Automatic analyzer

Abstract

To provide an automatic analyzer which can realize downsizing of a constitution unit for inventory management of reagents.SOLUTION: An automatic analyzer (1) includes: a reagent tray (20) capable of holding a plurality of reagent containers (10); a gripping mechanism (41) configured to move over the reagent tray and grip a reagent container; an imaging unit (44) configured to simultaneously read a plurality of reagent containers held on the reagent tray; and a control unit (52) configured to acquire position information of the plurality of reagent containers held on the reagent tray based on image information read by the imaging unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an automatic analyzer. [Background technology]

[0002] Automated analyzers for clinical testing are equipped with reagents such as reaction reagents and calibration reagents for each measurement item, and it is desirable to automate inventory management of the onboard reagents.

[0003] For example, an automatic analyzer is known that can acquire reagent identification information and reagent placement information using an identification code reading means and a sensor that determines the presence or absence of reagent containers when multiple reagent containers with first identification codes attached are loaded on multiple container trays (see Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2013-228307 Summary of the Invention [Problem to be solved by the invention]

[0005] However, if you want to determine the presence or absence of multiple reagent containers at the same time, you need the same number of sensors as the reagent containers. While it is possible to determine the presence or absence of multiple reagent containers with a single reflective photosensor, it takes time to make the determination. Therefore, the more reagent containers are installed, the more reflective photosensors are required, which increases the space required to install the hardware. The present invention aims to provide an automated analyzer that can achieve a compact configuration unit for managing reagent inventory. [Means for solving the problem]

[0006] The automatic analyzer of the present invention is characterized by having a reagent tray capable of holding multiple reagent containers, a gripping mechanism that is movable on the reagent tray and is used to grip the reagent containers, an imaging unit that can read multiple reagent containers held on the reagent tray at once, and a control unit that can acquire positional information for the multiple reagent containers held on the reagent tray based on the image information read by the imaging unit.

[0007] Furthermore, the automated analyzer according to the present invention preferably further comprises an illumination unit including a light source that illuminates the reagent container, and the imaging unit and the illumination unit are preferably arranged to be movable together with the gripping mechanism.

[0008] Furthermore, in the automatic analyzer of the present invention, it is preferable that the control unit extracts images of multiple specific areas in which reagent containers are placed on the reagent tray from the image information, and determines the presence or absence of reagent containers in each of the multiple specific areas based on a predetermined number of pixels contained in the multiple specific areas.

[0009] Furthermore, in the automatic analyzer according to the present invention, it is preferable that the control unit performs image distortion correction on the image information and then determines whether or not a reagent container is present.

[0010] Furthermore, in the automatic analyzer according to the present invention, it is preferable that the control unit changes the threshold value for determining the presence or absence of a reagent container depending on the position on the reagent tray.

[0011] Furthermore, in the automatic analyzer according to the present invention, it is preferable that the imaging unit has a function of reading reagent identification information for identifying the contents attached to the reagent container and / or a code placed on the reagent tray. [Effects of the Invention]

[0012] In the automatic analyzer according to the present invention, the component unit that manages the inventory of reagents can be made smaller. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a schematic diagram of an automatic analyzer 1. FIG. [Figure 2] FIG. 1 is a schematic plan view of an automatic analyzer 1. [Figure 3] 1(a) is a diagram showing an empty reagent container 10, and FIG. 1(b) is a diagram showing the reagent container 10 with a reagent sealed with an aluminum foil seal. [Figure 4] 1(a) is a plan view of an empty reagent tray 20, and FIG. 1(b) is a plan view of the reagent tray 20 holding reagent containers 10. FIG. [Figure 5] 10 is a flowchart showing an example of a scanning operation for checking the inventory of reagent containers. [Figure 6] FIG. 10 is a diagram showing an example of a reagent tray image acquired for reading a two-dimensional code. [Figure 7] 10 is a flowchart illustrating an example of an image processing operation. [Figure 8] (a) is a diagram showing an example of a jig 110 for acquiring projective transformation coordinates, (b) is a diagram showing an example of an image 120 of the jig 110 captured by the imaging unit 44, (c) is a diagram showing projective transformation coordinates in the captured image 120, (d) is a diagram showing an example of a captured reagent tray image 100, (e) is a diagram showing projective transformation coordinates in the captured reagent tray image 100, and (f) is a diagram showing an example of a reagent tray image 100' after distortion correction. [Figure 9] 10A and 10B are diagrams illustrating an example of a method for extracting a specific region and calculating the sum of the number of pixels equal to or less than a threshold in the specific region. [Figure 10] FIG. 10 is a diagram showing the determination result of whether or not a reagent container is present, based on the total number of pixels. [Figure 11] FIG. 10 is a distribution diagram showing all pixels included in a specific region as the number of pixels for each gradation. [Figure 12] 10A and 10B are diagrams showing an example of the illuminance of light at each position on the reagent tray. [Figure 13] FIG. 2 is a diagram showing an example of a specific area 105. [Figure 14] (a) is an image of the sample dilution reagent container captured using a white LED, and (b) is an image of the sample dilution reagent container captured using an LED with a color temperature of 1800K. [Figure 15] (a) shows an example of a reagent tray image under direct lighting, and (b) shows an example of a reagent tray image under indirect lighting. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, an automatic analyzer according to an embodiment of the present invention will be described with reference to the drawings. However, it should be noted that the technical scope of the present invention is not limited to these embodiments, but extends to the inventions set forth in the claims and their equivalents.

