, NFB610-70S, NFB660-120S, NFR660, NFR 685, NFR700, NFR710, NFY570, NFY590, NFY610, NFY660, NFY690, NFY700, NFY730, Pacific Orange, PE-Cy5.5, PE-eFluor 610, PE / Dazzle 594, PE / Fire 640, PE / Fire 700, PE / Fire 810, PerCP-eFluor 710, SB436, SB600, SB645, SB702, SB780, Spark Blue 550, Spark Blue 574, Spark NIR 685, Spark UV 387, Spark Violet 423, Spark Violet 538, Spark YG 581, Spark YG 593 and tdTomato.

[0095] In some cases, fluorescent dyes are polymer dyes (e.g., fluorescent polymer dyes). A variety of fluorescent polymer dyes are available for use in the subject methods and systems. In some cases of these methods, the polymer dyes include conjugated polymers. Conjugated polymers (CPs) are characterized by a delocalized electronic structure comprising a backbone of alternating unsaturated bonds (e.g., double and / or triple bonds) and saturated bonds (e.g., single bonds), where π electrons can move from one bond to another. Thus, the conjugated backbone can endow polymer dyes with extended linear structures where the bond angles between polymer repeating units are restricted. For example, proteins and nucleic acids, while also polymers, do not form extended rod-like structures in some cases, but rather fold into higher-order three-dimensional shapes. Furthermore, CPs can form “rigid rod” polymer backbones and undergo limited twist (e.g., torsion) angles between monomer repeating units along the polymer backbone. In some cases, polymer dyes include CPs with rigid rod-like structures. The structural features of polymer dyes can influence the fluorescence properties of the molecules.

[0096] Polymer dyes of interest include, but are not limited to, those described in U.S. Patent Nos. 7,270,956; 7,629,448; 8,158,444; 8,227,187; 8,455,613; 8,575,303; 8,802,450; 8,969,509; 9,139,869; 9,371,559; 9,547,008; 10,094,838; 10,302,648; 10,458,989; 10,641,775 and 10,962,546, the disclosures of which are incorporated herein by reference in their entirety; and Gaylord et al., J. Am. Chem. Soc., 2001, 123 (26), pp 6417-6418; Feng et al., Chem. Soc. Rev., 2010, 39, 2411-2419; and Traina et al., J. Am. Chem. Soc., 2011, 133 (32), pp 12600-12607, the entire contents of which are incorporated herein by reference. Specific polymer dyes that may be used include, but are not limited to, BD Horizon Brilliant™ dyes, such as BD Horizon Brilliant™ purple dyes (e.g., BV421, BV510, BV605, BV650, BV711, BV786); BD Horizon Brilliant™ ultraviolet dyes (e.g., BUV395, BUV496, BUV737, BUV805); and BD Horizon Brilliant™ blue dyes (e.g., BB515) (BD Biosciences, San Jose, CA). Any fluorescent dyes known to or not yet discovered by those skilled in the art may be used, including but not limited to the dyes described above.

[0097] In some implementations, the neural network includes a backpropagation neural network. In some cases, the neural network contains two or more stages, such as three or more stages, four or more stages, five or more stages, six or more stages, seven or more stages, eight or more stages, nine or more stages, ten or more stages, twelve or more stages, sixteen or more stages, twenty or more stages, and includes 24 or more stages. In some cases, the neural network contains 2 to 32 stages, such as 2 to 24 stages, or 2 to 16 stages, and includes 2 to 10 stages. In some cases, the neural network includes at least one dropout stage. In some cases, the dropout stage is configured to prevent overfitting during training.

[0098] In some implementations, each stage includes one or more hidden layers, such as two or more hidden layers, three or more hidden layers, four or more hidden layers, five or more hidden layers, six or more hidden layers, seven or more hidden layers, eight or more hidden layers, nine or more hidden layers, ten or more hidden layers, twelve or more hidden layers, sixteen or more hidden layers, twenty-four or more hidden layers, thirty-two or more hidden layers, and up to sixty-four or more hidden layers. In some implementations, each stage includes at least two hidden layers, such as at least four, eight, twelve, sixteen, twenty-four, or thirty-two hidden layers. In some cases, each stage includes eight to sixty-four hidden layers.

[0099] Figure 1A Neural networks according to certain implementations are depicted, which are applied to captured images to determine the classification of cancer cells in a sample. In some cases, the neural network is a backpropagation neural network. In others, the neural network is a forward propagation neural network. In some cases, the input layer consists of one or more parameters computed from the generated image data. The neural network may include multiple stages with hidden neurons (hidden layers) (e.g., four stages). The neural network also includes an output layer, for example, a probability estimate based on cell identity (e.g., the cell is a cancer cell, circulating tumor cell).

[0100] In some cases, neural networks apply sigmoid activation functions, Adam optimizers, adaptive learning rates, dynamic thresholding, binary classification (e.g., positive vs. negative), binary cross-entropy loss functions, or any combination thereof. In some cases, the memory contains instructions for normalizing the intensity of the captured image. In some cases, the neural network is applied to the normalized image. In some cases, the neural network includes an input layer consisting of one or more parameters computed from the generated image data.

[0101] In some implementations, the neural network is trained using sample images labeled with a fluorophore benchmark. In some cases, the memory contains instructions for training the neural network using a dataset reflecting the ratio of positive to negative categories of cancer cells to non-cancer cells. In some cases, the detection accuracy of the neural network is validated on multiple samples containing different types of cancer cells. In some cases, the neural network is configured to distinguish between cancer cells and non-cancer cells based on specific biomarkers. In some cases, the memory contains instructions for normalizing the intensity of captured images. In some cases, the neural network is applied to the normalized images. In some cases, normalized fluorescence intensity of cells improves detection accuracy, for example, by 5% or more, 10% or more, 25% or more, 50% or more, 75% or more, 90% or more, and including improvements of 99% or more.

[0102] In some embodiments, the memory contains instructions for generating one or more images of cells (e.g., cancer cells) in a sample. In some embodiments, the system includes a memory storing instructions for generating an image based on detected light absorption, detected light scattering, detected light emission, or any combination thereof. In some cases, the memory contains instructions for generating an image based on light absorption detected from the sample (e.g., from a bright-field photodetector). In some cases, the memory contains instructions for generating an image based on light scattering detected from the sample (e.g., from a side-scatter detector, a forward-scatter detector, or a combination of both). In some cases, the memory contains instructions for generating an image based on light emitted from the sample. In other cases, the memory contains instructions for generating an image based on a combination of detected light absorption and detected light scattering.

[0103] In some embodiments, the memory includes instructions for generating a single image for each cell (e.g., each cancer cell) in the sample from each detected light form. In other embodiments, the memory includes instructions for generating multiple images for each cell, such multiple images being, for example, two or more images, three or more images, five or more images, ten or more images, and including 25 or more images for each cell. For example, the memory includes instructions for generating a first image of the cell from fluorescence detected from labeled cells; instructions for generating a second image of the cell from detected light absorption; and instructions for generating a third image of the cell from detected light scattering. In other embodiments, the memory includes instructions for generating two or more images from each detected light form, such two or more images being, for example, three or more images, four or more images, five or more images, and including ten or more images or combinations thereof.

[0104] In some cases, the memory contains instructions for generating image data, such as digital waveforms generated in one or more photodetector channels. In some cases, the memory contains instructions for calculating image parameters directly from the waveform. In some cases, the memory contains instructions for calculating image parameters from image data, the generated image, or a combination thereof. In some cases, the memory contains instructions for determining cell parameters solely from the calculated image parameters (e.g., radial moment, eccentricity, etc.). In some cases, the memory contains instructions for determining cell parameters solely from the particle image and not from another data source (e.g., a data signal waveform). In some cases, the memory contains instructions for determining cell parameters using a combination of a generated particle image and a data signal waveform generated in response to measurement light from irradiated particles.

[0105] In some embodiments, the system includes a memory having instructions for generating frequency-coded data (e.g., frequency-coded spatial data) from measurement light from cells in a flowing stream sample. In some cases, the memory includes instructions for generating one or more images from the frequency-coded data. The frequency-coded data may be generated in one or more detection channels, such as two or more, three or more, four or more, five or more, six or more, and including eight or more detection channels. In some embodiments, the frequency-coded data includes data components acquired (or derived) from different detectors (e.g., detected light absorption or detected light scattering). In some cases, the memory includes instructions for performing phase correction on the frequency-coded data. In some cases, the memory includes instructions for generating a phase-corrected image of the cells by performing a transform on the frequency-coded data. In one example, the memory includes instructions for performing phase correction on the frequency-coded data by performing a Fourier transform (FT). In another example, the memory includes instructions for performing phase correction on the frequency-coded data by performing a discrete Fourier transform (DFT). In yet another example, the memory contains instructions to perform phase correction on the frequency-coded data by performing a short-time Fourier transform (STFT). In some implementations, the memory contains instructions for performing a transform on the frequency-coded data without performing any mathematical imaginary operations (i.e., only performing mathematical real number operations on the transform) to generate an image from the frequency-coded data.

[0106] In some implementations, the memory contains instructions for determining one or more sorting gates for sample cells, such as classifying cancer cells in the sample by assigning each cell to a particle population cluster. In some cases, the memory contains instructions for generating one or more sorting gates that capture cells in a target particle population cluster and exclude cells in non-target particle population clusters. In some cases, the memory contains instructions for determining sorting gates configured to maximize the inclusion rate of cells in a target particle population cluster (e.g., cells from a specific class of multiple classes). For example, the sorting gates are configured to make the inclusion rate of the generated target particle population cluster cells 50% or higher, such as 55% or higher, 60% or higher, 65% or higher, 70% or higher, 75% or higher, 80% or higher, 85% or higher, 90% or higher, 95% or higher, 97% or higher, 99% or higher, and include determining a sorting gate configured to make the inclusion rate of the generated target particle population cluster cells 99.9% or higher. In some cases, the memory contains instructions for determining a sorting gate configured to maximize the purity yield of target particle population clusters of cells. For example, the sorting gate is configured to achieve a purity yield of 50% or higher for the target particle population clusters of cells, such as 55% or higher, 60% or higher, 65% or higher, 70% or higher, 75% or higher, 80% or higher, 85% or higher, 90% or higher, 95% or higher, 97% or higher, or 99% or higher, and includes determining a sorting gate configured to achieve a purity yield of 99.9% or higher for the generated target particle population clusters of cells.

[0107] In some cases, the memory contains instructions for determining a sorting gating configured to maximize the exclusion of non-target particle swarm cells. In other cases, the memory contains instructions for determining a sorting gating configured to exclude 50% or more of non-target particle swarm cells, such as 55% or more, 60% or more, 65% or more, 70% or more, 75% or more, 80% or more, 85% or more, 90% or more, 95% or more, 97% or more, 99% or more, and includes determining a sorting gating configured to exclude 99.9% or more of non-target particle swarm clustered cells.

[0108] In some cases, the memory contains instructions for evaluating sorting gates of a gating strategy and adjusting one or more generated sorting gates. This adjustment can be based on a metric indicating the accuracy of the sorting gates, according to the desired sorting strategy. In some cases, the metric can be an accuracy metric (e.g., purity), where purity is compared to a predetermined threshold. This metric can be generated based on the confidence of a classifier included in the sorting strategy. In other cases, the metric can be a yield metric, where the yield of target cells in particle population clustering is compared to a predetermined threshold.

[0109] In some embodiments, the memory includes instructions for generating sorting decisions based on determined sorting gating of sample cells. In some cases, the memory includes instructions for generating particle sorting decisions using eight or more gating strategies, such as seven or fewer, six or fewer, five or fewer, four or fewer, three or fewer, and two or fewer. In some embodiments, the memory includes instructions for generating sorting gatings using a graphical display that shows one or more analysis algorithms for applying classification parameters to image parameters determined for the cells.

[0110] In some implementations, the memory includes instructions for generating sorting gating using a graphical display that shows one or more analysis algorithms for applying determined classification parameters to image parameters determined for the cells. In some implementations, the system includes a display for visualizing the gating strategy on a graphical user interface.

[0111] In some cases, the graphical user interface applies analytical algorithms to generate gating policies. For example, the analytical algorithm may be one or more of a spectral compensation matrix, a clustering algorithm, and a t-distributed random neighborhood embedding (t-SNE) algorithm. In some cases, the analytical algorithm is applied to a granular population cluster by dragging and dropping its icon onto that cluster. In other cases, the granular population cluster is selected, and the analytical algorithm is applied by selecting it from a drop-down menu. In some cases, the analytical algorithm is a spectral unmixing algorithm, such as the algorithm described in U.S. Patent No. 11,009,400 and International Patent Application No. PCT / US2021 / 46741, filed August 19, 2021, the disclosure of which is incorporated herein by reference.

[0112] In some cases, the system includes memory storing instructions for determining a gating strategy using a computational sorting algorithm, such as that described in U.S. Patent No. 11,513,054, the disclosure of which is incorporated herein by reference. In other cases, the system includes memory storing computer software for determining sorting gating, such as HyperFinder (e.g., as described in Bonavia et al. Frontiers in Immunology 2022; 13:1007016) and computational sorting using HyperFinder, FlowJo software, and BD FACSDiva software (Becton Dickinson, 2021), the disclosures of which are incorporated herein by reference. In some embodiments, the gating strategy is developed on a separate computing system (e.g., a different computer system or network) and transmitted to a flow cytometer to implement the gating strategy, for example, using a particle sorter of the flow cytometer (e.g., having a sorting decision module).

[0113] In some implementations, the system for generating the gating strategy is part of, or operationally coupled to, a particle analyzer system (e.g., a flow cytometer) to generate the flow cytometry data described herein.

[0114] In some embodiments, the system includes an integrated circuit device programmed to implement one or more of the methods described herein. In some embodiments, the integrated circuit device of interest includes a field-programmable gate array (FPGA). In other embodiments, the integrated circuit device includes an application-specific integrated circuit (ASIC). In still other embodiments, the integrated circuit device includes a complex programmable logic device (CPLD).

[0115] In some embodiments, the system includes a particle sorter component. In some embodiments, the sorting mechanism is configured to sort cells (e.g., cancer cells in the sample) into containers based on the presence, classification, or both of cells in the sample. The term "sorting" is used herein in its conventional sense to refer to the separation of components from a sample (e.g., target cells with mitochondrial morphology of interest, non-target cells with mitochondrial morphology of no interest, non-cellular particles such as biomacromolecules) and, in some cases, delivery of the separated components to one or more sample collection containers. For example, the system may be configured to sort samples having two or more components, such as three or more components, four or more components, five or more components, ten or more components, fifteen or more components, and includes sorting samples containing 25 or more components. One or more sample components can be separated from a sample and delivered to a sample collection container, such as two or more sample components, three or more sample components, four or more sample components, five or more sample components, ten or more sample components, and including fifteen or more sample components that can be separated from a sample and delivered to a sample collection container.

[0116] In some embodiments, the particle sorting system of interest is configured to sort particles using enclosed particle sorting modules, such as those described in U.S. Patent Publication No. 2017 / 0299493, filed March 28, 2017, the disclosure of which is incorporated herein by reference. In some embodiments, a sorting decision module having multiple sorting decision units is used to sort particles (e.g., cells) of a sample, such as those described in U.S. Patent Publication No. 2020 / 0256781, the disclosure of which is incorporated herein by reference. In some embodiments, the subject system includes a particle sorting module with deflection plates, such as those described in U.S. Patent Publication No. 2017 / 0299493, filed March 28, 2017, the disclosure of which is incorporated herein by reference.

[0117] Figure 1BA flowchart illustrating the classification of cancer cells in a sample according to certain embodiments is described. In step 101, the sample with the lowest concentration of cancer cells is irradiated with light. In step 102, light from the irradiated cells is detected using a light detection system with a photodetector, and an image of the cancer cells is captured. In some embodiments, image data is calculated from the captured cancer cell image (step 102a). In some embodiments, in step 102b, one or more parameters of the cells (e.g., morphology, punctiformity, radial moment, etc.) are calculated from the image data. In step 103, the presence of cancer cells with the lowest concentration is detected. In step 104, the classification of each cancer cell detected in the sample is determined among multiple classifications. In some cases, the multiple classifications include ovarian cancer, T-lymphocyte carcinoma, breast cancer, colon cancer, or epithelial carcinoma. In some cases, the multiple classifications include OVCAR, Jurkat, MCF-7, and adenocarcinoma. In some cases, a neural network applied to determine the classification is trained using sample images labeled with fluorescent dye benchmarks (step 104a). In some embodiments, a gating strategy for classifying cells is determined (step 105). The gating strategy can be used to generate a sorting decision in step 106 and to sort cancer cells with cell parameters of interest from the sample in step 107, for example, for further analysis or for cancer diagnosis or staging.

[0118] The system according to some embodiments may include a display and an operator input device. For example, the operator input device may be a keyboard, mouse, etc. The processing module includes a processor that can access memory storing instructions for performing the subject method steps. The processing module may include an operating system, a graphical user interface (GUI) controller, system memory, memory storage devices, and input / output controllers, caches, data backup units, and many other devices. The processor may be a commercially available processor or one of other existing or soon-to-be-released processors. The processor executes the operating system, which interfaces with firmware and hardware in a well-known manner and helps the processor coordinate and execute the functions of various computer programs written in various programming languages ​​(e.g., Java, Perl, C++, other high-level or low-level languages, and combinations thereof), as known in the art. The operating system typically works in cooperation with the processor to coordinate and execute the functions of other computer components. The operating system also provides scheduling, input / output control, file and data management, memory management, and communication control and related services, all conforming to known techniques. The processor may be any suitable analog or digital system. In some embodiments, the processor includes analog electronics that provide feedback control (e.g., negative feedback control).

