System, method and non-transitory computer-readable medium for determining a set of parameters to be used for cell categorization

Abstract

Provided is a system and a method for determining a set of parameters to be used for cell categorization at a cell analysis device, and a non-transitory, computer-readable medium comprising a program code that, when the program code is executed on a processor, a computer, or a programmable hardware component, causes the processor, computer, or programmable hardware component to perform the method. The system comprises storage circuitry and processor circuitry. The system is configured to obtain an image showing a plurality of cells, perform an image analysis algorithm on the image using an initial set of parameters to determine a categorization of objects shown in the image, generate indicia of the respective categorization of the objects overlaid over the image, provide a user interface for changing at least one of a parameter of the set of parameters and a categorization of an object, with the user interface comprising the image and the indicia, obtain an input from a user, the input including at least one of a change to at least one parameter of the set of parameters or a change to the categorization of at least one object, and repeat performing at least a portion of the image analysis algorithm and generating the indicia based on the input obtained from the user to update the initial set of parameters based on the input from the user.

Description

[0001] System, Method, and Non-Transitory Computer-Readable Medium for Determining a Set of Parameters to Be Used for Cell Categorization

[0002] Cross-Reference to Related Application

[0003] This application is being filed on February 19, 2026, as a PCT International application and claims the benefit of and priority to U.S. Application No. 63 / 760,458, filed on February 19, 2025. the disclosure of which is hereby incorporated by reference in its entirety.

[0004] Field

[0005] The present disclosure relates to a system and a method for determining a set of parameters to be used for cell categorization at a cell analysis device, and to a non-transitory. computer-readable medium comprising a program code that, when the program code is executed on a processor, a computer, or a programmable hardware component, causes the processor, computer, or programmable hardware component to perform the method.

[0006] Background

[0007] Cell viability analysis is a technique used to determine the health and functional state of cells, specifically whether they are alive or dead. This analysis is crucial in numerous biological and medical research contexts, such as drug development, cancer research, and cytotoxicity testing.

[0008] Cell viability analyzers work by measuring various indicators of cell health and metabolic activity. Common methods include colorimetric assays, which involve adding a reagent that changes color in response to cellular activity, fluorescence-based assays, which use fluorescent dyes that can penetrate live cells but not dead ones, or dyes that bind to DNA in dead cells (e.g., propidium iodide), and luminescence-based assays, in which the presence of ATP - an indicator of metabolically active cells, is determined using a luminescent reaction (e.g., ATP assay). The readings are obtained using specialized equipment, which is used to quantify the changes in color, fluorescence, or luminescence, correlating them with the number of viable cells.

[0009] Some cell viability analyzers, such as the Vi-CELL BLU cell viability analyzer by Beckman Coulter (a trademark of Beckman Coulter, Inc.), use Trypan Blue Dye Exclusion as a method for analyzing cell viability. Viable (live) cells exclude the dye, remaining unstained, while non-viable (dead) cells take up the dye and appear blue.

[0010] Many of the above-mentioned cell viability analysis methods are based on applying image processing on images showing the cells to distinguish viable from dead cells (and debris). To perform this analysis, parameters are defined that enable the image processingalgorithm to perform the categorization of the cells. These parameters are specific to the experiment at hand, as default parameters are not suitable for each experiment. While providers of cell vi ability analysis devices generally provide tools to manually define and optimize the parameters, this is generally considered an expert-level task.

[0011] Summary

[0012] There may be a desire to provide an improved concept for determining parameters for use in cell analysis.

[0013] This desire is addressed by the subject matter of the independent claims.

[0014] Various examples of the present disclosure are based on the finding that manually defining and optimizing parameters to be used for cell analysis is a complex task because it often requires users to understand the details of each parameter, including its effects, sensitivities, and interactions. Many users can visually identify objects as live cells, dead cells, or other objects, but are not familiar with the cell type parameters and how they affect the categorization of their cells. The proposed concept is based on providing a user interface that allows the user to manipulate the parameters, both directly and indirectly, while observing the effects of their manipulation. The user starts with an initial set of parameters and is provided with an interface to either directly manipulate the parameters (and observe the effect of the manipulation), or to re-categorize objects (such as viable cells, dead cells, or "other" objects, e.g. debris) via the user interface, with the system determining parameters that closely match the re-categorization performed by the user. Tn this way, the proposed concept assists the user in determining appropriate parameters for the task at hand, enabling most users of cell analyzers to determine and optimize appropriate parameters without requiring expert knowledge of the details of each parameter.

[0015] Some aspects of the present disclosure relate to a system for determining a set of parameters to be used for cell categorization at a cell analysis device. The system comprises storage circuitry and processor circuitry. The system is configured to obtain an image showing a plurality of cells. The system is configured to perform an image analysis algorithm on the image using an initial set of parameters to determine a categorization of objects shown in the image. The system is configured to generate indicia of the respective categorization (i.e., a graphical representation of the respective categorization) of the objects overlaid over the image. The system is configured to provide a user interface for changing at least one of a parameter of the set of parameters and a categorization of an object. The user interface comprises the image and the indicia. The system is configuredto obtain an input from a user, with the input including at least one of a change to at least one parameter of the set of parameters or a change to the categorization of at least one object. The system is configured to repeat performing at least a portion of the image analysis algorithm and generating the indicia based on the input obtained from the user to update the initial set of parameters based on the input from the user. This way, the system assists the user in determining suitable parameters for the task at hand, enabling most users of cell analyzers to determine and optimize suitable parameters without requiring expert knowledge on the details of each parameter.

[0016] In most cases, the user is assisted in determining suitable parameters if the user receives (possibly instantaneous) feedback regarding the impact of the changes the user makes to the parameters. Thus, the user interface may be configured to trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia upon (e.g., directly in response to) the user changing the at least one parameter. For example, the image analysis algorithm may be re-run, and the indicia may be updated in response to any change to the parameters performed by the user.

[0017] While repeating performing at least a portion of the image analysis algorithm and generating the indicia can be sped up by reducing the amount of computational effort to be performed when the parameters are changed, a delay may nonetheless be introduced by said compute-intensive tasks. To reduce the impact of said delay on the effort of the user performing the assisted determination of the parameters, these tasks may be performed in the background, such that the user can continue using the user interface (and making changes) while the impact of the previous changes are calculated in the background. In other words, the user interface may be configured to trigger repeating performing at least a portion of image analysis algorithm and generating the indicia such that performing at least the portion of the image analysis algorithm and generating the indicia is performed in the background while the user is operating the user interface. This can greatly speed up the process, as the user can continue refining without being interrupted by the computations.

