Formed classification apparatus, system, method and program, and reclassification apparatus

Abstract

This provides a classification technique for formed elements that is even superior to known techniques. In the classification of formed elements, an image of the formed elements contained in the sample is acquired (S10), each of the formed element images is classified in relation to the type of formed element (S20), and it is automatically determined, based on predefined conditions, whether or not to further reclassify the classified formed element images with higher accuracy (S40).

Description

Technical Field

[0001] The present invention relates to a technique for classifying formed elements contained in a biological specimen.

Background Art

[0002] An apparatus and method for classifying formed elements contained in a specimen fluid by imaging using a flow method are known. The images of the formed elements detected in the image of the specimen fluid are classified for each known type. By this classification, the number concentration of the formed elements contained in the specimen fluid can be measured (for example, Japanese Patent Application Laid-Open No. 2020-085535).

Summary of the Invention

Problems to be Solved by the Invention

[0003] There is a demand for providing a classification technique for formed elements that is more excellent than the above-described known technique.

Means for Solving the Problems

[0004] In a first aspect, a formed element classification apparatus includes a first acquisition unit that acquires a formed element image representing a formed element contained in a specimen, and a first classification unit that classifies each of the formed element images in association with the type of the formed element, and it is automatically determined whether the classified formed element images are reclassified with high accuracy based on pre-defined conditions.

[0005] In a second aspect, a reclassification apparatus (remote processing apparatus) acquires the classified formed element images by the formed element classification apparatus of the first aspect and reclassifies the formed element images.

[0006] In a third embodiment, the formed-subject classification system comprises a formed-subject classification device according to the first embodiment and a remote processing device according to the second embodiment configured to communicate with each other via a network line, wherein the remote processing device includes a second acquisition unit that acquires formed-subject images classified by the formed-subject classification device and a second classification unit that reclassifies the classified formed-subject images based on user input on an operation screen.

[0007] In a fourth aspect, the formed element classification method includes the steps of: acquiring formed element images representing formed elements contained in a sample; classifying each of the formed element images in relation to the type of formed element; and automatically determining, based on predefined conditions, whether or not to further reclassify the classified formed element images with higher accuracy.

[0008] In a fifth aspect, a program for classifying formed substances is provided, which includes instructions for causing a computer processor to execute the classifying formed substances method according to the fourth aspect.

[0009] Further embodiments are disclosed in the dependent claims, drawings and the following description. [Brief explanation of the drawing]

[0010] [Figure 1] A perspective view showing an example configuration of a urinary formed-molecular-weight analysis device applied to embodiments of the present invention. [Figure 2] Figure 1 shows a side view of the urine sediment analyzer. [Figure 3] A block diagram illustrating an embodiment of the formed classification system of the present invention. [Figure 4] A block diagram illustrating the functional configuration of the formed classification device shown in Figure 3. [Figure 5] A block diagram illustrating the functional configuration of the remote processing unit (reclassification device) shown in Figure 3. [Figure 6] A flowchart illustrating the processing of a formed classification device according to an embodiment of the present invention. [Figure 7] A diagram illustrating an image of formed elements. [Figure 8] Figure illustrating the status screen. [Figure 9] Figure illustrating the work list screen. [Figure 10] Figure illustrating the dashboard screen. [Figure 11] Figure illustrating the atlas screen. [Figure 12] Figure illustrating the approval screen. [Figure 13] Figure illustrating the tangible component display screen. [Figure 14] Flowchart illustrating the reclassification process in the remote processing device. [Figure 15] Flowchart illustrating the reprocessing in the tangible component classification device. [Figure 16] Figure illustrating the setting screen. [Figure 17] Figure illustrating the automatic review request determination screen. [Figure 18] Figure illustrating the flag condition setting screen. [Figure 19] Figure illustrating the tangible component condition setting screen. [Figure 20] Figure illustrating the qualitative condition setting screen.

Embodiments for Carrying Out the Invention

[0011] Hereinafter, an example of an embodiment for implementing the technology of the present disclosure will be described in detail with reference to the drawings. In addition, components and processes that perform the same functions in terms of operation, action, and function may be given the same reference numerals throughout the drawings, and redundant explanations may be omitted as appropriate. Each drawing only schematically shows the technology of the present disclosure to such an extent that it can be sufficiently understood. Therefore, the technology of the present disclosure is not limited only to the illustrated examples. Also, in this embodiment, descriptions of configurations that are not directly related to the present disclosure and well-known configurations may be omitted.

[0012] Before explaining the embodiment in detail with reference to FIG. 1, some general explanations will be given.

[0013] In the flow method described at the beginning, a formed element classification apparatus and method for acquiring an image of a specimen flowing through a flow cell are generally known. The formed element image is classified into known classification groups (types) based on the formed elements of the specimen detected in the image. For each of the thus classified types, the concentration of the formed elements in the specimen is measured.

[0014] When the value of the measured concentration of the formed elements indicates an abnormality, it is recognized that there may be a problem with the classification of the formed element image.

[0015] Therefore, some embodiments relate to a formed element classification apparatus, comprising a first acquisition unit that acquires a formed element image representing the formed elements contained in a specimen, and a first classification unit that classifies each of the formed element images in association with the type of the formed elements, and the classified formed element images are automatically determined as to whether to be reclassified with higher precision based on pre-defined conditions.

[0016] This formed element classification apparatus can be configured as a medical device used by a user (operator), or can be configured based on a standard (general-purpose) computer, or can also be configured based on other types of electronic devices that can execute the functions described in this embodiment.

[0017] The formed element classification apparatus may include one or more processors (for example, at least one of a central processing unit and a graphics processing unit), a field programmable gate array, an application-specific integrated circuit, etc. implemented in a standard computer, or other known electronic components.

[0018] A formed element image can be obtained based on at least one sample image. In other words, one sample image can represent one or more formed elements, and one or more formed element images can be extracted and generated from one image by algorithms based on, for example, pattern recognition, machine learning, or other techniques described in this embodiment and known to those skilled in the art, as will be explained in more detail below. In some embodiments, each formed element image represents one formed element detected in one image.

[0019] When multiple images of a sample are taken, formed element images are extracted from each of the captured sample images and sorted or grouped according to the type of formed element detected in the image. Therefore, in some embodiments, the number of formed element images representing a particular type of formed element may be associated with the concentration of that particular type of formed element in the sample.

[0020] By associating each formed element image with a type of formed element, the formed element images are classified by type.

[0021] The classification of formed elements includes, as described later, for example, red blood cells, white blood cells, non-squamous epithelial cells, squamous epithelial cells, bacteria, crystals, yeast, hyaline casts, other casts, mucus, sperm, leukocyte agglutinations, and other formed elements. Therefore, in some embodiments, the detected formed elements in the associated formed element image are classified into (e.g., a predetermined) classification. Here, the predetermined classification may be based on a predetermined set of classifications, for example, a set of predefined classes or classifications (where, for example, different classes or classifications correspond to different (types) of formed elements (or groups of formed elements, etc.)). Formed elements and associated formed element images that cannot be associated with a class or classification of the predetermined classification are unclassifiable in some embodiments and are therefore considered unclassified or defined as such.

[0022] The association between formed element images and the types of formed elements is performed based on the types of formed elements detected in the corresponding formed element images.

[0023] Furthermore, based on predefined conditions, it is automatically determined whether or not to reclassify the already classified formed element images. This reclassification is performed with higher classification accuracy than the original classification of the formed element images. Therefore, in some embodiments, this allows the operator or user to automatically examine the formed element images in more detail without having to determine whether the images are correctly classified or if there may be problems.

[0024] As mentioned above, and as will be explained in more detail later, the classification of segmented images is typically based on pattern recognition algorithms or machine learning-based algorithms. The accuracy of such machine learning algorithms usually depends on the quantity and quality of the training data. It may also depend on other factors, such as the kernel size used in the convolutional neural network, the number of neurons in the neural network, or other parameters of the machine learning algorithm, which will also be explained in more detail later.

[0025] In some embodiments, the classification of formed element images by a formed element classifier may be performed with a coarser accuracy than the accuracy of the reclassification. Reclassification may be performed by specially trained personnel, pattern recognition, machine learning algorithms, or other methods with higher accuracy than the relevant algorithms used for the (initial) classification. These will also be discussed in more detail later.

[0026] In some embodiments, predefined conditions are associated with the concentration of a type of formed element in the sample. The concentration derived from the formed element image may be associated with the concentration of a specific formed element in the sample.

[0027] In some embodiments, the concentration of a certain type of formed element in the sample is determined for each of these types based on the number of classified formed element images. Thus, in some embodiments, the concentration of a particular type of formed element in the sample can be indicated by the number of such formed elements.

[0028] In some embodiments, the concentration of a type of formed element is used as a predefined condition, for example, in the form of a threshold. Thus, the predefined condition can be defined to perform reclassification when the concentration of a type of formed element exceeds a predefined threshold for that type of formed element.

[0029] In some embodiments, the predefined conditions may be met even if a concentration of zero or a very low value below a lower threshold indicates a malfunction of the device or circuit or a failure in concentration measurement, for example, when the concentration is within a predetermined range of concentration values, or when the concentration is zero or below (or equal to) a predetermined threshold, for example, when at least a low concentration of a certain type of formed element in the sample is expected.

[0030] The concentration may be provided, for example, as a percentage per volume, but may also be provided as the number of components per image area, weight per volume, absolute number, or as other variables or numerical values ​​representing the concentration of a substance known to those skilled in the art.

[0031] In some embodiments, predefined conditions are associated with the classification accuracy of formed elements. For example, if the classification accuracy of a particular formed element falls below (or equals) a predefined threshold, it may be determined that reclassification is necessary. On the other hand, in some cases, if the classification accuracy is very high and exceeds (or equals) an upper threshold, it may indicate a malfunction in the device or concentration measurement.

[0032] As mentioned above, in some embodiments, the classification accuracy is specialized for classifying formed element images into specific types of formed elements. In other words, the classification accuracy may be low for a first type of formed element (image) and high for a second type of formed element (image). For example, the detection accuracy of formed elements may differ for each formed element, and as a result, the classification accuracy may also differ. In some cases, the mapping between the detected type of formed element and the associated classification (class) may be ambiguous, with different classifications being associated with the same type of formed element, or conversely, different types of formed elements being associated with the same classification, which may result in reduced classification accuracy.

[0033] In some embodiments, predefined conditions are associated with sample quality values. Quality values ​​indicate the quality of the measurement, such as the measurement of the sample, the image quality for generating formed element images, and the overall quality status of the instrument, which will be described in more detail later.

[0034] Quality values ​​can be obtained by measuring the quality of a sample. In some embodiments, the quality of a sample is measured, thereby obtaining quality values. Quality values ​​can be obtained based on the performance of a test, for example, in relation to a urinalysis that yields a urinalysis result showing one or more quality values, as will be explained in more detail below.

[0035] In some embodiments, predefined conditions are further associated with error information representing anomalies related to the quality measurement of a sample. For example, the device may detect at least one of the following: errors during sample testing, device malfunctions, or malfunctions of sensors or other electronic components. Error information may also include "flags" indicating the occurrence of error events, which will be described in more detail later.

[0036] In some embodiments, the predefined conditions can be configured by the user by setting one or more conditions, such as the occurrence of an event, error, threshold, etc., which serves as a criterion for automatically determining whether to reclassify the material formed element image, as described in this embodiment.

[0037] In some embodiments, the classified formed element images are further configured to be transmitted to a remote processing unit (reclassification unit) for reclassification, which will be described in more detail later. The formed element classification unit and the remote processing unit can be configured to communicate with each other, for example, via a network, the internet, wireless or wired, direct link or protocol (e.g., TCP / IP), or via other types of communication described in this embodiment, or other types of digital communication suitable for digital communication between electronic devices or devices and known to those skilled in the art.

[0038] In this embodiment, a formed sorting device can be configured as the first processing device and a remote processing device (re-classification device) as the second processing device.

[0039] The remote processing device may have a circuit configured to perform any of the methods, functions, and features described in this embodiment.

[0040] Furthermore, the remote processing unit (reclassification unit) may be located remotely from the device; that is, it may be separate from the device rather than being in the same enclosure. However, the disclosure is not limited to this, and in some embodiments, the remote processing unit may be located remotely at a functional level. Thus, the communication link between the formed classification unit and the remote processing unit can be adjusted to suit the settings, distance between the device and the remote processing unit, and the type of connection that is technically suitable for transmitting digital data, such as formed images or other data described in this embodiment, between the device and the remote processing unit.

[0041] In some embodiments, the system is further configured to determine a group of classified formed element images, and reclassification is performed based on this group of classified formed element images. Thus, in some embodiments, the group of formed element images is determined for reclassification and sent to the remote processing unit, reducing the total amount of data transmitted and the processing load of reclassifying the formed element images. Furthermore, the group of classified formed element images may be determined based on at least one of the following, for example, one or more predefined classifications or classes, types of formed elements, the number of formed element images, or predefined criteria. For example, it may be known in advance that the classification accuracy for a particular class or classification may be lower than that for other classes or classifications. The same applies to types of formed elements; it may be known in advance that the classification accuracy for a particular type of formed element may be lower than that for other types. In some cases, a small number of formed element images (for a particular formed element) may indicate low classification accuracy.