[0119] System memory can be any known or future memory storage device. Examples include any commonly used random access memory (RAM), magnetic media (such as resident hard disks or magnetic tapes), optical media (such as optical discs), flash memory devices, or other memory storage devices. Memory storage devices can be any known or future devices, including optical disc drives, magnetic tape drives, removable hard disk drives, or floppy disk drives. This type of memory storage device typically reads from and / or writes to program storage media (not shown), such as optical discs, magnetic tapes, removable hard disks, or floppy disks. Any of these program storage media, or other program storage media currently in use or that may be developed in the future, can be considered a computer program product. As will be understood, these program storage media typically store computer software programs and / or data. Computer software programs (also known as computer control logic) are typically stored in system memory and / or program storage devices used in conjunction with memory storage devices.

[0120] In some embodiments, a computer program product is described, comprising a computer-usable medium storing control logic (computer software program, including program code). When executed by a computer processor, the control logic causes the processor to perform the functions described herein. In other embodiments, some functions are implemented primarily in hardware, for example using a hardware state machine. It will be apparent to those skilled in the art that implementing a hardware state machine to perform the functions described herein is appropriate.

[0121] The memory can be any suitable device in which the processor can store and retrieve data, such as magnetic, optical, or solid-state storage devices (including disks, optical discs, magnetic tapes, RAM, or any other suitable device, whether fixed or portable). The processor can include a general-purpose digital microprocessor, which is suitably programmed from a computer-readable medium carrying the necessary program code. The programming can be provided to the processor remotely via a communication channel or pre-stored in a computer program product (e.g., memory or some other portable or fixed computer-readable storage medium) using any of these devices associated with the memory. For example, a disk or optical disc can carry the program and can be read by a disk writer / reader. The system of the present invention also includes a program, for example in the form of a computer program product, an algorithm for practicing the methods described above. The program according to the invention can be recorded on a computer-readable medium, such as any medium that can be directly read and accessed by a computer. Such media include, but are not limited to: magnetic storage media, such as floppy disks, hard disk storage media, and magnetic tape; optical storage media, such as CD-ROMs; electrical storage media, such as RAM and ROM; portable flash drives; and mixtures of these categories, such as magnetic / optical storage media.

[0122] The processor can also access communication channels to communicate with users in remote locations. A remote location refers to a location where the user does not directly interact with the system but instead passes input information from an external device (such as a computer connected to a wide area network (“WAN”), telephone network, satellite network, or any other suitable communication channel, including mobile phones (i.e., smartphones)) to the input manager.

[0123] In some embodiments, the system according to this disclosure may be configured to include a communication interface. In some embodiments, the communication interface includes a receiver and / or transmitter for communicating with a network and / or another device. The communication interface may be configured for wired or wireless communication, including but not limited to radio frequency (RF) communication (e.g., RFID, Zigbee communication protocol, WiFi, infrared, wireless universal serial bus (USB), ultra-wideband (UWB), Bluetooth® communication protocol) and cellular communication, such as Code Division Multiple Access (CDMA) or Global System for Mobile Communications (GSM).

[0124] In one implementation, the communication interface is configured to include one or more communication ports, such as physical ports or interfaces, such as USB ports, RS-232 ports, or any other suitable electrical connection ports, to allow data communication between the subject system and other external devices (such as computer terminals (e.g., in a doctor's office or hospital environment)) configured for similar complementary data communication.

[0125] In one implementation, the communication interface is configured for infrared communication, Bluetooth® communication, or any other suitable wireless communication protocol to enable the subject system to communicate with other devices, such as computer terminals and / or networks, communication-enabled mobile phones, personal digital assistants, or any other communication devices that the user can use in conjunction with them.

[0126] In one implementation, the communication interface is configured to provide data transmission connectivity using the Internet Protocol (IP) via a cellular telephone network, a short message service (SMS), a wireless connection to a personal computer (PC) connected to the Internet on a local area network (LAN), or a WiFi connection to the Internet at a WiFi hotspot.

[0127] In one implementation, the subject system is configured to wirelessly communicate with a server device via a communication interface, for example, using a common standard such as 802.11 or the Bluetooth® RF protocol, or the IrDA infrared protocol. The server device can be another portable device, such as a smartphone, personal digital assistant (PDA), or laptop computer; or a larger device, such as a desktop computer, appliance, etc. In some implementations, the server device has a display, such as a liquid crystal display (LCD), and input devices, such as buttons, a keyboard, a mouse, or a touchscreen.

[0128] In some implementations, the communication interface is configured to automatically or semi-automatically communicate data stored in the subject system (e.g., optional data storage unit) with network or server devices using one or more of the communication protocols and / or mechanisms described above.

[0129] The output controller may include any of a variety of known display devices for presenting information to a user (whether human or machine, local or remote). If a display device provides visual information, that information is typically logically and / or physically organized into an array of image elements. The graphical user interface (GUI) controller may include a variety of known or future software programs for providing a graphical input and output interface between the system and the user and for processing user input. The functional elements of the computer may communicate with each other via a system bus. In alternative implementations, some of these communications may be implemented using networks or other types of remote communication. The output manager may also provide information generated by the processing module to a user at a remote location, for example, via the Internet, telephone, or satellite networks, according to known technologies. The presentation of data by the output manager may be implemented according to a variety of known technologies. As some examples, the data may include SQL, HTML, or XML documents, emails, or other files or other forms of data. The data may include Internet URLs, allowing the user to retrieve other SQL, HTML, XML, or other documents or data from a remote source. One or more platforms present in the subject system may be any type of known computer platform or type to be developed in the future, although they typically belong to a class of computers commonly referred to as servers. However, they may also be mainframes, workstations, or other computer types. They can be connected via any known or future type of cable or other communication system (including wireless systems, whether networked or otherwise). They can be located together or physically separated. A variety of operating systems can be used on any computer platform, depending on the type and / or brand of the chosen platform. Applicable operating systems include Windows 10 and Windows NT. ®, Windows XP, Windows 7, Windows 8, iOS, Sun Solaris, Linux, OS / 400, Compaq Tru64 Unix, SGI IRIX, Siemens Reliant Unix, Ubuntu, Zorin OS, etc.

[0130] In some embodiments, the subject system includes one or more optical adjustment components for adjusting the light irradiated onto the sample (e.g., light from a laser) or the light collected from the sample (e.g., scattered light, fluorescence). For example, optical adjustment can be used to increase the size of the light, focus the light, or collimate the light. In some cases, optical adjustment is an amplification scheme to increase the size of the light (e.g., the spot size), for example, by 5% or more, such as 10% or more, 25% or more, 50% or more, and including an increase of 75% or more. In other embodiments, optical adjustment includes focusing the light to reduce the light size, such as by 5% or more, such as 10% or more, 25% or more, 50% or more, and including reducing the spot size by 75% or more. In some embodiments, optical adjustment includes collimating the light beam. The term "collimation," used in its conventional sense, refers to optically adjusting the collinearity of light propagation or reducing the divergence of light rays from a common propagation axis. In some cases, collimation includes reducing the spatial cross-section of the beam (e.g., reducing the beam profile of a laser).

[0131] In some embodiments, the optical adjustment element is a focusing lens with a magnification of 0.1 to 0.95, such as 0.2 to 0.9, 0.3 to 0.85, 0.35 to 0.8, 0.5 to 0.75, and including 0.55 to 0.7, such as 0.6. For example, in some cases, the focusing lens is a biachromatic reducing lens with a magnification of about 0.6. The focal length of the focusing lens can vary from 5 mm to 20 mm, such as from 6 mm to 19 mm, 7 mm to 18 mm, 8 mm to 17 mm, 9 mm to 16 mm, and includes a focal length range of 10 mm to 15 mm. In some embodiments, the focal length of the focusing lens is about 13 mm.

[0132] In other embodiments, the optical adjustment component is a collimator. The collimator can be any convenient collimation scheme, such as one or more mirrors or curved lenses, or combinations thereof. For example, in some cases, the collimator is a single collimating lens. In other cases, the collimator is a collimating mirror. In still other cases, the collimator includes two lenses. In still other cases, the collimator includes a mirror and a lens. When the collimator includes one or more lenses, the focal length of the collimating lens can vary from 5 mm to 40 mm, for example, from 6 mm to 37.5 mm, for example, from 7 mm to 35 mm, for example, from 8 mm to 32.5 mm, for example, from 9 mm to 30 mm, for example, from 10 mm to 27.5 mm, for example, from 12.5 mm to 25 mm, and includes focal lengths ranging from 15 mm to 20 mm.

[0133] In some embodiments, the subject system includes a flow cell nozzle having a nozzle orifice configured to allow a flow stream to pass through the flow cell nozzle. The subject flow cell nozzle has an orifice that delivers a fluid sample to a sample probing region. In some embodiments, the flow cell nozzle includes a proximal cylindrical portion defining a longitudinal axis and a distal truncated conical portion terminating on a flat surface having a nozzle orifice transverse to the longitudinal axis. The length of the proximal cylindrical portion (as measured along the longitudinal axis) can vary from 1 mm to 15 mm, for example, 1.5 mm to 12.5 mm, for example, 2 mm to 10 mm, for example, 3 mm to 9 mm, and includes 4 mm to 8 mm. The length of the distal truncated conical portion (as measured along the longitudinal axis) can also vary from 1 mm to 10 mm, for example, 2 mm to 9 mm, for example, 3 mm to 8 mm, and includes 4 mm to 7 mm. In some embodiments, the diameter of the flow pool nozzle chamber can vary from 1 mm to 10 mm, for example 2 mm to 9 mm, for example 3 mm to 8 mm, and includes 4 mm to 7 mm.

[0134] In some cases, the nozzle chamber does not include a cylindrical portion, and the entire flow pool nozzle chamber is a truncated cone. In these embodiments, the length of the truncated cone nozzle chamber (as measured along the longitudinal axis transverse to the nozzle orifice) can range from 1 mm to 15 mm, for example from 1.5 mm to 12.5 mm, for example from 2 mm to 10 mm, for example from 3 mm to 9 mm, and includes 4 mm to 8 mm. The diameter of the proximal portion of the truncated cone nozzle chamber can range from 1 mm to 10 mm, for example from 2 mm to 9 mm, for example from 3 mm to 8 mm, and includes 4 mm to 7 mm.

[0135] In some embodiments, the sample flow exits from an orifice at the distal end of the flow cell nozzle. Depending on the desired characteristics of the flow, the flow cell nozzle orifice can be of any suitable shape, with cross-sectional shapes of interest including, but not limited to: linear cross-sectional shapes, such as squares, rectangles, trapezoids, triangles, hexagons, etc.; curved cross-sectional shapes, such as circles, ellipses; and irregular shapes, such as parabolic bottoms coupled to the top of a plane. In some embodiments, the flow cell nozzle of interest has a circular orifice. The nozzle orifice size can vary, ranging from 1 μm to 20,000 μm in some embodiments, such as from 2 μm to 17,500 μm, such as from 5 μm to 15,000 μm, such as from 10 μm to 12,500 μm, such as from 15 μm to 10,000 μm, such as from 25 μm to 7,500 μm, such as from 50 μm to 5,000 μm, such as from 75 μm to 1,000 μm, such as from 100 μm to 750 μm, and including from 150 μm to 500 μm. In some embodiments, the nozzle orifice is 100 μm.

[0136] In some embodiments, the flow cell nozzle includes a sample injection port configured to supply a sample to the flow cell nozzle. In these embodiments, the sample injection system is configured to supply an appropriate sample flow to the flow cell nozzle chamber. Depending on the desired characteristics of the flow flow, the rate at which the sample is delivered from the sample injection port to the flow cell nozzle chamber can be 1 μL / sec or greater, for example, 2 μL / sec or greater, for example, 3 μL / sec or greater, for example, 5 μL / sec or greater, for example, 10 μL / sec or greater, for example, 15 μL / sec or greater, for example, 25 μL / sec or greater, for example, 50 μL / sec or greater, for example, 100 μL / sec or greater, for example, 150 μL / sec or greater, for example, 200 μL / sec or greater, for example, 250 μL / sec or greater, for example, 300 μL / sec or greater, for example, 350 μL / sec or greater, for example, 400 μL / sec or greater, for example, 450 μL / sec or greater, and includes 500 μL / sec or greater. For example, the sample fluid flow rate can range from 1 μL / sec to about 500 μL / sec, such as 2 μL / sec to about 450 μL / sec, such as 3 μL / sec to about 400 μL / sec, such as 4 μL / sec to about 350 μL / sec, such as 5 μL / sec to about 300 μL / sec, such as 6 μL / sec to about 250 μL / sec, such as 7 μL / sec to about 200 μL / sec, such as 8 μL / sec to about 150 μL / sec, such as 9 μL / sec to about 125 μL / sec, and includes 10 μL / sec to about 100 μL / sec.

[0137] The sample injection port can be an orifice located on the nozzle chamber wall or a conduit located near the proximal end of the nozzle chamber. When the sample injection port is an orifice located on the nozzle chamber wall, the orifice can be of any suitable shape, wherein the cross-sectional shapes of interest include, but are not limited to: straight cross-sectional shapes, such as squares, rectangles, trapezoids, triangles, hexagons, etc.; curved cross-sectional shapes, such as circles, ellipses, etc.; and irregular shapes, such as the bottom of a parabola coupled to the top of a plane. In some embodiments, the sample injection port has a circular orifice. The size of the sample injection port can vary depending on the shape, and in some cases, the opening ranges from 0.1 mm to 5.0 mm, for example 0.2 to 3.0 mm, for example 0.5 mm to 2.5 mm, for example 0.75 mm to 2.25 mm, for example 1 mm to 2 mm, and includes 1.25 mm to 1.75 mm, for example 1.5 mm.

[0138] In some cases, the sample injection port is a conduit located near the end of the flow cell nozzle chamber. For example, the sample injection port may be a conduit positioned such that its orifice aligns with the flow cell nozzle orifice. When the sample injection port is a conduit aligned with the flow cell nozzle orifice, the cross-sectional shape of the sample injection tube can be any suitable shape, including but not limited to: straight cross-sectional shapes, such as squares, rectangles, trapezoids, triangles, hexagons, etc.; curved cross-sectional shapes, such as circles, ellipses; and irregular shapes, such as parabolic bases coupled to the top of a plane. The orifice of the conduit may vary depending on its shape, and in some cases, the opening ranges from 0.1 mm to 5.0 mm, for example 0.2 to 3.0 mm, for example 0.5 mm to 2.5 mm, for example 0.75 mm to 2.25 mm, for example 1 mm to 2 mm, and includes 1.25 mm to 1.75 mm, for example 1.5 mm. The shape of the tip of the sample injection port may be the same as or different from the cross-sectional shape of the sample injection tube. For example, the orifice of the sample injection port may include a beveled tip with a bevel angle ranging from 1° to 10°, such as 2° to 9°, such as 3° to 8°, such as 4° to 7°, and including a bevel angle of 5°.

[0139] In some embodiments, the flow cell nozzle further includes a sheath fluid injection port configured to supply sheath fluid to the flow cell nozzle. In embodiments, the sheath fluid injection system is configured to supply a sheath fluid flow to the flow cell nozzle chamber, for example, along with a sample, to generate a sheath fluid layer flow around the sample flow. Depending on the desired characteristics of the flow flow, the rate of sheath fluid delivered to the flow cell nozzle chamber can be 25 μL / sec or faster, for example 50 μL / sec or faster, for example 75 μL / sec or faster, for example 100 μL / sec or faster, for example 250 μL / sec or faster, for example 500 μL / sec or faster, for example 750 μL / sec or faster, for example 1000 μL / sec or faster, and includes 2500 μL / sec or faster. For example, the range of sheath fluid flow rate can be from 1 μL / sec to about 500 μL / sec, such as 2 μL / sec to about 450 μL / sec, such as 3 μL / sec to about 400 μL / sec, such as 4 μL / sec to about 350 μL / sec, such as 5 μL / sec to about 300 μL / sec, such as 6 μL / sec to about 250 μL / sec, such as 7 μL / sec to about 200 μL / sec, such as 8 μL / sec to about 150 μL / sec, such as 9 μL / sec to about 125 μL / sec, and includes 10 μL / sec to about 100 μL / sec.

[0140] In some embodiments, the sheath fluid injection port is an orifice located in the nozzle chamber wall. The sheath fluid injection port orifice can be of any suitable shape, with cross-sectional shapes of interest including, but not limited to: straight cross-sectional shapes, such as squares, rectangles, trapezoids, triangles, hexagons, etc.; curved cross-sectional shapes, such as circles, ellipses; and irregular shapes, such as parabolic bottoms coupled to the top of a plane. The size of the sample injection port can vary depending on its shape, and in some cases, the opening ranges from 0.1 mm to 5.0 mm, for example 0.2 to 3.0 mm, for example 0.5 mm to 2.5 mm, for example 0.75 mm to 2.25 mm, for example 1 mm to 2 mm, and includes 1.25 mm to 1.75 mm, for example 1.5 mm.

[0141] In some cases, the system includes a sample probing region in fluid communication with the flow cell nozzle orifice. In these cases, the sample flow exits from the orifice at the distal end of the flow cell nozzle, and particles in the flow can be irradiated by a light source in the sample probing region. The size of the probing region can vary depending on the characteristics of the flow nozzle, such as the size of the nozzle orifice and the size of the sample inlet. In embodiments, the width of the probing region can be 0.01 mm or greater, for example 0.05 mm or greater, for example 0.1 mm or greater, for example 0.5 mm or greater, for example 1 mm or greater, for example 2 mm or greater, for example 3 mm or greater, for example 5 mm or greater, and includes 10 mm or greater. The length of the probe area can also vary, and in some cases, its range can be 0.01 mm or greater, such as 0.1 mm or greater, such as 0.5 mm or greater, such as 1 mm or greater, such as 1.5 mm or greater, such as 2 mm or greater, such as 3 mm or greater, such as 5 mm or greater, such as 10 mm or greater, such as 15 mm or greater, such as 20 mm or greater, such as 25 mm or greater, and includes 50 mm or greater.