[0018] An intuitive and efficient way of changing the categorization of objects relies on the provision of a tool that lets users “paint” objects with a painting tool to set or correct their categorization. In other words, the user interface may be configured to provide a coloring tool for annotating objects by color. Using the painting tool, the user can efficiently change or establish the categorization of one or multiple objects. Once the user is done with a round of categorization and wants to see the effect of the change, the user cantrigger repeating performing at least the portion of the image analysis algorithm and generating the indicia via the user interface. In other words, the user interface may be configured to trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia after the user has finished annotating one or more objects by color. This way, the user can edit the categorization of multiple objects in batch before the effect of the change in categorization is applied and can be observed by the user, which can improve the workflow of the user.

[0019] Another way to change the categorization of objects is by clicking on them in the user interface. For example, the user interface may be configured to, upon the user clicking on an object, cycle through different categorizations of the object. This way, the user does not have to preset the categorization being applied (as with a painting tool) but can perform changes on the spot. Again, once the user is done with a round of categorization and wants to see the effect of the change, the user can trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia via the user interface. In other words, the user interface may be configured to trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia after the user has finished selecting the categorization of one or more objects by cycling through the different categorizations. This way, the user can edit the categorization of multiple objects in batch before the effect of the change in categorization is applied and can be observed by the user, which can improve the workflow of the user.

[0020] The proposed concept supports at least one of the following changes - changes to the parameters themselves, and changes to the categorization of objects, which are then translated into changes to the parameters. In the latter case, the user reviews the existing categorizations (being provided based on the initial set of parameters) and changes them as needed. Once the user has approved the (changed) set of categorizations, an optimization process can be applied to determine parameters that would result in a set of categorizations that more closely matches the user-approved set of categorizations. In other words, the system may be configured to, after the user has changed one or more categorizations to arrive at a user-approved set of categorizations, change one or more parameters of the set of parameters so that the set of parameters, when used by the image analysis algorithm, improves a match between the categorizations provided by the image analysis algorithm and the user-approved set of categorizations. This way, the user can determine improved parameters merely by changing the categorization of objects. Note that, in the present context, the term “optimization” or “optimization process” does notnecessarily indicate that the result is the absolute optimum. The term “optimization'’ or “optimization process” is used to describe a process which aims to improve a quality of an item, such as the parameters, without necessarily arriving at the optimum.

[0021] In particular, the determination of improved parameters may be a numerical optimization problem. For example, the image analysis algorithm may be based on comparing numerical characteristics of the objects shown in the image to thresholds defined by the set of parameters. For example, the system being configured to change the one or more parameters of the set of parameters so that the set of parameters, when used by the image analysis algorithm in the comparison between the numerical characteristics of the respective objects and the thresholds defined by the set of parameters, improves the match. This way, an improved set of parameters can be determined efficiently, as the comparison between the numerical properties of objects and the parameters can be done efficiently by a processor.

[0022] In general, to arrive at an improved set of parameters, the optimization process may start with the initial set of parameters and iteratively tweak the respective parameters to determine the impact of the change on the quality of the match between the resulting set of categorizations and the user-improved set of categorizations. For example, the system may be configured to determine the one or more parameters by iteratively changing the one or more parameters until the match satisfies a criterion, e g., starting from the initial set of parameters. This way. the parameters can be improved iteratively as long as gains with respect to the quality of the match can be observed.

[0023] While the user can change the categorization of individual objects, it may be hard for the user to grasp, on an empirical basis, how well the categorizations provided based on the set of parameters match the user-approved set of categorizations. To enable the user to see progress in the process of determining the improved set of parameters, the user interface may output a measure that lets the user grasp the progress. Accordingly, the system may be configured to provide the user interface with a measure representing the match between the categorizations provided by the image analysis algorithm and the user-approved set of categorizations. For example, the measure may include a number or ratio of characterizations output based on the respective set of parameters that match the user-approved set of characterizations. For example, the user interface may be provided with information on how this measure has changed over multiple iterations of the set of parameters.In general, the user may operate on a single image to improve the set of parameters. To verify the result, i.e.. the set of parameters, the set of parameters may be used to process multiple images showing the same cell lines or experiment, e.g., over time, or multiple experiments with the same basic setup. For example, the system may be configured to obtain a plurality of images. The system may be configured to perform the image analysis algorithm on the plurality of images. The system may be configured to provide the user interface with statistical or numerical information on the categorizations provided by the image analysis algorithm across the plurality of images. This way, the user can easily verify the qualify of the set of parameters across a larger range of images.

[0024] Cell analyzers, such as cell viability analyzers, are often used to determine statistical measures of a sample, such as viable (i.e., living) cells per unit of volume, dead cells per unit of volume etc. Scientists, which are predominantly the users of cell analyzers, often have an intuitive grasp of whether the statistical measures derived using a set of parameters are plausible or not. For example, the system may be configured to provide the user interface with at least one of a statistical measure related to live cells per unit of volume, a statistical measure related to dead cells per unit of volume, and a statistical measure related to a viability of cells. These measures can be used by the user to determine whether the results being achieved using the present set of parameters are plausible. Statistical analysis can also be used to support the user in selecting appropriate values for the parameters. For example, the system may be configured to provide the user interface with at least one histogram showing the distribution of at least one numerical characteristic of the objects shown in the image or of objects shown across a plurality' of images. If the categorization of the respective objects is based on a comparison of the numerical characteristics of the objects with thresholds (defined by the set of parameters), the user can use the histograms to determine suitable parameters, e.g., by placing the respective parameters such that the parameter is placed between two peaks in the respective histogram that indicate different categorizations.

[0025] There are various (numerical) parameters that can be used to characterize a cell, e.g., to distinguish between living and dead cells. For example, the set of parameters may comprise at least one of a minimum cell diameter, a maximum cell diameter, a cell sharpness parameter, a minimum circularity7parameter, a viable spot area parameter, and a viable spot brightness parameter.

[0026] The proposed concept relies on a user interface that is suitable for providing the user with feedback regarding the effect of the changes input by the user. To lower the time requiredfor repeating performing at least the portion of the image analysis algorithm and generating the indicia, the image processing may be split into two portions - one portion to detect possible cells in the image (the objects, also called "blobs”) and to compute the numerical properties of the objects, and a second portion to perform the characterization based on the set of parameters. In other words, the image analysis algorithm may comprise a first portion and a second portion. For example, the first portion may comprise performing object detection on the image and determining numerical characteristics of the detected objects. The second portion may comprise comparing numerical characteristics of the detected objects to thresholds defined by the set of parameters. For example, repeating performing at least the portion of the image analysis algorithm and generating the indicia based on the input obtained from the user to update the set of parameters based on the input of the user may include performing the second portion and omit performing the first portion. This way, only the second portion of the image analysis algorithm might be performed upon a change of the parameters, which may reduce the time required for repeating the image analysis algorithm with new parameters.