[0042] In some embodiments, as described above, only the determined and classified group of formed element images is transmitted to the remote processing unit.

[0043] In some embodiments, the system is configured to additionally transmit classification information associated with the classified formed element image to the remote processing unit. This classification information can be used by the remote processing unit to perform reclassification.

[0044] In some embodiments, the system is further configured to determine whether predefined conditions are met. For example, if the predefined conditions are met, it is automatically decided to reclassify the formed element image.

[0045] Determining whether predefined conditions are met may include determining the relationship between the concentration of formed elements of a user-specified type and the threshold specified by the user, the relationship between the quality value representing the qualitative test result of the sample and the threshold, and the occurrence status of at least one of the error items specified by the user within the error information. These have been described above and will be explained in more detail below.

[0046] In some embodiments, the sample is urine.

[0047] Some embodiments relate to a remote processing device (reclassifier) ​​configured to acquire classified formed element images and reclassify the formed elements by associating them with types, wherein the reclassification is performed with higher classification accuracy than the classification of the classified formed element images.

[0048] The remote processing device (reclassification device) can be configured to perform any of the methods, functions, and features described based on the formed classification device according to this embodiment.

[0049] In some embodiments, the formed element classification system described in this embodiment is configured such that the formed element classification device and the remote processing device communicate with each other via a network. The remote processing device acquires the formed element images classified by the formed element classification device and reclassifies these acquired classified formed element images based on user input on the operation screen. This reclassification is performed with higher classification accuracy than the classification of the classified formed element images.

[0050] Furthermore, as mentioned above, the formed sorting device and the remote processing device can communicate with each other via any type of communication link.

[0051] The operator of the remote processing device may be a user who is an expert in the (re)classification of formed element images. The operation screen may consist of a human-machine interface that provides information to the operator and graphical elements for interacting with the remote processing device, as described in this embodiment.

[0052] In some embodiments, user input includes at least one of the following: selection of a formed element image, selection of a type of formed element, and selection of a reclassification method.

[0053] In some embodiments, the remote processing unit is further configured to transmit the results of the reclassification of the formed element image to the formed element classifier. This allows the formed element classifier to, for example, display the reclassification results to the user.

[0054] Some embodiments relate to a corresponding formed element classification method, which includes acquiring formed element images representing formed elements contained in a sample, classifying each formed element image in relation to a type of formed element, and automatically determining the reclassification of the classified formed element images based on predefined conditions, the reclassification being performed with higher classification accuracy than the classification of formed element images, as described in relation to this embodiment and the formed element classification device. Furthermore, all functions, methods, and features that can be performed in the formed element classification device may (at least in part) be part of the formed element classification method.

[0055] Some embodiments relate to computer programs for formed classification, which, when executed by a processor (or a set of processors, circuits, or computers), include instructions that cause the processor to perform the method described in these embodiments.

[0056] In some embodiments, a non-temporary computer-readable recording medium is also provided that, when executed by a processor (or a plurality of processors, circuits, or a computer), stores a computer program that causes the method described in this embodiment to be executed.

[0057] All units and entities described in this embodiment can be implemented as integrated circuit logic, for example, on a chip or within a circuit, unless otherwise specified, and the functions provided by such units and entities can be implemented by software unless otherwise specified.

[0058] In each embodiment, the formed element classification device includes a first acquisition unit that acquires formed element images representing formed elements contained in a sample, a first classification unit that classifies each of the formed element images in relation to the type of formed element, and an output unit that outputs a first status and a second status to be displayed on the operation screen. The first status indicates that the results obtained from the classified formed element images are unapproved, and the second status indicates that the classification of the formed element images is under review. The classified formed element images are associated with either the first status or the second status based on the corresponding classification. This will be explained in more detail below.

[0059] Here, the operation screen is configured as a graphical user interface (GUI) as a man-machine interface (MMI) to provide the user with interactive elements for presenting information and controlling the device.

[0060] The user interface includes at least two types of elements: an information element for presenting information and an interaction element for interacting with the device. In some cases, a single element may provide both functions simultaneously, namely the functions of an information element and an interaction element.

[0061] Interactive elements can be configured to interact with the user via an operation screen by receiving user input through input means such as a pointer device, keyboard, touchscreen, voice commands, or gesture recognition.

[0062] Furthermore, by providing the first and second situations, the embodiment allows even users with limited technical and medical knowledge to manage the device and quickly classify formed element images. This enables users to monitor the progress of the classification of formed element images while operating the device. The embodiment is superior to the prior art, which lacks automatic classification of formed element images, does not automatically associate the classification results with the first or second situation, and does not provide such information to the user.

[0063] In some embodiments, the operation screen includes an approval screen in which information about the classified formed element image is displayed when the user interacts with the display of a first status. This approval screen is configured to display associated information based on the classified formed element image. In some embodiments, the information is associated with or includes results obtained from the classified formed element image. In some embodiments, the user can approve the classification of the classified formed element image by interacting with the approval screen. In some embodiments, the results obtained based on the classification of the formed element image can also be approved.

[0064] In some embodiments, the information is at least one of the concentrations of formed elements and qualitative test results obtained from the classified formed element images. This information may be associated with or include results obtained based on the classification of the formed element images. Thus, in some embodiments, the results are at least one of the concentrations of formed elements and qualitative test results, or include both. This allows the user (operator) to determine the need to reclassify the classified formed element images based on at least one of the concentrations of formed elements and qualitative test results.

[0065] The approval screen is configured to accept user input, and based on this user input, the formed element image is transmitted via the network line and reclassified by the remote processing unit. This allows the user (or operator) to easily reclassify already classified material formed element images.

[0066] Furthermore, as will be explained below, the concentration of each type of formed element in the sample is determined based on the number of classified formed element images.

[0067] In some embodiments, "under review" in the second situation indicates that reclassification is being attempted. This allows the user to understand the status of the classified formed element images.

[0068] In some embodiments, the user interface displays a third status indicating that the reclassified formed element images are awaiting approval. This allows the user to see which reclassified formed element images (including, in some embodiments, results obtained based on the reclassified formed element images) are available for approval. The user can then approve the reclassified formed element images by interacting with the display of the third status. Furthermore (or alternatively), the results obtained based on the reclassified formed element images may be approved.

[0069] In some embodiments, reclassification is performed automatically based on predefined conditions, and the reclassification is performed with higher accuracy than the classification in the classification unit.

[0070] As mentioned above, predefined conditions may be configurable by the user.

[0071] In some cases, the user interface may include a settings screen configured to set predefined conditions based on user actions, and this settings screen may include at least one of the following: a setting for formed elements and a setting for qualitative elements. Thus, the user can set predefined conditions by setting at least one of the formed elements and qualitative conditions, and based on these conditions, it is determined whether or not to reclassify the formed element image.

[0072] Some embodiments relate to a formed object classification system comprising the formed object classification device described above and a remote processing device configured to communicate with each other via a network line, wherein the remote processing device includes a second acquisition unit that acquires formed object images classified by the formed object classification device, and a second classification unit that further reclassifies the classified formed object images with higher accuracy based on user input on an operation screen.

[0073] Some embodiments relate to a method for classifying formed elements, comprising the steps of: acquiring formed element images representing formed elements contained in a sample; classifying each formed element image in relation to the type of formed element; and outputting a first status and a second status to be displayed on an operation screen, wherein the first status indicates that the results obtained from the classified formed element images are unapproved, and the second status indicates that the classification of the formed element images is under review, and the classified formed element images are associated with either the first status or the second status based on the corresponding classification.

[0074] Some embodiments relate to a formed classification program that includes instructions causing a computer processor (or a set of processors, circuits, or a computer) to perform a formed classification method.

[0075] Figure 1 is a perspective view showing a urinary formed-molecular-weight element analyzer 70 applied to an embodiment of the present invention. In this embodiment, a formed-molecular-weight-element classifier 10 is applied to the urinary formed-molecular-weight-element analyzer 70.

[0076] As shown in Figure 1, the urine formed sediment analyzer 70 comprises a flow cell 40, a housing 72, a camera 74, and a light source 76. The arrow UP in Figure 1 indicates the upper side in the vertical direction of the urine formed sediment analyzer 70.

[0077] The flow cell 40 is used for urinary sediment testing, for example, by introducing a urine sample, which is an example of a specimen, along with a sheath fluid. The camera 74 then captures images of the formed elements in the urine sample, and various analyses are performed based on the shape of the formed elements in the captured images. This camera 74 is an example of an imaging unit. A urine sample contains several types of formed elements. Examples of these formed elements include red blood cells, white blood cells, epithelial cells, casts, and bacteria. In this embodiment, a urine sample is used as an example of a specimen, and the number concentration of the target formed elements in the urine is measured, with red blood cells, white blood cells, non-squamous epithelial cells, squamous epithelial cells, bacteria, crystals, yeast, hyaline casts, other casts (also called pathological casts), mucus filaments, sperm, and leukocyte clumps being the targets for measurement. Although the description focuses on urinary sediment testing, it can also be used for formed element testing of blood, cells, body fluids, etc.

[0078] The urine formed-element analyzer 70 comprises a housing 72 in which a flow cell 40 is arranged. The housing 72 has a recess 72A into which the flow cell 40 is inserted, and the area of ​​the housing 72 including the recess 72A is made of a transparent material (for example, glass). Inside the housing 72, a camera 74 is provided at a position opposite the flow cell 40. On the upper side of the housing 72, a light source 76 is provided at a position opposite the camera 74 and the flow cell 40, with the camera 74 in between. The camera 74 is positioned to capture images of the sample fluid flowing through the flow cell 40.

[0079] The urine formed sediment analyzer 70 includes a first supply device 78 that supplies sample fluid to the sample inlet 42 of the sample flow path (not shown) in the flow cell 40. The first supply device 78 includes a supply pipe 80, one end of which is connected to the sample inlet 42, and a pump 82 provided in the middle of the supply pipe 80. A test tube 84 in which the sample fluid is stored is positioned at the other end of the supply pipe 80. A barcode label displaying a barcode representing a sample ID for uniquely identifying the sample contained in the test tube 84 is affixed to the side of the test tube 84.

[0080] The urine formed element analyzer 70 includes a second supply device 86 that supplies sheath fluid to the sheath inlet 44 of the sheath channel (not shown) in the flow cell 40. The second supply device 86 includes a supply pipe 88, one end of which is connected to the sheath inlet 44, a pump 90 provided in the middle of the supply pipe 88, and a tank 92 connected to the other end of the supply pipe 88 and which stores the sheath fluid. In some embodiments of the formed element analyzer, or the apparatus or system for classifying formed elements, the second supply device 86 may be omitted, or a different fluid may be supplied to assist in classifying the formed elements of the sample. In some embodiments, two or more supply devices may be used in addition to the first supply device 78 that supplies the sample.

[0081] Furthermore, the flow cell 40 is provided with an outlet 46 between the sample inlet 42 and the sheath inlet 44. One end of a discharge pipe (not shown) is connected to the outlet 46, and the other end of this discharge pipe is connected to a waste tank (not shown). The flow cell 40 includes a confluence section (not shown) that combines the sample introduced from the sample inlet 42 and the sheath fluid introduced from the sheath inlet 44, and the combined fluid flows through the flow path. The camera 74 captures images of the formed elements in the sample flow. In other words, by imaging the sample flow with the camera 74, an image representing the formed elements in the sample flow is generated. In some embodiments, an image of formed elements showing, for example, a specific type of formed element is extracted from such an image. This will be explained in more detail below.

[0082] Figure 2 is a side view illustrating a urine sediment analyzer 70 applied to an embodiment of the present invention.

[0083] The urine formed element analyzer 70 shown in Figure 2 is equipped with the formed element classification device 10 according to this embodiment. The arrow UP in Figure 2 indicates the upper side of the urine formed element analyzer 70 in the vertical direction, similar to Figure 1.

[0084] The formed sorting device 10 functions as a control unit that controls the operation of the camera 74, the light source 76, the light source operating unit 77, the pump 82, and the pump 90, which are electrically connected to the camera 74 and the light source 76. The formed sorting device 10 causes the light source 76 to emit light at predetermined intervals by supplying a pulse signal to the light source operating unit 77. The formed sorting device 10 also drives the pump 82 to control the flow rate of the sample and drives the pump 90 to control the flow rate of the sheath fluid. Although not shown in the figure, the device may also include multiple cameras 74 and an optical system that guides light to each camera 74. Each optical system is adjusted so that each camera 74 is focused at a different position (depth) in the flow cell 40. In other words, multiple cameras 74 simultaneously capture multiple images that are focused at the same position in the horizontal plane but at different depth positions. The simultaneously captured images are stored in the storage unit 15 shown in Figure 3, which will be described later, in correspondence. The depth direction here refers to the direction perpendicular to the flow direction of the sample, and means the up and down direction in Figure 2. Each focal point and the distance to the wall on the side of the flow cell 40 closest to the camera 74 are different.

[0085] Figure 3 is a block diagram illustrating an embodiment of the formed classification system 100 of the present invention.

[0086] As shown in Figure 3, the formed sedimentation system 100 comprises a formed sedimentation device 10, a remote processing device 20 (reclassification device), a urine qualitative analyzer 30, and a server 35. The formed sedimentation device 10 is connected to the remote processing device 20 via a network line N. The urine qualitative analyzer 30 is an example of a qualitative analyzer that measures the qualitative properties of a sample and is connected to the server 35 via a network line N. The urine qualitative analyzer 30 is connected, for example, to a urine formed sediment analyzer 70.