[0142] The probe region can be configured to facilitate the irradiation of a planar cross-section of the outflow, or it can be configured to facilitate the irradiation of a predetermined length of diffused field (e.g., using a diffused laser or lamp). In some embodiments, the probe region includes a transparent window that facilitates the irradiation of a predetermined length of the outflow, such as 1 mm or longer, 2 mm or longer, 3 mm or longer, 4 mm or longer, 5 mm or longer, and including 10 mm or longer. Depending on the light source used to irradiate the outflow (described below), the probe region can be configured to allow light in the wavelength range of 100 nm to 1500 nm to pass through, such as 150 nm to 1400 nm, 200 nm to 1300 nm, 250 nm to 1200 nm, 300 nm to 1100 nm, 350 nm to 1000 nm, 400 nm to 900 nm, and including 500 nm to 800 nm.Therefore, the probe area can be formed of any transparent material capable of transmitting the desired wavelength range, including but not limited to optical glass, borosilicate glass, Pyrex glass, ultraviolet quartz, infrared quartz, sapphire, and plastics such as polycarbonate, polyvinyl chloride (PVC), polyurethane, polyether, polyamide, polyimide, or copolymers of these thermoplastics, such as PETG (ethylene glycol-modified polyethylene terephthalate), and other polymeric plastic materials, including polyesters, wherein the polyester of interest may include, but is not limited to, poly(alkylene terephthalate), such as poly(ethylene terephthalate) (PET), bottle-grade PET (… A copolymer based on monoethylene glycol, terephthalic acid and other comonomers such as isophthalic acid, cyclohexenedimethyl alcohol, etc.; poly(butylene terephthalate) (PBT) and poly(hexamethylene terephthalate); poly(alkylene adipate), such as poly(ethylene adipate), poly(1,4-butylene adipate) and poly(hexamethylene adipate); poly(alkylene octanoate), such as poly(ethylene octanoate); poly(alkylene sebacate), such as poly(ethylene sebacate); poly(caprolactone) and poly(propiolactone); poly(alkylene isophthalate), such as poly(ethylene isophthalate); Poly(2,6-naphthalene dicarboxylic acid alkylene ester), for example, poly(2,6-naphthalene dicarboxylic acid ethylene glycol ester); poly(sulfonyl-4,4′-dibenzoic acid alkylene ester), for example, poly(sulfonyl-4,4′-dibenzoic acid ethylene glycol ester); poly(p-phenylene alkylene dicarboxylic acid ester), for example, poly(p-phenylene vinyl dicarboxylic acid ester); poly(trans-1,4-cyclohexanediylalkylene dicarboxylic acid ester), for example, poly(trans-1,4-cyclohexanediylethylene dicarboxylic acid ester); poly(1,4-cyclohexane-dimethylene alkylene dicarboxylic acid ester), for example, poly(1,4-cyclohexane-dimethylene vinyl dicarboxylic acid ester); poly([2.2.2]-bicyclooctane -1,4-dimethylenealkylene dicarboxylate, such as poly([2.2.2]-bicyclooctane-1,4-dimethyleneethylene dicarboxylate); lactic polymers and copolymers, such as (S)-polylactic acid, (R,S)-polylactic acid, poly(tetramethylglycolic acid) and poly(lactic acid-co-glycolic acid); and polycarbonates of bisphenol A, 3,3′-dimethylbisphenol A, 3,3′,5,5′-tetrachlorobisphenol A, 3,3′,5,5′-tetramethylbisphenol A; polyamides, such as poly(p-phenylene terephthalamide); polyesters, such as polyethylene terephthalate, such as Myla™ polyethylene terephthalate; and so on. In some embodiments, the subject system includes a small pool located in the sample probing area.In the implementation, the wavelength range of light that can be transmitted through the small pool is 100 nm to 1500 nm, for example 150 nm to 1400 nm, for example 200 nm to 1300 nm, for example 250 nm to 1200 nm, for example 300 nm to 1100 nm, for example 350 nm to 1000 nm, for example 400 nm to 900 nm, and includes 500 nm to 800 nm.

[0143] In some embodiments, the optical detection system having multiple photodetectors as described above is part of or located within a particle analyzer (e.g., a particle sorter). In some embodiments, the subject system is a flow cytometry system that includes photodiodes and an amplifier assembly as part of the optical detection system for detecting light emitted from a sample in a flowing stream. Suitable flow cytometry systems may include, but are not limited to, Ormerod (ed.), FlowCytometry: A Practical Approach, Oxford Univ. Press (1997); Jaroszeski et al. (eds.), Flow Cytometry Protocols, Methods in Molecular Biology No. 91, HumanaPress (1997); Practical Flow Cytometry, 3rd Edition, Wiley-Liss (1995); Virgo et al. (2012) Ann Clin Biochem. Jan;49(pt 1):17-28; Linden et al., Semin Throm Hemost. 2004 Oct;30(5):502-11; Alison et al. J Pathol, 2010 Dec; 222(4):335-344; and Herbig et al. (2007) Crit Rev Ther Drug Carrier Syst. Those described in 24(3):203-255; the contents of which are incorporated herein by reference. In some cases, the flow cytometry system of interest includes BD BiosciencesFACSCanto TM Flow cytometer, BD Biosciences FACSCanto TM II flow cytometer, BD Accuri TM Flow cytometer, BD Accuri TM C6 Plus flow cytometer, BD Biosciences FACSCelesta TMFlow cytometer, BDBiosciences FACSLyric TM Flow cytometer, BD Biosciences FACSVerse TM Flow cytometer, BDBiosciences FACSymphony TM Flow cytometer, BD Biosciences LSRFortessa TM Flow cytometer, BDBiosciences LSRFortessa TM X-20 flow cytometer, BD Biosciences FACSPresto TM Flow cytometer, BD Biosciences FACSVia TM Flow cytometer and BD Biosciences FACSCalibur TM Cell sorter, BD Biosciences FACSCount TM Cell sorter, BD Biosciences FACSLyric TM Cell sorting instrument, BDBiosciences Via TM Cell sorting instruments, including BD Biosciences Influx™, BD Biosciences Jazz™, BD Biosciences Aria™, BD Biosciences FACSAria™ II, BD Biosciences FACSAria™ III, BD Biosciences FACSAria™ Fusion, BD Biosciences FACSMelody™, and BD Biosciences FACSymphony. TM S6 cell sorter, etc.

[0144] In some implementations, the subject system is a flow cytometry system, such as those described in U.S. Patent Nos. 10,663,476, 10,620,111, 10,613,017, 10,605,713, 10,585,031, 10,578,542, 10,578,469, 10,481,074, 10,302,545, 10,145,793, 10,113,967, 10,006,852, 9,952,076, 9,933,341, 9,726,527, 9,453,789, 9,200,334, and 9,097,640. The disclosures of those listed in ,095,494, 9,092,034, 8,975,595, 8,753,573, 8,233,146, 8,140,300, 7,544,326, 7,201,875, 7,129,505, 6,821,740, 6,813,017, 6,809,804, 6,372,506, 5,700,692, 5,643,796, 5,627,040, 5,620,842, 5,602,039, 4,987,086, and 4,498,766 are incorporated herein by reference in their entirety.

[0145] In some implementations, the flow cytometer is configured as an imaging flow cytometer. For example, in some cases, the subject system is a flow cytometry system configured to image particles in a flowing stream using fluorescence imaging with radio frequency labeled emission (FIRE), as described in Diebold, et al., Nature Photonics Vol. 7(10); 806-810 (2013) and in U.S. Patent Nos. 9,423,353; 9,784,661; 9,983,132; 10,006,852; 10,036,699; 10,078,045; 10,222,316; 10,288,546; 10,324,019; 10,408,758; The disclosures of those listed in 10,451,538; 10,620,111; 10,684,211; 10,845,295; 10,935,482; 10,935,485; 11,105,728; 11,280,718; 11,327,016; 11,366,052; 11,371,937; 11,692,926; 11,630,053; 11,774,343; 11,940,369; and 11,946,851 are incorporated herein by reference.

[0146] Figure 2A system 200 for flow cytometry according to an illustrative embodiment of the present disclosure is shown. System 200 includes a laser 201 configured to irradiate particles 211 in a flow stream 214 at a probe point 215 within a flow cell 210. Although Figure 2 The example shows a single laser, but it is understood that multiple lasers can also be used. The laser beam from laser 201 is directed to focusing lens 202, which focuses the beam onto the fluid flow portion of the sample containing particles 211 within flow cell 210. Flow cell 210 is part of a fluid system that directs particles (typically one at a time) in the flow to the focused laser beam for probing. Alternatively, when the flow cytometer is a gas flow cytometer, a nozzle top may be used.

[0147] like Figure 2 As shown, flow cell 210 is fluidly connected to a sheath fluid reservoir 203 containing sheath fluid and a sample flow reservoir 204 containing sample fluid. Sheath fluid from sheath fluid reservoir 203 is supplied to at least one sheath fluid injection port 208 via a conduit (i.e., sheath fluid line) 207. Furthermore, a sample flow containing particles 211 from sample flow reservoir 204 is supplied to sample injection port 206 via a conduit (i.e., sample flow line) 205. Sample injection port 206 is fluidly connected to a sample injector 213 (e.g., a sample injection needle) configured to introduce particles 211 into the interior of flow cell 210. Particles 211 are hydrodynamically focused by the sheath fluid entering from sheath fluid injection port 208, such that a flow stream 214 is formed downstream of the conical portion 212 of flow cell 210. Particles emitted at the distal end of flow cell 210 can be processed and / or collected by any suitable method. For example, depending on the type of flow cytometry performed, particles can be collected at the distal end of flow cell 210, for example, via a waste line. Alternatively, particles can be sorted.

[0148] Light from the laser beam interacts with particles 211 in the sample through diffraction, refraction, reflection, scattering, and absorption, and is re-emitted at various wavelengths depending on the characteristics of the particles (e.g., their size, internal structure, and the presence of one or more fluorescent molecules attached to or within the particles). The fluorescence emission, as well as the diffracted, refracted, reflected, and scattered light, can be routed to one or more detectors. Specifically, forward scattered light (FSC) is routed to forward scattered light detector 223. Forward scattered light detector 223 is positioned slightly off-center from the axis of the direct beam passing through flow cell 210 and is configured to detect diffracted light, i.e., excitation light that passes primarily forward or around the particles. The intensity of the light detected by forward scattered light detector 223 depends on the overall size of the particles. The forward scattered detector may include, for example, a photodiode. An optical filter 221a and a scattering strip 222 are provided between the forward scattered light detectors 223. Optical filter 221a can be configured to filter out non-FSC light of at least one wavelength, while scattering strip 222 can be configured to prevent forward scattering detector 223 from detecting the incident beam (i.e., non-scattered light) from laser 201.

[0149] Furthermore, side-scattered light (SSC) is detected by side-scattered light detector 224. In other words, side-scattered light detector 224 is configured to detect refracted and reflected light from the surface and internal structures of particle 211, which tend to increase with increasing particle structure complexity. Figure 2 In the example, the flow cytometer 200 includes a dichroic mirror 220a configured to reflect SSC light to a side-scatter light detector 224 while allowing non-SSC (e.g., fluorescence) light to pass through. An optical filter 221b is configured to prevent the side-scatter light detector 224 from detecting at least one wavelength of non-SSC light. Fluorescence detectors 225a-225c are also shown, each configured to detect fluorescence at a different wavelength. For example, the dichroic mirror 220b may be configured to reflect fluorescence (FL) corresponding to a first wavelength (or wavelength range) to the fluorescence detector 225a while allowing light of other wavelengths to pass through. The optical filter 221c may be configured to prevent the fluorescence detector 225a from detecting at least one wavelength that does not correspond to the first wavelength (or wavelength range). Similarly, the dichroic mirror 220c may be configured to reflect FL light corresponding to a second wavelength (or wavelength range) to the fluorescence detector 225b while allowing light of a third wavelength (or wavelength range) to pass through for detection by the fluorescence detector 225c. Optical filter 221d can be configured to prevent fluorescence detector 225b from detecting at least one light wavelength that does not correspond to the second wavelength (or wavelength range). Furthermore, optical filter 221e is configured to prevent fluorescence detector 225c from detecting at least one light wavelength that does not correspond to the third wavelength (or wavelength range).

[0150] Those skilled in the art will recognize that the flow cytometer according to embodiments of this disclosure is not limited to... Figure 2 The flow cytometer shown may include any flow cytometer known in the art. For example, a flow cytometer may have any number of lasers, beam splitters, filters, and detectors, with various wavelengths and different configurations. For example, although Figure 2 The implementation scheme shown for illustrative purposes uses three fluorescent detectors, but it should be understood that any suitable number of fluorescent detectors can be used.

[0151] During operation, the cytometer is controlled by controller / processor 290, and measurement data from the detector can be stored in memory 295 and processed by controller / processor 290. Although not explicitly shown, controller / processor 290 is coupled to the detector to receive output signals from the detector and can also be coupled to the electrical and electromechanical components of the flow cytometer to control laser 201, flow parameters, etc. Input / output (I / O) functionality 297 may also be provided in the system. Memory 295, controller / processor 290, and I / O 297 can be provided entirely as part of the flow cytometer. In such an embodiment, a display may also form part of I / O functionality 297 for presenting experimental data to the user of the cytometer 200. Alternatively, some or all of memory 295, controller / processor 290, and I / O functionality may be part of one or more external devices (e.g., a general-purpose computer). In some embodiments, some or all of memory 295 and controller / processor 290 may communicate wirelessly or wired with the cytometer 210. The controller / processor 290, combined with memory 295 and I / O 297, can be configured to perform various functions associated with the preparation and analysis of flow cytometry experiments.

[0152] Different fluorescent molecules in a fluorescent dye plate used in flow cytometry experiments will emit light in their own characteristic wavelength bands. Specific fluorescent labels and their associated fluorescence emission bands can be selected for the experiment, typically coinciding with the detector's filter window. I / O 297 can be configured to receive data about a flow cytometry experiment having a set of fluorescent labels and multiple cell populations with multiple markers, each cell population having a subset of multiple markers. I / O 297 can also be configured to receive biological data assigning one or more markers to one or more cell populations, marker density data, emission spectral data, data assigning labels to one or more markers, and flow cytometry configuration data. Flow cytometry experiment data (e.g., label spectral characteristics and flow cytometry configuration data) can also be stored in memory 295. Controller / processor 290 can be configured to evaluate the assignment of one or more labels to markers.

[0153] In some embodiments, the subject system is a particle sorting system configured to sort particles using a closed particle sorting module, such as the system described in U.S. Patent Publication No. 2017 / 0299493, filed March 28, 2017, the disclosure of which is incorporated herein by reference. In some embodiments, a sorting decision module having multiple sorting decision units is used to sort particles (e.g., cells) of a sample, such as the system described in U.S. Patent Publication No. 2020 / 0256781, filed December 23, 2019, the disclosure of which is incorporated herein by reference. In some embodiments, the system for sorting sample components includes a particle sorting module with deflection plates, such as the system described in U.S. Patent Publication No. 2017 / 0299493, filed March 28, 2017, the disclosure of which is incorporated herein by reference.

[0154] In some implementations, the system is an image-assisted particle sorter that uses frequency-coded data, such as... Figure 3AAs shown. The particle sorter 300 includes an optical radiation assembly 300a, which includes a light source 301 (e.g., a 488 nm laser) that generates an output beam 301a, which is split into beams 302a and 302b by a beam splitter 302. Beam 302a propagates through an acousto-optic device (e.g., an acousto-optic deflector, AOD) 303 to generate an output beam 303a with one or more angled deflections. In some cases, the output beam 303a generated from the acousto-optic device 303 includes a local oscillator beam and multiple radio frequency comb beams. Beam 302b propagates through an acousto-optic device (e.g., an acousto-optic deflector, AOD) 304 to generate an output beam 304a with one or more angled deflections. In some cases, the output beam 304a generated from the acousto-optic device 304 includes a local oscillator beam and multiple radio frequency comb beams. Output beams 303a and 304a, generated from acousto-optic devices 303 and 304 respectively, are combined with beam splitter 305 to generate output beam 305a, which is delivered through optical assembly 306 (e.g., objective lens) to irradiate particles in flow cell 307. In some embodiments, acousto-optic device 303 (AOD) splits a single laser beam into an array of smaller beams, each with a different optical frequency and angle. A second AOD 304 adjusts the optical frequency of a reference beam and then overlaps it with the array of smaller beams at beam combiner 305. In some embodiments, the light irradiation system having a light source and acousto-optic devices may also include those described in Schraivogel, et al. (“High-speed fluorescence image-enabled cell sorting” Science (2022), 375(6578): 315-320) and U.S. Patent Publication No. 2021 / 04049433, the disclosure of which is incorporated herein by reference.

[0155] Output beam 305a irradiates sample particles 308 propagating through flow cell 307 (e.g., using sheath fluid 309) at radiation region 310. As shown in radiation region 310, multiple beams (e.g., angular deflection RF shift beams depicted as points on radiation region 310) overlap with a reference local oscillator beam (depicted as shaded lines on radiation region 310). Due to their different optical frequencies, the overlapping beams exhibit beat frequency behavior, resulting in each sub-beam operating at a different frequency f. 1-n Carrying sinusoidal modulation.

[0156] Light from the radiation sample is transmitted to a light detection system 300b comprising multiple photodetectors. The light detection system 300b includes a forward-scattering photodetector 311 for generating a forward-scattering image 311a and a side-scattering photodetector 312 for generating a side-scattering image 312a. The light detection system 300b also includes a bright-field photodetector 313 for generating a light loss image 313a. In some embodiments, the forward-scattering detector 311 and the side-scattering detector 312 are photodiodes (e.g., avalanche photodiodes, APDs). In some cases, the bright-field photodetector 313 is a photomultiplier tube (PMT). Fluorescence detectors 314-317 are also used to detect fluorescence from the radiation sample. In some cases, photodetectors 314-317 are photomultiplier tubes. Light from the radiation sample is directed through a beamsplitter 320 to the side-scattering detection channel 312 and the fluorescence detection channels 314-317. The optical detection system 300b includes bandpass optics 321, 322, 323, and 324 (e.g., dichroic mirrors) for propagating light of a predetermined wavelength to photodetectors 314-317. In some cases, optics 321 has a 534 nm / 40 nm bandpass. In some cases, optics 322 has a 586 nm / 42 nm bandpass. In some cases, optics 323 has a 700 nm / 54 nm bandpass. In some cases, optics 324 has a 783 nm / 56 nm bandpass. The first number indicates the center of the spectral band. The second number provides the range of the spectral band. Thus, the 510 / 20 filter extends 10 nm on each side of the center of the spectral band, or from 500 nm to 520 nm.