[0027] The process of improving the set of parameters can be sped up and made easier if the starting point, i.e., the initial set of parameters, already results in a categorization that is mostly correct. This can be achieved by using an initial set of parameters that is closely matched to the sample at hand. For example, the system may be configured to determine a species of a sample being shown in the image, and to load the initial set of parameters based on the species of the sample. This way, the effort required for improving the parameters can be reduced.

[0028] Cell analyzers generally are laboratory devices that are used in laboratory settings. While cell analyzers often have a display to review the images taken and the statistical results, the display often is rather small (as it is not the main feature of the respective cell analyzer), which makes it less suitable for the task of determining a set of parameters. Therefore, the process of determining the set of parameters can be moved to any computing device (being more convenient to use outside the laboratory or providing a larger screen space) by providing the user interface as a web application, e.g., as a HTML (HyperText Markup Language)-based website that is served by a web server functional ity of the system. In other words, the system may be configured to provide the user interface as a web application. This may make the process of determining the set of parameters easier for the user, as the user is not restricted to the laboratory and can use computing devices with arbitrarily sized (e.g., larger) displays.The proposed concept may be used in the context of cell viability analysis. Accordingly, the categorization may differentiate between an object being a live cell (a viable cell), a dead cell, and another type of object (e.g., debris).

[0029] Some further aspects of the present disclosure relate to a corresponding method for determining a set of parameters to be used for cell categorization at a cell analysis device. The method comprises obtaining an image showing a plurality of cells. The method comprises performing an image analysis algorithm on the image using an initial set of parameters to determine a categorization of cells shown in the image. The method comprises generating indicia of the respective categorization of the cells overlaid over the image. The method comprises providing a user interface for changing at least one of a parameter of the set of parameters and a categorization of a cell. The user interface comprises the image and the indicia. The method comprises obtaining an input from a user, with the input comprising at least one of a change to at least one parameter of the set of parameters and a change to the categorization of at least one cell. The method comprises repeating performing at least a portion of the image analysis algorithm and generating the indicia based on the input obtained from the user to update the initial set of parameters based on the input from the user.

[0030] Another aspect of the present disclosure relates to a non-transitory. computer-readable medium comprising a program code that, when the program code is executed on a processor. a computer, or a programmable hardware component, causes the processor, computer, or programmable hardware component to perform the above method.

[0031] Brief description of the Figures

[0032] Some examples of apparatuses and / or methods will be described in the following by way of example only, and with reference to the accompanying figures, in which

[0033] Fig. 1 shows a schematic diagram of an example of a system for determining a set of parameters to be used for cell categorization at a cell analysis device; Fig. 2 shows an image of cells and other objects with and without indicia representing a categorization of the respective cells;

[0034] Fig. 3 shows a first view of a user interface for determining parameters to be used for cell categorization at a cell analysis device;

[0035] Fig. 4 shows a second view of a user interface for determining parameters to be used for cell categorization at a cell analysis device;

[0036] Fig. 5 shows a flow chart of an example of a method for determining a set of parameters to be used for cell categorization at a cell analysis device; andFig. 6 shows a diagram of a system comprising a cell analysis device or microscope and a computer system.

[0037] Detailed Description

[0038] Some examples are now described in more detail with reference to the enclosed figures. However, other possible examples are not limited to the features of these embodiments described in detail. Other examples may include modifications of the features as well as equivalents and alternatives to the features. Furthermore, the terminology used herein to describe certain examples should not be restrictive of further possible examples.

[0039] Throughout the description of the figures same or similar reference numerals refer to same or similar elements and / or features, which may be identical or implemented in a modified form while providing the same or a similar function. The thickness of lines, layers and / or areas in the figures may also be exaggerated for clarification.

[0040] When two elements A and B are combined using an “or”, this is to be understood as disclosing all possible combinations, i.e. only A, only B as well as A and B, unless expressly defined otherwise in the individual case. As an alternative wording for the same combinations, "at least one of A and B" or "A and / or B" may be used. This applies equivalently to combinations of more than two elements.

[0041] If a singular form, such as “a”, “an” and “the” is used and the use of only a single element is not defined as mandatory either explicitly or implicitly, further examples may also use several elements to implement the same function. If a function is described below as implemented using multiple elements, further examples may implement the same function using a single element or a single processing entity. It is further understood that the terms "include", "including", "comprise" and / or "comprising", when used, describe the presence of the specified features, integers, steps, operations, processes, elements, components and / or a group thereof, but do not exclude the presence or addition of one or more other features, integers, steps, operations, processes, elements, components and / or a group thereof.

[0042] Fig. 1 shows a schematic diagram of an example of a system 110 for determining a set of parameters to be used for cell categorization at a cell analysis device 120. Fig. 1 further shows the cell analysis device 120 being coupled with the system 110. For example, the cell analysis device 120 may include the system 110 or may be coupled with the system 110 via a computer network. In this case, the set of parameters to be used for cell categorization at the cell analysis device 120 may be provided by the system 110 to the cell analysis device 120 for use in the cell analysis device. Alternatively, the system 110 maybe entirely separate from the cell analysis device 120. In this case, transferring the set of parameters to be used for categorization at the cell analysis device 120 may be up to the user. For example, the system 110 may be a web server or may be part of a web server providing a service for determining a set of parameters to be used for cell categorization at a cell analysis device 120 that is not connected to the cell analysis device 120.

[0043] In general, the system 110 may be a computer system. The system 110 comprises one or more processors 114 and one or more storage devices 116. Optionally, the system 110 further comprises one or more interfaces 112. The one or more processors 114 are coupled to the one or more storage devices 116 and to the one or more interfaces 112. In general, the functionality of the system 110 may be provided by the one or more processors 114. in conjunction with the one or more interfaces 112 (for exchanging data / infor-mation with one or more other entities, such as the cell analysis device 120 or a client device (not shown) being used by a user of the system 110), and with the one or more storage devices 116 (for storing information, such as machine-readable instructions of a computer program being executed by the one or more processors). In general, the functionality of the one or more processors 114 may be implemented by one or more processors 114 executing machine-readable instructions. Accordingly, any feature ascribed to the one or more processors 114 may be defined by one or more instructions of a plurality7of machine-readable instructions. The system 110 may comprise the machine-readable instructions, e.g., within the one or more storage devices 116.

[0044] The system 110 is configured to obtain an image showing a plurality of cells, e.g., an image generated by an optical imaging sensor of the cell analysis device 120. The system 110 is configured to perform an image analysis algorithm on the image using an initial set of parameters to determine a categorization of objects shown in the image (e.g., with the categorization differentiating between an object being a live cell, a dead cell, and another type of object, such as debris). The system 110 is configured to generate indicia of the respective categorization of the objects overlaid over the image. The system 110 is configured to provide a user interface for changing at least one of a parameter of the set of parameters and a categonzation of an object. The user interface comprises the image and the indicia. The system 110 is configured to obtain an input from a user. The input includes at least one of a change to at least one parameter of the set of parameters or a change to the categorization of at least one object. The system 110 is configured to repeat performing at least a portion of the image analysis algorithm and generating theindicia based on the input obtained from the user to update the initial set of parameters based on the input from the user.

