Automatic focusing with synthetic particle images

Abstract

Synthetic particles may be used to assess a focal quality in a biological analyzer. Such a method may comprise providing a plurality of synthetic particles, and receiving images of the plurality of synthetic particles. It may also include, for each of the plurality of synthetic particles, analyzing pixel data and calculating a predicted focal position from the pixel data.

Description

[0001] AUTOMATIC FOCUSING WITH SYNTHETIC PARTICLE

[0002] IMAGES

[0003] BACKGROUND

[0004] Blood cell analysis is one of the most commonly performed medical tests for providing an overview of a patient's health status. A blood sample can be drawn from a patient's body and stored in a test tube containing an anticoagulant to prevent clotting. A whole blood sample normally comprises three major classes of blood cells including red blood cells (erythrocytes), white blood cells (leukocytes) and platelets (thrombocytes). Each class can be further divided into subclasses of members. For example, five major types or subclasses of white blood cells (WBCs) have different shapes and functions. White blood cells may include neutrophils, lymphocytes, monocytes, eosinophils, and basophils. There are also subclasses of the red blood cell types. The appearances of particles in a sample may differ according to pathological conditions, cell maturity and other causes. Red blood cell subclasses may include reticulocytes and nucleated red blood cells.

[0005] This analysis may involve capturing images of a sample comprising blood cells, and the higher the quality of these images, the more suitable they are for analysis. However, capturing high quality images presents many problems. For example, blood cells and other particles may be out of focus for many reasons, such as a defective flowcell, improper installation of the flowcell and / or its subcomponents, improper fluidic configurations, and so on. Accordingly, there is a need for technology which can automatically determine whether imaged particles are in focus and which therefore can be used for identifying and troubleshooting focusing issues in a biological analyzer.

[0006] - 1 -

[0007] 0133788.0810417 4912-7631-1663v6 SUMMARY

[0008] Aspects of the present disclosure may be used in determining the focusing positions of particles in hematology flow imaging analyzers.

[0009] According to a first aspect, the disclosed technology may be used to implement a method of assessing a focal quality in a biological analyzer. Such a method may comprise providing a plurality of synthetic particles, and receiving images of the plurality of synthetic particles. It may also include, for each of the plurality of synthetic particles, analyzing pixel data and calculating a predicted focal position from the pixel data.

[0010] In some examples, for at least one of the plurality of synthetic particles: analyzing the pixel data comprises extracting a plurality of features from the pixel data; and calculating the predicted focal position from the pixel data comprises providing one or more of the plurality of features as input to a focus prediction function.

[0011] In some examples, the focus prediction function comprises a trained machine learning model.

[0012] In some examples, the trained machine learning model is a support vector regression model.

[0013] In some examples, the focus prediction function comprises a polynomial curve fitting equation.

[0014] In some examples, each of the one or more features from the plurality of features which are provided as input to the focus prediction function is a ratio of two other features from the plurality of features from the pixel data.

[0015] In some examples, for the at least one of the plurality of synthetic particles, extracting the plurality of features comprises: organizing pixels comprised by the pixel data into a plurality of groups based on the pixels’ locations relative to a border of that synthetic particle; and generating an achromatic characteristic curve, wherein the achromatic characteristic curve is in a space having index value for groups from the plurality of groups as a first dimension, and

[0016] - 2 -

[0017] 0133788.0810417 4912-7631-1663v6 average achromatic characteristic values of pixels of the groups from the plurality of groups as a second dimension.

[0018] In some examples, for the at least one of the plurality of synthetic particles, extracting the plurality of features comprises: generating a smoothed curve by applying a smoothing function to the achromatic characteristic curve; generating a first derivative curve by taking a derivative of the smoothed curve; and generating a second derivative curve by taking a derivative of the first derivative curve.

[0019] In some examples, for the at least one of the plurality of synthetic particles, the plurality of features extracted from the pixel data comprise a set of features selected from: a first feature, wherein extracting the first feature is an amplitude of a peak in the smoothed curve; a second feature, wherein the second feature is a position of the peak in the smoothed curve; a third feature, wherein the third feature is a position of a peak in the first derivative curve; a fourth feature, wherein the fourth feature is a difference between the third feature and the second feature; a fifth feature, wherein the fifth feature is a position of a valley on a right side of a peak in the second derivative curve; a sixth feature, wherein the sixth feature is an amplitude of a peak in the first derivative curve; a seventh feature, wherein the seventh feature is an amplitude of a peak in the second derivative curve; an eighth feature, wherein the eighth feature is a sum of absolute values in a region of the first derivative curve between a valley on a left side of the peak in the first derivative curve and zero; a ninth feature, wherein the ninth feature is a sum of absolute values in a region of the second derivative curve between: a zero crossing between a valley on a right side of the peak in the second derivative curve and the peak in the second derivative curve; and a maximum index of any group from the plurality of groups; and a tenth feature, wherein the tenth feature is half of a sum of absolute values in a region of the second derivative curve between: a zero crossing between a valley on a left side of the peak in the second derivative curve and the peak in the second derivative curve; and zero.

[0020] - 3 -

[0021] 0133788.0810417 4912-7631-1663v6 In some examples, for the at least one of the plurality of synthetic particles, the one or more features provided as input to the focus prediction function comprise: a ratio of the ninth feature to the first feature; a ratio of the sixth feature to the first feature; a ratio of the eighth feature to the first feature; a ratio of the fourth feature to the second feature; a ratio of the tenth feature to the fifth feature; and a ratio of the seventh feature to the first feature.

[0022] In some examples, the method may comprise: providing a flow cell for receiving the plurality of synthetic particles.

[0023] In some examples a system may be provided which comprises at least one processor; and a non-transitory computer readable medium having stored thereon instruction which, when executed by the at least one processor, cause the system to perform the method of any of the preceding examples.

[0024] In some examples, a biological analyzer may be provided which comprises at least one processor and a non-transitory computer readable medium stored thereon instruction which, when executed by the at least one processor, cause the biological analyzer to perform the method of any of the preceding examples.

[0025] According to a second aspect, a method of assessing a focal quality in a biological analyzer is provided, where the method comprises: providing a plurality of particles; receiving images of the plurality of particles; and for each of the plurality of particles: extracting a plurality of features from pixel data in an image of that particle; and calculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a trained machine learning model.

[0026] In some examples, the trained machine learning model is a support vector regression model.

[0027] In some examples, each of the one or more features from the plurality of features which are provided as input to the focus prediction function is a ratio of two other features from the plurality of features.

[0028] - 4 -

[0029] 0133788.0810417 4912-7631-1663v6 In some examples, for the at least one of the particles, extracting the plurality of features from the pixel data in the image of that particles comprises: organizing pixels comprised by the pixel data into a plurality of groups based on the pixels’ locations relative to a border of that particle; and generating an achromatic characteristic curve, wherein the achromatic characteristic curve is in a space having index value for groups from the plurality of groups as a first dimension, and average achromatic characteristic values of pixels of the groups from the plurality of groups as a second dimension.

[0030] In some examples, for the at least one of the plurality of particles, extracting the plurality of features comprises: generating a smoothed curve by applying a smoothing function to the achromatic characteristic curve; generating a first derivative curve by taking a derivative of the smoothed curve; and generating a second derivative curve by taking a derivative of the first derivative curve.