[0157] Data signals generated in response to light detected in scattered light detection channels 311 and 312, bright field light detection channel 313, and fluorescence detection channels 314-317 are digitally processed in real time by processors 350 and 351. Based on the data signals generated in processors 350 and 351, images 311a-317a can be generated in each light detection channel. Image-assisted sorting is performed in response to a sorting signal generated in sorting trigger 352. Sorting assembly 300c includes deflection plate 331 for deflecting particles into sample container 332 or waste stream 333. In some cases, sorting assembly 300c is configured to sort particles using a closed particle sorting module, such as those described in U.S. Patent Publication No. 2017 / 0299493, filed March 28, 2017, the disclosure of which is incorporated herein by reference. In some embodiments, the sorting component 300c includes a sorting decision module having multiple sorting decision units, such as those described in U.S. Patent Publication No. 2020 / 0256781, the disclosure of which is incorporated herein by reference.

[0158] Figure 3B Image-assisted particle sorting data processing according to certain implementation schemes is described. In some cases, image-assisted particle sorting data processing is a low-latency data processing flow. Each photodetector generates a pulse with high-frequency modulation, which encodes an image (waveform). Fourier analysis is performed to reconstruct the image from the modulated pulse. The image processing flow produces a set of image features (image analysis), which are combined with features derived from the pulse processing flow (event packets). Real-time sorting and classification electronics then classify the particles based on the image features, generating a sorting decision that selectively charges the droplets.

[0159] In some implementations, the system is a particle analyzer, wherein the particle analysis system 401 ( Figure 4A It can be used to analyze and characterize particles with or without physically sorting them into collection containers. Figure 4A A functional block diagram of a particle analysis system for computation-based sample analysis and particle characterization is shown. In some embodiments, particle analysis system 401 is a flow system. Figure 4A The particle analysis system 401 shown can be configured to perform, in whole or in part, the methods described herein. The particle analysis system 401 includes a fluid system 402. The fluid system 402 may include or be coupled to a sample tube 405 and a moving fluid column within the sample tube, wherein particles 403 (e.g., cells) of the sample move along a common sample path 409.

[0160] The particle analysis system 401 includes a detection system 404 configured to collect a signal from each particle as it passes through one or more detection stations along a common sample path. Detection station 408 typically refers to a monitored area 407 of the common sample path. In some embodiments, detection may include detecting light or one or more other characteristics of the particle 403 as it passes through the monitored area 407. Figure 4A The image shows a detection station 408 with a monitored area 407. Some embodiments of the particle analysis system 401 may include multiple detection stations. Furthermore, some detection stations can monitor multiple areas.

[0161] Each signal is assigned a signal value to form a data point for each particle. As mentioned above, this data can be referred to as event data. The data points can be multi-dimensional data points and include values ​​for various characteristics measured for the particle. The detection system 404 is configured to collect a series of such data points within a first time interval.

[0162] The particle analysis system 401 may also include a control system 406. The control system 406 may include one or more processors, amplitude control circuitry, and / or frequency control circuitry. The control system shown may be operatively associated with the fluid system 402. The control system may be configured to generate a calculated signal frequency for at least a portion of the first time interval based on the number of data points collected by the detection system 404 during the first time interval. The control system 406 may be further configured to generate an experimental signal frequency based on the number of data points in a portion of the first time interval. The control system 406 may further compare the experimental signal frequency with the calculated signal frequency or a predetermined signal frequency.

[0163] Figure 4B A system 400 for flow cytometry according to an illustrative embodiment of the present invention is shown. System 400 includes a flow cytometer 410, a controller / processor 490, and a memory 495. The flow cytometer 410 includes one or more excitation lasers 415a-415c, a focusing lens 420, a flow chamber 425, a forward scattering detector 430, a side scattering detector 435, a fluorescence collecting lens 440, one or more beam splitters 445a-445g, one or more bandpass filters 450a-450e, one or more long-pass (“LP”) filters 455a-455b, and one or more fluorescence detectors 460a-460f.

[0164] The excitation lasers 415a-415c emit light in the form of a laser beam. Figure 4B In the example system, the laser beams emitted by the excitation lasers 415a-415c have wavelengths of 488 nm, 633 nm, and 325 nm, respectively. The laser beams are first guided through one or more beamsplitters 445a and 445b. Beamsplitter 445a transmits the 488 nm light and reflects the 633 nm light. Beamsplitter 445b transmits ultraviolet light (wavelengths ranging from 10 to 400 nm) and reflects both the 488 nm and 633 nm light.

[0165] The laser beam is then guided to a focusing lens 420, which focuses the beam onto the fluid flow portion containing the sample particles within the flow chamber 425. The flow chamber is part of a fluid system that guides particles (typically one at a time) in the flow to the focused laser beam for probing. The flow chamber may include a flow cell in a benchtop cytometer or a nozzle tip in a gas flow cytometer.

[0166] Light from the laser beam interacts with particles in the sample through diffraction, refraction, reflection, scattering, and absorption, and is re-emitted at various wavelengths depending on the characteristics of the particles (e.g., their size, internal structure, and the presence of one or more fluorescent molecules attached to, on, or naturally present in the particles). Fluorescence emission, along with diffracted, refracted, reflected, and scattered light, is routed via one or more beam splitters 445a-445g, bandpass filters 450a-450e, longpass filters 455a-455b, and a fluorescence collecting lens 440 to one or more forward scattering detectors 430, side scattering detectors 435, and one or more fluorescence detectors 460a-460f.

[0167] A fluorescence collecting lens 440 collects light emitted from the interaction of the particle laser beam and routes the light to one or more beamsplitters and filters. Bandpass filters, such as bandpass filters 450a-450e, allow a narrow range of wavelengths to pass through the filter. For example, bandpass filter 450a is a 510 / 20 filter. The first number indicates the center of the spectral band. The second number provides the range of the spectral band. Thus, a 510 / 20 filter extends 10 nm on each side of the center of the spectral band, or from 500 nm to 520 nm. Short-pass filters transmit light with wavelengths equal to or shorter than a specified wavelength. Long-pass filters, such as long-pass filters 455a-455b, transmit light with wavelengths equal to or longer than a specified wavelength. For example, long-pass filter 455a, as a 670 nm long-pass filter, transmits light with wavelengths equal to or longer than 670 nm. Filters are typically selected to optimize the detector's specificity for a particular fluorescent dye. Filters can be configured such that the spectral band of the light transmitted to the detector is close to the emission peak of the fluorescent dye.

[0168] Beam splitters direct light of different wavelengths in different directions. Beam splitters may be characterized by filter properties such as short-pass and long-pass characteristics. For example, beam splitter 445g is a 620 SP beam splitter, meaning that beam splitter 445g transmits light with wavelengths of 620 nm or shorter and reflects light with wavelengths greater than 620 nm in different directions. In one embodiment, beam splitters 445a-445g may include optical mirrors, such as dichroic mirrors.

[0169] A forward scattering detector 430 is positioned slightly off-axis from the direct beam passing through the flow cell and is configured to detect diffracted light, i.e., excitation light that primarily travels forward through or around the particle. The intensity of the light detected by the forward scattering detector depends on the overall size of the particle. The forward scattering detector may include a photodiode. A side scattering detector 435 is configured to detect refracted and reflected light from the particle surface and internal structure, and tends to increase with increasing particle structural complexity. Fluorescence emission of fluorescent molecules associated with the particle can be detected by one or more fluorescence detectors 460a-460f. The side scattering detector 435 and the fluorescence detector may include photomultiplier tubes. The signals detected at the forward scattering detector 430, the side scattering detector 435, and the fluorescence detector can be converted into electronic signals (voltages) by the detectors. This data can provide information about the sample.

[0170] Those skilled in the art will recognize that the flow cytometer according to embodiments of the present invention is not limited to... Figure 4B The flow cytometer shown may include any flow cytometer known in the art. For example, a flow cytometer may have any number of lasers, beam splitters, filters, and detectors, with various wavelengths and different configurations.

[0171] During operation, the cytometer is controlled by a controller / processor 490, and measurement data from the detector can be stored in memory 495 and processed by the controller / processor 490. Although not explicitly shown, the controller / processor 490 is coupled to the detector to receive output signals from it and can also be coupled to the electrical and electromechanical components of the flow cytometer 410 to control the laser, fluid flow parameters, etc. Input / output (I / O) functionality 497 may also be provided in the system. Memory 495, controller / processor 490, and I / O 497 can be provided entirely as integral parts of the flow cytometer 410. In such embodiments, a display may also form part of the I / O functionality 497 for presenting experimental data to the user of the cytometer 410. Alternatively, some or all of memory 495, controller / processor 490, and I / O functionality may be part of one or more external devices (e.g., a general-purpose computer). In some embodiments, some or all of memory 495 and controller / processor 490 may communicate wirelessly or wiredly with the cytometer 410. The controller / processor 490, combined with memory 495 and I / O 497, can be configured to perform various functions related to the preparation and analysis of flow cytometry experiments.

[0172] Figure 4BThe system shown contains six different detectors that detect fluorescence in six different wavelength bands (which may be referred to herein as “filter windows” for a given detector), defined by the configuration of filters and / or spectrometers in the beam path from flow cell 425 to each detector. Different fluorescent molecules used in flow cytometry experiments will emit light in their own characteristic wavelength bands. Specific fluorescent labels and their associated fluorescence emission bands can be selected for the experiment to generally coincide with the filter windows of the detectors. However, as more detectors are provided and more labels are used, a perfect correspondence between filter windows and fluorescence emission spectra is not possible. Typically, although the peak of the emission spectrum of a particular fluorescent molecule may lie within the filter window of a particular detector, some emission spectra of that label may also overlap with the filter windows of one or more other detectors. This can be referred to as overflow. I / O 497 can be configured to receive data on a flow cytometry experiment having a set of fluorescent labels and multiple cell populations with multiple markers, each cell population having a subset of multiple markers. I / O 497 can also be configured to receive biological data, marker density data, emission spectral data, data on label assignment to one or more markers, and cytometer configuration data, including data on assigning one or more markers to one or more cell populations. Flow cytometry experimental data (e.g., label spectral properties and flow cytometry configuration data) can also be stored in memory 495. Controller / processor 490 can be configured to evaluate the assignment of one or more labels to markers.

[0173] Figure 5 A functional block diagram of an example particle analyzer control system, such as an analysis controller 500, for analyzing and displaying biological events is shown. The analysis controller 500 can be configured to implement various processes for controlling the graphical display of biological events.

[0174] The particle analyzer or sorting system 502 can be configured to acquire biological event data. For example, a flow cytometer can generate flow cytometry event data. The particle analyzer 502 can be configured to provide biological event data to the analysis controller 500. A data communication channel can be included between the particle analyzer or sorting system 502 and the analysis controller 500. Biological event data can be provided to the analysis controller 500 via the data communication channel.

[0175] Analysis controller 500 can be configured to receive bioevent data from particle analyzer or sorting system 502. The bioevent data received from particle analyzer or sorting system 502 may include flow cytometry event data. Analysis controller 500 can be configured to provide a graphical display of a first graph including the bioevent data to display device 506. Analysis controller 500 can be further configured to render regions of interest as gating around the bioevent data cluster shown by display device 506, for example, overlaying it on the first graph. In some embodiments, gating may be a logical combination of one or more graphical regions of interest plotted on a single parameter histogram or bivariate graph. In some embodiments, the display may be used to display particle parameters or saturation detector data.

[0176] The analysis controller 500 can be further configured to display bio-event data on the display device 506 that differs from other events in the bio-event data outside the gate. For example, the analysis controller 500 can be configured to render the colors of the bio-event data contained within the gate differently than the colors of the bio-event data outside the gate. The display device 506 can be implemented as a monitor, tablet computer, smartphone, or other electronic device configured to present a graphical interface.

[0177] The analysis controller 500 can be configured to receive a gating selection signal from a first input device for gating authentication. For example, the first input device can be implemented as a mouse 510. The mouse 510 can initiate a gating selection signal to the analysis controller 500 to authenticate a gating to be displayed on or manipulated by the display device 506 (e.g., providing a click on or in a desired door when the cursor is over the desired door). In some embodiments, the first device can be implemented as a keyboard 508 or other means for providing input signals to the analysis controller 500, such as a touchscreen, stylus, optical detector, or voice authentication system. Some input devices may include multiple input functions. In such embodiments, each input function can be considered an input device. For example, such as... Figure 5 As shown, the mouse 510 may include a right mouse button and a left mouse button, and each button can generate a trigger event.

[0178] Triggering events can cause the analysis controller 500 to change how data is displayed, which parts of the data are actually displayed on the display device 506, and / or provide input for further processing, such as the selection of groups of interest for particle sorting.

[0179] In some implementations, the analysis controller 500 can be configured to detect when the mouse 510 initiates gating selection. The analysis controller 500 can be further configured to automatically modify the plotting visualization to facilitate the gating process. The modification can be based on a specific distribution of the biological event data received by the analysis controller 500.

[0180] The analysis controller 500 can be connected to a storage device 504. The storage device 504 can be configured to receive and store biological event data from the analysis controller 500. The storage device 504 can also be configured to receive and store flow cytometry event data from the analysis controller 500. The storage device 504 can be further configured to allow the analysis controller 500 to retrieve biological event data, such as flow cytometry event data.

[0181] Display device 506 can be configured to receive display data from analysis controller 500. The display data may include graphs of biological event data and gating of portions of the graphs. Display device 506 can be further configured to change the presented information based on input received from analysis controller 500, along with input from particle analyzer 502, storage device 504, keyboard 508, and / or mouse 510.

[0182] In some implementations, the analysis controller 500 may generate a user interface to receive sample events for sorting. For example, the user interface may include controls for receiving sample events or sample images. Sample events, images, or sample gating may be provided before collecting event data for the sample, or based on an initial set of events for a portion of the sample.

[0183] Figure 6A This is a schematic diagram of a particle sorting system 600 (e.g., a particle analyzer or sorting system 502) according to one embodiment described herein. In some embodiments, the particle sorting system 600 is a cell sorting system. Figure 6A As shown, a droplet-forming transducer 602 (e.g., a piezoelectric oscillator) is coupled to a fluid conduit 601, which may be coupled to, contain, or be the nozzle 603 itself. Within the fluid conduit 601, a sheath fluid 604 hydrodynamically focuses a sample fluid 606 containing particles 609 into a moving fluid column 608 (e.g., a flow). Within the moving fluid column 608, particles 609 (e.g., cells) are arranged in a single file, passing through a monitored area 611 (e.g., the intersection of laser and flow), and irradiated by a radiation source 612 (e.g., a laser). Vibration of the droplet-forming transducer 602 causes the moving fluid column 608 to break into multiple droplets 610, some of which contain particles 609.

[0184] In operation, a detection station 614 (e.g., an event detector) identifies when a particle (or cell) of interest crosses a monitoring area 611. The detection station 614 is fed into a timing circuit 628, which in turn feeds into a flash charge circuit 630. At the droplet splitting point, a timing droplet delay (Δt) informs the flash charge that it can be applied to the moving fluid column 608, causing the droplet of interest to become charged. The droplet of interest may contain one or more particles or cells to be sorted. The charged droplet can then be deflected into a container (e.g., a collection tube or a porous or microporous sample plate) for sorting by activating a deflection plate (not shown), where pores or micropores can be associated with a specific droplet of interest. Figure 6A As shown, the droplets can be collected in the drainage container 638.

[0185] A detection system 616 (e.g., a droplet boundary detector) is used to automatically determine the phase of the droplet drive signal as a particle of interest passes through a monitoring region 611. U.S. Patent No. 7,679,039 describes an exemplary droplet boundary detector, the entire contents of which are incorporated herein by reference. The detection system 616 enables the instrument to accurately calculate the position of each detected particle within the droplet. The detection system 616 may be fed an amplitude signal 620 and / or a phase signal 618, which in turn (via amplifier 622) are fed into an amplitude control circuit 626 and / or a frequency control circuit 624. The amplitude control circuit 626 and / or the frequency control circuit 624, in turn, control the droplet forming transducer 602. The amplitude control circuit 626 and / or the frequency control circuit 624 may be included in a control system.

[0186] In some embodiments, sorting electronics (e.g., detection system 616, detection station 614, and processor 640) may be coupled to a memory configured to store detected events and sorting decisions based on those events. The sorting decisions may be included in the event data of the particles. In some embodiments, detection system 616 and detection station 614 may be implemented as a single detection unit or communication coupling, such that event measurements can be collected by one of detection system 616 or detection station 614 and provided to non-collecting elements.

[0187] Figure 6B This is a schematic diagram of a particle sorting system according to one embodiment described herein. Figure 6B The particle sorting system 600 shown includes deflection plates 652 and 654. Charge can be applied via a current-charged wire in the barbs. This generates a droplet stream 610 containing particles 609 for analysis. The particles can be irradiated with one or more light sources (e.g., lasers) to produce light scattering and fluorescence information. Particle information can be obtained through sorting electronics or other detection systems. Figure 6B(Not shown in the image) Analysis is performed. Deflection plates 652 and 654 can be independently controlled to attract or repel charged droplets, thereby guiding the droplets to a target collection container (e.g., one of 672, 674, 676, or 678). Figure 6B As shown, deflectors 652 and 654 can be controlled to direct particles along a first path 662 to container 674, or along a second path 668 to container 678. If a particle is not of interest (e.g., no scattering or irradiation information is displayed within a specified sorting area), the deflectors can allow the particle to continue flowing along path 664. These uncharged droplets can enter the waste container via, for example, an aspirator 670.