[0087] The formed classification device 10 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, an input / output interface (I / O) 14, a storage unit 15, a display unit 16, an operation unit 17, a communication unit 18, and a connection unit 19. The CPU 11 may be, for example, a processor such as a GPU (Graphics Processing Unit). A GPU may also be additionally provided to perform specific graphics calculations or calculations of machine learning algorithms such as neural networks. The formed classification device 10 may have fewer hardware components, different hardware components, or more hardware components than those exemplified (in some embodiments, these may constitute circuits).

[0088] The formed sedimentation device 10 according to this embodiment can be a general-purpose computer device such as a personal computer (PC). Alternatively, a portable computer device such as a smartphone or tablet may be used for the formed sedimentation device 10. Furthermore, the formed sedimentation device 10 may be divided into multiple units. For example, it may include a first unit that controls a measurement system such as a camera 74, a light source 76, a pump 82, and a pump 90, and a second unit that processes and analyzes images captured by the camera 74. The formed sedimentation device 10 may also be externally connected to a urine formed sediment analyzer 70.

[0089] The control unit 10A is comprised of a CPU 11, ROM 12, RAM 13, and I / O 14. The control unit 10A has functions for controlling measurement systems such as a camera 74, light source 76, pump 82, and pump 90, and functions for processing and analyzing images captured by the camera 74. These components, CPU 11, ROM 12, RAM 13, and I / O 14, are connected to each other via a bus.

[0090] I / O 14 is connected to various functional units, including a storage unit 15, a display unit 16, an operation unit 17, a communication unit 18, and a connection unit 19. These functional units are capable of communicating with the CPU 11 via I / O 14.

[0091] The control unit 10A may be configured as a sub-control unit that controls the operation of a part of the formed sorting device 10, or as part of the main control unit that controls the operation of the entire formed sorting device 10. Some or all of the blocks of the control unit 10A may use integrated circuits such as LSIs (Large Scale Integration) or IC (Integrated Circuit) chipsets. Individual circuits may be used for each of the above blocks, or circuits that integrate some or all of them may be used. The above blocks may be provided as a single unit, or some blocks may be provided separately. Furthermore, parts of each of the above blocks may be provided separately. For the integration of the control unit 10A, dedicated circuits or general-purpose processors may be used, not just LSIs.

[0092] For example, the storage unit 15 can be an HDD (Hard Disk Drive), an SSD (Solid State Drive), or flash memory. The storage unit 15 stores an information processing program 15A for performing the measurement and re-measurement processes, which will be explained later. This information processing program 15A may also be stored in the ROM 12. The storage unit 15 may also have external memory or additional memory added later.

[0093] The information processing program 15A may, for example, be pre-installed in the formed sorting device 10. The information processing program 15A may also be implemented by storing it on a non-volatile non-temporary storage medium or distributing it via a network line N, and then installing or upgrading it in the formed sorting device 10 as appropriate. Examples of non-volatile non-temporary storage mediums include CD-ROMs (Compact Disc Read Only Memory), magneto-optical disks, HDDs, DVD-ROMs (Digital Versatile Disc Read Only Memory), flash memory, and memory cards.

[0094] The display unit 16 may include, for example, a liquid crystal display (LCD) or an organic EL (electroluminescence) display. The display unit 16 may also have an integrated touch panel. The operation unit 17 is equipped with, for example, a keyboard or mouse for inputting information. The user communicates instructions to the formed sorting device 10 through the operation of the operation unit 17. The display unit 16 displays various information such as the results of processing performed in response to the instructions received from the user, and notifications regarding the processing.

[0095] The communication unit 18 is connected to a network line N such as the Internet, LAN (Local Area Network), or WAN (Wide Area Network), and is capable of communicating with the remote processing unit 20 via the network line N.

[0096] The connection section 19 is connected to, for example, a camera 74, a light source 76, a pump 82, a pump 90, and the like. These measurement systems, including the camera 74, light source 76, pump 82, and pump 90, are controlled by the control unit 10A described above. The connection section 19 also functions as an input port for receiving the image output by the camera 74.

[0097] On the other hand, the remote processing unit 20 according to this embodiment includes a CPU 21, a ROM 22, a RAM 23, an input / output interface (I / O) 24, a storage unit 25, a display unit 26, an operation unit 27, and a communication unit 28. The CPU 21 may be a processor such as a GPU. In addition, a GPU may be provided to perform specific graphical calculations or calculations of machine learning algorithms such as neural networks. The remote processing unit 20 may include fewer hardware components, different hardware components, or more hardware components than those exemplified (which may constitute circuits depending on the embodiment).

[0098] In this embodiment, the remote processing unit 20 can be a general-purpose computer device such as a server computer. Since the remote processing unit 20 performs more data processing than the formed classification device 10, it is desirable that the memory of the remote processing unit 20 has a faster access speed than the memory of the formed classification device 10, and that the CPU 21 of the remote processing unit 20 has a faster processing speed than the CPU 11 of the formed classification device 10.

[0099] The control unit 20A is composed of a CPU 21, ROM 22, RAM 23, and I / O 24 (which may be part of the circuit of a formed sorting device). These components, CPU 21, ROM 22, RAM 23, and I / O 24, are connected to each other via a bus.

[0100] The I / O 24 is connected to various functional units, including a storage unit 25, a display unit 26, an operation unit 27, and a communication unit 28. These functional units are capable of communicating with the CPU 21 via the I / O 24.

[0101] For example, the storage unit 25 can be an HDD, SSD, flash memory, etc. The storage unit 25 stores a data management program 25A for performing the reclassification process, which will be explained later. This data management program 25A may also be stored in the ROM 22. The storage unit 25 may have external memory or additional memory added later.

[0102] The data management program 25A may, for example, be pre-installed on the remote processing unit 20. The data management program 25A may also be implemented by storing it on a non-volatile, non-temporary storage medium, or by distributing it via a network line N, and then installing or upgrading it on the remote processing unit 20 as needed. Examples of non-volatile, non-temporary storage media include CD-ROMs, magneto-optical disks, HDDs, DVD-ROMs, flash memory, and memory cards.

[0103] The display unit 26 may include, for example, a liquid crystal display (LCD) or an organic EL display. The display unit 26 may also have an integrated touch panel. The operation unit 27 is equipped with, for example, a device for operation input such as a keyboard or mouse. The user transmits instructions to the remote processing unit 20 through the operation of the operation unit 27. The display unit 26 displays various information such as the results of processing performed in response to the instructions received from the user, and notifications regarding the processing.

[0104] The communication unit 28 is connected to a network line N such as the Internet, LAN, or WAN, and is capable of communicating with the formed sorting device 10 via the network line N.

[0105] Furthermore, the urine qualitative analyzer 30 is connected to the urine formed sediment analyzer 70 via a urine sample transport path. The urine qualitative analyzer 30 is a device for performing a urine qualitative test on a urine sample. A urine qualitative test is a test that determines whether a target component is present in a urine sample, or measures the concentration of a target component in a urine sample, by applying urine to a test strip called a test tape, which changes color in reaction with the target component in the urine sample, and measuring the color change. The urine qualitative analyzer 30 is equipped with a barcode reader (not shown) for reading the sample ID of the sample to be measured from a barcode label attached to the side of the test tube 84, and transmits the urine qualitative test result of the urine sample tested by the urine qualitative analyzer 30 and the sample ID of the urine sample linked (associated) to each other via the network line N to the server 35. The transmitted sample ID and the urine qualitative test result linked to the sample ID are stored in the server 35. Furthermore, if an error occurs when measuring a urine sample, the urine qualitative analyzer 30 transmits the error information of the urine sample, linked to the sample ID of the urine sample, to the server 35 via the network line N.

[0106] Next, with reference to Figure 4, the functional configuration of the formed sorting device 10 shown in Figure 3 will be described in detail.

[0107] The CPU 11 of the formed sorting device 10 according to this embodiment functions as the various parts shown in Figure 4 by writing the information processing program 15A stored in the storage unit 15 to the RAM 13 and executing it.

[0108] As shown in Figure 4, the CPU 11 of the formed classification device 10 functions as a first acquisition unit 11A, a first classification unit 11B, a calculation unit 11C, a transmission unit 11D, a reception unit 11E, an output unit 11F, and a receiving unit 11G.

[0109] The memory unit 15 stores the first trained model 15B. The first trained model 15B is a model used for image classification processing by the first classification unit 11B.

[0110] The first acquisition unit 11A captures the sample flowing through the flow cell 40 with the camera 74 and extracts from multiple images (hereinafter also referred to as "sample images"; for example, 300 images, 1000 images) the various types of formed elements contained in the sample, as formed element images 3, and acquires one or more extracted formed element images. Specifically, the first classification unit 11B extracts formed element images 3 from each sample image using various known techniques such as image processing such as binarization and contour extraction, machine learning methods, and pattern matching methods. Each formed element image 3 contains one formed element. In other words, each formed element image 3 contains one formed element.

[0111] The first classification unit 11B classifies the formed element images 3 acquired by the first acquisition unit 11A into detection components according to predetermined classifications (e.g., type, size, shape, presence or absence of a nucleus, etc.). The set of formed element images 3 classified according to predetermined classifications by the first classification unit 11B, i.e., the formed element image group, is temporarily stored in the storage unit 15 separately for each sample. Therefore, the storage unit 15 can store the classified formed element images for each sample, and each classified formed element image is associated with a specific class or classification, and the class or classification is associated with a specific type of formed element. As a method for classifying the formed element images 3, various known techniques can be applied, such as a machine learning method or a pattern matching method. In this embodiment, the formed element image group is classified, for example, using a first trained model 15B. The first trained model 15B is a model generated by machine learning training data obtained by associating detection components for each predetermined classification with each of the formed element images 3 obtained in the past. In other words, the training data is supervised data. The first trained model 15B takes the formed elemental image 3 as input and outputs the detection component for each predetermined classification. For machine learning training models, for example, a CNN (Convolutional Neural Network) is used. For machine learning techniques, for example, deep learning is used. Since the group of formed elemental images is composed of individual formed elemental images 3, it is also called the group of formed elemental images 3, using the same code as the formed elemental image 3.

[0112] The main classifications of formed elements include, for example, red blood cells, white blood cells, non-squamous epithelial cells, squamous epithelial cells, bacteria, crystals, yeast, hyaline casts, other casts, mucous filaments, sperm, and leukocyte clumps, as well as other formed elements (hereinafter also referred to as unclassified) such as formed elements formed by the aggregation of different types of tangible objects. Red blood cells are represented as RBC, white blood cells as WBC, non-squamous epithelial cells as NSE, squamous epithelial cells as SQEC, other casts as NHC, and bacteria as BACT. Crystals are represented as CRYS, yeast as YST, hyaline casts as HYST, mucous filaments as MUCS, sperm as SPRM, and leukocyte clumps as WBCC. Furthermore, formed elements other than red blood cells, white blood cells, non-squamous epithelial cells, squamous epithelial cells, bacteria, crystals, yeast, hyaline casts, other casts, mucous filaments, sperm, and leukocyte clumps are represented as UNCL (unclassified) or "other formed elements." In other words, the detected components for each predetermined classification classified by the first classification unit 11B correspond to the respective formed elements and the classifications defined as unclassified.

[0113] Furthermore, when classifying the formed element images 3, the first classification unit 11B calculates the degree of fit based on the image classification method being used (for example, machine learning, pattern matching, etc.). The first classification unit 11B classifies the formed element images to the classification with the highest degree of fit. The degree of fit here indicates the classification accuracy of the classified image, and for each predetermined classification, a higher value is assigned the higher the percentage of matching with the ground truth image or predetermined feature points. If there is a perfect match with the ground truth image or feature points, the degree of fit is 100%. In other words, formed element images 3 with a relatively low degree of fit are likely to have not been properly classified. The degree of fit may also be expressed as precision. In some embodiments, the degree of fit is used as a predefined criterion for deciding whether to reclassify a classified formed element image (for example, one associated with a specific specimen).

[0114] The goodness of fit can vary depending on how the formed elements are depicted in formed element image 3. Specifically, in images where the formed elements are in focus, classification using machine learning is easier, resulting in a high goodness of fit for accurate classification and a low goodness of fit for inaccurate classification. However, in images where the formed elements are out of focus, i.e., blurred, the goodness of fit for accurate classification will be low, and the difference between the goodness of fit for accurate and inaccurate classification will also be small. Furthermore, in images where multiple formed elements overlap, the goodness of fit may be low. In addition, even if an item in a rare sample is not learned by the first trained model 15B and should accurately be classified as unclassified, it will be classified into one of the formed elements, resulting in a low goodness of fit in this case.

[0115] The calculation unit 11C calculates the number concentration of formed elements in the sample based on the number of formed element images classified for each predetermined classification by the first classification unit 11B. The concentration may be a number concentration (for example, the number of images classified as a specific formed element, or as described in detail below), a percentage per volume of the sample or a portion of the sample, or other concentration indicators.