[0031] In some examples, for the at least one of the plurality of particles, the plurality of features extracted from the pixel data comprise a set of features selected from: a first feature, wherein extracting the first feature is an amplitude of a peak in the smoothed curve; a second feature, wherein the second feature is a position of the peak in the smoothed curve; a third feature, wherein the third feature is a position of a peak in the first derivative curve; a fourth feature, wherein the fourth feature is a difference between the third feature and the second feature; a fifth feature, wherein the fifth feature is a position of a valley on a right side of a peak in the second derivative curve; a sixth feature, wherein the sixth feature is an amplitude of a peak in the first derivative curve; a seventh feature, wherein the seventh feature is an amplitude of a peak in the second derivative curve; an eighth feature, wherein the eighth feature is a sum of absolute values in a region of the first derivative curve between a valley on a left side of the peak in the first derivative curve and zero; a ninth feature, wherein the ninth feature is a sum of absolute values in a region of the second derivative curve between: a zero crossing between a valley on a right side of the peak in the second derivative curve and the peak in the second derivative curve; and a maximum index of any group from the plurality of groups; and a tenth

[0032] - 5 -

[0033] 0133788.0810417 4912-7631-1663v6 feature, wherein the tenth feature is half of a sum of absolute values in a region of the second derivative curve between: a zero crossing between a valley on a left side of the peak in the second derivative curve and the peak in the second derivative curve; and zero.

[0034] In some examples, for the at least one of the plurality of particles, the one or more features provided as input to the trained machine learning model comprise: a ratio of the ninth feature to the first feature; a ratio of the sixth feature to the first feature; a ratio of the eighth feature to the first feature; a ratio of the fourth feature to the second feature; a ratio of the tenth feature to the fifth feature; and a ratio of the seventh feature to the first feature.

[0035] In some examples, a method of any of the preceding examples may comprise providing a flow cell for receiving the plurality of particles.

[0036] In some examples, a system may be provided which comprises at least one processor and a non-transitory computer readable medium having stored thereon instructions which, when executed by the at least one processor, cause the system to perform the method of any of the preceding examples.

[0037] In some examples, a biological analyzer may be provided which comprises at least one processor; and a non-transitory computer readable medium having stored thereon instructions which, when executed by the at least one processor, cause the system to perform the method of any of the preceding examples.

[0038] According to a third aspect, a method of assessing a focal quality in a biological analyzer may be provided which comprises providing a plurality of particles; receiving images of the plurality of particles; and for each of the plurality of particles: extracting a plurality of features from pixel data in an image of that particle; and calculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a focus prediction function, wherein each of the one or more features which is provided as input to the focus prediction function is a ratio of two other features from the plurality of features.

[0039] - 6 -

[0040] 0133788.0810417 4912-7631-1663v6 In some examples, for the at least one of the particles, extracting the plurality of features from the pixel data in the image of that particles comprises: organizing pixels comprised by the pixel data into a plurality of groups based on the pixels’ locations relative to a border of that particle; and generating an achromatic characteristic curve, wherein the achromatic characteristic curve is in a space having index value for groups from the plurality of groups as a first dimension, and average achromatic characteristic values of pixels of the groups from the plurality of groups as a second dimension.

[0041] In some examples, for the at least one of the plurality of particles, extracting the plurality of features comprises: generating a smoothed curve by applying a smoothing function to the achromatic characteristic curve; generating a first derivative curve by taking a derivative of the smoothed curve; and generating a second derivative curve by taking a derivative of the first derivative curve.

[0042] In some examples, for the at least one of the plurality of particles, the plurality of features extracted from the pixel data comprise a set of features selected from: a first feature, wherein extracting the first feature is an amplitude of a peak in the smoothed curve; a second feature, wherein the second feature is a position of the peak in the smoothed curve; a third feature, wherein the third feature is a position of a peak in the first derivative curve; a fourth feature, wherein the fourth feature is a difference between the third feature and the second feature; a fifth feature, wherein the fifth feature is a position of a valley on a right side of a peak in the second derivative curve; a sixth feature, wherein the sixth feature is an amplitude of a peak in the first derivative curve; a seventh feature, wherein the seventh feature is an amplitude of a peak in the second derivative curve; an eighth feature, wherein the eighth feature is a sum of absolute values in a region of the first derivative curve between a valley on a left side of the peak in the first derivative curve and zero; a ninth feature, wherein the ninth feature is a sum of absolute values in a region of the second derivative curve between: a zero crossing between a valley on a right side of the peak in the second derivative curve and the peak in the second derivative curve; and a maximum index of any group from the plurality of groups; and a tenth

[0043] - 7 -

[0044] 0133788.0810417 4912-7631-1663v6 feature, wherein the tenth feature is half of a sum of absolute values in a region of the second derivative curve between: a zero crossing between a valley on a left side of the peak in the second derivative curve and the peak in the second derivative curve; and zero.

[0045] In some examples, for the at least one of the plurality of particles, the one or more features provided as input to the focus prediction function comprise: a ratio of the ninth feature to the first feature; a ratio of the sixth feature to the first feature; a ratio of the eighth feature to the first feature; a ratio of the fourth feature to the second feature; a ratio of the tenth feature to the fifth feature; and a ratio of the seventh feature to the first feature.

[0046] In some examples, a method such as described in the preceding examples may comprise providing a flow cell for receiving the plurality of synthetic particles.

[0047] In some examples, a system may be provided which comprises: at least one processor; and a non-transitory computer readable medium having stored thereon instruction which, when executed by the at least one processor, cause the system to perform the method of any of the preceding examples.

[0048] In some examples, a biological analyzer may be provided which comprises at least one processor; and a non-transitory computer readable medium stored thereon instruction which, when executed by the at least one processor, cause the biological analyzer to perform the method of any of the preceding examples.

[0049] According to a fourth aspect, a system may be provided which comprises: a flowcell configured to receive a plurality of synthetic particles; a camera configured to capture a plurality of images of the synthetic particles; at least one processor; and a computer-readable medium storing instructions that are configured to. when executed by the at least one processor: receive images of the plurality of synthetic particles; and for each of the plurality of synthetic particles: analyze pixel data; and calculate a predicted focal position from the pixel data.

[0050] - 8 -

[0051] 0133788.0810417 4912-7631-1663v6 In some examples, a method may be provided which comprises performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of the preceding example is to perform when executed.

[0052] According to a fifth aspect, a biological analyzer may be provided which comprises: a flowcell configured to receive a plurality of synthetic particles; a camera configured to capture a plurality of images of the synthetic particles; at least one processor; and a computer-readable medium storing instructions that are configured to, when executed by the at least one processor: receive images of the plurality of synthetic particles; and for each of the plurality of synthetic particles: analyze pixel data; and calculate a predicted focal position from the pixel data.

[0053] In some examples, a method may be provided which comprises performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of the preceding example is to perform when executed.