[0188] The sorting electronics may include functions for initiating measurement data collection, receiving fluorescence signals from particles, and determining how to adjust deflection plates to sort the particles. Figure 6B Example implementations of the illustrated scheme include the BD FACSAria™ series of flow cytometers commercially available from Becton, Dickinson and Company (Franklin Lakes, NJ).

[0189] A method for classifying cancer cells in whole blood samples in a flowing stream.

[0190] This disclosure also includes methods for classifying cancer cells. According to some embodiments, the method includes irradiating a whole blood sample containing cells in a flowing stream with a light source, capturing an image of the light emitted from the irradiated cells using a light detection system with a photodetector, and applying a neural network to the captured image to detect the presence of cancer cells in the whole blood sample with a minimum concentration of cancer cells and to determine the classification of the cancer cells among multiple categories. In some cases, the method includes training the neural network using images of samples labeled with a fluorophore benchmark truth marker.

[0191] When implementing the subject method according to certain embodiments, a light detection system with a photodetector measures light from a sample containing cells in a flowing stream (e.g., irradiating the sample with a light source). A “biological sample” can refer to a whole organism, a plant, fungus, or a subset of animal tissues, cells, or components, which in some cases may be present in blood, mucus, lymph, synovial fluid, cerebrospinal fluid, saliva, bronchoalveolar lavage fluid, amniotic fluid, amniotic cord blood, urine, vaginal secretions, and semen. Therefore, a “biological sample” refers both to a natural organism or a subset of its tissues and to homogenates, lysates, or extracts prepared from an organism or a subset of its tissues, including but not limited to, for example, plasma, serum, cerebrospinal fluid, lymph, skin sections, respiratory tract, gastrointestinal tract, cardiovascular and genitourinary tract, tears, saliva, breast milk, blood cells, tumors, and organs. A biological sample can be any type of biological tissue, including healthy tissue and diseased tissue (e.g., cancerous, malignant, necrotic, etc.). In some implementations, the biological sample is a liquid sample, such as blood or its derivatives, such as plasma, tears, urine, semen, etc. In some cases, the sample is a blood sample, including whole blood, such as blood obtained from venipuncture or finger prick (where the blood may or may not be mixed with any reagents such as preservatives, anticoagulants, etc., before testing). In some cases, the sample is a whole blood sample.

[0192] In some embodiments, a sample containing cells (e.g., in a flow stream of a flow cytometer) is irradiated with light from a light source. In some embodiments, the light source is a broadband light source that emits light with a wide range of wavelengths, for example, spanning 50 nm or greater, such as 100 nm or greater, such as 150 nm or greater, such as 200 nm or greater, such as 250 nm or greater, such as 300 nm or greater, such as 350 nm or greater, such as 400 nm or greater, and including 500 nm or greater. For example, a suitable broadband light source emits light with wavelengths from 200 nm to 1500 nm. Another example of a suitable broadband light source includes a light source that emits light with wavelengths from 400 nm to 1000 nm. If the method involves irradiation with a broadband light source, the broadband light source schemes of interest may include, but are not limited to, halogen lamps, deuterium arc lamps, xenon arc lamps, stable fiber-coupled broadband light sources, broadband LEDs with a continuous spectrum, superluminescent diodes, semiconductor light-emitting diodes, broadband LED white light sources, multi-LED integrated white light sources, and other broadband light sources or any combination thereof.

[0193] In other embodiments, the method includes irradiation with a narrowband light source that emits a specific wavelength or a narrow wavelength range, such as a light source emitting light in a narrow wavelength range, such as 50 nm or less, for example 40 nm or less, for example 30 nm or less, for example 25 nm or less, for example 20 nm or less, for example 15 nm or less, for example 10 nm or less, for example 5 nm or less, for example 2 nm or less, and includes a light source emitting light of a specific wavelength (i.e., monochromatic light). If the method includes irradiation with a narrowband light source, the narrowband light source scheme of interest may include, but is not limited to, narrow-wavelength LEDs, laser diodes, or broadband light sources coupled with one or more optical bandpass filters, diffraction gratings, monochromators, or any combination thereof.

[0194] In some embodiments, the method includes irradiating a sample with two or more lasers. In some cases, the type and number of lasers will depend on the sample and the light to be collected, and may be gas lasers, such as helium-neon lasers, argon lasers, krypton lasers, xenon lasers, nitrogen lasers, CO2 lasers, CO lasers, argon-fluorine (ArF) excimer lasers, krypton-fluorine (KrF) excimer lasers, xenon-chlorine (XeCl) excimer lasers, or xenon-fluorine (XeF) excimer lasers, or combinations thereof. In other cases, the method includes irradiating a flow stream with a dye laser, such as a stilbene, coumarin, or rhodamine laser. In still other cases, the method includes irradiating a flow stream with a metal vapor laser, such as a helium-cadmium (HeCd) laser, a helium-mercury (HeHg) laser, a helium-selenium (HeSe) laser, a helium-silver (HeAg) laser, a strontium laser, a neon-copper (NeCu) laser, a copper laser, or a gold laser, or combinations thereof. In other cases, the method includes irradiating the flow with a solid-state laser, such as a ruby ​​laser, an Nd:YAG laser, an NdCrYAG laser, an Er:YAG laser, an Nd:YLF laser, an Nd:YVO4 laser, an Nd:yCa4O(BO3)3 laser, an Nd:YCOB laser, a titanite sapphire laser, a thulium YAG laser, a ytterbium YAG laser, a ytterbium₂O₃ laser, or a cerium-doped laser, or combinations thereof.

[0195] The sample can be irradiated with one or more of the above-described light sources, such as two or more light sources, three or more light sources, four or more light sources, five or more light sources, and including ten or more light sources. The light sources can include combinations of any type of light source. For example, in some embodiments, the method includes irradiating the sample in the flow with a laser array (e.g., an array having one or more gas lasers, one or more dye lasers, and one or more solid-state lasers).

[0196] The sample can be irradiated with wavelengths ranging from 200 nm to 1500 nm, such as 250 nm to 1250 nm, 300 nm to 1000 nm, 350 nm to 900 nm, and including 400 nm to 800 nm. For example, if the light source is a broadband light source, the sample can be irradiated with wavelengths from 200 nm to 900 nm. In other cases, if the light source comprises multiple narrowband light sources, the sample can be irradiated with specific wavelengths in the range of 200 nm to 900 nm. For example, the light source can be multiple narrowband LEDs (1 nm - 25 nm) that each independently emit light in the wavelength range of 200 nm to 900 nm. In other embodiments, the narrowband light source comprises one or more lasers (e.g., a laser array), and the sample can be irradiated with light in the range of 200 nm to 700 nm, for example, using a laser array having gas lasers, excimer lasers, dye lasers, metal vapor lasers, and solid-state lasers as described above.

[0197] When using more than one light source, the sample can be irradiated with the light sources simultaneously, sequentially, or in combination. For example, the sample can be irradiated simultaneously with each light source. In other embodiments, the flow stream is irradiated sequentially with each light source. When multiple light sources are used to sequentially irradiate the sample, the irradiation time of each light source can be independently 0.001 microseconds or longer, for example, 0.01 microseconds or longer, for example, 0.1 microseconds or longer, for example, 1 microsecond or longer, for example, 5 microseconds or longer, for example, 10 microseconds or longer, for example, 30 microseconds or longer, and includes 60 microseconds or longer. For example, the method can include irradiating the sample with a light source (e.g., a laser) for a duration ranging from 0.001 microseconds to 100 microseconds, for example, 0.01 microseconds to 75 microseconds, for example, 0.1 microseconds to 50 microseconds, for example, 1 microsecond to 25 microseconds, and includes 5 microseconds to 10 microseconds. In embodiments where two or more light sources are used to sequentially irradiate the sample, the irradiation duration of each light source can be the same or different.

[0198] The time interval between irradiations from each light source can also vary as needed, with independent intervals of 0.001 microseconds or longer, such as 0.01 microseconds or longer, 0.1 microseconds or longer, 1 microsecond or longer, 5 microseconds or longer, 10 microseconds or longer, 15 microseconds or longer, 30 microseconds or longer, and including 60 microseconds or longer. For example, the time interval between irradiations from each light source can range from 0.001 microseconds to 60 microseconds, such as 0.01 microseconds to 50 microseconds, such as 0.1 microseconds to 35 microseconds, such as 1 microsecond to 25 microseconds, and including 5 microseconds to 10 microseconds. In some embodiments, the time interval between irradiations from each light source is 10 microseconds. In embodiments where the sample is sequentially irradiated by more than two (i.e., three or more) light sources, the delay between irradiations from each light source can be the same or different.

[0199] The sample can be irradiated continuously or at discontinuous intervals. In some cases, the method involves continuously irradiating the sample with a light source. In other cases, the sample is irradiated with a light source at discontinuous intervals, such as every 0.001 milliseconds, every 0.01 milliseconds, every 0.1 milliseconds, every 1 millisecond, every 10 milliseconds, every 100 milliseconds, and including every 1000 milliseconds or other intervals.

[0200] Depending on the light source, the sample can be irradiated from different distances, such as 0.01 mm or more, 0.05 mm or more, 0.1 mm or more, 0.5 mm or more, 1 mm or more, 2.5 mm or more, 5 mm or more, 10 mm or more, 15 mm or more, 25 mm or more, and including 50 mm or more. Furthermore, the irradiation angle can also vary, ranging from 10° to 90°, such as 15° to 85°, 20° to 80°, 25° to 75°, and including 30° to 60°, such as 90°.

[0201] In some embodiments, the method includes irradiating a cell sample with a frequency-modulated light beam. In some cases, the method includes irradiating the sample with two or more frequency-shifted light beams. As described above, a beam generator assembly having a laser and an acousto-optic device can be used to frequency-shift the laser. In these embodiments, the method includes irradiating the acousto-optic device with a laser. Depending on the desired wavelength of the light generated in the output laser beam (e.g., for irradiating a sample in a flowing stream), the laser can have a specific wavelength varying from 200 nm to 1500 nm (e.g., 250 nm to 1250 nm, e.g., 300 nm to 1000 nm, e.g., 350 nm to 900 nm, and including 400 nm to 800 nm). The acousto-optic device can be irradiated with one or more lasers, e.g., two or more lasers, e.g., three or more lasers, e.g., four or more lasers, e.g., five or more lasers, and includes ten or more lasers. The lasers can include any type of laser combination. For example, in some embodiments, the method includes irradiating an acousto-optic device with a laser array (e.g., an array having one or more gas lasers, one or more dye lasers, and one or more solid-state lasers).

[0202] When using more than one laser, the acousto-optic device can be irradiated with the laser simultaneously, sequentially, or a combination of both. For example, the acousto-optic device can be irradiated with each laser simultaneously. In other embodiments, the acousto-optic device is irradiated with each laser sequentially. When using more than one laser to irradiate the acousto-optic device sequentially, the irradiation time of each laser can be independently 0.001 microseconds or longer, for example, 0.01 microseconds or longer, for example, 0.1 microseconds or longer, for example, 1 microsecond or longer, for example, 5 microseconds or longer, for example, 10 microseconds or longer, for example, 30 microseconds or longer, and includes 60 microseconds or longer. For example, the method can include irradiating the acousto-optic device with lasers for durations ranging from 0.001 microseconds to 100 microseconds, for example, 0.01 microseconds to 75 microseconds, for example, 0.1 microseconds to 50 microseconds, for example, 1 microsecond to 25 microseconds, and including 5 microseconds to 10 microseconds. In an embodiment where the acousto-optic device is irradiated sequentially by two or more lasers, the duration of irradiation of the acousto-optic device by each laser can be the same or different.

[0203] The time interval between irradiations by each laser can also vary as needed, independently spaced by delays of 0.001 microseconds or longer, such as 0.01 microseconds or longer, 0.1 microseconds or longer, 1 microsecond or longer, 5 microseconds or longer, 10 microseconds or longer, 15 microseconds or longer, 30 microseconds or longer, and including 60 microseconds or longer. For example, the time interval between irradiations by each light source can range from 0.001 microseconds to 60 microseconds, such as 0.01 microseconds to 50 microseconds, such as 0.1 microseconds to 35 microseconds, such as 1 microsecond to 25 microseconds, and including 5 microseconds to 10 microseconds. In some embodiments, the time interval between irradiations by each laser is 10 microseconds. In embodiments where the acousto-optic device is sequentially irradiated by more than two (i.e., three or more) lasers, the delay between irradiations by each laser can be the same or different.

[0204] Acousto-optic devices can be irradiated continuously or at discontinuous intervals. In some cases, the method involves continuously irradiating the acousto-optic device with a laser. In other cases, the acousto-optic device is irradiated with a laser at discontinuous intervals, such as every 0.001 milliseconds, every 0.01 milliseconds, every 0.1 milliseconds, every 1 millisecond, every 10 milliseconds, every 100 milliseconds (including every 1000 milliseconds), or other intervals.

[0205] Depending on the laser used, the irradiation distance of the acousto-optic device can vary: 0.01 mm or more, for example 0.05 mm or more, for example 0.1 mm or more, for example 0.5 mm or more, for example 1 mm or more, for example 2.5 mm or more, for example 5 mm or more, for example 10 mm or more, for example 15 mm or more, for example 25 mm or more, and including 50 mm or more. Furthermore, the irradiation angle can also vary, ranging from 10° to 90°, for example 15° to 85°, for example 20° to 80°, for example 25° to 75°, and including 30° to 60°, for example 90°.

[0206] In one embodiment, the method includes applying radio frequency (RF) drive signals to an acousto-optic device to generate an angle-deflected laser beam. Two or more RF drive signals can be applied to the acousto-optic device to generate an output laser beam having a desired number of angle-deflected laser beams, such as three or more RF drive signals, four or more RF drive signals, five or more RF drive signals, six or more RF drive signals, seven or more RF drive signals, eight or more RF drive signals, nine or more RF drive signals, ten or more RF drive signals, fifteen or more RF drive signals, twenty-five or more RF drive signals, fifty or more RF drive signals, and including one hundred or more RF drive signals.

[0207] Each angle-deflecting laser beam generated by a radio frequency (RF) drive signal has an intensity based on the amplitude of the applied RF drive signal. In some embodiments, the method includes applying an RF drive signal with an amplitude sufficient to generate an angle-deflecting laser beam with a desired intensity. In some cases, the RF drive signal for each application independently has an amplitude of about 0.001 V to about 500 V, for example about 0.005 V to about 400 V, for example about 0.01 V to about 300 V, for example about 0.05 V to about 200 V, for example about 0.1 V to about 100 V, for example about 0.5 V to about 75 V, for example about 1 V to 50 V, for example about 2 V to 40 V, for example about 3 V to about 30 V, and includes about 5 V to about 25 V. In some implementations, the radio frequency drive signal for each application has a frequency of about 0.001 MHz to about 500 MHz, for example about 0.005 MHz to about 400 MHz, for example about 0.01 MHz to about 300 MHz, for example about 0.05 MHz to about 200 MHz, for example about 0.1 MHz to about 100 MHz, for example about 0.5 MHz to about 90 MHz, for example about 1 MHz to about 75 MHz, for example about 2 MHz to about 70 MHz, for example about 3 MHz to about 65 MHz, for example about 4 MHz to about 60 MHz, and includes about 5 MHz to about 50 MHz.

[0208] In these embodiments, the angle-deflecting laser beams in the output laser beam are spatially separated. Depending on the applied RF drive signal and the desired irradiance distribution of the output laser beam, the angle-deflecting laser beams may be spaced apart by 0.001 μm or more, for example, 0.005 μm or more, for example, 0.01 μm or more, for example, 0.05 μm or more, for example, 0.1 μm or more, for example, 0.5 μm or more, for example, 1 μm or more, for example, 5 μm or more, for example, 10 μm or more, for example, 100 μm or more, for example, 500 μm or more, for example, 1000 μm or more, and including 5000 μm or more. In some embodiments, the angle-deflecting laser beams overlap, for example, with adjacent angle-deflecting laser beams along the horizontal axis of the output laser beam. The overlap between adjacent angle-deflected laser beams (e.g., overlap of beam points) can be 0.001 μm or greater, such as 0.005 μm or greater, such as 0.01 μm or greater, such as 0.05 μm or greater, such as 0.1 μm or greater, such as 0.5 μm or greater, such as 1 μm or greater, such as 5 μm or greater, such as 10 μm or greater, and includes 100 μm or greater overlap.

[0209] In some cases, a sample in a flowing stream is irradiated with a multi-beam frequency-shifting beam, and images of cellular mitochondria in the flowing stream are generated as described below, for example, Diebold et al., Nature Photonics Vol. 7(10); U.S. Patents 806-810 (2013), and U.S. Patents 9,423,353, 9,784,661, 9,983,132, 10,006,852, 10,078,045, 10,036,699, 10,222,316, 10,288,546, 10,324,019, 10,408,758, 10,451,538, 10,620,111 and U.S. Patent Publications 2017 / 0133857, 2017 / 0328826, 2017 / 0350803, 2018 / 0275042, 2019 / 0376895 and 2019 / 0376894, the disclosures of which are incorporated herein by reference.

[0210] Light from irradiated cells in the sample is transmitted to a light detection system described in more detail below and measured by multiple photodetectors. In some embodiments, the method includes measuring light collected within a certain wavelength range (e.g., 200 nm–1000 nm). For example, the method may include the spectrum of light collected in one or more wavelength ranges within the 200 nm–1000 nm range. In other embodiments, the method includes measuring light collected at one or more specific wavelengths. For example, light collected at one or more of the following wavelengths may be measured: 450 nm, 518 nm, 519 nm, 561 nm, 578 nm, 605 nm, 607 nm, 625 nm, 650 nm, 660 nm, 667 nm, 670 nm, 668 nm, 695 nm, 710 nm, 723 nm, 780 nm, 785 nm, 647 nm, 617 nm, and any combination thereof.