[0116] As will be explained later, if it is necessary to remeasure the number concentration of formed elements in the sample, the transmitting unit 11D controls the communication unit 18 to transmit the formed element image 3 to the remote processing unit 20 via the network line N. The formed element image 3 transmitted to the remote processing unit 20 may be all of the classified formed element image 3 or only some of the images. In addition, the transmitting unit 11D transmits the classification result of the formed element image 3 classified by the first classification unit 11B along with the formed element image 3.

[0117] The receiving unit 11E controls the communication unit 18 to receive the reclassification results from the remote processing unit 20 (reclassification device) after the formed element image 3 has been reclassified by the remote processing unit 20.

[0118] The output unit 11F outputs at least one of the first, second, and third situations regarding the reclassification of the formed element image 3. This output may be displayed on the display unit 16 or printed from a printer (not shown). The first situation represents the state after the first classification unit 11B has classified the formed element image 3 into predetermined categories, and is waiting for an instruction to send the formed element image 3 to the remote processing unit 20. The second situation represents the state waiting for the reclassification result to be received from the remote processing unit 20. The third situation represents the state after the reclassification result has been received from the remote processing unit 20.

[0119] The reception unit 11G receives operations entered by the user via the operation unit 17.

[0120] Next, with reference to Figure 5, the functional configuration of the remote processing device 20 applied to this embodiment will be described in detail. As mentioned above, depending on the embodiment, the remote processing device constitutes part of the formed classification system.

[0121] The CPU 21 of the remote processing unit 20 applied to this embodiment functions as the various parts shown in Figure 5 by writing the data management program 25A stored in the storage unit 25 to the RAM 23 and executing it.

[0122] Figure 5 is a block diagram showing an example of the functional configuration of the remote processing device 20 (reclassification device) applied to this embodiment.

[0123] As shown in Figure 5, the CPU 21 of the remote processing device 20 applied to this embodiment functions as a second acquisition unit 21A, a second classification unit 21B, a display control unit 21C, a reply unit 21D, and a receiving unit 21E.

[0124] The memory unit 25 stores the second trained model 25B. The second trained model 25B is a model used for image classification by the second classification unit 21B, and depending on the embodiment, it may have higher classification accuracy than the first trained model 15B.

[0125] The receiving unit 21E controls the communication unit 28 to receive the formed element image 3 from the formed element classification device 10. The (classified) formed element image 3 received from the formed element classification device 10 is temporarily stored in the storage unit 25 as a group of images to be classified.

[0126] The second acquisition unit 21A acquires the formed element image 3 to be classified from the group of images to be classified stored in the storage unit 25.

[0127] The second classification unit 21B (re)classifies the formed element image 3 acquired by the second acquisition unit 21A as detection components according to predetermined classifications (e.g., type, size, shape, presence or absence of nuclei, etc.) of the formed element image. The formed element image 3 classified according to predetermined classifications by the second classification unit 21B is sent to the reply unit 21D. As an example of a method for classifying the formed element image, a machine learning method is applied. In this case, the formed element image 3 is classified using, for example, a second trained model 25B. The second trained model 25B is a model generated by machine learning using the same machine learning algorithm as the first trained model 15B, on separate training data to which more detection components are associated than the training data of the first trained model 15B. The amount of training data trained on the second trained model 25B is greater than the amount of training data trained on the first trained model 15B. In other words, the second trained model 25B is trained to have higher classification performance than the first trained model 15B.

[0128] Furthermore, the second pre-trained model 25B may be a model generated by machine learning the training data of the first pre-trained model 15B using a different algorithm that has higher classification performance than the machine learning algorithm used in the first pre-trained model 15B. In addition to the CNN mentioned above, there are various other machine learning algorithms, such as linear regression, regularization, decision trees, random forests, k-nearest neighbors (k-NN), logistic regression, and support vector machines (SVM). If the classification performance of the pre-trained models is higher with SVM than with CNN, for example, CNN will be used for the first pre-trained model 15B and SVM will be used for the second pre-trained model 25B. Conversely, if the classification performance of the pre-trained models is higher with CNN than with SVM, SVM will be used for the first pre-trained model 15B and CNN will be used for the second pre-trained model 25B. Note that to compare the classification performance of the pre-trained models, a method may be used in which metrics representing model performance (e.g., accuracy, precision, etc.) are calculated using pre-prepared test data and compared. In some embodiments, such index values ​​are used as predetermined criteria for determining whether or not reclassification of the formed element image is necessary.

[0129] Furthermore, the second trained model 25B may be a model generated by machine learning using a different algorithm that has higher classification performance than the machine learning algorithm used for the first trained model 15B, on a different set of training data to which more detection components are associated than the training data for the first trained model 15B.

[0130] Furthermore, if the version of the second pre-trained model 25B is being managed, it is desirable that the version of the second pre-trained model 25B be kept up to date.

[0131] Here, the second classification unit 21B may classify the formed element images 3 according to the user's classification operation. In other words, the second classification unit 21B performs classification according to the user's instructions. The user referred to here is preferably a medical technologist or similar person who is familiar with the classification of formed element images 3. Hereafter, the user operating the remote processing device 20 will be referred to as a "medical technologist" to distinguish him from the user operating the formed element classification device 10.

[0132] The display control unit 21C controls the display unit 26 to display the formed element images 3 to be classified, corresponding to the classification results from the classification unit 11B. The user reclassifies the formed element images 3 that have been incorrectly classified among the formed element images 3 displayed on the display unit 26 to the appropriate classification. In this case, the second classification unit 21B classifies and displays the formed element images 3 on the display unit 26 according to the classification operation performed by the medical technologist.

[0133] <Measurement processing by the formed classification device 10> Next, with reference to Figure 6, the operation of the formed sorting device 10 according to this embodiment will be described.

[0134] Figure 6 is a flowchart showing an example of the flow of measurement processing performed by the formed semantics classifier 10 when the reception unit 11G receives a measurement instruction for a sample from the user. Depending on the embodiment, this measurement processing may be part of the formed semantics classifier method. The CPU 11 of the formed semantics classifier 10 reads the information processing program 15A stored in the storage unit 15 and executes the measurement processing.

[0135] First, in step S10, the control unit 10A drives a transport unit (not shown) to transport the test tube 84 containing the sample, which is placed in a predetermined position on the transport unit, to the sample collection position. A barcode reader (not shown) is attached to the sample collection position, and the control unit 10A uses the barcode reader to read the barcode label attached to the side of the test tube 84. The barcode label displays, for example, a barcode representing a sample ID for uniquely identifying the sample, and the control unit 10A obtains the sample ID of the sample to be measured by reading the barcode label.

[0136] The control unit 10A controls an actuator (not shown) that moves the supply tube 80 vertically within the urine sediment analyzer 70, causing the tip of the supply tube 80 (the tip opposite to the tip connected to the sample intake port 42), which is positioned above the opening of the spit tube 84 that has been transported to the sample collection position, to descend into the interior of the spit tube 84 through the opening. After the control unit 10A has lowered the tip of the supply tube 80 to a position where it can reach the sample, it drives the pump 82. As a result, the sample inside the spit tube 84 is introduced into the flow cell 40 from the sample intake port 42 at a predetermined flow rate, and a specified volume of sample flows into the flow cell 40.

[0137] Meanwhile, the control unit 10A also drives the pump 90 in conjunction with the drive of the pump 82. As a result, the sheath fluid stored in the tank 92 is introduced into the flow cell 40 from the sheath inlet 44 at a predetermined flow rate, and the sample and the sheath fluid merge within the flow cell 40.

[0138] The control unit 10A controls the camera 74 to capture sample images of the sample in the flow cell 40 and store them, for example, in the storage unit 15. There is no limit to the number of sample images that can be captured; the control unit 10A will capture, for example, only the number of sample images pre-stored in the storage unit 15. The user can change the number of sample images stored in the storage unit 15 via the operation unit 17.

[0139] Since the captured sample image contains various types of formed elements, the first acquisition unit 11A extracts images of each formed element contained in the sample image, i.e., formed element images 3, for each formed element.

[0140] Figure 7 shows an example of a formed element image 3 extracted by the first acquisition unit 11A. The formed element image 3 is a rectangular image that includes the entire formed element. Therefore, the size of the formed element image 3 changes depending on the size of the formed element.

[0141] The first acquisition unit 11A assigns a formed element image ID to each formed element image 3 extracted from the sample image. The formed element image ID is an identifier that uniquely identifies each formed element image 3 and is used, for example, as the file name of the formed element image 3. The first acquisition unit 11A also generates a classification list that associates each formed element image 3 with the sample ID of the sample from which the formed element image 3 was taken, and stores it, for example, in the storage unit 15. An example of a classification list is shown in Table 1. Since each formed element image 3 is an image obtained from the same sample, as shown in Table 1, the same sample ID is associated with each formed element image ID.

[0142] [Table 1]

[0143] In step S20, the first classification unit 11B uses the first trained model 15B, which is pre-stored in the memory unit 15, to classify each formed element image 3 according to the type of formed element.

[0144] As already explained, the first pre-trained model 15B is an example of a classification model for an elemental image 3 generated by machine learning using training data that takes an elemental image 3, in which the types of elements contained are known, as input and outputs the types of elements contained in the elemental image 3. The number of nodes in the output layer of the first pre-trained model 15B is set to match the number of types of elements that can be classified by the elemental classification device 10, and each node in the output layer of the first pre-trained model 15B is associated one-to-one with a type of element.

[0145] When an image of formed elements 3 is input to the first pre-trained model 15B, the first pre-trained model 15B outputs a fitness score from each node in the output layer. Since each node in the output layer is associated with a type of formed element, the first classification unit 11B classifies the type of formed element associated with the output layer node that outputs the highest fitness score into a type of formed element contained in the image of formed elements 3 input to the first pre-trained model 15B. In this way, the first classification unit 11B sequentially inputs all the images of formed elements 3 extracted from the sample image into the first pre-trained model 15B, thereby classifying each image of formed elements 3 contained in the sample image according to the type of formed element. In other words, by classifying the images of formed elements 3 into images of formed elements 3 that show the formed elements of the target of measurement and images of formed elements 3 that show formed elements other than the formed elements of the target of measurement (i.e., other formed elements), the first classification unit 11B identifies the images of formed elements 3 that show the formed elements of the target of measurement from among the images of formed elements 3.

[0146] The first classification unit 11B associates each formed element image ID in the classification list shown in Table 1 with the type of formed element contained in the formed element image 3, which is represented by the formed element image ID, as classified using the first trained model 15B. An example of a classification list with the types of formed elements associated is shown in Table 2. Note that the values ​​in the classification column of the classification list in Table 2 do not have to be formed element names, but may be symbols that represent the formed element names. The classification list with the types of formed elements associated with each formed element image ID is an example of the classification result of the formed element images.

[0147] [Table 2]

[0148] In step S30 of Figure 6, the calculation unit 11C refers to the classification list shown in Table 2 obtained by the processing in step S20 and calculates the number concentration of formed elements contained in the sample based on the number of formed element images 3 classified for each type of formed element. The number concentration of formed elements is an index that represents the concentration of formed elements contained in the sample by the number of formed elements contained in a predetermined unit volume, such as 1 μL.

[0149] Specifically, the calculation unit 11C calculates the number concentration of each formed element contained in the sample using a concentration calculation formula pre-stored in the storage unit 15. Table 3 shows an example of a concentration calculation formula for each formed element.

[0150] [Table 3]

[0151] In the concentration calculation formulas shown in Table 3, the operator "*" represents multiplication. The number concentration y of each formed element is expressed by a linear function where, for example, the number of formed element images 3 for each type of formed element is the explanatory variable x. In the concentration calculation formula, "an" (where n is an integer) is the slope determined for each type of formed element, and bn is the intercept determined for each type of formed element. "Xn" is the number of formed element images 3 for each of the n types of formed elements, and "Yn" is the number concentration for each of the n types of formed elements. The concentration calculation formula for each formed element is a formula prepared in advance through experiments or computer simulations that determine the relationship between the number of formed element images 3 taken of formed elements contained in a specified volume of sample and the number concentration of the formed elements, and is stored in the memory unit 15.

[0152] Note that the concentration calculation formulas shown in Table 3 are just examples, and the concentration calculation formulas for each formed element are not limited to linear functions. Also, although Table 3 shows concentration calculation formulas corresponding to 13 types of formed elements, the number of formed elements classified by the formed element classification device 10 is also just one example.

[0153] In step S40 of Figure 6, the control unit 10A refers to the number concentration of each formed element calculated by the calculation unit 11C in step S30 and determines whether the predetermined items related to the test of the sample (hereinafter referred to as "judgment items") meet the review conditions. The review conditions are conditions set by the user via the operation unit 17, which indicate that recalculation of the number concentration is recommended. The judgment items and review conditions are stored in advance in the storage unit 15, for example, and the user can change the judgment items and review conditions via the operation unit 17. Details of the judgment items and review conditions will be explained later.

[0154] If the review conditions are not met (resulting in a negative judgment), recalculating the number concentration of formed components is not necessarily required, so the process proceeds to step S50.

[0155] In step S50, the output unit 11F displays the measurement status of the number concentration of formed elements contained in the sample (hereinafter referred to as "measurement status of formed element concentration in urine") on the display unit 16.