[0054] According to a sixth aspect, a system may be provided which comprises a flowcell configured to receive a plurality of synthetic particles; a camera configured to capture a plurality of images of the synthetic particles; at least one processor; and a computer-readable medium storing instructions that are configured to, when executed by the at least one processor: receive images of the plurality of synthetic particles; and for each of the plurality of particles: extracting a plurality of features from pixel data in an image of that particle; and calculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a trained machine learning mode.

[0055] In some examples, a method may be provided which comprises performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of the preceding example is to perform when executed.

[0056] According to a seventh aspect, a biological analyzer may be provided which comprises: a flowcell configured to receive a plurality of synthetic particles: a camera configured to capture

[0057] - 9 -

[0058] 0133788.0810417 4912-7631-1663v6 a plurality of images of the synthetic particles; at least one processor; and a computer-readable medium storing instructions that are configured to, when executed by the at least one processor: receive images of the plurality of synthetic particles; and for each of the plurality of particles: extracting a plurality of features from pixel data in an image of that particle; and calculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a trained machine learning model.

[0059] In some examples, a method may be provided which comprises performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of the preceding example is to perform when executed.

[0060] According to an eighth aspect, a system may be provided which comprises: a flowcell configured to receive a plurality of synthetic particles; a camera configured to capture a plurality of images of the synthetic particles; at least one processor; and a computer-readable medium storing instructions that are configured to, when executed by the at least one processor: receive images of the plurality of synthetic particles; and for each of the plurality of particles: extracting a plurality of features from pixel data in an image of that particle; and calculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a focus prediction function, wherein each of the one or more features which is provided as input to the focus prediction function is a ratio of two other features from the plurality of features.

[0061] In some examples, a method may be provided which comprises performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of the preceding example is to perform when executed.

[0062] According to a ninth aspect, a biological analyzer may be provided which comprises a camera configured to capture a plurality of images of the synthetic particles; at least one processor; and a computer-readable medium storing instructions that are configured to, when executed by the at least one processor: receive images of the plurality of synthetic particles; and for each of

[0063] - 10 -

[0064] 0133788.0810417 4912-7631-1663v6 the plurality of particles: extracting a plurality of features from pixel data in an image of that particle; and calculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a focus prediction function, wherein each of the one or more features which is provided as input to the focus prediction function is a ratio of two other features from the plurality of features.

[0065] In some examples, a method may be provided which comprises performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of the preceding example is to perform when executed.

[0066] It should be noted that any of the various features of the aspects disclosed herein can be included or combined in each of those aspects. Other aspects and implementations are described herein. Accordingly, the exemplary aspects described in this summary should be understood as being illustrative only and should not be treated as limiting.

[0067] BRIEF DESCRIPTION OF THE DRAWINGS

[0068] While the specification concludes with claims which particularly point out and distinctly claim the invention, it is believed the present invention will be better understood from the following description of certain examples taken in conjunction with the accompanying drawings, in which like reference numerals identify the same elements and in which:

[0069] FIG. 1 is a schematic illustration, partly in section and not to scale, showing operational aspects of an exemplary flow cell which may be used in some implementations.

[0070] FIG. 2 illustrates a method which may be used for assessing the focus of particles flowing through a flowcell.

[0071] FIG. 3 illustrates a configuration which may be used in imaging particles.

[0072] FIG. 4 illustrates actions which could be performed in the analysis of synthetic particle pixel data and calculation of a predicted focal position.

[0073] - 11 -

[0074] 0133788.0810417 4912-7631-1663v6 FIG. 5 illustrates actions which may be performed in extracting features from a particle image.

[0075] FIG. 6 illustrates how rings may relate to a particle boundary in a particle image.

[0076] FIG. 7 illustrates actions which may be performed in extracting features from a particle image.

[0077] FIG. 8 illustrates example intensity curves.

[0078] FIG. 9 illustrates histograms of predicted focusing positions.

[0079] FIG. 10 illustrates a method to define a function for predicting focal positions.

[0080] The drawings are not intended to be limiting in any way, and it is contemplated that various embodiments of the invention may be carried out in a variety of other ways, including those not necessarily depicted in the drawings. The accompanying drawings incorporated in and forming a part of the specification illustrate several aspects of the present invention, and together with the description serve to explain the principles of the invention; it being understood, however, that this invention is not limited to the precise arrangements shown.

[0081] DETAILED DESCRIPTION

[0082] The present disclosure relates to articles, systems, and methods for assessing focal quality in a biological analyzer. In some aspects, the analyzer may be a visual analyzer comprising one or more processors to facilitate automated conversion and / or analysis of images. Such analyzers may be useful, for example, in characterizing particles in biological fluids, such as detecting and quantifying erythrocytes, reticulocytes, nucleated red blood cells, platelets, and white blood cells, including white blood cell differential counting, categorization and subcategorization and analysis. Other similar uses such as characterizing blood cells from other fluids (serum, bone marrow, lavage fluid, effusions, exudates, cerebrospinal fluid, pleural fluid, peritoneal fluid, and amniotic fluid) are also contemplated.

[0083] - 12 -

[0084] 0133788.0810417 4912-7631-1663v6 Turning now to the drawings, FIG. 1 schematically shows an exemplary flow cell 22 which may be used in an analyzer for conveying a sample fluid through a viewing zone 23 of a high optical resolution imaging device 24 (e.g., a camera) in a configuration for imaging microscopic particles in a sample flow stream 32 using digital image processing. Flow cell 22 is coupled to a source 25 of sample fluid which may have been subjected to processing, such as contact with a particle contrast agent composition and heating. Flow cell 22 is also coupled to one or more sources 27 of a particle and / or intracellular organelle alignment liquid (PIOAL) / sheath fluid, such as a clear glycerol solution having a viscosity that is greater than the viscosity of the sample fluid, an example of which is disclosed in U.S. Pat. Nos. 9,316,635 and 10,451,612, the disclosures of which are hereby incorporated by reference in their entirety.

[0085] The sample fluid is injected through a flattened opening at a distal end 28 of a sample feed tube 29, and into the interior of the flow cell 22 at a point where the PIOAL flow has been substantially established resulting in a stable and symmetric laminar flow of the PIOAL above and below (or on opposing sides of) the ribbon-shaped sample stream. The sample and PIOAL streams may be supplied by precision metering pumps that move the PIOAL with the injected sample fluid along a flow path that narrows substantially. The PIOAL envelopes and compresses the sample fluid in the zone 21 where the flow path narrows. Hence, the decrease in flow path thickness at zone 21 can contribute to a geometric focusing of the sample flow stream 32. The sample flow stream 32 is enveloped and carried along with the PIOAL downstream of the narrowing zone 21, passing in front of, or otherwise through the viewing zone 23 of. the high optical resolution imaging device 24 where images are collected, for example, using a Charge-Coupled Device (CCD) 48 observing the sample stream as illuminated using an illumination source 42 through a viewing port 57. Processor 18 can receive, as input, pixel data from CCD 48. The sample fluid ribbon flows together with the PIOAL to a discharge 33.