[0045] In the following, the functionality of the system 110 will be introduced in more detail with reference to a concrete implementation of the processing functionality and user interface discussed in connection with Figs. 2 to 4.

[0046] Cell analyzers, particularly cell viability analyzers such as the Beckman Coulter Vi-CELL BLU Cell Viability Analyzer, automate the quantification of cell concentration and viability. Such a cell analyzer takes a sample from a cell culture, processes the sample, takes images of the processed sample with a microscope, and analyzes the images. Such instruments are used for research, development and manufacturing in the biopharmaceutical industry, academia, etc. Cell analyzers use cell types, a set of image analysis parameters, to classify and quantify cells in specific ways based on visual characteristics of each cell line.

[0047] An example of such an analysis is show n in Fig. 2. Fig. 2 shows an image of cells and other objects with and without labels representing a categorization of the respective cells. On the left, the image is shown without indicia representing the categorization of the respective objects (cells). The image shows live / viable cells 200, dead cells 210, and debris 220. On the right side of Fig. 2, the image is overlaid with annotations. Living cells (200) are overlaid with a thin, light, single-line circle. Dead cells 210 are overlaid with a thicker, three-line circle (black on the inner and outer edges, white in the center). Unknown / other objects (such as debris) are either not highlighted at all or are highlighted with a thicker, black, single-line circle 220. On the right side, tw o objects 230 are shown that are misclassified - they are categorized as living cells (thin, light, single-line circle), whereas they are only debris.

[0048] This illustrates a challenge in cell viability analysis: Standard cell types sometimes do not work well for all cell lines. Custom cell types can be created through cell type optimization, which is the process of determining improved parameters for categorizing objects on a cell analyzer. Such cell type optimization can be a multi-step process. For example, according to the manual cell type optimization protocol provided with the Vi-Cell BLU, if the cells are not correctly annotated with the default cell ty pe parameters, load the sample into a cell type reanalysis screen and create a new7cell type by copying the default cell type. If any small or large cells are circled blue (indicating debris), adjust a minimum or maximum cell diameter parameter until all cells are circled either green or red. If bright cells are circled red (indicating dead cells) and / or dark cells are circledgreen (indicating viable cells), adjust a viable cell spot brightness parameter until the cells are properly circled green or red. If clusters of cells are not properly declustered, adjust a declustering level (e.g., in addition to sample preparation changes). If further adjustments to properly annotated cells are required, a sharpness parameter, circularity parameter, and / or viable cell spot area parameter may be adjusted until the cells are properly annotated. It is understood that this is a time-consuming expert-level operation that may be required each time cell viability analysis is performed on a new cell line. Users can benefit from the ability to quickly create custom cell types without requiring an expert-level understanding of cell type parameters and manual optimization procedures.

[0049] The proposed concept, as provided by the system 110 of Fig. 1, provides an Automated Cell Type Optimizer (ACTO) to generate a custom cell type (i.e., a set of parameters to be used for cell analysis). As input, ACTO takes image(s) from a cell analysis device, e g., together with manual annotations of cells, with ACTO providing optimized or improved cell type parameters as output, e.g., along with an image showing annotations (i.e., indicia) being generated using the optimized (i.e., improved) cell t pe and / or with population-level statistics from analysis of multiple images.

[0050] The Automated Cell Type Optimizer (ACTO) is a software tool that facilitates the generation of custom cell types for use on a cell analysis device, such as Vi-CELL BLU. This tool enables users that are unfamiliar with the cell t pe parameters and / or the manual optimization workflow to create custom cell types rapidly and easily. Users manually change annotations (i.e., indicia) on an image (e.g., an image generated using the cell analysis device) to their expected values and ACTO may be used to generate a cell type that approaches this target data set, e.g., to achieve a categorization that best matches the target data set.

[0051] The workflow for utilization of ACTO may comprise one or more of the following operations. First, the user may upload one or more images (i.e., so that the system 110 obtains the image). Then ACTO runs the image analysis algorithm with an initial cell type (e.g., a default cell type, or a cell type that is selected based on the species of the sample) initially and displays the image with annotations. For this purpose, the system 110 performs the image analysis algorithm on the image using the initial set of parameters to determine a categorization of objects shown in the image and generates indicia of the respective categorization of the objects overlaid over the image. In the example of Figs. 2, 3 and 4, circular indicia are used that use different line strokes that surround aperimeter of the respective objects. Alternatively, the indicia may comprise or correspond to colorized perimeters that surround the respective objects. In another example, the indicia may correspond to colored semi-transparent area overlays (circular or according to the shape of the respective objects). In some other examples, rectangular boxes (with colored borders, with borders having different line sty les, or with differently colored area overlays) may be used as indicia etc. In more general terms, any kind of indicia may be used that provide a graphical representation of the categorization of the respective object.

[0052] In general, the user can improve the parameters by directly changing the respective cell type parameters (first case), or by changing the categorizations of individual objects (second case). For this purpose, the system 110 provides the user interface for changing at least one of a parameter of the set of parameters and a categorization of an object. In the first case, the user can (directly) adjust the cell type parameters (such as at least one of a minimum cell diameter, a maximum cell diameter, a cell sharpness parameter, a minimum circularity parameter, a viable spot area parameter, and a viable spot brightness parameter) to iterate towards the desired result. The categorizations and / or the indicia may be updated in response to any changes. In other words, the user interface may be configured to trigger repeating performing at least a portion of image analysis algorithm and generating the indicia upon the user changing the at least one parameter, e.g., in response to the user changing any parameter. As a result, users can observe the effects of changing cell type parameters in real time. This is significantly faster and more user-friendly than the existing offline reanalysis tool that can take several minutes to show the results of a single new cell type configuration. This significant change in performance can be achieved by a dissection of the image analysis algorithm. In particular, a computationally expensive action can be broken down into several discrete, simpler tasks that can be run more efficiently. For example, the image analysis algorithm may comprise a first portion and a second portion, with the first portion being run (only) once (or once per declustering setting) at the beginning, and with the second portion being run any time the parameters are changed. The first portion includes the computationally expensive portions of detecting the objects (or '‘blobs”) and determining their numerical characteristics (such as one or more of a diameter of the respective object, a sharpness of the respective object, a circumference of the respective object, a spot area of the respective object, and a spot brightness of the respective object) . In other words, the first portion may comprise performing object detection on the image and determining numericalcharacteristics of the detected objects. Known image processing algorithms and machinelearning models may be used for these purposes. After the first portion has been completed, initial indicia (and later refined / updated indicia) of the respective categorization of the objects can be generated using the second portion of the image analysis algorithm. The second portion comprises the vastly less computationally expensive task of comparing the numerical characteristics of the detected objects to thresholds defined by the set of parameters. As the first portion of the image analysis algorithm does not depend on the parameters of the set of parameters, it can be omitted when performing at least a portion of the image analysis algorithm and generating the indicia is repeated. In other words, repeating performing at least the portion of the image analysis algorithm and generating the indicia based on the input obtained from the user to update the set of parameters based on the input of the user may include performing the second portion and omit performing the first portion. To further improve performance, the user interface may be configured to trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia such that the process is performed in the background while the user is operating the user interface.