[0156] Here, the screens that the output unit 11F displays on the display unit 16 will be described. The screens that the output unit 11F displays on the display unit 16 include, for example, a status screen 61, a worklist screen 62, a dashboard screen 63, and an atlas screen 64. One or more screens 61-64 may form an operation screen as a man-machine interface for a user to operate the formed sorting device. At least one of the different screens 61-64 may form an element of the operation screen. In some embodiments, the following elements may be at least partially part of the operation screen.

[0157] Figure 8 shows an example of the status screen 61, Figure 9 shows an example of the worklist screen 62, Figure 10 shows an example of the dashboard screen 63, and Figure 11 shows an example of the atlas screen 64.

[0158] The output unit 11F displays the status screen 61 on the display unit 16 when the user selects the status button 2A with a mouse or the like. The output unit 11F displays the worklist screen 62 on the display unit 16 when the user selects the worklist button 2B with a mouse or the like. The output unit 11F displays the dashboard screen 63 on the display unit 16 when the user selects the dashboard button 2C with a mouse or the like. The output unit 11F displays the atlas screen 64 on the display unit 16 when the user selects the atlas image button 2D with a mouse or the like.

[0159] The status screen 61 displays information about the user operating the formed classification device 10, i.e., the operator; the status of the formed classification device 10, such as the connection status between the formed classification device 10 and other devices and the remaining amount of consumables used for measuring samples; and the number of samples measured and the calibration results of the formed classification device 10. The status screen 61 also displays information about the previous scheduled maintenance, the cleaning status of components that require cleaning, such as the supply pipe 80, information about the shutdown of the formed classification device 10, and information about the startup process performed when the formed classification device 10 is started.

[0160] The worklist screen 62 displays comprehensive information related to the measurement of each sample in a list format, such as the measurement time of the sample.

[0161] The dashboard screen 63 displays the measurement status of urinary formed element concentration for each sample in a panel format. A sample panel 5 is associated with each sample, and the sample panel 5 displays, for example, the sample ID of the sample associated with the sample panel 5. The dashboard screen 63 is provided with display areas for, for example, ordered, unapproved, under review, awaiting approval, awaiting microscopic examination, and requiring confirmation, and the display area where the sample panel 5 is displayed shows the measurement status of urinary formed element concentration in the sample associated with the sample panel 5. In some embodiments, the "unapproved" element is the first status on the operation screen, the "under review" element is the second status on the operation screen, and the "awaiting approval" element is the third status on the operation screen.

[0162] Atlas screen 64 displays atlas image 4, which is a standard component image for each formed element. In other words, atlas image 4 is a sample image for each formed element.

[0163] Additionally, the operation bar 7 located at the bottom of the status screen 61, worklist screen 62, dashboard screen 63, and atlas screen 64 displays various buttons corresponding to each screen.

[0164] At step S50 in Figure 6, since the urine formed element concentration has not yet been approved, the control unit 10A sets the measurement status of the urine formed element concentration to Not Approved. Therefore, the output unit 11F displays a dashboard screen 63 on the display unit 16, showing the sample panel 5 corresponding to the sample to be measured in the approval waiting area.

[0165] In step S60, the reception unit 11G determines whether or not it has received a user selection for any of the sample panels 5 displayed on the dashboard screen 63. If no selection for a sample panel 5 has been received (negative determination), the determination process in step S60 is repeatedly executed until a sample panel 5 is selected, thereby monitoring the user's selection status for the sample panel 5. On the other hand, if a selection for a sample panel 5 has been received (positive determination), the system proceeds to steps S80 and S70.

[0166] In step S70, the control unit 10A determines whether the urine qualitative test result for the sample associated with the selected sample panel 5 is stored in the server 35. Specifically, the control unit 10A determines whether the urine qualitative test result associated with the same sample ID as the sample ID associated with the selected sample panel 5 is stored in the server 35. If the urine qualitative test result is stored in the server 35 (positive determination), the process proceeds to step S80. For the sake of explanation, the sample associated with the selected sample panel 5 is referred to as the "selected sample".

[0167] In step S80, the control unit 10A obtains the urine qualitative test result of the selected sample from the server 35 and proceeds to step S90.

[0168] In the judgment process of step S70, if it is determined that the urine qualitative test result for the selected sample is not stored in the server 35 (in the case of a negative judgment), the process of step S80 is not executed and the process proceeds to step S90.

[0169] In step S90, the output unit 11F displays an approval screen 65 on the display unit 16 for approving the concentration of formed elements in the urine of the selected sample.

[0170] If the urine qualitative test results for the selected sample have been obtained through the processing in step S80, the output unit 11F displays the concentration of formed elements in the urine of the selected sample and the urine qualitative test results for the selected sample on the approval screen 65. On the other hand, if the urine qualitative test results for the selected sample have not been stored in the server 35, the output unit 11F outputs only the concentration of formed elements in the urine of the selected sample on the approval screen 65.

[0171] Figure 12 shows an example of an approval screen 65 that displays the concentration of formed elements in urine and the results of a qualitative urine test. In the approval screen 65 of Figure 12, the table on the left displays the results of the qualitative urine test, and the table on the right displays the concentration of formed elements in urine. Note that the approval screen 65 is a pop-up screen that is overlaid on the dashboard screen 63.

[0172] Furthermore, the approval screen 65 displays a selection button 6 consisting of an approval button 6A, a review button 6B, a display button 6C, and an exit button 6D.

[0173] The approval button 6A is used to approve the urine formed element concentration displayed on the approval screen 65, that is, the measurement result of the urine formed element concentration in the selected sample. By approving the urine formed element concentration, the measurement result of the urine formed element concentration in the selected sample is finalized.

[0174] Review button 6B is used to recalculate the urinary formed element concentration in the selected sample. Users select review button 6B if the urinary formed element concentration displayed on approval screen 65 differs from the trend of urinary formed element concentration estimated from the results of qualitative urine tests, etc., or if they want to calculate the urinary formed element concentration in more detail.

[0175] Display button 6C is used to display image 3 of the formed elements of the selected sample, which was used to calculate the concentration of formed elements in the urine. The user selects display button 6C when they want to check the formed elements contained in the selected sample.

[0176] The Exit button 6D is used to close the approval screen 65 and display the dashboard screen 63.

[0177] In step S100 of Figure 6, the reception unit 11G determines whether or not it has received instructions from the user based on the selection of a selection button 6 via the operation unit 17. If no instructions have been received from the user (negative determination), the determination process in step S100 is repeatedly executed until any selection button 6 is selected, thereby monitoring the selection status of the selection buttons 6 by the user. On the other hand, if instructions have been received from the user (positive determination), the reception unit 11G notifies the control unit 10A of the received instructions and proceeds to step S110.

[0178] In step S110, the control unit 10A determines whether or not it has received a review instruction, which is notified by the selection of the review button 6B. If the review instruction has not been received (negative determination), the process proceeds to step S120.

[0179] In step S120, the control unit 10A determines whether or not it has received the instruction to display the formed element image 3, which is notified by the selection of the display button 6C. If the instruction to display the formed element image 3 has not been received (negative determination), the process proceeds to step S130.

[0180] In step S130, the control unit 10A determines whether or not it has received the approval instruction notified by the selection of the approval button 6A. If the approval instruction has not been received (negative determination), the user has selected the exit button 6D. When the exit button 6D is selected, an instruction to end the display is notified. Therefore, the output unit 11F closes the approval screen 65 in accordance with the instruction from the control unit 10A and proceeds to step S50. As a result of the processing in step S50, the dashboard screen 63 is displayed on the display unit 16, and the measurement status of the urinary formed element concentration for each sample is displayed.

[0181] On the other hand, if the judgment process in step S130 determines that an approval instruction has been received (in the case of a positive judgment), the process proceeds to step S140.

[0182] In this case, the user has approved the measurement result of the urinary formed element concentration in the selected sample. Therefore, in step S140, the control unit 10A transmits the measurement result of the urinary formed element concentration, associated with the sample ID of the selected sample, to the server 35 via the transmission unit 11D. As a result, the urinary formed element concentration of the sample measured by the urinary formed element analyzer 70 is registered in the server 35, and the measurement process shown in Figure 6 is terminated. Following the registration of the urinary formed element concentration to the server 35, the output unit 11F deletes the sample panel 5 associated with the sample for which the urinary formed element concentration has been approved from the dashboard screen 63.

[0183] On the other hand, if the determination process in step S120 determines that the instruction to display the formed element image 3 has been received (in the case of a positive determination), the process proceeds to step S150.

[0184] In this case, the user requests confirmation of the shape and size of formed elements contained in the selected sample. Therefore, in step S150, the output unit 11F displays the formed elements display screen 66 on the display unit 16.

[0185] Figure 13 shows an example of the formed element display screen 66. The formed element display screen 66 displays formed element images 3 of the formed elements contained in the selected sample, categorized by type of formed element. Region 60A of the formed element display screen 66 displays the formed element images 3 of the formed elements contained in the selected sample, and region 60B of the formed element display screen 66 displays atlas images 4 of the same type as the formed element images 3 displayed in region 60A.

[0186] Furthermore, the formed element display screen 66 includes a first item button group 52 and a second item button group 53. The first item button group 52 consists of buttons for each type of formed element contained in the selected sample. The second item button group 53 consists of buttons for each of all types of formed elements that can be classified by the formed element classification device 10.

[0187] The output unit 11F displays the formed element image 3 of the type of formed element associated with the button selected by the user from the first item button group 52 in area 60A. On the other hand, if any button from the second item button group 53 is selected, the output unit 11F displays a reclassification work screen (not shown) on the display unit 16. The reclassification work screen provides the user with an interface to reclassify the formed element image 3 selected from the formed element images 3 displayed in area 60A of the formed element display screen 66 into the type of formed element corresponding to any button selected from the second item button group 53.

[0188] The output unit 11F may also display the urine correction test result of the selected sample in area 66A of the formed sediment display screen 66. Furthermore, if the urine sediment measurement result of the selected sample is stored in the server 35, the control unit 10A may acquire the urine sediment measurement result of the selected sample from the server 35, and the output unit 11F may display the urine sediment measurement result acquired by the control unit 10A together with the urine qualitative test result in area 66A.

[0189] In step S160, the reception unit 11G determines whether or not it has received a signal to end the display on the formed element display screen 66. If it has not received a signal to end the display (negative determination), it monitors for the presence or absence of a signal to end the display by repeatedly executing the determination process in step S160 until it receives a signal to end the display. On the other hand, if it has received a signal to end the display (positive determination), it proceeds to step S50. As a result of the processing in step S50, the dashboard screen 63 is displayed on the display unit 16, and the measurement status of the formed element concentration in urine for each sample is displayed.

[0190] On the other hand, if the judgment process in step S110 determines that a review instruction has been received (in the case of a positive judgment), the process proceeds to step S180.

[0191] In this case, the user requests a recalculation of the urinary formed element concentration in the selected sample. Therefore, in step S180, the control unit 10A transmits the formed element image 3 obtained from the selected sample, along with the sample ID, to the remote processing unit 20, which provides a classification service for the formed element image 3, via the transmission unit 11D. The control unit 10A also transmits information other than the sample ID and formed element image 3 to the remote processing unit 20 in response to instructions from the user. For example, the control unit 10A transmits to the remote processing unit 20 the sample ID and formed element image 3 of the selected sample, a classification list (see Table 2) associating the type of formed element with each formed element image 3 in the selected sample, and the measurement results of the urinary formed element concentration for each type of formed element contained in the selected sample.

[0192] The formed element images 3 transmitted by the control unit 10A to the remote processing unit 20 are preferably all formed element images 3 obtained from the selected sample, but may also be a group of formed element images 3 consisting of some (or all) of the obtained formed element images 3. The user can select the formed element images 3 to be transmitted to the remote processing unit 20.

[0193] Furthermore, if the urine qualitative test results for the selected sample have been obtained from the server 35, the control unit 10A may also transmit the urine qualitative test results for the selected sample to the remote processing unit 20.

[0194] This completes the request to the remote processing unit 20 to reclassify the formed element image 3. This process of sending the formed element image 3 to the remote processing unit 20 and having the remote processing unit 20 reclassify the formed element image 3 is called "review".

[0195] Since the remote processing unit 20 is being asked to reclassify the formed element image 3 obtained from the selected sample, in step S190, the control unit 10A sets the measurement status of the formed element concentration in the urine of the selected sample to "under review". The measurement status of the formed element concentration in the urine of each sample is managed by a measurement status list. The measurement status list is, for example, a list stored in the storage unit 15. An example of a measurement status list is shown in Table 4.

[0196] [Table 4]

[0197] In the example measurement status list shown in Table 4, the measurement status of the formed element concentration in the urine of 16 samples, each represented by sample IDs "#A0001" to "#A0016", is set.

[0198] Upon receiving the setting for the measurement status of the formed element concentration in the urine sample, the output unit 11F displays the dashboard screen 63 shown in Figure 10 on the display unit 16, and displays the sample panel 5 associated with the selected sample in the display area corresponding to the measurement status of the formed element concentration in the urine sample. In this case, since the measurement status of the formed element concentration in the urine sample is "under review," the sample panel 5 associated with the selected sample is displayed in the "under review" area of ​​the dashboard screen 63. With this, the measurement process shown in Figure 6 is completed.

[0199] On the other hand, if the judgment item is determined to satisfy the review conditions in the judgment process of step S40 in Figure 6 (i.e., a positive judgment), the process proceeds to step S170.