[0086] As shown here, the narrowing zone 21 can have a proximal flow path portion 21a having a proximal thickness PT and a distal flow path portion 21b having a distal thickness DT, such

[0087] - 13 -

[0088] 0133788.0810417 4912-7631-1663v6 that distal thickness DT is less than proximal thickness PT. The sample fluid can therefore be injected through the distal end 28 of sample tube 29 at a location that is distal to the proximal portion 21a and proximal to the distal portion 21b. Hence, the sample fluid can enter the PIO AL envelope as the PIO AL stream is compressed by the zone 21, wherein the sample fluid injection tube has a distal exit port through which sample fluid is injected into flowing sheath fluid, the distal exit port bounded by the decrease in flow path size of the flow cell.

[0089] The digital high optical resolution imaging device 24 with objective lens 46 is directed along an optical axis that intersects the ribbon-shaped sample flow stream 32. The relative distance between the objective 46 and the flow cell 33 is variable by operation of a motor drive 54, for resolving and collecting a focused digitized image on a photosensor array. Additional information regarding the construction and operation of an exemplary flow cell such as shown in FIG. 1 is provided in U.S. Patent 9,322,752, entitled “Flow cell Systems and Methods for Particle Analysis in Blood Samples,” filed on March 17, 2014, the disclosure of which is hereby incorporated by reference in its entirety.

[0090] In operation, an analyzer such as an analyzer incorporating a flowcell based imaging system as illustrated in FIG. 1 may assess the focus of particles (e.g., synthetic particles used in calibrating the analyzer) flowing through a flowcell, such as by using a process of the type shown in FIG. 2. Initially, in the method of FIG. 2, a plurality of synthetic particles may be provided 201. This may be done, for example, by flowing rigid synthetic spheres having sizes analogous to sizes of particles which would be expected to be imaged in patient samples (e.g., 5 pm - 15 pm in diameter) through a flowcell such as flowcell 22 depicted in FIG. 1. These synthetic particles may be, for example opaque sulfate latex beads, and the particles flowed through the flowcell may be all one size (e.g., all 5 pm diameter beads) or may be a mix of bead sizes. It is also possible that other types of beads (e.g., beads other than sulfate latex beads) or beads of different sizes than noted (e.g., beads ranging from 0.5 to 20 pm diameter) may be used in some cases, and so the examples of synthetic particles provided should be understood as being illustrative only, and should not be treated as limiting.

[0091] - 14 -

[0092] 0133788.0810417 4912-7631-1663v6 The analyzer may also receive 202 images of the plurality of synthetic particles. To achieve this, as particles flow through a flowcell of a system such as shown in FIG. 1, non-overlapping images, each of which may be referred to herein as a “frame.” may be captured. The frames may then be subjected to some additional processing, such as filtering them to remove frames and segments of frames that don’t include particles, organizing frame patches that include particles into images such that each of the organized images includes a single particle, and then converting those organized images into a format (e.g., HSV or HSI format) which could then be subjected to further processing, thereby resulting in a set of processable images which each depict a single synthetic particle being “received” 202 for purposes of the method of FIG. 2.

[0093] In some examples, the synthetic particles can be provided as part of a control sample used to help set a relative focus position of a flowcell region for image analysis (e.g., one or more of a camera, flowcell, or objective being moved to adjust the relative position between a camera and flowcell). There would be a control procedure to help set the relative position to establish good focal quality (e.g., once daily). The synthetic particles, in one example, are within a fluid sample (e.g., saline) within a control sample tube.

[0094] In some examples, a control sample process unique to identifying and counting control samples is used as part of a validation procedure run to confirm the machine is functioning correctly (e.g., assessing control particle counts or concentrations correctly). This process can be done in various ways, for instance a scanning of the barcode which identifies the sample as a control sample and then enables the running of a count process unique to control samples (e.g., a control sample / particle counting algorithm, which is trained or optimized to control material). Alternatively, a first process is used to establish a type of particle (e.g., assessing whether at least some of the sample material is control or patient sample), and then will result in a second process used to count that particle type (e.g., a patient sample counting algorithm, or a control sample counting algorithm will be used). Alternatively, a single counting algorithm which is trained on or programmed to count both control and patient samples is used, and then makes a determination whether a control sample or patient sample is involved based on the number of

[0095] - 15 -

[0096] 0133788.0810417 4912-7631-1663v6 control particles or patient sample particles identified. Further examples of how control and patient samples is provided in international patent application PCT / US2023 / 015038 filed on March 11, 2023 for “Controls and their use in Analyzers,” the disclosure of which is hereby incorporated by reference in its entirety.

[0097] In the method of FIG. 2, once the images of synthetic particles had been received 202, the pixel data in those images could potentially be analyzed 203 and used to calculate 204 a predicted focal position which can then be in focusing the system’s imaging device. For example, in the configuration shown in FIG. 3, a predicted focal position of +4 pm could mean that the camera was positioned 4 pm too far in the positive z direction, and therefore that if it was moved 4 pm in the negative z direction (or if its focal plane was moved 4 pm in the negative z direction through the camera’s imaging optics being manipulated while the camera remained stationary, or the flowcell is moved 4 pm in the positive z direction such that the sample stream would be coincident with the camera’s focal plane) the particles passing through the flowcell would be in focus. To make the calculation 204 of a predicted focal position, as well as to perform the analysis 203 that would underly that calculation, a variety of approaches may be used, examples of which are illustrated and discussed below in the context of FIGS. 4-9.

[0098] Turning next to FIG. 4, that figure illustrates actions which could be performed in the analysis of synthetic particle pixel data and calculation of a predicted focal position in a method such as described above in the context of FIG. 2. As shown in FIG. 4, analyzing 203 synthetic particle pixel data may comprise extracting 301 features from that data. As shown in FIG. 5, this extraction may include organizing 401 the pixels into groups, and generating 402 an intensity curve based on those groups. For example, in some implementations, the boundary of a particle in a received 202 image may be identified 403, such as by identifying pixels in the image which exhibit a relatively abrupt (e.g., significantly greater magnitude) transition between dark and light pixels, and treating an ellipse generated based on those pixels as the boundary of the particle depicted in the received 202 image. Once the boundary had been identified 403, a plurality of rings can be generated 404 based on that boundary. For instance,

[0099] - 16 -

[0100] 0133788.0810417 4912-7631-1663v6 morphological image erosion may be applied to create rings inside the boundary, while morphological image dilation is applied to create rings outside the boundary. As another option which may be used in some implementations, rings may be generated by increasing (for rings outside the particle) or decreasing (for rings inside the boundary) a radius of the boundary by a set increment (e.g., a measurement such as 1 pm, 0.1 pm; a pixel value, such as 1 pixel, 5 pixels, 10 pixels; a proportion such as 10% of the radius of the particle, 15% of the radius of the particle; etc.) for each new ring to be created. An illustration of how these types of rings might relate to an identified boundary for a particle in a particle image is shown in FIG. 6, though other relationships (e.g., greater number of rings, narrower or broader rings, etc.) are also possible, and so the illustration of FIG. 6 should not be treated as implying limitations on how rings may be generated 404 in implementations of the disclosed technology. The rings can be construed as bins (each ring being a bin) in a pixel binning approach to analyzing pixels within particular region(s) of the image.