[0211] The collected light can be measured continuously or at discontinuous intervals. In some cases, the method involves measuring the light continuously. In others, the light is measured at discontinuous intervals, such as every 0.001 milliseconds, every 0.01 milliseconds, every 0.1 milliseconds, every 1 millisecond, every 10 milliseconds, every 100 milliseconds, and including every 1000 milliseconds or some other interval.

[0212] The collected light can be measured once or multiple times during the main method, for example, two or more times, three or more times, five or more times, and including ten or more times. In some embodiments, the light from the sample is measured two or more times, and in some cases the data are averaged.

[0213] Measuring light from cells in a sample at one or more wavelengths (e.g., 5 or more different wavelengths, 10 or more different wavelengths, 25 or more different wavelengths, 50 or more different wavelengths, 100 or more different wavelengths, 200 or more different wavelengths, 300 or more different wavelengths), and including measuring the collected light at 400 or more different wavelengths.

[0214] In some implementations, the method includes further adjusting the light from the sample before detection. For example, light from the sample source may pass through one or more lenses, mirrors, pinholes, slits, gratings, light refractors, and any combination thereof. In some cases, the collected light passes through one or more focusing lenses, for example, to reduce the light profile. In other cases, the emitted light from the sample passes through one or more collimators to reduce beam divergence.

[0215] In some embodiments, image data is generated from measured light. In some cases, image data includes data waveforms generated in one or more photodetector channels. In some embodiments, one or more images of cells in a sample are generated. Images can be generated from detected fluorescence, detected light absorption, detected light scattering, detected light emission, or any combination thereof. In some cases, images are generated from fluorescence from one or more fluorescent labels. In some cases, images are generated from light absorption detected from the sample, such as light absorption from a bright-field photodetector. In these cases, images are generated based on bright-field image data from particles in a flowing stream. In other cases, images are generated from light scattering detected from the sample, such as light scattering from a side-scatter detector, a forward-scatter detector, or a combination of side-scatter and forward-scatter detectors. In these cases, images are generated based on scattered light image data. In still other cases, images are generated from light emitted from the sample. In other cases, images are generated from a combination of detected light absorption, light scattering, and light emission. In some cases, images of cell mitochondria are generated from fluorescence from one or more fluorescent labels and one or more light losses, side-scattered light, and forward-scattered light.

[0216] One or more images can be generated from the measured light. In some embodiments, a single image is generated for each cell in the sample based on each form of detected light. In other embodiments, multiple images are generated for each cell, such as two or more, three or more, five or more, ten or more, and including generating 25 or more images for each cell. For example, a first image of the cell is generated from fluorescence; a second image of the cell is generated from detected light absorption; and a third image of the cell is generated from detected light scattering. In other embodiments, two or more images are generated from each form of detected light, such as three or more, four or more, five or more, and including ten or more images or combinations thereof.

[0217] In some implementations, frequency-coded data (e.g., frequency-coded spatial data) is generated from measurement light on fluorescently labeled cells in a flowing stream. In some cases, one or more images are generated from the frequency-coded data. The frequency-coded data can be generated in one or more detection channels, such as two or more, three or more, four or more, five or more, six or more, and including eight or more detection channels. In some implementations, the frequency-coded data includes data components acquired (or derived) from different detectors (e.g., fluorescence detection channels, detected light absorption, or detected light scattering). In some cases, the frequency-coded data undergoes phase correction. In some cases, the frequency-coded data used to generate cell images is phase-corrected by performing a transform on the frequency-coded data. For example, phase correction is performed on the frequency-coded data by performing a Fourier Transform (FT). In another example, phase correction is performed on the frequency-coded data by performing a Discrete Fourier Transform (DFT). In yet another example, phase correction is performed on the frequency-coded data by performing a Short-Time Fourier Transform (STFT). In some implementations, the method includes performing a transformation on the frequency-coded data without performing any mathematical imaginary operations (i.e., performing only the mathematical real operations of the transformation) to generate an image from the frequency-coded data.

[0218] In some embodiments, the minimum concentration of cancer cells present in a whole blood sample is at least about 0.01%, at least about 0.05%, at least about 0.1%, at least about 0.5%, at least about 1%, at least about 2%, at least about 3%, at least about 4%, at least about 5%, at least about 6%, at least about 7%, at least about 8%, at least about 9%, and includes at least about 10%. In some cases, the cancer cells in the sample constitute 1% or less of the cells in the sample (e.g., a whole blood sample), such as 0.5% or less, 0.1% or less, 0.05% or less, 0.01% or less, 0.005% or less, 0.001%, and include cases where the cancer cells constitute 0.0001% or less of the cells in the sample.

[0219] In some embodiments, the method includes applying a neural network to captured images to detect the presence of cancer cells in a whole blood sample with a minimum concentration of cancer cells. In some embodiments, the method also includes determining the classification of cancer cells. In some embodiments, the method includes classifying cancer cells in the sample based on image data (e.g., waveforms generated in each photodetector channel), generated cell images, calculated image parameters, or a combination thereof. In some embodiments, multiple classifications include two or more classifications, such as three or more classifications, such as four or more classifications, such as five or more classifications, such as six or more classifications, such as seven or more classifications, such as eight or more classifications, such as nine or more classifications, and including ten or more classifications. In some cases, multiple classifications include ovarian cancer classification, T-lymphocytic carcinoma classification, breast cancer classification, colon cancer classification, or epithelial carcinoma classification. In some cases, multiple classifications include OVCAR classification, Jurkat classification, MCF-7 classification, and adenocarcinoma classification.

[0220] In some implementations, the method involves applying machine learning algorithms to captured images to determine the classification of cancer cells. Any suitable machine learning algorithm can be implemented, including but not limited to linear regression, logistic regression, Naive Bayes, k-nearest neighbors (kNN), random forest, decision tree, support vector machine, gradient boosting, and clustering algorithms. In some implementations, the method includes applying neural networks, such as artificial neural networks, convolutional neural networks, or recurrent neural networks. In some implementations, the method includes implementing a Python script. In some implementations, the method includes applying artificial neural networks (e.g., convolutional neural networks, etc.), Bayesian statistics, learning automata, hidden Markov models, linear classifiers, quadratic classifiers, association rule learning, and / or similar methods.

[0221] In some implementations, a neural network comprises a network of nodes. Nodes can be organized into layers, where the first layer is the input layer, from which data flows. The neural network can also include output layers from which transformed data flows out. Each node can have multiple inputs and a single output (e.g., input layer nodes have only a single input). The output of a node represents a linear combination of its inputs. In other words, the input can be multiplied by an associated constant. The product can be accumulated along a node path with a constant offset. This constant offset, or “bias,” can represent another degree of freedom that can be adjusted during training. For example, the constant offset could be a threshold, as it can reduce node values ​​below zero, causing the activation function to output zero.

[0222] The obtained value is evaluated using an activation function and used as the node's output. Nodes in a given layer of a neural network are connected to each node in their neighboring layers. A neural network can be trained by using gradient descent and an error function to compare the network's expected output with its actual output, thereby minimizing network error. The weights of one or more nodes can be adjusted to simulate the desired results produced by the network.

[0223] In some cases, the method includes classifying cells based on one or more parameters computed from the generated image data. In some cases, the method includes applying a classifier neural network to classify cells. In some cases, the neural network includes an input layer consisting of one or more parameters computed from the generated image data. In some cases, the input layer contains only imaging parameters. In some cases, the neural network includes an output layer that is a probability estimate based on cell identity (e.g., the cell is a cancer cell, circulating tumor cell). In some cases, the neural network includes a learning algorithm configured to learn from classification errors. In some cases, the method includes updating classification parameters based on one or more parameters computed from the generated image data. In some cases, the updating of classification parameters is based on a comparison between one or more parameters computed from the generated image data and one or more parameters computed from benchmark ground truth image data. In some cases, the neural network includes a learning algorithm configured to learn from classification errors. In some cases, the method includes generating benchmark ground truth image data. In some cases, the method includes updating classification parameters based on one or more parameters computed from the captured image. In some cases, the classification parameters are updated based on a comparison between one or more parameters calculated from the captured image and one or more parameters calculated from the baseline ground truth image data.

[0224] In some cases, samples are labeled with fluorophore-based truth markers. In some embodiments, specific fluorophores are used as truth markers to identify different cancer cell types during training. In some embodiments, the method includes contacting cells with one or more fluorophores to fluorescently label the cells, such as the fluorophores described above. The term "label" is used herein in its conventional sense to refer to the coupling or binding of one or more components of a sample cell to one or more detection components (e.g., components detectable by luminescence (e.g., fluorescence, phosphorescence, etc.), contrast agents, radioisotopes, etc.). For example, labeling may include coupling particles to detection components via non-covalent interactions (e.g., hydrogen bonds, dipole-dipole bonds, ionic bonds) or via one or more covalent bonds.

[0225] In some cases, neural networks apply sigmoid activation functions, Adam optimizers, adaptive learning rates, dynamic thresholding, binary classification (e.g., positive vs. negative), binary cross-entropy loss functions, or any combination thereof. In some cases, the method involves normalizing the intensity of the captured image. In some cases, the neural network is applied to normalize the image. In some cases, the neural network includes an input layer consisting of one or more parameters computed from the generated image data.

[0226] In some implementations, the neural network is trained using images of samples labeled with a fluorophore benchmark. In some cases, the method includes training the neural network with a dataset reflecting the ratio of positive to negative categories of cancer cells to non-cancer cells. In some cases, the detection accuracy of the neural network is validated on multiple samples containing different types of cancer cells. In some cases, the neural network is configured to distinguish between cancer cells and non-cancer cells based on specific biomarkers. In some cases, the method includes normalizing the intensity of the captured images. In some cases, the neural network is applied to the normalized images. In some cases, normalized fluorescence intensity of cells improves detection accuracy, for example, by 5% or more, 10% or more, 25% or more, 50% or more, 75% or more, 90% or more, and including improvements of 99% or more.

[0227] In some implementations, the method includes determining whether a cell belongs to a rare cell population. The term "rare cell" is used herein in its conventional sense, referring to cells in a sample where the abundance of that cell type is 0.1% or less of the cell population, for example, 0.05% or less, 0.01% or less, 0.005% or less, 0.001% or less, 0.0005% or less, and includes cases where the abundance of that cell type is 0.0001% or less of the cell population. In some cases, calculating cell parameters includes determining that the rare cell is a stem cell. In some cases, calculating cell parameters includes determining that the rare cell is a circulating endothelial cell. In some cases, calculating cell parameters includes determining that the rare cell is a circulating tumor (malignant or benign) cell. In some cases, calculating cell parameters includes determining that the rare cell is a residual lesion cell, such as a blood or bone marrow disease cell.

[0228] In some implementations, the method includes validating the detection accuracy of the neural network on multiple different samples, such as samples containing different types of cancer cells. In some cases, the detection accuracy of the neural network is validated on two or more different samples, such as three or more, four or more, five or more, six or more, twelve or more, sixteen or more, twenty-four or more, and including 32 or more different samples. The samples may contain two or more different types of cancer cells, such as three or more, four or more, five or more, six or more, seven or more, eight or more, nine or more, and including ten or more different types of cancer cells.

[0229] In some embodiments, the method includes determining one or more sorting gates for classifying cells in a sample. The term "gating," used herein in its conventional sense, refers to a classifier boundary that identifies a subset of data of interest. In some cases, gating can restrict a set of events of particular interest. Furthermore, performing "gating" can refer to the process of classifying data using a gating defined for a given dataset, where the gating can be one or more regions of interest combined with Boolean logic. In some embodiments, gating identifies particles exhibiting the same image parameters. Examples of gating methods have been described, for example, in U.S. Patent Nos. 4,845,653; 5,627,040; 5,739,000; 5,795,727; 5,962,238; 6,014,904; 6,944,338; and 8,990,047; the disclosures of which are incorporated herein by reference. In some embodiments, gating restricts a clustering of particle populations from one or more different samples, which has been previously determined (e.g., by a user) to correspond to an attribute of interest.

[0230] In some cases, one or more sorting gates capture cells of the target particle population cluster and exclude particles that are not in the target particle population cluster. In some cases, the sorting gate maximizes the inclusion rate of cells in the target particle population cluster. In some cases, the sorting gate is configured to maximize the inclusion rate of the target particle population cluster, for example, wherein the sorting gate is configured to generate an inclusion rate of 50% or greater of the cells in the target particle population cluster, such as 55% or greater, 60% or greater, 65% or greater, 70% or greater, 75% or greater, 80% or greater, 85% or greater, 90% or greater, 95% or greater, 97% or greater, 99% or greater, and includes determining a sorting gate configured to generate an inclusion rate of 99.9% or greater of the cells in the target particle population cluster. In some cases, the sorting gate is configured to maximize the purity yield of cells from the target particle population cluster, for example, wherein the sorting gate is configured to generate a purity yield of 50% or higher of the target particle population cluster cells, such as 55% or higher, 60% or higher, 65% or higher, 70% or higher, 75% or higher, 80% or higher, 85% or higher, 90% or higher, 95% or higher, 97% or higher, 99% or higher, and includes determining a sorting gate configured to generate a purity yield of 99.9% or higher of the target particle population cluster cells.

[0231] In some cases, sorting gates are configured to maximize the exclusion of particles from non-target particle groups. In some cases, sorting gates are configured to exclude 50% or more of the particles in the non-target particle group, such as 55% or more, 60% or more, 65% or more, 70% or more, 75% or more, 80% or more, 85% or more, 90% or more, 95% or more, 97% or more, 99% or more, and include determining sorting gates configured to exclude 99.9% or more of the particles in the non-target particle group cluster.

[0232] In some cases, the method includes evaluating the sorting gating of the sorting strategy and adjusting one or more generated sorting gatings. This adjustment can be based on a metric that indicates the accuracy of the sorting gating according to the desired sorting strategy. In some cases, the metric can be an accuracy (e.g., purity) metric, where purity is compared to a predetermined threshold. This metric can be generated based on the confidence of the classifiers included in the sorting strategy. In other cases, the metric can be a yield metric, where the yield of target cells in particle population clustering is compared to a predetermined threshold.

[0233] In some embodiments, the method includes generating a sorting decision based on determined sorting gating for sample cells. In some cases, generating particle sorting decisions according to this disclosure includes a gating strategy with eight or fewer sorting gatings, such as seven or fewer, six or fewer, five or fewer, four or fewer, three or fewer, and includes two or fewer. In some embodiments, the method includes generating sorting gatings using a graphical display (described below) that displays one or more analysis algorithms for applying classification parameters to image parameters determined for cells.

[0234] In some implementations, the method includes displaying a gating strategy on a graphical user interface. For example, the analysis algorithm may be one or more of a spectral compensation matrix, a clustering algorithm, and a t-Distributed Stochastic Neighbor Embedding (t-SNE) algorithm. In some cases, the analysis algorithm is applied to a swarm cluster by dragging its icon onto the swarm cluster. In other cases, the swarm cluster is selected and the analysis algorithm is applied by selecting it from a drop-down menu. In some cases, the analysis algorithm is a spectral unmixing algorithm, such as those described in U.S. Patent No. 11,009,400, filed August 19, 2021, and International Patent Application No. PCT / US2021 / 46741, the disclosures of which are incorporated herein by reference. In some cases, the gating strategy is determined using a computational sorting algorithm, such as those described in U.S. Patent No. 11,513,054, the disclosure of which is incorporated herein by reference. In some cases, sorting gating can be determined using computer software, such as HyperFinder (e.g., as described in Bonavia et al. Frontiers in Immunology 2022; 13: 1007016) and computational sorting using HyperFinder, FlowJo software, and BD FACSDiva software (Becton Dickinson, 2021), the disclosures of which are incorporated herein by reference. In some embodiments, the gating strategy is developed on a separate computing system (e.g., a different computer system or network) and transmitted to a flow cytometer to implement the gating strategy, for example, using a particle sorter of the flow cytometer (e.g., with a sorting decision module).

[0235] In some embodiments, the method includes sorting one or more cells of a sample. For example, the method may include sorting two or more components of the sample, such as three or more components, four or more components, five or more components, ten or more components, fifteen or more components, and may include sorting 25 or more components of the sample.

[0236] In particle sorting, methods include data acquisition, analysis, and recording (e.g., using a computer), where multiple data channels record data from each detector used to obtain overlapping spectra of multiple fluorophores associated with the particle. In these embodiments, the analysis includes spectral resolution of the light from the multiple fluorophores with overlapping spectra associated with the particle (e.g., by calculating a spectral unmixing matrix) and particle identification based on the estimated abundance of each fluorophore associated with the particle. This analysis can be passed to a sorting system configured to generate a set of digitized parameters according to particle classification.

[0237] In some embodiments, methods for sorting sample components include sorting cells in a sample using a particle sorting module with deflection plates, such as those described in U.S. Patent Publication No. 2017 / 0299493, filed March 28, 2017, the disclosure of which is incorporated herein by reference. In some embodiments, cells in a sample are sorted using a sorting decision module having multiple sorting decision units, such as those described in U.S. Patent Publication No. 2020 / 0256781, the disclosure of which is incorporated herein by reference. In some embodiments, the subject system includes a particle sorting module with deflection plates, such as those described in U.S. Patent Publication No. 2017 / 0299493, filed March 28, 2017, the disclosure of which is incorporated herein by reference.