[0053] In the second case, the user teaches ACTO the correct classification of individual objects and tasks ACTO with determining a set of parameters that results in the classification being provided by the user. In this case, the user can manually change the object type for any object identified on the image. For example, the user can change any blob to one of the following categories: ‘live cell’’ (green colorized perimeter, or thin, bright, single-line line style), “dead cell” (red colorized perimeter, or broader, triple-line (black-white-black) line style), “other” (blue colorized perimeter, or broader, black, single-line line style). One way to change the object type is to click the blob and cycle through the three types. In other words, the user interface may be configured to, upon the user clicking on an object, cycle through different categorizations of the object. Another way to change the object type is to hold the mouse buttons and “paint” over blobs to change their type. In other words, the user interface may be configured to provide a coloring tool for annotating objects by color.

[0054] The user then initiates the optimization procedure. In other words, the user interface may be configured to trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia after the user has finished selecting the categorization of one or more objects, e.g., by cycling through the different categorizations or by annotating one or more objects by color. In this case, repeating performing at least theportion of the image analysis algorithm and generating the indicia further includes a determination of suitable parameters based on the categorization selected by the user. The user’s manually adjusted blobs are taken as the target result. The software iteratively optimizes a cell type parameter configuration that yields a result that most closely matches the target result. In other words, the system may be configured to, after the user has changed one or more categorizations to arrive at a user-approved set of categorizations, change one or more parameters of the set of parameters so that the set of parameters, when used by the image analysis algorithm, improves a match between the categorizations provided by the image analysis algorithm and the user-approved set of categorizations (as compared to the categorization provided based on the previously used set of parameters).

[0055] In the present case, the optimization process being performed is a numerical optimization process, in which the parameters are varied such that, when applied to categorize the previously detected objects, they lead to a categorization that, if possible, most closely matches the user-approved set of categorizations. In various examples, in which the image analysis algorithm is based on comparing numerical characteristics of the objects shown in the image to thresholds defined by the set of parameters, this can be done by setting the thresholds in a way that leads to, or at least approaches, the user-approved set of categorizations. In other words, the system may be configured to change the one or more parameters of the set of parameters (e.g., at least one of a minimum cell diameter, a maximum cell diameter, a cell sharpness parameter, a minimum circularity parameter, a viable spot area parameter, and a viable spot brightness parameter of the set of parameters) so that the set of parameters, when used by the image analysis algorithm in the comparison between the numerical characteristics of the respective objects and the thresholds defined by the set of parameters, improves the match (as compared with a previously used set of parameters).

[0056] This numerical optimization process may be an iterative process that starts from the initial set of parameters or. if the user has manually changed at least one parameter, from the present set of parameters. The system may be configured to determine the one or more parameters by iteratively changing the one or more parameters until the match satisfies a criterion. For example, an evolutionary algorithm may be used to iteratively change the one or more parameters. For example, the criterion may be based on the quality of the match, with the quality of the match being based on a number or ratio of characterizations output based on the respective set of parameters that match the user-approved set of characterizations. For example, the system may be configured to change a parameter into one direction (e.g.. increase or decrease) until the quality of the match decreases (e.g., one or more iterations in a row). For example, an algorithm, such as the evolutionary algorithm, may be used to determine the parameter(s) to be changed during an algorithm and / or the criterion.

[0057] As this optimization process may take some time (e.g., a few seconds), again, the user interface may be configured to trigger repeating performing at least the portion of the image analysis algorithm (including determination of the set of parameters) and generating the indicia such that performing at least the portion of the image analysis algorithm and generating the indicia is performed in the background while the user is operating the user interface.

[0058] Once a suitable set of parameters has been established and the second portion of the image analysis algorithm (that is based on the set of parameters) has been run, the optimized cell type parameters and resulting annotations may be displayed to the user. In addition, the system may be configured to provide the user interface with a measure representing the match between the categorizations provided by the image analysis algorithm and the user-approved set of categorizations, to enable the user to determine whether the set of parameters selected as part of the optimization process performs better than previous versions of the set of parameters. The user can choose to apply the optimized cell type parameters and observe its performance on the uploaded image. Finally, the user can choose to repeat the cycle of manual annotation and optimization or accept the result. The user then creates anew cell type (i.e., set of parameters) with the optimized parameters, e.g., on their cell analysis device.

[0059] In Fig. 3, a first view of a user interface for determining parameters to be used for cell categorization at a cell analysis device is shown, to illustrate how the user can manipulate the parameters directly or change the categorization of individual cells, followed by the optimization procedure, to arrive at a suitable set of parameters. In Fig. 3, the user can use buttons 300, 310, to switch between the first view shown in Fig. 3 and a second view shown in Fig. 4. In the center of the first view, the image is shown with corresponding indicia (as discussed in connection Fig. 2). A box 320 is used to highlight a currently selected object. On the bottom, the identifier (“172”), the categorization (“Live cell” 321) and the numerical properties (diameter 322, sharpness 323, circularity 324, spot area 325 and sport brightness 326) of the currently selected object are shown. On the right, an upload button 330 is shown, where the user can upload an image. Below the uploadbutton 330, a checkbox 340 is shown that lets the user toggle on / off the indicia. Alternatively, a keyboard shortcut may be used for this purpose. Dropdown menu 350 is provided to select the declustering degree, and button 360 can be used to trigger the user interface to run the optimization process (after the user has, at least temporarily, finished categorizing objects). Below, optimized parameters 370 are shown, including minimum diameter 371, maximum diameter 372, sharpness 373, circularity 374, spot area 375 and spot brightness 376. To the right, a cell mismatch percentage 377 and a viable mismatch percentage 378 are shown as measure representing the match between the categorizations provided by the image analysis algorithm and the user-approved set of categorizations. Above, a button 379 is provided that lets the user use the suggested (i.e., optimized parameters). Below this section of the first view, a section is provided in which the user can manually change the parameters, including minimum diameter (in pm) 380, maximum diameter (in pm) 381, cell sharpness 382, minimum circularity 383, viable spot area (in %) 384 and viable spot brightness (in %) 385. Further, a checkbox 386 (“Apply Cell Type”) is provided to enable automatically updating the indicia upon changes to the parameters, and a button ("Reset to Mammalian”) to reset the parameters to a default or initial set of parameters.