[0101] Continuing with the discussion of FIG. 5, once the rings had been generated 404, they may be used to determine 405 intensity values. For instance, for each pair of adjacent rings, an average may be calculated for the intensity values of the pixels between that pair of rings. These intensity values may then be used to determine 406 an intensity curve. This may be done, for example, by treating the intensity value for the smallest pair of adjacent rings as having an index of 0, the intensity value for the next smallest pair of adjacent rings as having an index of 1, etc. and then treating each intensity-index pair as a coordinate pair with intensity and index treated as defining vertical and horizontal positions, respectively. The intensity curve could then be defined 406 by fitting a curve to those coordinates.

[0102] In an implementation which utilizes a method such as shown in FIG. 5, once the intensity curve had been generated 402, it may be used to generate additional curves. For example, as shown in FIG. 7, the intensity curve may be used to generate 501 a smoothed curve, such as by applying a smoothing filter, or by taking a moving average of the points on the intensity curve. Additionally, a first derivative curve can then be generated 502 by taking the first derivative

[0103] - 17 -

[0104] 0133788.0810417 4912-7631-1663v6 of the smoothed curve (and potentially smoothing the curve representing the first derivative). A second derivative can then be generated 503 in the same manner. These curves can then be used to define a plurality of features for the particle image from which they were derived. Examples of such features are provided below in tables 1-6, with FIG. 8 illustrating where certain of those features may be found on the smoothed curve (labeled as I_mean), the first derivative curve (labeled as I_mean_lst) and the second derivative curve (labeled as I_mean_2nd).

[0105]

[0106] - 18 -

[0107] 0133788.0810417 4912-7631-1663v6

[0108]

[0109] Table 1: Exemplary peaks and amplitudes of curve extrema

[0110]

[0111] Table 2: Exemplary distances between extrema

[0112]

[0113] - 19 -

[0114] 0133788.0810417 4912-7631-1663v6

[0115]

[0116] - 20 -

[0117] 0133788.0810417 4912-7631-1663v6

[0118]

[0119] - 21 -

[0120] 0133788.0810417 4912-7631-1663v6

[0121]

[0122] Table 3: Exemplary ratios using means, areas under curves and amplitudes

[0123]

[0124] - 22 -

[0125] 0133788.0810417 4912-7631-1663v6

[0126]

[0127] Table 4: Ratios to the position of the maximum of the first derivative curve

[0128]

[0129] - 23 -

[0130] 0133788.0810417 4912-7631-1663v6

[0131]

[0132] - 24 -

[0133] 0133788.0810417 4912-7631-1663v6

[0134]

[0135] Table 5: Ratios to the amplitude of the maximum of the smoothed curve.

[0136]

[0137] - 25 -

[0138] 0133788.0810417 4912-7631-1663v6

[0139]

[0140] Table 6: Ratios to the position of the maximum of the smoothed curve.

[0141] It should be understood that the features listed in tables 1-6 are intended to be illustrative only, and that other types of features may be used in different embodiments of the disclosed technology. To illustrate, consider that, in some cases, different instances of the types of feature listed in the various tables may be used, either in addition to, or instead of, the features listed on tables 1-6. For example, in some cases, either in addition to, or as alternatives to, the ratios listed in table 3, other ratios using means, areas under curve, and amplitudes (e.g., a ratio of the absolute value of the amplitude of the local minimum to the left of the maximum of the second derivative curve, to the absolute value of the amplitude of the maximum of the second derivative curve) could be used. It is also possible that features to be extracted may be defined based on different types of characteristics, rather than intensity as described above. For example, in some cases, instead of using intensity values (e.g., values which could be calculated as the average intensities of the R, G and B channels of an RGB image) other types of values, such as brightness values (e.g., as could be calculated as the maximum intensity of the intensities of the R, G and B channels in an RGB image) could alternatively be used to create curves for particle images, and those curves could be used to extract features such as those set forth in tables 1-6. Other variations (e.g., combinations, in which features are extracted using curves based on intensity values and features are extracted using curves based

[0142] - 26 -

[0143] 0133788.0810417 4912-7631-1663v6 on brightness values) are also possible and will be immediately apparent to those of skill in the art. Accordingly, the above examples of features which may be extracted from particle images should be understood as being illustrative only, and should not be treated as limiting.

[0144] Whatever features are used in a particular case, once they have been extracted 301, they may be used to calculate 204 a predicted focal position for the image to which they correspond, such as by being provided 302 to a focus prediction function. Such a focus prediction function may be, for example, a machine learning model (e.g., a support vector regression model) trained to predict focal positions using features from ground truth images whose actual focal positions were known. However, other types of focus prediction functions are also possible and may be used in some cases. For example, in some cases, a focus prediction function may be a polynomial curve fitting equation determined based on similar ground truth images with known focal positions such as could be used to train a machine learning model. Other types of focus prediction functions are also possible, and so the exemplary function types mentioned above should be understood as being illustrative only, and should not be treated as limiting.

[0145] Variations are also possible in aspects other than the type of function used for focus prediction. For example, in some cases, a focus prediction function may use less than all of the features to predict a focal position for the corresponding image. To illustrate how this may take place, consider FIG. 10, which illustrates a method for defining a function to predict focal position using a limited number of parameters. As shown in FIG. 10, this type of function definition could begin with selecting 1001 a parameter to include in the function, this may be done in a variety of ways. For example, in some cases, the parameters could be tested to identify which parameter which was not already included in the function (which, on the first iteration, would be all of the parameters) had the strongest correlation with predicted focal position, and then add the parameter with the strongest correlation to the function. A function which included the selected parameter could then be fit 1002 to the data (e.g., a set of ground truth images having known focal positions). This may be done, for example, by minimizing the least square distances from a curve defined using the selected parameter (and any other parameters which

[0146] - 27 -

[0147] 0133788.0810417 4912-7631-1663v6 had already been selected on previous iterations) to the focal positions in the ground truth data. A check 1003 may then be performed to determine if the fit was sufficiently close (e.g., if the root mean squared error was below a threshold value). If it was, then the process could terminate 1004 with the function which was most recently fit 1002 to the data being treated as the function for predicting focal position. Otherwise, as discussed in more detail below, further complexity could be added to the function so that’s focal position predictions would more accurately match the focal positions from the ground truth data.

[0148] Continuing with the discussion of FIG. 10, in a case where a function’s fit was not sufficient, one step which can be taken is to determine 1005 whether to divide the function’s domains into one or more segments (or, if the function’s domain had already been split, determining whether one or more of those segments should be subdivided). This may be done, for example, by checking if errors in a particular portion of the function’s domain were biased in one direction or another and, if so, dividing 1006 the domain according to those portions so that the different portions of the domain could be fit separately. For instance, if errors in the first half of the domain were biased towards predicted focal positions which were too small, while errors in the second half of the domain were biased towards focal positions which were too large, then the domain could be split in half, thereby allowing the two groups of errors to be independently addressed.