[0238] Non-transitory computer-readable storage medium

[0239] This disclosure further includes a non-transitory computer-readable storage medium having instructions for practicing the subject matter methods. The computer-readable storage medium can be used on one or more computers to fully or partially automate the system used for practicing the methods described herein. In some embodiments, the instructions according to the methods described herein can be encoded in a “programmed” form onto a computer-readable medium, wherein the term “computer-readable medium” as used herein refers to any non-transitory storage medium that participates in providing instructions and data to a computer for execution and processing. Examples of suitable non-transitory storage media include floppy disks, hard disks, optical disks, magneto-optical disks, CD-ROMs, CD-Rs, magnetic tapes, non-volatile memory cards, ROMs, DVD-ROMs, Blu-ray discs, solid-state drives, and network attached storage (NAS), whether these devices are internal or external to a computer. Files containing information can be “stored” on a computer-readable medium, where “stored” means recording the information so that the computer can later access and retrieve it. The computer implementation methods described herein can be executed using a program, which can be written in one or any number of computer programming languages. Such languages ​​include, for example, Python, Java, JavaScript, C, C#, C++, Go, R, Swift, PHP, and many other languages.

[0240] A non-transitory computer-readable storage medium is also provided, having instructions with an algorithm for classifying cells in a sample (e.g., a whole blood sample). According to certain embodiments, the non-transitory computer-readable storage medium has an algorithm for irradiating a whole blood sample containing cells in a flowing stream with a light source, an algorithm for capturing an image of light emitted from the irradiated cells using a light detection system with a photodetector, and an algorithm for applying a neural network to the captured image to detect the presence of cancer cells in a whole blood sample with a minimum concentration of cancer cells and to determine the classification of cancer cells among multiple categories. In some cases, the neural network is trained using images of samples labeled with fluorophore benchmark ground truth markers.

[0241] In some implementations, the minimum concentration of cancer cells present in a whole blood sample is 0.5% or more, for example, 1% or more. In other cases, the minimum concentration of cancer cells present in a whole blood sample is 0.5% to 0.5%.

[0242] In the implementation plan, cancer cells in the sample are classified. In some cases, multiple classifications include two or more categories, such as three or more categories, and include four or more categories. In some cases, multiple classifications include ovarian cancer classification, T-lymphocytic carcinoma classification, breast cancer classification, colon cancer classification, or epithelial carcinoma classification. In some cases, multiple classifications include OVCAR classification, Jurkat classification, MCF-7 classification, and adenocarcinoma classification.

[0243] In some cases, the non-transitory computer-readable storage medium contains algorithms for classifying cells based on one or more parameters computed from generated image data. In some cases, the non-transitory computer-readable storage medium contains algorithms for applying a classifier neural network to classify cells. In some cases, the neural network includes an input layer consisting of one or more parameters computed from the generated image data. In some cases, the neural network includes an output layer that is a probability estimate based on cell identity (e.g., the cell is a cancer cell, circulating tumor cell). In some cases, the neural network includes a learning algorithm configured to learn from classification errors. In some cases, the non-transitory computer-readable storage medium contains algorithms for updating classification parameters based on one or more parameters computed from the generated image data. In some cases, the updating of classification parameters is based on a comparison between one or more parameters computed from the generated image data and one or more parameters computed from benchmark ground truth image data. In some cases, the non-transitory computer-readable storage medium contains algorithms for generating benchmark ground truth image data.

[0244] In some implementations, the neural network includes a backpropagation neural network. In some cases, the neural network includes two or more stages, such as four or more stages. In some cases, the neural network includes at least one dropout stage. In some cases, each stage includes at least eight hidden layers, such as 32 or more hidden layers. In some cases, each stage includes eight to 64 hidden layers. In some cases, the neural network applies a sigmoid activation function, an Adam optimizer, an adaptive learning rate, a dynamic threshold, binary classification, a binary cross-entropy loss function, or any combination thereof. In some cases, the non-transitory computer-readable storage medium contains an algorithm for normalizing the intensity of the captured image. In some cases, the neural network is applied to normalize the image. In some cases, the neural network includes an input layer consisting of one or more parameters computed from the generated image data.

[0245] In some implementations, the image data includes one or more waveforms generated in response to measurement light. In some cases, the non-transitory computer-readable storage medium contains algorithms for generating images of cells from the image data. In some cases, the non-transitory computer-readable storage medium contains algorithms for calculating image parameters based on one or more image data and the cell image. In some cases, the non-transitory computer-readable storage medium contains algorithms for evaluating cell morphology. In some cases, the non-transitory computer-readable storage medium contains algorithms for evaluating cell morphology by determining one or more of cell size, cell shape, and the degree of punctiformity of the cell. In some cases, the non-transitory computer-readable storage medium contains algorithms for calculating cell parameters based on image parameters. In some cases, the image parameters are quantized image parameters. In some cases, the parameters are selected from centroid, centroid change, diffusivity, eccentricity, major axis moment, maximum intensity, radial moment, minor axis moment, size, total intensity, light loss from the cell nucleus, forward scattered light from the cell nucleus, side scattered light from the cell nucleus, and combinations thereof. In some cases, a non-transitory computer-readable storage medium contains an algorithm for calculating image parameters based on an image generated from fluorescence measured from cells, light loss measured from cells, forward scattered light measured from cells, side scattered light measured from cells, and combinations thereof.

[0246] In some embodiments, the non-transitory computer-readable storage medium includes an algorithm for classifying cells in a sample based on computed cell parameters. In some cases, the non-transitory computer-readable storage medium includes an algorithm for classifying cells based on a waveform generated from measurement light, a generated cell image, image parameters computed from the generated cell image, or a combination thereof. In some cases, the non-transitory computer-readable storage medium includes an algorithm for classifying cells by assigning cells to one or more particle population clusters. In some cases, the non-transitory computer-readable storage medium includes an algorithm for determining one or more sorting gates for classifying cells in a sample. In some cases, the non-transitory computer-readable storage medium includes an algorithm for computing one or more sorting gates that capture clusters of circulating tumor cells and exclude clusters of normal cell populations. In some cases, the non-transitory computer-readable storage medium includes an algorithm for computing a sorting gate that maximizes the inclusion rate of clusters of circulating tumor cells. In some cases, the non-transitory computer-readable storage medium includes an algorithm for computing a sorting gate that maximizes the exclusion of clusters of normal cell populations.

[0247] In some cases, the non-transitory computer-readable storage medium contains algorithms for sorting cells from a sample into multiple sample containers. In some cases, the non-transitory computer-readable storage medium contains algorithms for sorting cells based on the presence, classification, or both. In some cases, the non-transitory computer-readable storage medium contains algorithms for sorting cells based on generated cell images. In some cases, the non-transitory computer-readable storage medium contains algorithms.

[0248] Non-transitory computer-readable storage media can be used in one or more computer systems having a display and operator input devices. Operator input devices may be a keyboard, mouse, etc. The processing module includes a processor that can access memory storing instructions for performing the subject method steps. The processing module may include an operating system, a graphical user interface (GUI) controller, system memory, memory storage devices, and input / output controllers, caches, data backup units, and many other devices. The processor may be a commercially available processor or one of other existing or soon-to-be-commercial processors. The processor executes the operating system, which interfaces with firmware and hardware in a well-known manner and assists the processor in coordinating and executing the functions of various computer programs written in various programming languages ​​(such as those mentioned above, other high-level or low-level languages, and combinations thereof), as known in the art. The operating system typically works in cooperation with the processor to coordinate and execute the functions of other computer components. The operating system also provides scheduling, input / output control, file and data management, memory management, and communication control and related services, all consistent with known techniques.

[0249] Reagent test kit

[0250] This disclosure also includes a kit containing one or more integrated circuits described herein. In some embodiments, the kit may also include instructions for programming the subject system, for example in the form of a computer-readable medium (e.g., a flash drive, USB storage, optical disc, DVD, Blu-ray disc, etc.), or for downloading the programming from an Internet network protocol or cloud server. The kit may further include instructions for practicing the subject method. These instructions may exist in a variety of forms within the subject kit, one or more of which may be present in the kit. One form of these instructions may be as printed information on a suitable medium or substrate (e.g., one or more sheets of paper with information printed on them), kit packaging, packaging inserts, etc. Another form of these instructions is as a computer-readable medium on which information is recorded, such as a floppy disk, optical disc (CD), portable flash drive, etc. Yet another form of these instructions is a website address that can be used to access information on a remote site via the Internet.

[0251] use

[0252] The subject matter systems, methods, and computer systems can be used to evaluate cancer cells in whole blood samples. In some cases, this disclosure provides high-quality, high-purity cancer cells isolated from heterogeneous whole blood samples. Furthermore, the subject matter systems and methods can be used in a variety of applications where it is desirable to analyze and sort particulate components in a sample in a fluid medium. In some embodiments, the systems and methods described herein can be used for flow cytometry characterization of cell isolates. Embodiments of this disclosure can be used in situations where it is desirable to provide flow cytometry with improved cell sorting accuracy, enhanced particle collection, particle charging efficiency, more accurate particle charging, and enhanced particle deflection during cell sorting.

[0253] The embodiments of this disclosure can also be used in applications where cells prepared from biological samples are desired for research, laboratory testing, or therapeutic purposes. In some embodiments, the subject methods and apparatus can facilitate the acquisition of individual cells prepared from a target fluid or tissue biological sample. For example, the subject methods and systems facilitate the acquisition of specific cells in a fluid or tissue sample for use as a research or diagnostic sample. The methods and apparatus of this disclosure allow for the separation and collection of cells from biological samples (e.g., organs, tissues, tissue fragments, fluids) with greater efficiency and lower cost compared to conventional flow cytometry systems.

[0254] Example

[0255] The following embodiments are provided to provide a complete disclosure and description of how the invention can be made and used to those skilled in the art, and are not intended to limit the scope of the inventors’ ideas.

[0256] Example 1 - Deep Learning for Circulating Tumor Cell (CTC) Classification

[0257] method

[0258] A classifier neural network was developed as a label-free discriminator for distinguishing cancer cells (from healthy normal cells). The input was data from whole blood cells containing less than 1% circulating tumor cells. Four different cell types were investigated: ovarian adenocarcinoma (OVCAR) cells, colorectal adenocarcinoma (HT-29) cells, breast adenocarcinoma (MCF-7) cells, and T lymphocytes (Jurkat) cells. The output of the neural network is a probabilistic classification of the cell as a CTC compared to healthy cells. The input layer contains only imaging parameters from flow cytometry data. The output layer provides cell classification based on a probability estimate of any event belonging to cancer. The dataset was incorporated with 1% of the four CTC types, each with a baseline ground truth marker.

[0259] result

[0260] Figure 8A Captured images and accuracy evaluations were depicted when classifying cancer cells in samples according to certain implementation schemes. Cancer cells were present at a concentration of 1% in each sample. The neural network achieved a classification accuracy of 92.8% for ovarian cancer cells (OVCAR classification); 99.77% for breast cancer cells (MCF-7 classification); 99.6% for colorectal adenocarcinoma cells (HT-29 classification); and 94.1% for T lymphocyte cancer cells (Jurkat classification). Figure 8B The classification of T lymphocyte cancer cells using a neural network trained on samples labeled with fluorophore-based truth markers according to certain implementation schemes is described. Figure 8B The clustering plot showed a ratio of 1:99 for T lymphocyte CTCs to peripheral blood mononuclear cells, indicating that the applied neural network could classify T lymphocytes (Jurkat classification) with 99% accuracy. The baseline truth label was set in the BV605-A fluorophore channel.

[0261] in conclusion

[0262] A novel neural network-based classifier was developed to classify circulating tumor cells (CTCs) from normal cells in whole blood samples with a CTC to normal cell ratio of 1:99. The neural network achieved an accuracy exceeding 99% for all MCF-7, Jurkat, HT-29, and OVCAR cell lines. Based on the confusion matrix, the accuracy was 92.8% for the OVCAR cell line, 99.7% for the MCF-7 cell line, 99.6% for the HT-29 cell line, and 94.1% for the Jurkat cell line.

[0263] Example 2 - Using a neural network classifier to perform label-free classification of circulating tumor cells (CTCs) based on image-derived features Record the test results.

[0264] Circulating tumor cells (CTCs) are cells that detach from the primary tumor and enter the circulatory system. The presence and frequency of these cells in peripheral blood can serve as prognostic indicators of tumor progression, metastatic potential, and treatment. CTCs can exist in blood at extremely low frequencies and exhibit a wide range of phenotypes, posing a significant challenge to detection. Current CTC detection methods rely on next-generation sequencing (NGS) technology, which is costly. This embodiment demonstrates a label-free detection method utilizing the light scattering properties of CTCs, simultaneously achieving counting and separation. This method can be implemented using BDFACSDiscovery. TMImaging features measured at a rate of 10,000 cells per second on an S8 cell sorter demonstrate a novel neural network-based classifier capable of separating CTCs from normal cells in heterogeneous samples. Whole blood samples were co-pollinated with cancer cell lines exhibiting a range of cell size, morphology, and phenotype. These included MCF-7, OVCAR, HT-29, and Jurkat. Whole blood samples were co-pollinated with each cancer cell line at frequencies as low as 1%. Validation was performed using benchmark ground truth surface markers. The neural network demonstrated >99% accuracy on each of the four datasets, and CTC recall fractions of 95.9% (OVCAR), 99.6% (MCF-7), 97.9% (HT-29), and 91.9% (Jurkat). This detection method requires minimal sample preparation, is independent of known surface marker phenotypes, and is applicable to a wide range of cancer types.

[0265] method

[0266] Whole blood was mixed with cancer cell lines exhibiting a range of cell size, morphology, and phenotype. Cell lines included MCF-7, OVACAR, HT-29, and Jurkat. Whole blood samples were mixed with each cancer cell line at a frequency as low as 1%. Validation was performed using benchmark truth surface markers. BD Pharm Lyse was used. TM Whole blood was lysed and washed. Cell lines (MCF-7, OVCAR, HT-29, or Jurkat) were collected and washed. Whole blood was stained with CD45 BV605 or CD45 BV421 (in the case of Jurkat mixed cells). Cells were then washed twice and counted. Cell lines were stained with baseline true surface markers that varied depending on the cell line: OVCAR (Her-NeuFITC), HT-29 and MCF7 (CD326 Epcam BB515), and Jurkat (CD45 BV605). Cell lines were then washed twice after staining and counted. Samples were prepared according to Table 1 below. Each sample was prepared at a concentration of 2 million cells / mL in 500 µL. BD FACSDiscover was used. TM The S8 cell sorter generates flow cytometry data. The generated data is analyzed using FlowJo analysis software.

[0267] Table 1

[0268]

[0269] A novel neural network architecture was designed to predict the probability of an event belonging to a circulating tumor cell (CTC) category based on imaging features derived from flow cytometry data. A general network architecture was designed and demonstrated that can be applied to detect four CTC types, including ovarian cancer, breast cancer, colon cancer, and T-cell cancer. Hyperparameter tuning was performed, and network architecture decisions were applied to the general architecture to capture all four CTC types. Performance evaluation through confusion matrix visualization showed that CTC identification using label-free morphological features has high recall and high precision.

[0270] result

[0271] A series of standard cleanup gating mechanisms were used to quantify circulating tumor cells incorporated into whole blood samples. These cleanup gating mechanisms were applied to each heterogeneous sample preparation to exclude duplexes. Figure 9A Charts illustrating the gating of whole blood samples with cancer cell lines according to certain implementation schemes are presented. As shown in Table 1 above, whole blood samples were mixed with one of four cancer cell lines at frequencies ranging from 1% to 50%. Figure 9A The charts shown represent samples co-occurring with 50% OVCAR, HT-29, MCF7, or Jurkat cells. A series of cleanup gates were applied to each sample to exclude duplexes that might interfere with the analysis (the cleanup gate for HT-29 is shown). The gates for each of the four cell lines after cleanup are shown on the right. Viability markers (FVS780) are included to exclude dead cells. The markers used to establish the baseline truth vary by cell line (see the notes on the right side of each chart). The baseline truth gates were established using the 50% co-occurring sample and then applied to each subsequent lower percentage sample. This baseline truth amount of CTCs within each sample was used to evaluate the performance of the neural network classifier.

[0272] Figure 9B-9E A diagram illustrating gating of circulating tumor cells according to certain implementation schemes is depicted. Figure 9B Cells derived from the ovarian adenocarcinoma cell line (OVCAR) were depicted. OVCAR cells were incorporated into whole blood sample preparations at an approximately 1:1 ratio. OVCAR cells expressed Her2-Neu, while PBMCs did not. A graph of CD45 (expressed by blood cells) versus Her2-Neu (expressed by OVCAR cells) was used to identify and quantify the OVCAR cell content in the samples. This gating established a benchmark ground truth for comparison with a neural network classifier of the OVCAR cell line. Figure 9CCells from the colorectal adenocarcinoma cell line (HT-29) were depicted. HT-29 cells were incorporated into whole blood sample preparations at an approximately 1:1 ratio. HT-29 cells expressed CD326 (Epcam), while blood cells did not. A graph of CD45 (expressed by blood cells) versus CD326 (expressed by HT-29 cells) was used to identify and quantify the HT-29 cell content in the samples. This gating established a benchmark ground truth for comparison with a neural network classifier of the HT-29 cell line. Figure 9D Cells from the mammalian adenocarcinoma cell line (MCF7) were depicted. MCF7 cells were incorporated into whole blood sample preparations at an approximately 1:1 ratio. MCF7 cells expressed CD326 (Epcam), while blood cells did not. A graph of CD45 (expressed by blood cells) versus CD326 (expressed by MCF7 cells) was used to identify and quantify the MCF7 cell content in the samples. This gating established a benchmark ground truth for comparison with a neural network classifier of the MCF7 cell line. Figure 9E Cells derived from the T-cell leukemia cell line (Jurkat cells) were depicted. Jurkat cells were incorporated into a whole blood sample preparation at an approximately 1:1 ratio. Both Jurkat cells and blood cells expressed CD45. To identify and quantify the incorporated Jurkat cells, they were stained with CD45 BV605 prior to mixing, while blood cells were stained with CD45 BV421. Icons for CD45a (BV421) and CD45b (BV605) were used to identify and quantify the Jurkat cell content in the sample. This gating established a benchmark ground truth for comparison with a neural network classifier of the Jurkat cell line.