[0060] Fig. 4 shows the second view of the user interface. While the first view show n in Fig. 3 is focused on a single image, the second view is a population-level view that is based on applying the set of parameters to a plurality of images. In other words, the system may be configured to obtain a plurality of images, perform the image analysis algorithm on the plurality of images. This population-level view lets the user perform a plausibility check, as it contains the results 410 of the analysis (including cell count 413, average cells per image 414, total (x 10A6) cells / mL 415, viable (x 10A6) cells / mL 416, viability (%) 416, average diameter (pm) 417), at a declustering degree 411 selectable via dropdow n menu 412, over a plurality of images. Thus, the system may be configured to provide the user interface with statistical or numerical information on the categorizations provided by the image analysis algorithm across the plurality of images. For example, as shown in section 410 of the second view, the system may be configured to provide the user interface with at least one of a statistical measure related to live cells per unit of volume, a statistical measure related to dead cells per unit of volume, and a statistical measure related to the viability of cells. Button 420 can be used to upload an archive of images, on which the population analysis is to be performed.In addition to the population level statistics 410, the second view further includes a representation of one of the images, with a selected object 320’ being highlighted and properties 321’ -326 being shown on the bottom. The second view further includes the buttons 300, 310 to switch between the first and second view.

[0061] The second view also provides further population-level insights, shown on the right, which can further be used to select appropriate parameters for the set of parameters. To visually highlight the impact of the selected parameters, next to user interface elements 380-385 for changing the parameters minimum diameter 380, maximum diameter 381, cell sharpness 382, minimum circularity 383, viable spot area 384 and viable spot brightness 386, user interface control elements (sliders, partially with multiple selectors) 430, 440, 450, 460. 470 and corresponding histograms 435. 445, 455, 465. 475 (with the y-axis showing the frequency) are shown. In other words, the system may be configured to provide the user interface with at least one histogram showing the distribution of at least one numerical characteristic of the objects shown in the image or of objects shown across a plurality of images. Inside the histogram, the selected parameter(s) is / are shown as dashed vertical hne(s). The user can use the histogram to determine where to place the parameter, e.g., such that the vertical line is located between two peaks (corresponding to viable and dead cells, for example) of the respective histogram.

[0062] In the examples shown in Fig. 3 and Fig. 4, the user interface is provided as a w eb application. In this case, the system 110, which is configured to provide the user interface, is a web server or being included in a web server. On the user’s client device, the user interface may be provided via a w eb browser, or as part of an application that leverages web browser technology.

[0063] The proposed concept provides an automatic cell type optimizer, for which no understanding of the cell type parameters is required. The user need only be able to identity live cells, dead cells, and other objects in cell analysis device images. Even if the optimizer algorithm is not used, this system provides a simple interface for rapidly executing a manual cell type optimization. Furthermore, in the examples shown in Figs. 3 and 4, the software is web-based, so no software needs to be downloaded by the user. Furthermore, users are not required to share large amounts of data to utilize the tool. The user can choose to analyze as little as one image for analysis, with no metadata being required. As a result, there is no or only a minimal reliance on sample metadata for completion of the image analysis and optimization using ACTO. The input requirements are minimal -only image(s) from a cell analyzer are required. Existing workflows involving offlineanalysis tools involve exporting a large amount of information about the instrument, sample set. and individual samples from the instrument, of which some might be proprietary or sensitive. In existing workflows, this data is compressed into a .zip file and can then be loaded into the offline analysis tool. ACTO does not require this extra data. In contrast, when using ACTO, only one or more images are used, such that metadata input can be reduced or minimized. Moreover, the proposed concept provides real-time sample analysis, which is an improvement over other approaches.

[0064] Fig. 5 shows a flow chart of an example of a corresponding method for determining a set of parameters to be used for cell categorization at a cell analysis device. The method comprises obtaining 500 an image showing a plurality of cells. Optionally, at 510, the method further comprises determining a species of a sample of a sample being shown in the image, and loading an initial set of parameters based on the species of the sample. The method comprises performing an image analysis algorithm (e.g., performing 520 a first portion and performing 525 a second portion of the image analysis algorithm) on the image using an initial set of parameters to determine a categorization of cells shown in the image. The method comprises generating 530 indicia of the respective categorization of the cells overlaid over the image. The method comprises providing 540 a user interface for changing at least one of a parameter of the set of parameters and a categorization of a cell, with the user interface comprising the image and the indicia. The method comprises obtaining 550 an input from a user, the input comprising at least one of a change to at least one parameter of the set of parameters and a change to the categorization of at least one cell, and repeating performing 525 at least a portion of the image analysis algorithm and generating 530 the indicia based on the input obtained from the user to update the initial set of parameters based on the input from the user. Optionally, e.g., in case the user has changed the categorization of one or more objects, the method may further comprise changing 560 one of more parameters, e.g., using a numerical optimization process. Further optionally, the method may comprise, at 570, obtaining a plurality of images, and providing the user interface with statistical or numerical information on the categorizations provided by the image analysis algorithm across the plurality of images.

[0065] For example, the method of Fig. 5 may be implemented by the system 110 introduced in connection with one of the Figs. 1 to 4. Features introduced in connection with the system 110 of Figs. 1 to 4 may likewise be included in the corresponding method.Some embodiments relate to a microscope or cell analysis device (e.g., a cell viability analysis device) comprising a system as described in connection with one or more of the Figs. 1 to 5. Alternatively, a microscope or cell analysis device may be part of or connected to a system as described in connection with one or more of the Figs. 1 to 5. Fig. 6 shows a schematic illustration of a system 600 configured to perform a method described herein. The system 600 comprises a cell analysis device or microscope 610 and a computer system 620. The cell analysis device or microscope 610 is configured to take images and is connected to the computer system 620. The computer system 620 is configured to execute at least a part of a method described herein. The computer system 620 may be configured to execute a machine learning algorithm. The computer system 620 and cell analysis device or microscope 610 may be separate entities but can also be integrated together in one common housing. The computer system 620 may be part of a central processing system of the cell analysis device or microscope 610 and / or the computer system 620 may be part of a subcomponent of the cell analysis device or microscope 610, such as a sensor, an actor, a camera or an illumination unit, etc. of the cell analysis device or microscope 610.