[0200] In this case, the control unit 10A has determined that it is necessary to recalculate the concentration of formed elements in the urine.

[0201] Therefore, in step S170, the control unit 10A determines whether or not automatic transmission to the remote processing unit 20 has been set. If automatic transmission has not been set (in the case of a negative determination), the control unit 10A cannot send the formed element image 3 obtained from the sample to the remote processing unit 20 and request reclassification of the formed element image 3 without the user's permission, so the process proceeds to step S50. That is, the control unit 10A displays the dashboard screen 63 on the display unit 16 and leaves the decision of whether or not recalculation of the formed element concentration in the urine is necessary to the user. In this case, the control unit 10A sets the measurement status of the formed element concentration in the sample urine of the sample to "awaiting approval", so the sample panel 5 of the sample to be measured is displayed in the awaiting approval area of ​​the dashboard screen 63.

[0202] On the other hand, if automatic transmission is enabled (in the case of a positive judgment), the process proceeds to step S180. As already explained, in step S180, the control unit 10A transmits the formed element image 3 obtained from the sample to be measured to the remote processing unit 20 via the transmission unit 11D. As a result, if the review conditions in the judgment item are met, the formed element image 3 of the sample is automatically transmitted from the formed element classifier 10 to the remote processing unit 20 without the user having to give a review instruction to the formed element classifier 10. Whether or not automatic transmission is enabled can be set by the user.

[0203] <Reclassification of formed element image 3 by remote processing device 20> Next, the operation of the remote processing unit 20 will be explained. Figure 14 is a flowchart showing an example of the reclassification process performed by the remote processing unit 20 when it receives a formed element image 3 of a sample, represented by a sample ID, from the formed element classifier 10. The CPU 21 of the remote processing unit 20 reads the data management program 25A stored in the storage unit 25 and executes the reclassification process. Hereafter, an example will be described in which the remote processing unit 20 receives a formed element image 3 of a sample, represented by a sample ID, and a classification list, i.e., the classification result by the classification unit 11B, from the formed element classifier 10.

[0204] For example, in the remote processing unit 20, a specialist technician who examines the formed element image 3 and determines what type of formed element is contained in the formed element image 3 operates the remote processing unit 20 to reclassify the formed element image 3.

[0205] First, in step S200, the display control unit 21C displays the formed element display screen 66 shown in Figure 13 on the display unit 26. The formed element display screen 66 displays the formed element images 3 received from the formed element classification device 10, based on the classification of the classification list also received from the formed element classification device 10.

[0206] In step S210, the control unit 20A determines, via the operation unit 27, whether or not any of the buttons in the first item button group 52 on the formed substance display screen 66 have been selected. If none of the buttons in the first item button group 52 have been selected (negative determination), the determination process in step S210 is repeatedly executed until any of the buttons in the first item button group 52 have been selected, thereby monitoring the selection status of the first item button group 52 by the medical technologist. On the other hand, if any of the buttons in the first item button group 52 have been selected (positive determination), the process proceeds to step S220.

[0207] In step S220, the display control unit 21C displays the formed element image 3 of the type of formed element associated with the selected button in area 60A of the formed element display screen 66.

[0208] In step S230, the control unit 20A determines, via the operation unit 27, whether or not any of the buttons in the second item button group 53 on the formed substance display screen 66 have been selected. If none of the buttons in the second item button group 53 have been selected (negative determination), the determination process in step S230 is repeatedly executed until any of the buttons in the second item button group 53 have been selected, thereby monitoring the selection status of the second item button group 53 by the medical technologist. On the other hand, if any of the buttons in the second item button group 53 have been selected (positive determination), the process proceeds to step S240.

[0209] In step S240, the display control unit 21C displays the reclassification work screen on the display unit 26. The medical technologist reclassifies the formed element image 3, which has been found to be incorrectly classified, to the specified type of formed element through the reclassification work screen. If the results of the urine qualitative test of the specimen have been transmitted from the formed element classification device 10, the medical technologist may reclassify the formed element image 3 by referring to the results of the urine qualitative test.

[0210] In step S250, the control unit 20A determines whether or not it has received any instructions from the medical technologist. If no instructions have been received (negative determination), the control unit waits until it receives instructions from the medical technologist by repeatedly executing the determination process in step S250. On the other hand, if instructions have been received (positive determination), the control unit proceeds to step S260.

[0211] From this point onward, the control unit 20A grasps the instructions from the medical technologist. First, in step S260, the control unit 20A determines whether or not it has received a reclassification instruction from the medical technologist. If a reclassification instruction has been received (positive determination), the system proceeds to step S270.

[0212] In step S270, the second classification unit 21B, which is an example of a reclassification unit, reclassifies the types of formed elements contained in the formed element image 3 selected by the medical technologist into the types of formed elements specified by the medical technologist, and proceeds to step S280. Specifically, the control unit 20A updates the classification column of the classification list received from the remote processing unit 20. Table 5 shows an example of a classification list in which the formed element image 3 represented by the formed element image ID "#B00001" has been reclassified from red blood cells to yeast, relative to the classification list shown in Table 2. The updated classification list is an example of the reclassification result of the formed element image 3 by the second classification unit 21B.

[0213] [Table 5]

[0214] If the control unit 20A has not received a classification list from the formed element classification device 10, it should create a classification list that associates the types of formed elements contained in the formed element image 3 selected by the medical technologist with the formed element image ID.

[0215] On the other hand, if no reclassification instruction is received in the determination process of step S260 (i.e., a negative determination is made), the process of step S270 is not executed, and the process proceeds to step S280.

[0216] In step S280, the control unit 20A determines whether or not it has received a microscopic examination instruction from the medical technologist. A microscopic examination instruction is an instruction to examine the sample in detail using a microscopic method, for example, in which a person visually inspects the type and number of formed elements in the sample using a physical microscope. If a microscopic examination instruction is received (positive determination), the process proceeds to step S290.

[0217] In step S290, the control unit 20A adds a microscopic examination status to the sample ID, indicating that a microscopic examination instruction has been received from a medical technologist, and proceeds to step S300. On the other hand, if no microscopic examination instruction has been received in the judgment process of step S280 (in the case of a negative judgment), the process of step S290 is not executed, and the process proceeds to step S300.

[0218] In step S300, the control unit 20A sends a classification list reflecting the sample ID and reclassification results back to the formed segregator 10 via the reply unit 21D. If microscopic examination is instructed, the microscopic examination status is added to the sample ID sent back to the formed segregator 10. This completes the reclassification process shown in Figure 14.

[0219] The above describes an example in which the formed element image 3 is reclassified according to the reclassification instructions from the medical technologist. However, the remote processing unit 20 may reclassify the formed element image 3 even if the medical technologist does not specify the reclassification destination. Specifically, the second classification unit 21B may use the second trained model 25B, which is pre-stored in the memory unit 25, to classify the formed element image 3 specified by the medical technologist according to the type of formed element.

[0220] As already explained, the second pre-trained model 25B is a classification model with higher classification performance than the first pre-trained model 15B. Therefore, the second pre-trained model 25B can classify the formed element image 3 with greater accuracy than the first pre-trained model 15B, and thus can correct the errors in the classification of the formed element image 3 made by the first pre-trained model 15B.

[0221] Furthermore, when classifying formed element images 3 using the second trained model 25B, the control unit 20A may reclassify all formed element images 3 received from the formed element classification device 10 according to the type of formed element, even if the formed element images 3 to be reclassified are not specified by the medical technologist.

[0222] <Remeasurement of urine formed element concentration using formed element classification device 10> Next, we will explain the operation of the formed element classifier 10, which receives the sample ID and a classification list reflecting the reclassification results of the formed element image 3 from the remote processing device 20 (reclassification device).

[0223] Figure 15 is a flowchart showing an example of the remeasurement process performed by the formed element classifier 10 when it receives a classification list from the remote processing unit 20 that reflects the sample ID and the reclassification results of the formed element image 3. The CPU 11 of the formed element classifier 10 reads the information processing program 15A stored in the storage unit 15 and executes the remeasurement process.

[0224] The difference between the flowchart shown in Figure 15 and the measurement process flowchart shown in Figure 6 is that steps S10 to S40 and step S170 have been removed, and step S45 has been added. Also, the process in step S50 has been replaced by the process in step S50A. Since the other processes are the same as in Figure 6, the following explanation will focus on the processes in steps S45 and S50A to describe the re-measurement process in the formed classification device 10.

[0225] If the remote processing device 20 receives a classification list that reflects the sample ID and the reclassification result of the formed element image 3, step S45 is executed.

[0226] In step S45, the calculation unit 11C recalculates the number of formed element images 3 for each formed element by referring to the classification list received from the remote processing device 20, and recalculates the urine formed element concentration for each formed element by substituting the recalculated number into the concentration calculation formula shown in Table 3.

[0227] In step S50A, the control unit 10A refers to the sample ID received from the remote processing unit. If the sample ID has a microscopic examination status attached, it sets the measurement status of the formed element concentration in the sample urine of the sample represented by the sample ID to "Waiting for microscopic examination" in the measurement status list shown in Table 4. If the sample ID does not have a microscopic examination status attached, the control unit 10A sets the measurement status of the formed element concentration in the sample urine of the sample represented by the sample ID to "Waiting for approval" in the measurement status list (Table 4).

[0228] Furthermore, the output unit 11F displays a dashboard screen 63 on the display unit 16, showing the sample panel 5 associated with each sample in a display area that matches the measurement status of the formed element concentration in the sample urine set in the measurement status list. As a result, the display position of the sample panel 5 on the dashboard screen 63 is updated to match the latest measurement status of the formed element concentration in the sample urine.

[0229] Subsequently, the user selects one of the sample panels 5 from the updated dashboard screen 63, and the processes from step S60 onwards, as previously described, are executed. Specifically, for the sample corresponding to the selected sample panel 5, the following are repeatedly performed: approval of the measurement result of the urinary formed element concentration, review of the measurement result of the urinary formed element concentration, display of the formed element image 3, etc. With this, the re-measurement process shown in Figure 15 is completed.

[0230] <Review Criteria> Up to this point, we have described the flow of measuring the concentration of formed elements in urine using the formed elements classification system 100. In the judgment process of step S40 in Figure 6, the control unit 10A of the formed elements classification device 10 determined whether the predetermined judgment items met the review conditions. Hereafter, we will explain in detail the judgment items and review conditions that the control unit 10A refers to in the judgment process of step S40 in Figure 6.

[0231] When the user selects the setting button 7A on the operation bar 7 of the status screen 61 shown in Figure 8, the output unit 11F displays the setting screen 55 on the display unit 16.

[0232] Figure 16 shows an example of the settings screen 55. The settings screen 55 is a screen for setting the operation of various functions provided in the urinary formed sediment analyzer 70. For example, the settings screen 55 includes an operator account button for registering and deleting users in the formed sediment classifier 10. When the user selects the automatic review request determination button 55A on the settings screen 55, the output unit 11F displays the automatic review request determination screen 56 on the display unit 16.

[0233] Figure 17 shows an example of the automatic review request determination screen 56 of the operation screen. This screen is part of the condition setting screen of the operation screen, or may be part of it in some embodiments. The automatic review request determination screen 56 is a screen for selecting the type of determination item to refer to when the control unit 10A makes an automatic review request. An automatic review request is a review request made by the formed classification device 10 when the determination item satisfies the review conditions, regardless of the user's intention.

[0234] As shown in Figure 17, the types of judgment items include flags, formed element items, and qualitative test items.

[0235] A flag is a monitored event that occurs during the inspection process of a sample. The occurrence status of an event is represented using a flag indicating whether it has occurred or not. In some embodiments, this flag may be a predefined condition, and based on this condition, it is determined whether or not to reclassify the material component image. For this reason, the monitored event is called a "flag," and the occurrence of a monitored event is called "flag occurrence." In this case, in some embodiments, it is determined that the formed component image should be reclassified.

[0236] Formed element items refer to the various types of formed elements that can be analyzed by the urinary formed element analyzer 70.

[0237] Qualitative test items refer to the various qualitative test items that can be analyzed by the urine qualitative analyzer 30.

[0238] For each type of judgment item, there are selection lists 56A, 56B, and 56C that set whether or not to include the corresponding judgment item in the review criteria. Selection lists 56A, 56B, and 56C include a "Judge" option to include the corresponding judgment item in the review criteria and a "Do not judge" option to exclude it from the review criteria. The user sets the options for selection lists 56A, 56B, and 56C via the operation unit 17. In the example of the automatic review request judgment screen 56 shown in Figure 17, all of the judgment items—flag, formed element item, and qualitative test item—are set to be included in the review criteria. In addition, at least one of the flag, material component formed element item, and qualitative test item becomes a predefined condition for automatically determining the reclassification of the formed element image.

[0239] Furthermore, the automatic review request determination screen 56 has setting buttons for each type of determination item to set the review conditions for the determination items. The flag setting button 56D is for setting the review conditions for the flag. The threshold setting button 56E is for setting the review conditions for formed component items. The threshold setting button 56F is for setting the review conditions for qualitative test items.

[0240] The user creates review conditions through the review condition setting screen 57, which is displayed on the display unit 16 when the user selects a setting button corresponding to the judgment item used to determine the review conditions. After creating the review conditions, the user selects the apply button 56G and then the save button 56H. The application button 56G is selected, and the control unit 10A updates the review conditions. The save button 56H is selected, and the control unit 10A stores the updated review conditions in the storage unit 15.