[0273] Detecting circulating tumor cell types using a novel general neural network architecture. Figures 10A-10E Image confusion matrices for circulating tumor cell types detected by a neural network architecture according to certain implementation schemes are depicted. Light loss and lateral scattering images for each cell type are shown. Figure 10A Images of whole blood cells and a confusion matrix guide are provided. Figure 10B-10E Images of ovarian adenocarcinoma, colorectal adenocarcinoma, breast adenocarcinoma, and T-cell leukemia cell lines are shown. The confusion matrix was calculated based on the performance of a novel neural network architecture in identifying each CTC type incorporated into whole blood samples at a 1% group frequency. The confusion matrix shows that the general neural network architecture can detect four CTC types with recall fractions of 95.9% (OVCAR), 99.6% (MCF-7), 97.9% (HT-29), and 91.9% (Jurkat).

[0274] in conclusion

[0275] This study presents a novel label-free detection method that utilizes the inherent light scattering properties of CTCs, enabling both counting and separation without requiring known surface marker phenotypes. Label-free detection offers high accuracy and recall. A neural network-based classifier demonstrates high accuracy (>99%) and high recall scores across multiple cancer cell lines, indicating its effectiveness in distinguishing CTCs from normal cells in heterogeneous samples. This detection method requires minimal sample preparation and is applicable to various cancer types, making it a versatile tool for CTC detection in clinical settings.

[0276] Notwithstanding the appended claims, this disclosure is further defined by the following:

[0277] 1. A system comprising:

[0278] A light source configured to irradiate cells in a whole blood sample in a flowing stream;

[0279] A light detection system comprising a photodetector that captures an image of light emitted from irradiated cells; and

[0280] A processor having memory operatively coupled to the processor, wherein the memory contains instructions stored thereon that, when executed by the processor, cause the processor to apply a neural network to the captured image, such that:

[0281] The presence of cancer cells was detected in the whole blood sample containing the lowest concentration of cancer cells; and

[0282] The cancer cells are classified among multiple categories.

[0283] The neural network is trained using images of samples labeled with fluorophore benchmarks.

[0284] 2. The system according to item 1, wherein the minimum concentration is about 0.5% or greater.

[0285] 3. The system according to item 2, wherein the minimum concentration is about 1% or greater.

[0286] 4. The system according to any one of items 2 to 3, wherein the minimum concentration is 0.5% to 5%.

[0287] 5. The method according to any one of items 1-4, wherein the plurality of classifications includes three or more classifications.

[0288] 6. The system according to item 5, wherein the plurality of categories includes four or more categories.

[0289] 7. The method according to any one of items 1-6, wherein the plurality of classifications includes ovarian cancer classification, T-lymphocytic carcinoma classification, breast cancer classification, colon cancer classification, or epithelial carcinoma classification.

[0290] 8. The method according to any one of items 1-7, wherein the plurality of classifications includes the OVCAR classification, the Jurkat classification, the MCF-7 classification, or the adenocarcinoma classification.

[0291] 9. The system according to any one of items 1-8, wherein the neural network comprises a backpropagation neural network.

[0292] 10. The system according to any one of items 1-9, wherein the neural network comprises two or more stages.

[0293] 11. The system according to item 10, wherein the neural network comprises four or more stages.

[0294] 12. The system according to any one of items 10-11, wherein at least one stage includes a discarding stage.

[0295] 13. The system according to any one of items 10-12, wherein each stage comprises at least 8 hidden layers.

[0296] 14. The system according to item 13, wherein each stage comprises at least 32 hidden layers.

[0297] 15. The system according to any one of items 13-14, wherein each stage comprises 8 to 64 hidden layers.

[0298] 16. The system according to any one of items 1-15, wherein the neural network applies a sigmoid activation function, an Adam optimizer, an adaptive learning rate, a dynamic threshold, a binary classification, a binary cross-entropy loss function, or any combination thereof.

[0299] 17. The system according to any one of items 1-16, wherein the memory contains instructions for normalizing the intensity of the captured image, and wherein the neural network is applied to the normalized image.

[0300] 18. The system according to any one of items 1-17, wherein the neural network includes an input layer consisting of one or more parameters computed from the generated image data.

[0301] 19. The system according to any one of items 1-18, wherein the captured image includes one or more waveforms generated in response to the measuring light.

[0302] 20. The system according to any one of items 1-19 further includes a sorting mechanism for sorting cells into a container based on the presence, classification, or both of the cells.

[0303] 21. The system according to item 20, wherein the sorting mechanism includes a droplet deflector.

[0304] 22. The system according to any one of items 1-21, wherein the light source includes a beam generator assembly configured to generate at least a first frequency-shifted beam and a second frequency-shifted beam.

[0305] 23. The system according to item 22, wherein the beam generator includes an acousto-optic deflector.

[0306] 24. The system according to any one of items 22-23, wherein the beam generator comprises a direct digital synthesizer (DDS) radio frequency comb generator.

[0307] 25. The system according to any one of items 22 to 24, wherein the beam generator assembly is configured to generate a local oscillator beam.

[0308] 26. The system according to any one of items 22 to 25, wherein the beam generator assembly is configured to generate a plurality of frequency-shifted comb beams.

[0309] 27. The system according to any one of items 1-26, wherein the light source comprises a laser.

[0310] 28. The system according to item 27, wherein the laser is a continuous wave laser.

[0311] 29. The system according to any one of items 1-28, wherein the system is an imaging flow cytometer.

[0312] 30. The system according to any one of items 1-29, wherein the system comprises an integrated circuit device.

[0313] 31. A method comprising:

[0314] Irradiate a whole blood sample containing cells in a flowing stream with a light source;

[0315] Capturing images of light emitted from irradiated cells using a light detection system incorporating a photodetector; and

[0316] The neural network is applied to the captured image so that:

[0317] The presence of cancer cells was detected in the whole blood sample containing the lowest concentration of cancer cells; and

[0318] The cancer cells are classified among multiple categories.

[0319] The neural network is trained using images of samples labeled with fluorophore benchmarks.

[0320] 32. The method according to item 31, wherein the minimum concentration is about 0.5% or greater.

[0321] 33. The method according to item 32, wherein the minimum concentration is about 1% or greater.

[0322] 34. The method according to any one of items 32 to 33, wherein the minimum concentration is 0.5% to 5%.

[0323] 35. The method according to any one of items 31-34, wherein the plurality of classifications includes three or more classifications.

[0324] 36. The method according to item 35, wherein the plurality of categories includes four or more categories.

[0325] 37. The method according to any one of items 31-36, wherein the plurality of classifications includes ovarian cancer classification, T-lymphocytic carcinoma classification, breast cancer classification, colon cancer classification, or epithelial carcinoma classification.

[0326] 38. The method according to any one of items 31-37, wherein the plurality of classifications includes the OVCAR classification, the Jurkat classification, the MCF-7 classification, or the adenocarcinoma classification.

[0327] 39. The method according to any one of items 31-38, wherein the neural network comprises a backpropagation neural network.

[0328] 40. The method according to any one of items 31-39, wherein the neural network comprises two or more stages.

[0329] 41. The method according to item 40, wherein the neural network comprises four or more stages.

[0330] 42. The method according to any one of items 40-41, wherein at least one stage includes a discarding stage.

[0331] 43. The method according to any one of items 40-42, wherein each stage comprises at least 8 hidden layers.

[0332] 44. The method according to item 43, wherein each stage comprises at least 32 hidden layers.

[0333] 45. The method according to any one of items 43-44, wherein each stage comprises 8 to 64 hidden layers.

[0334] 46. ​​The method according to any one of items 31-45, wherein the neural network applies a sigmoid activation function, an Adam optimizer, an adaptive learning rate, a dynamic threshold, a binary classification, a binary cross-entropy loss function, or any combination thereof.

[0335] 47. The method according to any one of items 31-46, wherein the method includes normalizing the intensity of the captured image, and wherein the neural network is applied to the normalized image.

[0336] 48. The method according to any one of items 31-47, wherein the neural network includes an input layer consisting of one or more parameters computed from the generated image data.

[0337] 49. The method according to any one of items 31-48, wherein the captured image includes one or more waveforms generated in response to the measuring light.

[0338] 50. The method according to any one of items 31-49 further includes sorting cells into containers based on the presence, classification, or both of the cells.

[0339] 51. The method according to item 20, wherein the sorting mechanism includes a droplet deflector.

[0340] 52. The method according to any one of items 31-51, wherein the light source includes a beam generator assembly configured to generate at least a first frequency-shifted beam and a second frequency-shifted beam.

[0341] 53. The method according to item 52, wherein the beam generator includes an acousto-optic deflector.

[0342] 54. The method according to any one of items 52-53, wherein the beam generator comprises a direct digital synthesizer (DDS) radio frequency comb generator.

[0343] 55. The method according to any one of entries 52 to 54, wherein the beam generator assembly is configured to generate a local oscillating beam.

[0344] 56. The method according to any one of entries 52 to 55, wherein the beam generator assembly is configured to generate a plurality of frequency-shifted comb beams.

[0345] 57. The method according to any one of items 31-56, wherein the light source comprises a laser.

[0346] 58. The method according to item 57, wherein the laser is a continuous wave laser.

[0347] 59. A non-transitory computer-readable storage medium comprising instructions stored thereon, wherein the non-transitory computer-readable storage medium comprises:

[0348] An algorithm for irradiating a cell-containing whole blood sample in a flowing stream with a light source;

[0349] An algorithm for capturing images of light emitted from irradiated cells using a light detection system including a photodetector; and

[0350] An algorithm for applying a neural network to the captured image, so as to:

[0351] The presence of cancer cells was detected in the whole blood sample containing the lowest concentration of cancer cells; and

[0352] The cancer cells are classified among multiple categories.

[0353] The neural network is trained using images of samples labeled with fluorophore benchmarks.

[0354] 60. The non-transitory computer-readable storage medium according to item 59, wherein the minimum concentration is about 0.5% or greater.

[0355] 61. The non-transitory computer-readable storage medium according to item 60, wherein the minimum concentration is about 1% or more.

[0356] 62. The non-transitory computer-readable storage medium according to any one of entries 60 to 61, wherein the minimum concentration is 0.5% to 5%.

[0357] 63. The non-transitory computer-readable storage medium according to any one of entries 59 to 62, wherein the plurality of categories includes three or more categories.

[0358] 64. The non-transitory computer-readable storage medium as described in item 63, wherein the plurality of categories includes four or more categories.

[0359] 65. The non-transitory computer-readable storage medium according to any one of items 59-64, wherein the plurality of classifications includes ovarian cancer classification, T-lymphocytic carcinoma classification, breast cancer classification, colon cancer classification, or epithelial carcinoma classification.

[0360] 66. The non-transitory computer-readable storage medium according to any one of items 59-65, wherein the plurality of classifications includes the OVCAR classification, the Jurkat classification, the MCF-7 classification, or the adenocarcinoma classification.

[0361] 67. The non-transitory computer-readable storage medium according to any one of entries 59 to 66, wherein the neural network comprises a backpropagation neural network.

[0362] 68. The non-transitory computer-readable storage medium according to any one of items 59 to 67, wherein the neural network comprises two or more stages.

[0363] 69. The non-transitory computer-readable storage medium according to item 68, wherein the neural network comprises four or more stages.

[0364] 70. The non-transitory computer-readable storage medium according to any one of entries 68-69, wherein at least one stage includes a discard stage.

[0365] 71. The non-transitory computer-readable storage medium according to any one of items 68-70, wherein each stage comprises at least eight hidden layers.

[0366] 72. The non-transitory computer-readable storage medium as described in item 71, wherein each stage comprises at least 32 hidden layers.

[0367] 73. The non-transitory computer-readable storage medium according to any one of items 71-72, wherein each stage comprises 8 to 64 hidden layers.

[0368] 74. The non-transitory computer-readable storage medium according to any one of items 59-73, wherein the neural network applies a sigmoid activation function, an Adam optimizer, an adaptive learning rate, a dynamic threshold, a binary classification, a binary cross-entropy loss function, or any combination thereof.

[0369] 75. The non-transitory computer-readable storage medium according to any one of items 59-74, wherein the non-transitory computer-readable storage medium includes an algorithm for normalizing the intensity of the captured image, and wherein the neural network is applied to the normalized image.

[0370] 76. The non-transitory computer-readable storage medium according to any one of items 59-75, wherein the neural network includes an input layer consisting of one or more parameters computed from the generated image data.

[0371] 77. The non-transitory computer-readable storage medium according to any one of entries 59-76, wherein the captured image includes one or more waveforms generated in response to the measuring light.

[0372] 78. The non-transitory computer-readable storage medium according to any one of items 59-77, wherein the non-transitory computer-readable storage medium includes an algorithm for generating at least a first frequency-shifted beam and a second frequency-shifted beam with the light source.

[0373] 79. The non-transitory computer-readable storage medium according to item 78, wherein the non-transitory computer-readable storage medium includes an algorithm for generating a native oscillating beam with the light source.

[0374] 80. The non-transitory computer-readable storage medium according to any one of items 78-79, wherein the non-transitory computer-readable storage medium includes an algorithm for generating a plurality of frequency-shifted comb beams with the light source.

[0375] Although the foregoing invention has been described in some detail by way of illustration and example for the purpose of clarity, it will be apparent to those skilled in the art, based on the teachings of the invention, that certain changes and modifications may be made without departing from the spirit or scope of the appended claims.

[0376] Therefore, the foregoing only illustrates the principles of the invention. It should be understood that those skilled in the art will be able to conceive of various arrangements, although not explicitly described or shown herein, which embody the principles of the invention and are included within its spirit and scope. Furthermore, all examples and conditional language recorded herein are primarily intended to assist the reader in understanding the principles of the invention and the concepts contributed by the inventors to the field, and should be interpreted as not being limited to such specifically recorded examples and conditions. Moreover, all statements herein recounting the principles, aspects, and embodiments of the invention and their specific examples are intended to cover their structural and functional equivalents. Additionally, such equivalents are intended to include both currently known equivalents and future development equivalents, i.e., any element developed that performs the same function regardless of its structure. Furthermore, nothing disclosed herein is intended to be offered to the public, whether or not such disclosure is expressly recited in the claims.

[0377] Therefore, the scope of the invention is not intended to be limited to the exemplary embodiments shown and described herein. Rather, the scope and spirit of the invention are embodied in the appended claims. In the claims, 35 USC §112(f) or 35 USC §112(6) is explicitly defined as being invoked only when the exact phrase “means for” or the exact phrase “step for” is recited at the beginning of a definition in the claim; if such an exact phrase is not used in the definition in the claim, then 35 USC §112(f) or 35 USC §112(6) is not invoked.

Claims

1. A system comprising: A light source configured to irradiate cells in a whole blood sample in a flowing stream; A light detection system comprising a photodetector that captures an image of light emitted from irradiated cells; as well as A processor, the processor including memory operatively coupled to the processor, wherein the memory contains instructions stored thereon, the instructions, when executed by the processor, causing the processor to apply a neural network to the captured image, such that: The presence of cancer cells was detected in the whole blood sample containing the lowest concentration of cancer cells; as well as The cancer cells are classified among multiple categories. The neural network is trained using images of samples labeled with fluorophore benchmarks.

2. The system according to claim 1, wherein, The minimum concentration is approximately 0.5% or greater.

3. The method according to any one of claims 1-2, wherein, The multiple categories include three or more categories.

4. The method according to any one of claims 1-3, wherein, The various classifications include ovarian cancer, T-lymphocyte cancer, breast cancer, colon cancer, or epithelial cancer.

5. The method according to any one of claims 1-4, wherein, The multiple classifications include OVCAR classification, Jurkat classification, MCF-7 classification, or adenocarcinoma classification.

6. The system according to any one of claims 1-5, wherein, The neural network includes a backpropagation neural network.

7. The system according to any one of claims 1-6, wherein, The neural network comprises two or more stages.

8. The system according to claim 7, wherein, At least one stage includes a discarding stage.

9. The system according to any one of claims 7-8, wherein, Each stage includes at least 8 hidden layers.

10. The system according to any one of claims 1-9, wherein, The neural network uses the sigmoid activation function, Adam optimizer, adaptive learning rate, dynamic threshold, binary classification, binary cross-entropy loss function, or any combination thereof.

11. The system according to any one of claims 1-10, wherein, The memory contains instructions for normalizing the intensity of the captured image, and the neural network is applied to the normalized image.

12. The system according to any one of claims 1-11, wherein, The neural network includes an input layer, which consists of one or more parameters calculated from the generated image data.

13. The system according to any one of claims 1-12, wherein, The captured image includes one or more waveforms generated in response to the measured light.

14. The system according to any one of claims 1-12, further comprising a sorting mechanism for sorting cells into a container based on the presence, classification, or both of the cells.

15. The system according to any one of claims 1-14, wherein, The light source includes a beam generator assembly configured to generate at least a first frequency-shifted beam and a second frequency-shifted beam.

16. The system according to any one of claims 1-15, wherein, The light source includes a laser.

17. The system according to any one of claims 1-16, wherein, The system is an imaging flow cytometer.

18. The system according to any one of claims 1-17, wherein, The system includes integrated circuit devices.

19. A method comprising: Irradiate a whole blood sample containing cells in a flowing stream with a light source; A light detection system containing a photodetector captures images of light emitted from irradiated cells; as well as The neural network is applied to the captured image so that: The presence of cancer cells was detected in the whole blood sample containing the lowest concentration of cancer cells; as well as The cancer cells are classified among multiple categories. The neural network is trained using images of samples labeled with fluorophore benchmarks.

20. A non-transitory computer-readable storage medium comprising instructions stored thereon, wherein, The non-transitory computer-readable storage medium includes: An algorithm for irradiating a cell-containing whole blood sample in a flowing stream with a light source; An algorithm for capturing images of light emitted from irradiated cells using a light detection system that includes a photodetector; as well as An algorithm for applying a neural network to the captured image, so as to: The presence of cancer cells was detected in the whole blood sample containing the lowest concentration of cancer cells; as well as The cancer cells are classified among multiple categories. The neural network is trained using images of samples labeled with fluorophore benchmarks.

Images