[0066] The computer system 620 may be a local computer device (e.g. personal computer, laptop, tablet computer or mobile phone) with one or more processors and one or more storage devices or may be a distributed computer system (e.g. a cloud computing system with one or more processors and one or more storage devices distributed at various locations, for example, at a local client and / or one or more remote server farms and / or data centers). The computer system 620 may comprise any circuit or combination of circuits. In one embodiment, the computer system 620 may include one or more processors which can be of any type. As used herein, processor may mean any type of computational circuit, such as but not limited to a microprocessor, a microcontroller, a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a graphics processor, a digital signal processor (DSP), multiple core processor, a field programmable gate array (FPGA), for example, of a microscope or a microscope component (e.g. camera) or any other type of processor or processing circuit. Other types of circuits that may be included in the computer system 620 may be a custom circuit, an application-specific integrated circuit (ASIC), or the like, such as, for example, one or more circuits (such as a communication circuit) for use in wireless devices like mobile telephones, tablet computers, laptop computers, two-way radios, and similar electronic systems. The computer system620 may include one or more storage devices, which may include one or more memory elements suitable to the particular application, such as a main memory in the form of random access memory (RAM), one or more hard drives, and / or one or more drives that handle removable media such as compact disks (CD), flash memory cards, digital video disk (DVD), and the like. The computer system 620 may also include a display device, one or more speakers, and a keyboard and / or controller, which can include a mouse, trackball, touch screen, voice-recognition device, or any other device that permits a system user to input information into and receive information from the computer system 620.

[0067] Some or all of the method steps may be executed by (or using) a hardware apparatus, like for example, a processor, a microprocessor, a programmable computer or an electronic circuit. In some embodiments, some one or more of the most important method steps may be executed by such an apparatus.

[0068] Depending on certain implementation requirements, embodiments of the invention can be implemented in hardware or in software. The implementation can be performed using a non-transitory storage medium such as a digital storage medium, for example a floppy disc, a DVD, a Blu-Ray, a CD, a ROM, a PROM, and EPROM, an EEPROM or a FLASH memory, having electronically readable control signals stored thereon, which cooperate (or are capable of cooperating) with a programmable computer system such that the respective method is performed. Therefore, the digital storage medium may be computer readable.

[0069] Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which are capable of cooperating with a programmable computer system, such that one of the methods described herein is performed.

[0070] Generally, embodiments of the present invention can be implemented as a computer program product with a program code, the program code being operative for performing one of the methods when the computer program product runs on a computer. The program code may, for example, be stored on a machine-readable carrier.

[0071] Other embodiments comprise the computer program for performing one of the methods described herein, stored on a machine-readable carrier.

[0072] In other words, an embodiment of the present invention is, therefore, a computer program having a program code for performing one of the methods described herein, when the computer program runs on a computer.A further embodiment of the present invention is, therefore, a storage medium (or a data carrier, or a computer-readable medium) comprising, stored thereon, the computer program for performing one of the methods described herein when it is performed by a processor. The data carrier, the digital storage medium or the recorded medium are typically tangible and / or non-transitionary. A further embodiment of the present invention is an apparatus as described herein comprising a processor and the storage medium.

[0073] A further embodiment of the invention is, therefore, a data stream or a sequence of signals representing the computer program for performing one of the methods described herein. The data stream or the sequence of signals may, for example, be configured to be transferred via a data communication connection, for example, via the internet.

[0074] A further embodiment comprises a processing means, for example, a computer or a programmable logic device, configured to, or adapted to, perform one of the methods described herein.

[0075] A further embodiment comprises a computer having installed thereon the computer program for performing one of the methods described herein.

[0076] A further embodiment according to the invention comprises an apparatus or a system configured to transfer (for example, electronically or optically) a computer program for performing one of the methods described herein to a receiver. The receiver may, for example, be a computer, a mobile device, a memory device or the like. The apparatus or system may. for example, comprise a file server for transferring the computer program to the receiver.

[0077] In some embodiments, a programmable logic device (for example, a field programmable gate array) may be used to perform some or all the functionalities of the methods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor to perform one of the methods described herein. Generally, the methods are preferably performed by any hardware apparatus.

[0078] The aspects and features described in relation to a particular one of the previous examples may also be combined with one or more of the further examples to replace an identical or similar feature of that further example or to additionally introduce the features into the further example.

[0079] It is further understood that the disclosure of several steps, processes, operations or functions disclosed in the description or claims shall not be construed to imply that these operations are necessarily dependent on the order described, unless explicitly stated in the individual case or necessary for technical reasons. Therefore, the previous descriptiondoes not limit the execution of several steps or functions to a certain order. Furthermore, in further examples, a single step, function, process or operation may include and / or be broken up into several sub-steps, -functions, -processes or -operations.

[0080] If some aspects have been described in relation to a device or system, these aspects should also be understood as a description of the corresponding method. For example, a block, device or functional aspect of the device or system may correspond to a feature, such as a method step, of the corresponding method. Accordingly, aspects described in relation to a method shall also be understood as a description of a corresponding block, a corresponding element, a property or a functional feature of a corresponding device or a corresponding system.

[0081] The following claims are hereby incorporated in the detailed description, wherein each claim may stand on its own as a separate example. It should also be noted that although in the claims a dependent claim refers to a particular combination with one or more other claims, other examples may also include a combination of the dependent claim with the subject matter of any other dependent or independent claim. Such combinations are hereby explicitly proposed, unless it is stated in the individual case that a particular combination is not intended. Furthermore, features of a claim should also be included for any other independent claim, even if that claim is not directly defined as dependent on that other independent claim.Reference Numerals 110 System

[0082] 112 Interface

[0083] 114 Processor

[0084] 116 Storage device

[0085] 120 Cell analysis device

[0086] 200 Viable cell

[0087] 210 Dead cell

[0088] 220 Debris

[0089] 230 Mis-categorized objects

[0090] 300 Button to activate first view of user interface 310 Button to activate second view of user interface 320 Box to highlight selected object

[0091] 321 Categorization of the selected object

[0092] 322 Diameter of the selected object

[0093] 323 Sharpness of the selected object

[0094] 324 Circularity of the selected object

[0095] 325 Spot area of the selected object

[0096] 326 Spot brightness of the selected object

[0097] 330 Upload button

[0098] 340 Checkbox to toggle on / off annotations (indicia) 350 Dropdown menu to select declustering degree 360 Button to run the optimization process

[0099] 370 Optimized parameters

[0100] 371 Minimum diameter

[0101] 372 Maximum diameter

[0102] 373 Sharpness

[0103] 374 Circularity

[0104] 375 Spot area

[0105] 376 Spot brightness

[0106] 377 Cell mismatch percentage

[0107] 378 Viable mismatch percentage

[0108] 379 Button to use the optimized parameters

[0109] 380 UI element to change minimum diameter381 UI element to change maximum diameter