[0149] Another step which can be taken when a function’s fit is not sufficiently close is to determine 1007 whether to add a parameter to the function. This may be done, for example, by randomly selecting whether to add a parameter, with the likelihood that a new parameter would be added determined using a function whose value would decrease with the number of parameters that were already in the function (e.g., a logarithmic or exponential decay function). If it was determined 1007 that a new parameter should be added, then the new parameter could be selected 1001 using approaches such as described previously for the initial performance of that step. Alternatively, if it was determined 1007 that a new parameter should not be added, then the degree of the function may be increased 1008. For example, if the function prior to

[0150] - 28 -

[0151] 0133788.0810417 4912-7631-1663v6 increasing 1008 the degree had the form set forth below as equation 1 , then, after increasing the degree, the function may have the form set forth below as equation 2.

[0152] n

[0153] F = C + Ci * pi

[0154] i = 0

[0155] Equation 1

[0156]

[0157] Equation 2

[0158] In the above equations, F is the predicted focal position, C is an offset whose value would be set as part of fitting 1002 the function to the data, m is the degree of the function, n is the total number of parameters in the function, Ci is the coefficient for parameter pi whose value would be set as part of fitting 1002 the function to the data, pi is the ithparameter of the function, Cij is the coefficient for parameter p whose value would be set as part of fitting 1002 the function to the data, and pp is the jthpower of parameter p,.

[0159] It should be understood that, while FIG. 10 illustrated a method which can be used to define a function for predicting focal position, the method is intended to be illustrative only, and variations are possible in how such a focal prediction function may be defined. For example, while the above description explained how a method could determine 1003 if a function’s fit was sufficient based on whether the root mean squared error was below a threshold, in some cases, different checks may be performed on different iterations of the method. For instance, in some cases, a threshold for determining 1003 if a fit was sufficient may increase (i.e., get easier to satisfy) as the complexity of the function increased, such that a function which included two or more of r2-2nd_rt_neg_auc, r2-lst_md_pk_amp, r3-dist_lst_md_2_0th_rt, r2- lst_lt_neg_auc, ratio_auc_2nd_lt_hneg_2_rt_neg, and / or r2-2nd_md_pk_amp, may be treated as having a sufficient fit, even if the function’s root mean squared error would have been too high for a function with fewer parameters. It is also possible that other types of approaches,

[0160] - 29 -

[0161] 0133788.0810417 4912-7631-1663v6 such as principal component regression, least angle regression, or Theil-Sen estimators, may also be used in defining functions for predicting focal position. Accordingly, the above discussion of variations on the method of FIG. 10, like the discussion of FIG. 10 itself, should be understood as being illustrative only, and should not be treated as limiting.

[0162] Once predicted focal positions have been calculated 204, these positions can be used to assess the overall focusing quality of a sample run (potentially allowing a sample to be rerun if the focusing quality is sufficiently poor). However, such predicted focal positions can also be used for other purposes, such as gaining detailed information on an analyzer’s sample stream and fluidics. For example, in some cases, histograms such as shown in FIG. 9 can be created of predicted focusing positions in terms of acquisition time and X and Y positions. If an instrument is operating and calibrated properly, all of these histograms should be flat and the predicted focusing positions should be close to 0.0 pm. However, as shown in FIG. 9, this may not necessarily be the case. For example, in FIG. 9, the predicted focusing position is low at low end of Y and high at the high end of Y. This indicates that the sample stream along Y direction is probably flat but not completely perpendicular to the camera (aka the sample stream is tilting along Y direction). This information can then be used to take appropriate remedial action (e.g., adjusting the flowcell in the Y direction). Similarly, in FIG. 9, the predicted focusing position moves up along with acquisition time, which indicates that the sample stream is not stable during acquisition, and may give early warning of a potential issue (e.g., incipient component failure) before it becomes critical and leads to unplanned downtime for the analyzer.

[0163] Each of the calculations or operations described herein may be performed using a computer or other processor having hardware, software, and / or firmware. The various method steps may be performed by modules, and the modules may comprise any of a wide variety of digital and / or analog data processing hardware and / or software arranged to perform the method steps described herein. The modules optionally comprising data processing hardware adapted to perform one or more of these steps by having appropriate machine programming code

[0164] - 30 -

[0165] 0133788.0810417 4912-7631-1663v6 associated therewith, the modules for two or more steps (or portions of two or more steps) being integrated into a single processor board or separated into different processor boards in any of a wide variety of integrated and / or distributed processing architectures. These methods and systems will often employ a tangible media embodying machine-readable code with instructions for performing the method steps described above. Suitable tangible media may comprise a memory (including a volatile memory and / or a non-volatile memory), a storage media (such as a magnetic recording on a floppy disk, a hard disk, a tape, or the like; on an optical memory such as a CD, a CD-R / W, a CD-ROM, a DVD, or the like; or any other digital or analog storage media), or the like.

[0166] All patents, patent publications, patent applications, journal articles, books, technical references, and the like discussed in the instant disclosure are incorporated herein by reference in their entirety for all purposes.

[0167] Different arrangements of the components depicted in the drawings or described above, as well as components and steps not shown or described are possible. For example, in some cases, aspects of processing described herein (e.g„ determination of a morphology score, use of such a score to evaluate a patient’s health, use of such a score to assess cell alignment in a flowcell) may be performed in various configurations - for instance, using a processor which is comprised by (or local to) an analyzer, a parallel-processing arrangement, or processing being performed remotely from the analyzer which captures the processed images (such as using a cloud based platform, or using a remotely linked computer or system to process the analyzer results). Similarly, some features and sub-combinations are useful and may be employed without reference to other features and sub-combinations. Embodiments of the invention have been described for illustrative and not restrictive purposes, and alternative embodiments will become apparent to readers of this patent. In certain cases, method steps or operations may be performed or executed in differing order, or operations may be added, deleted or modified. It can be appreciated that, in certain aspects of the invention, a single component may be replaced by multiple components, and multiple components may be replaced by a single component, to

[0168] - 31 -

[0169] 0133788.0810417 4912-7631-1663v6 provide an element or structure or to perform a given function or functions. Except where such substitution would not be operative to practice certain embodiments of the invention, such substitution is considered within the scope of the invention. Accordingly, the claims should not be treated as limited to the examples, drawings, embodiments and illustrations provided above, but instead should be understood as having the scope provided when their terms are given their broadest reasonable interpretation as provided by a general-purpose dictionary, except that when a term or phrase is indicated as having a particular meaning under the heading Explicit Definitions, it should be understood as having that meaning when used in the claims.

[0170] Explicit Definitions

[0171] It should be understood that, in the above examples and the claims, a statement that something is “based on” something else should be understood to mean that it is determined at least in part by the thing that it is indicated as being based on. To indicate that something must be completely determined based on something else, it is described as being “based EXCLUSIVELY on” whatever it must be completely determined by.

[0172] It should be understood that, in the above examples and claims, the term “set” should be understood as one or more things which are grouped together. Similarly, “superset” and “subset” should be each be understood as being synonymous with “set,” with the alternative terms “superset” and “subset” being used only for ease of reading. For the avoidance of doubt, a “superset” of a type of item should not be understood as necessarily referring to all items of that type, unless such is explicitly stated. For example, a reference to capturing a “superset” of cell images should not be understood as requiring that the “superset” of cell images are all of the captured cell images. While the “superset” may be all of the captured cell images, this is not necessary, and should not be treated as being implied by the term “superset.”