[0241] If the user selects the exit button 56I, the output unit 11F closes the automatic review request determination screen 56 and displays the settings screen 55 on the display unit 16.

[0242] Figure 18 shows an example of the flag condition setting screen 57A, which is the review condition setting screen 57 for flags. As shown in Figure 18, the flag condition setting screen 57A displays a list of error items that may occur in, for example, the urine qualitative analyzer 30. If the list of error items cannot be fully displayed on the flag condition setting screen 57A, the user can move the scroll bar 57X up and down to scroll the display of the flag condition setting screen 57A, and all error items will be displayed on the flag condition setting screen 57A.

[0243] Each error item is associated with a corresponding field 57D. The user can set field 57D to either "Enabled" or "Disabled". When field 57D is set to "Enabled," a review condition is created that is met when the corresponding error item occurs. When field 57D is set to "Disabled," no review condition is created for the corresponding error item. In other words, the user sets the criteria for review condition evaluation by editing field 57D.

[0244] For example, if the validity column 57D in the error item "Qualitative item abnormality: abnormal color development" is set to "valid," a review condition is created that states the condition is met if the urine qualitative test results of the sample include a test result indicating abnormal color development.

[0245] If the Confirm button 57Y is selected, the control unit 10A temporarily stores the review conditions created on the flag condition setting screen 57A in the RAM 13. If the Exit button 57Z is selected, the output unit 11F closes the flag condition setting screen 57A and displays the automatic review request determination screen 56 on the display unit 16.

[0246] In step S40 of Figure 6, the control unit 10A retrieves error information from the server 35, which stores error information recorded by the urine qualitative analyzer 30 when the urine qualitative analyzer 30 performs qualitative analysis of a sample, and which is associated with the sample ID of the sample. The error information retrieves the same sample ID as the sample ID obtained in step S10. If the retrieved error information includes information that at least one error item has occurred for which the "enabled" column 57D in the flag condition setting screen 57A has been set to "enabled", the control unit 10A determines that the review conditions are met.

[0247] The output unit 11F may also display a list of error items that may occur in each device included in the formed sediment classification system 100 on the review condition setting screen 57. Specifically, the output unit 11F displays a list of error items that may occur in each of the formed sediment classification device 10, the urine qualitative analyzer 30, the server 35, and the urine formed sediment analyzer 70 on the flag condition setting screen 57A. In this case, the control unit 10A creates review conditions for flags for at least one of the formed sediment classification device 10, the urine qualitative analyzer 30, the server 35, and the urine formed sediment analyzer 70 based on the user's settings on the flag condition setting screen 57A.

[0248] In step S40 of Figure 6, the control unit 10A retrieves error information from the server 35, which stores error information for each of the formed sediment classifier 10, the urine qualitative analyzer 30, the server 35, and the urine formed sediment analyzer 70, linked to the sample ID, for each device associated with the same sample ID as the sample ID obtained in step S10. If the retrieved error information includes information that at least one error item has occurred for which the "enabled" column 57D in the flag condition setting screen 57A has been set to "enabled", the control unit 10A determines that the review conditions are met.

[0249] If an error occurs in the review condition setting screen 57 where the valid field 57D is set to "valid", it is recommended to recalculate the calculated urinary formed element concentration. Therefore, if the control unit 10A determines that the judgment item in at least one of the formed element classification device 10, the urine qualitative analyzer 30, the server 35, and the urinary formed element analyzer 70 satisfies the review conditions, it sends the formed element image 3 of the sample to the remote processing device 20 to request a review.

[0250] It goes without saying that the control unit 10A may also directly acquire error information generated by the formed sedimentation device 10, the urine qualitative analyzer 30, the server 35, and the urine formed sediment analyzer 70.

[0251] On the other hand, Figure 19 shows an example of the formed element condition setting screen 57B, which is the review condition setting screen 57 for formed element items.

[0252] As shown in Figure 19, the formed component condition setting screen 57B displays the Valid field 57D, Item field 57E, Threshold field 57F, Rank field 57G, and Display value field 57H.

[0253] Item column 57E displays, for example, all types of formed elements that can be analyzed by the urine formed element analyzer 70.

[0254] In the threshold field 57F, the user sets the threshold for the number concentration of each type of formed element corresponding to the row. The threshold field 57F is editable by the user, and the number concentration threshold is set for each type of formed element. The number concentration threshold also includes comparison information with the threshold. Comparison information with the threshold indicates the relationship between the number concentration and the threshold, such as whether the number concentration "matches the threshold," "greater than or equal to the threshold," "less than or equal to the threshold," "less than the threshold," or "greater than the threshold." The set threshold is displayed in the display value field 57H.

[0255] In the rank column 57G, the user sets classification information for the number concentration of the type of formed element corresponding to the row. The rank column 57G is editable by the user, and classification information for the number concentration is set for each type of formed element. Classification information refers to each group when the number concentration is divided into a predetermined number of groups, for example, "Level 1," "Level 2," and "Level 3," from lowest to highest number concentration. For the same type of formed element, the user sets a value in either the threshold column 57F or the rank column 57G.

[0256] For example, if the threshold for RBCs is set to "1.0 μL or more" and the RBC validity column 57D is set to "valid," a review condition is created that satisfies the condition if the number concentration of RBCs in the sample is 1.0 μL or more. Also, for example, if the rank of RBCs is set to "level 1" and the RBC validity column 57D is set to "valid," a review condition is created that satisfies the condition if the number concentration of RBCs in the sample falls within the level 1 range.

[0257] If the Confirm button 57Y is selected, the control unit 10A temporarily stores the review conditions created on the formed element condition setting screen 57B in the RAM 13. If the Exit button 57Z is selected, the output unit 11F closes the formed element condition setting screen 57B and displays the automatic review request determination screen 56 on the display unit 16.

[0258] If the "Valid" field 57D is set to "Invalid," review conditions based on the number concentration of the corresponding formed element type will not be created.

[0259] In step S40 of Figure 6, the control unit 10A refers to the number concentration of each formed element calculated by the calculation unit 11C in step S30. If the number concentration of at least one type of formed element for which the effective column 57D in the formed element condition setting screen 57B is set to "effective" satisfies the conditions set in the threshold column 57F or the rank column 57G, the control unit 10A determines that the review conditions are met.

[0260] On the other hand, Figure 20 shows an example of the qualitative condition setting screen 57C, which is the review condition setting screen 57 for qualitative test items.

[0261] As shown in Figure 20, the qualitative condition setting screen 57C displays the Valid field 57D, the Item field 57J, and the Rank field 57K.

[0262] Item column 57J displays, for example, all qualitative items that can be analyzed by the urine qualitative analyzer 30.

[0263] In the rank column 57K, users can set thresholds and classification information for corresponding qualitative items in the row direction. The rank column 57K is editable by the user, and thresholds or classification information can be set for the corresponding qualitative items.

[0264] For example, if the URO rank is set to "NORMAL" and the URO validity column 57D is set to "Valid," a review condition is created that the condition is met if the URO value in the sample falls within the range associated with "NORMAL." Also, for example, if the creatinine (CRE) threshold is set to "10 mg / dL or higher" and the CRE validity column 57D is set to "Valid," a review condition is created that the condition is met if the CRE value in the sample is 10 mg / dL or higher.

[0265] Furthermore, on the qualitative condition setting screen 57C, the names in the rank column 57K may be replaced with names that allow the user to intuitively understand the settings, such as "hue" and "density," depending on the type of qualitative item.

[0266] If the Confirm button 57Y is selected, the control unit 10A temporarily stores the review conditions created on the qualitative condition setting screen 57C in the RAM 13. If the Exit button 57Z is selected, the output unit 11F closes the qualitative condition setting screen 57C and displays the automatic review request determination screen 56 on the display unit 16.

[0267] If the "Enabled" field 57D is set to "Disabled," review conditions based on the values ​​of the corresponding qualitative items will not be created.

[0268] In step S40 of Figure 6, the control unit 10A refers to the urine qualitative test results associated with the same sample ID as the sample ID obtained in step S10, among the sample IDs stored in the server 35 and the urine qualitative test results associated with the said sample ID. In the qualitative condition setting screen 57C, if the value of at least one qualitative item whose validity column 57D is set to "valid" satisfies the condition set in the rank column 57K, the control unit 10A determines that the review condition is met.

[0269] In this manner, if the review conditions received by the reception unit 11G through the flag condition setting screen 57A, the formed element condition setting screen 57B, and the qualitative condition setting screen 57C are met, the control unit 10A transmits the formed element image 3 of the sample to the remote processing unit 20 and requests the remote processing unit 20 to review it.

[0270] In step S40 of Figure 6, the control unit 10A may proceed to step S170 if at least one of the judgment items satisfies the review conditions, or it may proceed to step S170 only if all of the predetermined judgment items satisfy their respective review conditions. For example, the control unit may proceed to step S170 if both the number concentration of RBCs and DRBCs (deformed red blood cells) in the formed element condition setting screen 57B shown in Figure 19 satisfy the review conditions. The combination of multiple judgment items may be a combination within the same type of judgment item, or a combination spanning different types of judgment items.

[0271] Furthermore, even if review conditions have been created for the judgment items in the flag condition setting screen 57A, the formed substance condition setting screen 57B, and the qualitative condition setting screen 57C, and the judgment items satisfy the review conditions, if the judgment items that satisfy the review conditions are set to "Do not judge" by the selection lists 56A, 56B, and 56C of the automatic review request judgment screen 56, that is, if automatic transmission settings have not been made for the judgment items that satisfy the review conditions, the control unit 10A will not send a review request to the remote processing unit 20. In other words, it will proceed to step S50 without proceeding to step S180. Therefore, the user can disable the judgment target of the review conditions for each type of judgment item simply by setting the selection lists 56A, 56B, and 56C, without having to change the enabled field 57D, which was previously set to "Enabled", back to "Disabled".

[0272] In this embodiment, in step S40, it is determined whether the determination item satisfies the review conditions, and in step S170, it is determined whether automatic transmission to the remote processing unit 20 has been set for the determination item that satisfies the review conditions. In another embodiment, after step S30, it is determined whether there are any determination items for which automatic transmission to the remote processing unit 20 has been set, and if there are determination items for which automatic transmission has been set, it is determined whether the review conditions are satisfied only for the determination items for which automatic transmission has been set, and if the review conditions are satisfied, the process proceeds to step S180. In this case, it is only necessary to check the review conditions for the determination items for which automatic transmission has been set, so the necessity of automatic transmission can be determined efficiently.

[0273] In each of the embodiments described above, the term "processor" refers to a processor in a broad sense, and includes general-purpose processors (e.g., CPU: Central Processing Unit, etc.) and dedicated processors (e.g., GPU: Graphics Processing Unit, ASIC: Application Specific Integrated Circuit, FPGA: Field Programmable Gate Array, programmable logic device, etc.).

[0274] In addition, the operations of the processor in each of the above embodiments may be performed not only by one processor but also by a plurality of physically separated processors cooperating with each other. Also, the order of each operation of the processor is not limited to the order described in each of the above embodiments, and may be changed as appropriate.

[0275] As described above, the component classification device 10 according to the embodiment has been illustrated and described. The embodiment may be in the form of a program for causing a computer to execute the functions of each part included in the component classification device 10. The embodiment may be in the form of a computer-readable non-temporary storage medium storing this program.

[0276] In addition, the configuration of the component classification device 10 described in the above embodiment is an example, and may be changed according to the situation within the scope not departing from the gist. The display of the component image 3 is not limited to the above embodiment, and may be displayed, for example, side by side horizontally. The display position of each button can also be changed as appropriate.

[0277] Also, the flow of the program processing described in the above embodiment is an example, and unnecessary steps may be deleted, new steps may be added, or the processing order may be changed within the scope not departing from the gist.

[0278] Also, in the above embodiment, the case where the processing according to the embodiment is realized by software configuration using a computer by executing a program has been described, but it is not limited to this. The embodiment may be realized, for example, by a hardware configuration or a combination of a hardware configuration and a software configuration.

[0279] The following is an appendix according to this embodiment.

[0280] [[ID=​​The first embodiment of the formed element classification device includes: a first acquisition unit that acquires formed element images representing formed elements contained in a sample; a first classification unit that classifies each of the formed element images in relation to the type of formed element; a calculation unit that calculates the concentration of the formed element in the sample for each of the types based on the number of classified formed element images; and a transmission unit that transmits the classified formed element images to a remote processing device via a network line in order to reclassify the classified formed element images when predefined conditions related to the concentration are met.

[0281] (Second aspect) According to the second embodiment, in the formed element classification device according to the first embodiment, if at least one of the qualitative test result of the sample and the error information of the qualitative analyzer satisfies the predefined conditions, the classified formed element image is transmitted to the network line and the remote processing device (reclassification device) is instructed to perform the reclassification.

[0282] (Third aspect) According to the third aspect, in the formed element classifier according to the second aspect, the determination of whether the predefined conditions are met is made using at least one of the following: the relationship between the concentration of a type of formed element specified by the user and a threshold specified by the user; the relationship between the quality value representing the result of a qualitative test of the sample and a threshold; and the occurrence status of an error item specified by the user from among the error items in the error information.