[0110] 382 UI element to change cell sharpness

[0111] 383 UI element to change minimum circularity

[0112] 384 UI element to change viable spot area

[0113] 385 UI element to change viable spot brightness

[0114] 386 Checkbox to enable / disable automatically updating annotations (indicia) 410 Statistical result of population-level analysis

[0115] 411 Declustering degree

[0116] 412 Selected declustering degree

[0117] 413 Cell count

[0118] 414 Average cells per image

[0119] 415 Total cells per mL

[0120] 416 Viable cells per mL

[0121] 417 Average diameter

[0122] 420 Upload button for archive containing multiple images

[0123] 430 Slider to control minimum and maximum diameter

[0124] 435 Histogram of diameter of the cells

[0125] 440 Slider to control cell sharpness

[0126] 445 Histogram of cell sharpness

[0127] 450 Slider to control minimum circularity

[0128] 455 Histogram of circularity

[0129] 460 Slider to control viable spot area

[0130] 465 Histogram of viable spot area

[0131] 470 Slider to control viable spot brightness

[0132] 475 Histogram of viable spot brightness

[0133] 500 Obtaining an image

[0134] 510 Determining a species and loading an initial set of parameters

[0135] 520 Performing a first portion of an image analysis algorithm

[0136] 525 Performing a second portion of the image analysis algorithm

[0137] 530 Generating indicia

[0138] 540 Providing a user interface

[0139] 550 Obtaining an input

[0140] 560 Changing one or more parameters

[0141] 570 Obtaining a plurality of imagesSystem

[0142] Imaging device, such as cell analysis device or microscope Computer system

Claims

WHAT IS CLAIMED IS:

1. A system (110) for determining a set of parameters to be used for cell categorization at a cell analysis device (120), the system (110) comprising storage circuitry (116) and processor circuitry (114), wherein the system (110) is configured to:obtain an image showing a plurality of cells;perform an image analysis algorithm on the image using an initial set of parameters to determine a categorization of objects shown in the image;generate indicia of the respective categorization of the objects overlaid over the image;provide a user interface for changing at least one of a parameter of the set of parameters and a categorization of an object, with the user interface comprising the image and the indicia;obtain an input from a user, the input including at least one of a change to at least one parameter of the set of parameters or a change to the categorization of at least one object; andrepeat performing at least a portion of the image analysis algorithm and generating the indicia based on the input obtained from the user to update the initial set of parameters based on the input from the user.

2. The system according to claim 1, wherein the user interface is configured to trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia upon the user changing the at least one parameter.

3. The system according to any one of claims 1-2, wherein the user interface is configured to trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia such that performing at least the portion of the image analysis algorithm and generating the indicia is performed in the background while the user is operating the user interface.

4. The system according to any one of claims 1-3, wherein the user interface is configured to provide a coloring tool for annotating objects by color, with the user interface being configured to trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia after the user has finished annotating one or more objects by color.

5. The system according to any one of claims 1-4, wherein the user interface is configured to, upon the user clicking on an object, cycle through different categorizations of the object, with the user interface being configured to trigger repeating performing at least the portion of the image analysis algorithm and generating the indicia after the user has finished selecting the categorization of one or more objects by cycling through the different categorizations.

6. The system according to any one of claims 1-5, wherein the system is configured to, after the user has changed one or more categorizations to arrive at a user-approved set of categorizations, change one or more parameters of the set of parameters so that the set of parameters, when used by the image analysis algorithm, improves a match between the categorizations provided by the image analysis algorithm and the user-approved set of categorizations.

7. The system according to any one of claims 1-6, wherein the image analysis algorithm is based on comparing numerical characteristics of the objects shown in the image to thresholds defined by the set of parameters, with the system being configured to change the one or more parameters of the set of parameters so that the set of parameters, when used by the image analysis algorithm in the comparison between the numerical characteristics of the respective objects and the thresholds defined by the set of parameters, improves the match.

8. The system according to claim 7, wherein the system is configured to determine the one or more parameters by iteratively changing the one or more parameters until the match satisfies a criterion.

9. The system according to any one of claims 1-8, wherein the system is configured to provide the user interface with a measure representing the match betw een the categorizations provided by the image analysis algorithm and the user-approved set of categorizations.

10. The system according to any one of claims 1-9, wherein the system is configured to obtain a plurality of images, perform the image analysis algorithm on the plurality’ of images, and to provide the user interface with statistical or numerical information on the categorizations provided by the image analysis algorithm across the plurality of images.

11. The system according to any one of claims 1-10, wherein the system is configured to provide the user interface with at least one of a statistical measure related to live cells per unit of volume, a statistical measure related to dead cells per unit of volume, and a statistical measure related to a viability of cells.

12. The system according to any one of claims 1-11, wherein the system is configured to provide the user interface with at least one histogram showing the distribution of at least one numerical characteristic of the objects shown in the image or of objects shown across a plurality of images.

13. The system according to any one of claims 1-12, wherein the set of parameters comprises at least one of a minimum cell diameter, a maximum cell diameter, a cell sharpness parameter, a minimum circularity parameter, a viable spot area parameter, and a viable spot brightness parameter.

14. The system according to any one of claims 1-13, wherein the image analysis algorithm comprises a first portion and a second portion, the first portion comprising performing object detection on the image and determining numerical characteristics of the detected objects, the second portion comprising comparing numerical characteristics of the detected objects to thresholds defined by the set of parameters, wherein repeating performing at least the portion of the image analysis algorithm and generating the indicia based on the input obtained from the user to update the set of parameters based on the input of the user includes performing the second portion and omits performing the first portion.

15. The system according to any one of claims 1-14, wherein the system is configured to determine a species of a sample being shown in the image, and to load the initial set of parameters based on the species of the sample.

16. The system according to any one of claims 1-15, wherein the system is configured to provide the user interface as web application.

17. The system according to any one of claims 1-16, wherein the categorization differentiates between an object being a live cell, a dead cell, and another type of object.

18. A method for determining a set of parameters to be used for cell categorization at a cell analysis device (120), the method comprising:obtaining (500) an image showing a plurality7of cells;performing (520; 525) an image analysis algorithm on the image using an initial set of parameters to determine a categorization of cells shown in the image;generating (530) indicia of the respective categorization of the cells overlaid over the image;providing (540) a user interface for changing at least one of a parameter of the set of parameters and a categorization of a cell, with the user interface comprising the image and the indicia;obtaining (550) an input from a user, the input comprising at least one of a change to at least one parameter of the set of parameters and a change to the categorization of at least one cell; andrepeating performing (525) at least a portion of the image analysis algorithm and generating (530) the indicia based on the input obtained from the user to update the initial set of parameters based on the input from the user.

19. A non-transitory, computer-readable medium comprising a program code that, when the program code is executed on a processor, a computer, or a programmable hardware component, causes the processor, computer, or programmable hardware component to perform the method of claim 18.

Images