[0173] - 32 -

[0174] 0133788.0810417 4912-7631-1663v6

Claims

What is claimed is:

1. A method of assessing a focal quality in a biological analyzer, comprising:providing a plurality of synthetic particles;receiving images of the plurality of synthetic particles; andfor each of the plurality of synthetic particles:analyzing pixel data; andcalculating a predicted focal position from the pixel data.

2. The method of claim 1, wherein, for at least one of the plurality of synthetic particles: analyzing the pixel data comprises extracting a plurality of features from the pixel data; andcalculating the predicted focal position from the pixel data comprises providing one or more of the plurality of features as input to a focus prediction function.

3. The method of claim 2, wherein the focus prediction function comprises a trained machine learning model.

4. The method of claim 3, wherein the trained machine learning model is a support vector regression model.

5. The method of claim 2, wherein, the focus prediction function comprises a polynomial curve fitting equation.

6. The method of claim 2, wherein each of the one or more features from the plurality of features which are provided as input to the focus prediction function is a ratio of two other features from the plurality of features from the pixel data.- 33 -0133788.0810417 4912-7631-1663v67. The method of claim 2, wherein, for the at least one of the plurality of synthetic particles, extracting the plurality of features comprises:organizing pixels comprised by the pixel data into a plurality of groups based on the pixels’ locations relative to a border of that synthetic particle; andgenerating an achromatic characteristic curve, wherein the achromatic characteristic curve is in a space having index value for groups from the plurality of groups as a first dimension, and average achromatic characteristic values of pixels of the groups from the plurality of groups as a second dimension.

8. The method of claim 7, wherein, for the at least one of the plurality of synthetic particles, extracting the plurality of features comprises:generating a smoothed curve by applying a smoothing function to the achromatic characteristic curve;generating a first derivative curve by taking a derivative of the smoothed curve; and generating a second derivative curve by taking a derivative of the first derivative curve.

9. The method of claim 8, wherein, for the at least one of the plurality of synthetic particles, the plurality of features extracted from the pixel data comprise a set of features selected from: a first feature, wherein extracting the first feature is an amplitude of a peak in the smoothed curve;a second feature, wherein the second feature is a position of the peak in the smoothed curve; a third feature, wherein the third feature is a position of a peak in the first derivative curve; a fourth feature, wherein the fourth feature is a difference between the third feature and the second feature;a fifth feature, wherein the fifth feature is a position of a valley on a right side of a peak in the second derivative curve;a sixth feature, wherein the sixth feature is an amplitude of a peak in the first derivative curve;- 34 -0133788.0810417 4912-7631-1663v6a seventh feature, wherein the seventh feature is an amplitude of a peak in the second derivative curve;an eighth feature, wherein the eighth feature is a sum of absolute values in a region of the first derivative curve between a valley on a left side of the peak in the first derivative curve and zero;a ninth feature, wherein the ninth feature is a sum of absolute values in a region of the second derivative curve between:a zero crossing between a valley on a right side of the peak in the second derivative curve and the peak in the second derivative curve; anda maximum index of any group from the plurality of groups;anda tenth feature, wherein the tenth feature is half of a sum of absolute values in a region of the second derivative curve between:a zero crossing between a valley on a left side of the peak in the second derivative curve and the peak in the second derivative curve; andzero.

10. The method of claim 9, wherein, for the at least one of the plurality of synthetic particles, the one or more features provided as input to the focus prediction function comprise:a ratio of the ninth feature to the first feature;a ratio of the sixth feature to the first feature;a ratio of the eighth feature to the first feature;a ratio of the fourth feature to the second feature;a ratio of the tenth feature to the fifth feature; anda ratio of the seventh feature to the first feature.

11. The method of claim 1, comprising: providing a flow cell for receiving the plurality of synthetic particles.- 35 -0133788.0810417 4912-7631-1663v612. A system comprising:at least one processor; anda non-transitory computer readable medium having stored thereon instruction which, when executed by the at least one processor, cause the system to perform the method of any of claims 1-11.

13. A biological analyzer comprising:at least one processor; anda non-transitory computer readable medium stored thereon instruction which, when executed by the at least one processor, cause the biological analyzer to perform the method of any of claims 1-11.

14. A method of assessing a focal quality in a biological analyzer, comprising:providing a plurality of particles;receiving images of the plurality of particles; andfor each of the plurality of particles:extracting a plurality of features from pixel data in an image of that particle; and calculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a trained machine learning model.

15. The method of claim 14, wherein the trained machine learning model is a support vector regression model.

16. The method of claim 14, wherein each of the one or more features from the plurality of features which are provided as input to the focus prediction function is a ratio of two other features from the plurality of features.- 36 -0133788.0810417 4912-7631-1663v617. The method of claim 14, wherein:for the at least one of the particles, extracting the plurality of features from the pixel data in the image of that particles comprises:organizing pixels comprised by the pixel data into a plurality of groups based on the pixels’ locations relative to a border of that particle; andgenerating an achromatic characteristic curve, wherein the achromatic characteristic curve is in a space having index value for groups from the plurality of groups as a first dimension, and average achromatic characteristic values of pixels of the groups from the plurality of groups as a second dimension.

18. The method of claim 17, wherein, for the at least one of the plurality of particles, extracting the plurality of features comprises: generating a smoothed curve by applying a smoothing function to the achromatic characteristic curve;generating a first derivative curve by taking a derivative of the smoothed curve; and generating a second derivative curve by taking a derivative of the first derivative curve.

19. The method of claim 18, wherein, for the at least one of the plurality of particles, the plurality of features extracted from the pixel data comprise a set of features selected from:a first feature, wherein extracting the first feature is an amplitude of a peak in the smoothed curve;a second feature, wherein the second feature is a position of the peak in the smoothed curve; a third feature, wherein the third feature is a position of a peak in the first derivative curve; a fourth feature, wherein the fourth feature is a difference between the third feature and the second feature;a fifth feature, wherein the fifth feature is a position of a valley on a right side of a peak in the second derivative curve;a sixth feature, wherein the sixth feature is an amplitude of a peak in the first derivative curve;- 37 -0133788.0810417 4912-7631-1663v6a seventh feature, wherein the seventh feature is an amplitude of a peak in the second derivative curve;an eighth feature, wherein the eighth feature is a sum of absolute values in a region of the first derivative curve between a valley on a left side of the peak in the first derivative curve and zero;a ninth feature, wherein the ninth feature is a sum of absolute values in a region of the second derivative curve between:a zero crossing between a valley on a right side of the peak in the second derivative curve and the peak in the second derivative curve; anda maximum index of any group from the plurality of groups; and a tenth feature, wherein the tenth feature is half of a sum of absolute values in a region of the second derivative curve between:a zero crossing between a valley on a left side of the peak in the second derivative curve and the peak in the second derivative curve; andzero.

20. The method of claim 19, wherein, for the at least one of the plurality of particles, the one or more features provided as input to the trained machine learning model comprise:a ratio of the ninth feature to the first feature;a ratio of the sixth feature to the first feature;a ratio of the eighth feature to the first feature;a ratio of the fourth feature to the second feature;a ratio of the tenth feature to the fifth feature; anda ratio of the seventh feature to the first feature.