[0283] (Fourth aspect) According to the fourth aspect, in the formed classification device according to the above-described aspect, the formed image is transmitted to the network line only when the predefined conditions are met and the reclassification by the remote processing device (reclassification device) is permitted in advance.

[0284] (Fifth aspect) According to the fifth embodiment, in the formed element classification device according to the above embodiment, the formed element image and its classification result are transmitted to the remote processing device (reclassification device) via the network line.

[0285] (Sixth aspect) According to the sixth aspect, in the formed sorting device according to the above aspect, the sample is urine.

[0286] (Seventh aspect) According to the seventh aspect, the formed classification system comprises a formed classification device as described above and a remote processing device (reclassification device) configured to communicate with each other via the network line, wherein the remote processing device includes a second classification unit that reclassifies the classified formed image acquired from the formed classification device and a reply unit that returns the reclassification result by the second classification unit to the formed classification device.

[0287] (Eighth aspect) According to the eighth aspect, the formed element classification method includes the steps of: acquiring formed element images representing formed elements contained in a sample; classifying each of the formed element images in relation to the type of formed element; calculating the concentration of the formed element in the sample for each of the types based on the number of classified formed element images; and transmitting the formed element images to a remote processing device via a network line for reclassification if predefined conditions related to the concentration are met.

[0288] (Ninth aspect) According to the ninth aspect, an information processing program that causes a processor to perform processing acquires formed element images representing formed elements contained in a sample, classifies each of the formed element images in relation to the type of formed element, calculates the concentration of the formed element in the sample for each of the types based on the number of classified formed element images, and transmits the information to a remote processing device via a network line to reclassify the formed element images if predefined conditions related to the concentration are met.

[0289] (Tenth aspect)

[0290] According to the tenth aspect, a non-temporary storage medium storing a program executable by a processor includes instructions to cause the processor to: acquire a formed element image representing a formed element contained in a specimen; classify each of the formed element images in association with the type of the formed element; calculate the concentration of the formed element in the specimen for each of the types based on the number of classified formed element images; and transmit, via a network line, the formed element image to a remote processing device to reclassify the formed element image when a predefined condition associated with the concentration is satisfied.

[0291] In the first, seventh to tenth aspects, when determining the necessity of reclassifying the classified formed element images, it has the effect of assisting the user.

[0292] In the second aspect, it has the effect of being able to determine whether reclassification of the formed element image is necessary by using at least one of the qualitative inspection result of the specimen and the error information in the qualitative analysis device.

[0293] In the third aspect, it has the effect that the user can determine the criterion for whether to reclassify according to the situation.

[0294] In the fourth aspect, it has the effect of preventing the formed element image of the specimen from being transmitted to the remote processing device without the permission of the user.

[0295] In the fifth aspect, in the remote processing device, it has the effect that it can be confirmed how the formed element image was classified in the formed element classification device.

[0296] In the sixth aspect, it has the effect of assisting in determining whether reclassification of the component image of the formed element in urine is necessary. [[ID=?]]

[0297] Furthermore, the present invention also extends to the following ranges that combine any aspects disclosed in this embodiment.

[0298] A1. The formed element classification device comprises a first acquisition unit that acquires formed element images representing formed elements contained in a sample, and a first classification unit that classifies each of the formed element images in relation to the type of formed element, and the classified formed element images are automatically determined to be further reclassified with higher accuracy based on predefined conditions.

[0299] In the formed-matter sorting apparatus of A2.A1, the predefined conditions are associated with the concentration of the formed-matter in each of the categories.

[0300] In the formed element classification device of A3.A2, the concentration is determined for each of the types based on the number of classified formed element images.

[0301] A4. In any of the formed component classifiers from A1 to A3, the predefined conditions are associated with the classification accuracy of the formed component.

[0302] In the formed element classification device of A5.A4, the classification accuracy is specialized for classifying the formed element images into a specific type.

[0303] In any of the formed classification devices from A1 to A5, the predefined conditions are associated with the quality values ​​of the sample.

[0304] In the formed semantic classifier of A7.A6, the quality value is obtained by measuring the quality of the sample.

[0305] In the formed sorting apparatus of A8.A6 or A7, the quality is measured and the quality value is obtained.

[0306] In the formed sorting apparatus of A9.A7 or A8, the predefined conditions are associated with error information representing anomalies related to the measurement of the quality.

[0307] In the formed classification devices A10.A1 to A9, the predefined conditions can be set by the user.

[0308] In any of the formed element classification devices from A1 to A10, the reclassification is performed by transmitting the classified formed element images to a remote processing device.

[0309] In the formed element classification device of A12.A11, a group of classified formed element images is determined, which serves as the basis for performing the reclassification.

[0310] In the formed element classification device of A13.A12, the group of classified formed element images determined are transmitted to the remote processing device.

[0311] A14. In any of the formed element classification devices from A11 to A13, the classification information associated with the classified formed element image is additionally transmitted to the remote processing device.

[0312] In any of the formed classification devices A15.A1 to A14, it is determined whether or not the predefined conditions are met.

[0313] In the formed-matter classifier of A16.A15, the determination is based on at least one of the following: the relationship between the concentration of the formed-matter in the type specified by the user and the threshold; the relationship between the quality value representing the qualitative test result of the sample and the threshold; and the occurrence status of error information specified by the user.

[0314] A17. In any of the formed classification devices from A1 to A16, the sample is urine.

[0315] A18. The reclassification device comprises a second acquisition unit that acquires the formed element images that have been classified by any of the formed element classification devices A1 to A3, and a second classification unit that reclassifies the formed element images.

[0316] A19. A formed subdivision system comprising a formed subdivision device according to A1 to A17 and a remote processing device configured to communicate with each other via a network line, wherein the remote processing device includes a second acquisition unit that acquires formed subdivision images classified by the formed subdivision device, and a second classification unit that reclassifies the classified formed subdivision images based on user input on an operation screen.

[0317] In the formed element classification system of A20.A19, the user input includes at least one of the following: the selection of the formed element image, the selection of the type, and the selection of the reclassification method.

[0318] In the formed classification system of A21.A19 or A20, the result of the reclassification is transmitted to the formed classification device.

[0319] A22. A method for classifying formed elements, comprising the steps of: acquiring formed element images representing formed elements contained in a sample; classifying each of the formed element images in relation to the type of formed element; and automatically determining, based on predefined conditions, whether or not to further reclassify the classified formed element images with higher accuracy.

[0320] A23. A formed classification program includes instructions that cause the computer's processor to execute the formed classification method of A22.

[0321] Furthermore, the present invention also extends to the following scope, combining any aspects disclosed in this embodiment.

[0322] B1. The formed element classification device comprises: a first acquisition unit that acquires formed element images representing formed elements contained in a sample; a first classification unit that classifies each of the formed element images in relation to the type of formed element; and an output unit that outputs a first status and a second status to be displayed on an operation screen, wherein the first status indicates that the results obtained from the classified formed element images are unapproved, and the second status indicates that the classification of the formed element images is under review, and the classified formed element images are associated with the first status or the second status based on the corresponding classification.

[0323] In the formed element classification device of B2.B1, the operation screen includes an approval screen in which, when the display of the first status is operated, information of the classified formed element image is displayed.

[0324] In the formed element classification device of B3.B2, the information refers to the concentration of the formed element obtained from the classified formed element image and at least one of the results of a qualitative test.

[0325] In the B4.B2 or B3 formed classification device, the approval screen can accept user input, and based on the user input, the formed element image is transmitted via a network line and reclassified by a remote processing device.

[0326] In any of the formed element classification devices from B5.B2 to B4, the concentration is determined for each of the types of formed elements contained in the sample, based on the number of classified formed element images.

[0327] In any of the formed classification devices from B6.B1 to B5, the review in the second situation refers to the reclassification.

[0328] In any of the formed element classification devices from B7.B1 to B6, the operation screen displays a third status indicating that the reclassified formed element image is awaiting approval.

[0329] In the formed element classification device of B8.B7, the user can perform an operation to display the third status, thereby obtaining approval for the reclassified formed element image.

[0330] In any of the formed classification devices described in B9.B1 to B8, the reclassification is performed automatically based on predefined conditions, and the reclassification is performed with higher accuracy than the classification in the classification unit.

[0331] In the formed classification device of B10.B9, the predefined conditions can be set by the user.

[0332] In the formed sorting device of B11.B10, the operation screen includes a condition setting screen configured to set the predefined conditions based on user operations.

[0333] In the formed-substance classification device of B12.B11, the condition setting screen includes at least one of the conditions for the formed-substances and the qualitative conditions.

[0334] B13. The device comprises a formed element classification device according to one of B1 to B12 and a remote processing device configured to communicate with each other via a network line, wherein the remote processing device includes a second acquisition unit that acquires the formed element images classified by the formed element classification device, and a second classification unit that further reclassifies the classified formed element images with higher accuracy based on user input on the operation screen.

[0335] B14. A method for classifying formed elements, comprising the steps of: acquiring a formed element image representing a formed element contained in a sample; classifying each of the formed element images in relation to the type of formed element; and outputting a first status and a second status to be displayed on an operation screen, wherein the first status indicates that the results obtained from the classified formed element image are unapproved, and the second status indicates that the classification of the formed element image is under review, and the classified formed element image is associated with the first status or the second status based on the corresponding classification.

[0336] B15. A program for classifying formed substances includes instructions that cause the computer's processor to execute the classifying formed substances method of B14. [Industrial applicability]

[0337] According to the present invention, it is possible to provide a formed classification device, a formed classification system, a formed classification method, and a formed classification program.

Claims

1. A first acquisition unit that acquires an image of formed elements representing formed elements contained in the sample, The system comprises a first classification unit that classifies each of the formed element images in relation to the type of formed element, A formed element classification device that automatically determines whether or not to further reclassify the classified formed element images with higher accuracy based on predefined conditions.

2. In the formed sorting device according to claim 1, The predefined conditions are associated with the concentration of the formed elements in each of the types of formed elements in the formed elements classification device.

3. In the formed sorting device according to claim 2, The concentration is determined for each of the types based on the number of classified formed element images in the formed element classification device.

4. In the formed sorting device according to any one of claims 1 to 3, The aforementioned predefined conditions are associated with the classification accuracy of the formed elements in the formed elements classification device.

5. In the formed sorting device according to claim 4, The classification accuracy is a formed element classifier specialized for classifying the formed element images into specific types.

6. In the formed sorting device according to any one of claims 1 to 3, The aforementioned predefined conditions are associated with the quality value of the sample in the formed classification device.

7. In the formed sorting device according to claim 6, The aforementioned quality value is obtained by measuring the quality of the sample, and is a formed semantic classifier.

8. In the formed sorting device according to claim 7, A formed sorting device configured to measure the quality and obtain the quality value.

9. In the formed sorting device according to claim 7, The predefined conditions are associated with error information representing anomalies related to the measurement of the quality of the formed classifier.

10. In the formed sorting device according to any one of claims 1 to 3, The aforementioned predefined conditions are a form-sorting device that can be set by the user.

11. In the formed sorting device according to any one of claims 1 to 3, The aforementioned reclassification is performed by a formed element classification device that transmits the classified formed element images to a remote processing device.

12. In the formed sorting device according to claim 11, A formed element classification device that determines a group of classified formed element images, which serves as the basis for performing the aforementioned reclassification.

13. In the formed sorting device according to claim 12, The determined group of classified formed element images is transmitted to the remote processing device by a formed element classifier.

14. In the formed sorting device according to claim 11, A formed element classification device to which the classification information associated with the classified formed element images is additionally transmitted to the remote processing device.

15. In the formed sorting device according to any one of claims 1 to 3, A formed classification device that determines whether or not the aforementioned predefined conditions are met.

16. In the formed sorting device according to claim 15, The determination is made by a formed element classifier based on at least one of the following: the relationship between the concentration of the formed element in the type specified by the user and a threshold; the relationship between the quality value representing the qualitative test result of the sample and a threshold; and the occurrence status of error information specified by the user.

17. In the formed sorting device according to any one of claims 1 to 3, A formed segregator in which the aforementioned sample is urine.

18. A reclassification device comprising: a second acquisition unit for acquiring the classified formed element images in a formed element classification device according to any one of claims 1 to 3; and a second classification unit for reclassifying the classified formed element images.

19. A formed sorting device according to any one of claims 1 to 3, A remote processing unit configured to communicate with each other via a network line, The remote processing unit is A second acquisition unit that acquires images of formed elements that have been classified by the formed element classification device, A formed element classification system comprising: a second classification unit that reclassifies the classified formed element images based on user input on an operation screen.

20. In the formed classification system according to claim 19, The user input is a formed element classification system comprising at least one of the following: selection of the formed element image, selection of the type, and selection of the reclassification method.

21. In the formed classification system according to claim 19, A formed classification system in which the results of the reclassification are transmitted to the formed classification device.

22. A step of obtaining an image of formed elements that represent the formed elements contained in the sample, The steps include classifying each of the aforementioned formed element images in relation to the type of the aforementioned formed element, A method for classifying formed elements, comprising the step of automatically determining, based on predefined conditions, whether or not to further reclassify the classified formed element images with higher accuracy.

23. A program for classifying formed substances, comprising instructions for causing a computer processor to execute the classifying formed substances method described in claim 22.

Images