21. The method of claim 14, comprising: providing a flow cell for receiving the plurality of particles.- 38 -0133788.0810417 4912-7631-1663v6l. K system comprising:at least one processor; anda non-transitory computer readable medium having stored thereon instruction which, when executed by the at least one processor, cause the system to perform the method of any of claims 14-21.

23. A biological analyzer comprising:at least one processor; anda non-transitory computer readable medium stored thereon instruction which, when executed by the at least one processor, cause the biological analyzer to perform the method of any of claims 14-21.

24. A method of assessing a focal quality in a biological analyzer, comprising:providing a plurality of particles;receiving images of the plurality of particles; andfor each of the plurality of particles:extracting a plurality of features from pixel data in an image of that particle; and calculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a focus prediction function, wherein each of the one or more features which is provided as input to the focus prediction function is a ratio of two other features from the plurality of features.

25. The method of claim 24, wherein:for the at least one of the particles, extracting the plurality of features from the pixel data in the image of that particles comprises:organizing pixels comprised by the pixel data into a plurality of groups based on the pixels’ locations relative to a border of that particle; and- 39 -0133788.0810417 4912-7631-1663v6generating an achromatic characteristic curve, wherein the achromatic characteristic curve is in a space having index value for groups from the plurality of groups as a first dimension, and average achromatic characteristic values of pixels of the groups from the plurality of groups as a second dimension.

26. The method of claim 25, wherein, for the at least one of the plurality of particles, extracting the plurality of features comprises:generating a smoothed curve by applying a smoothing function to the achromatic characteristic curve;generating a first derivative curve by taking a derivative of the smoothed curve; and generating a second derivative curve by taking a derivative of the first derivative curve.

27. The method of claim 26, wherein, for the at least one of the plurality of particles, the plurality of features extracted from the pixel data comprise a set of features selected from:a first feature, wherein extracting the first feature is an amplitude of a peak in the smoothed curve;a second feature, wherein the second feature is a position of the peak in the smoothed curve; a third feature, wherein the third feature is a position of a peak in the first derivative curve; a fourth feature, wherein the fourth feature is a difference between the third feature and the second feature;a fifth feature, wherein the fifth feature is a position of a valley on a right side of a peak in the second derivative curve:a sixth feature, wherein the sixth feature is an amplitude of a peak in the first derivative curve:a seventh feature, wherein the seventh feature is an amplitude of a peak in the second derivative curve;- 40 -0133788.0810417 4912-7631-1663v6an eighth feature, wherein the eighth feature is a sum of absolute values in a region of the first derivative curve between a valley on a left side of the peak in the first derivative curve and zero;a ninth feature, wherein the ninth feature is a sum of absolute values in a region of the second derivative curve between:a zero crossing between a valley on a right side of the peak in the second derivative curve and the peak in the second derivative curve; anda maximum index of any group from the plurality of groups; anda tenth feature, wherein the tenth feature is half of a sum of absolute values in a region of the second derivative curve between:a zero crossing between a valley on a left side of the peak in the second derivative curve and the peak in the second derivative curve; andzero.

28. The method of claim 27, wherein, for the at least one of the plurality of particles, the one or more features provided as input to the focus prediction function comprise:a ratio of the ninth feature to the first feature;a ratio of the sixth feature to the first feature;a ratio of the eighth feature to the first feature;a ratio of the fourth feature to the second feature;a ratio of the tenth feature to the fifth feature; anda ratio of the seventh feature to the first feature.

29. The method of claim 24, comprising: providing a flow cell for receiving the plurality of particles.

30. A system comprising:at least one processor; and- 41 -0133788.0810417 4912-7631-1663v6a non-transitory computer readable medium having stored thereon instruction which, when executed by the at least one processor, cause the system to perform the method of any of claims 24-29,31. A biological analyzer comprising:at least one processor; anda non-transitory computer readable medium stored thereon instruction which, when executed by the at least one processor, cause the biological analyzer to perform the method of any of claims 24-29.

32. A system comprising:a flowcell configured to receive a plurality of synthetic particles;a camera configured to capture a plurality of images of the synthetic particles;at least one processor; anda computer-readable medium storing instructions that are configured to, when executed by the at least one processor:receive images of the plurality of synthetic particles; andfor each of the plurality of synthetic particles:analyze pixel data; andcalculate a predicted focal position from the pixel data.

33. A method comprising performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of claim 32 is to perform when executed.

34. A biological analyzer, comprising:a flowcell configured to receive a plurality of synthetic particles;a camera configured to capture a plurality of images of the synthetic particles;at least one processor; and- 42 -0133788.0810417 4912-7631-1663v6a computer-readable medium storing instructions that are configured to, when executed by the at least one processor:receive images of the plurality of synthetic particles; andfor each of the plurality of synthetic particles:analyze pixel data; andcalculate a predicted focal position from the pixel data.

35. A method comprising performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of claim 33 is to perform when executed.

36. A system comprising:a flowcell configured to receive a plurality of synthetic particles;a camera configured to capture a plurality of images of the synthetic particles;at least one processor; anda computer-readable medium storing instructions that are configured to, when executed by the at least one processor:receive images of the plurality of synthetic particles; andfor each of the plurality of particles:extracting a plurality of features from pixel data in an image of that particle; andcalculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a trained machine learning model.

37. A method comprising performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of claim 36 is to perform when executed.

38. A biological analyzer, comprising:- 43 -0133788.0810417 4912-7631-1663v6a flowcell configured to receive a plurality of synthetic particles;a camera configured to capture a plurality of images of the synthetic particles;at least one processor; anda computer-readable medium storing instructions that are configured to, when executed by the at least one processor:receive images of the plurality of synthetic particles; andfor each of the plurality of particles:extracting a plurality of features from pixel data in an image of that particle; andcalculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a trained machine learning model.

39. A method comprising performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of claim 38 is to perform when executed.

40. A system comprising:a flowcell configured to receive a plurality of synthetic particles;a camera configured to capture a plurality of images of the synthetic particles;at least one processor; anda computer-readable medium storing instructions that are configured to. when executed by the at least one processor:receive images of the plurality of synthetic particles; andfor each of the plurality of particles:extracting a plurality of features from pixel data in an image of that particle; andcalculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a focus prediction function,- 44 -0133788.0810417 4912-7631-1663v6wherein each of the one or more features which is provided as input to the focus prediction function is a ratio of two other features from the plurality of features.

41. A method comprising performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of claim 40 is to perform when executed.

42. A biological analyzer, comprising:a flowcell configured to receive a plurality of synthetic particles;a camera configured to capture a plurality of images of the synthetic particles;at least one processor; anda computer-readable medium storing instructions that are configured to. when executed by the at least one processor:receive images of the plurality of synthetic particles; andfor each of the plurality of particles:extracting a plurality of features from pixel data in an image of that particle; andcalculating a predicted focal position for that particle by providing one or more features from the plurality of features as input to a focus prediction function, wherein each of the one or more features which is provided as input to the focus prediction function is a ratio of two other features from the plurality of features.

43. A method comprising performing the set of acts the instructions stored on the one or more non-transitory computer readable medium of claim 42 is to perform when executed.- 45 -0133788.0810417 4912-7631-1663v6

Images