[0015] Fig. 1 is a schematic diagram of an automatic analyzer 1, and Fig. 2 is a schematic plan view of the automatic analyzer 1. The automatic analyzer 1 is configured to include a storage unit 30 that stores a reagent tray 20 on which reagent containers 10 are placed, a transfer unit 40 that grasps and transfers the reagent containers 10, and an image processing unit 50. The automatic analyzer 1 also includes a transport unit 60 that transports the reagent containers 10 transferred from the storage unit 30 by the transfer unit 40 to a sample dispensing position, a tip supply unit 70 that supplies tips for dispensing samples into the reagent containers 10 transferred by the transport unit 60, a sample supply unit 80 for supplying samples, a sample dispensing unit 90 that aspirates the samples and then dispenses the samples into the reagent containers 10 on the reagent container transport unit, and a reaction / detection unit 95 that transports the samples from the transport unit 60 by the transfer unit 40 after dispensing, and performs reaction and detection.

[0016] The storage unit 30 is configured to include a three-tiered reagent tray table 31, a reagent tray table drive unit 32 for operating each tier of the reagent tray table 31, and the like. The reagent tray table 31 can accommodate up to four reagent trays 20, and can be moved in the Y-axis direction by the reagent tray table drive unit 32. The reagent tray 20 can store a total of 32 reagent containers 10 in four rows and eight columns. Note that the above-mentioned numbers of reagent tray tables 31, reagent trays 20, and reagent containers 10 are merely an example, and other numbers may be used.

[0017] The transfer unit 40 is configured to include a gripping mechanism 41, a gripping mechanism drive unit 42, a gripping arm 43, an imaging unit 44, an illumination unit 45, etc. The gripping mechanism 41 can be operated in the X and Z directions by the gripping mechanism drive unit 42. The gripping arm 43 can grip the reagent container 10 by inserting its tip into the opening 33 of the storage unit 30. The imaging unit 44 is a device, such as a camera, for capturing images of the reagent tray 20 and the reagent container 10. In addition to the imaging function, the imaging unit 44 also has a two-dimensional code reading function. The two-dimensional code reading function refers to a function of analyzing data contained in the read two-dimensional code and transmitting the analyzed data to the control unit 52.

[0018] The lighting unit 45 is configured with various light sources such as LEDs, fluorescent lamps, etc., for the purpose of illuminating the imaging target. These light sources may be configured to be adjustable in color temperature, light wavelength, light intensity, and illumination angle in order to provide optimal lighting conditions depending on the imaging target.

[0019] In this embodiment, the gripping arm 43, the image capturing unit 44, and the illumination unit 45 are disposed in the gripping mechanism 41, and therefore when the gripping mechanism 41 is moved by the gripping mechanism driving unit 42, the gripping arm 43, the image capturing unit 44, and the illumination unit 45 are also moved together. However, the gripping arm 43, the image capturing unit 44, and the illumination unit 45 do not necessarily have to be disposed in the gripping mechanism 41, and may be configured to move independently. Furthermore, the illumination unit 45 may further include a diffuser or reflector to uniformly irradiate the light from the light source onto the imaging target.

[0020] The image processing unit 50 is a PC (personal computer), data terminal, etc., which is composed of an interface unit 51, a control unit 52, a memory 53, an output unit 54, and a data bus 55 that interconnects the above-mentioned components.

[0021] The interface unit 51 is connected to the reagent tray table driving unit 32, the gripping mechanism driving unit 42, the imaging unit 44, the lighting unit 45, etc., and transmits and receives various control signals and image data captured by the imaging unit 44. The interface unit 51 is also connected to a keyboard, mouse, touch panel, etc. (not shown), and receives data signals input from them.

[0022] The image data that the interface unit 51 receives from the imaging unit 44 is an 8-bit grayscale image, but the image data may be an image with a bit number (gradation) other than 8 bits or a color image such as RGB. The image data may also be in BMP, PNG, JPEG, etc., or may be a single frame image obtained from a moving image.

[0023] The control unit 52 executes various processes including various controls of the automatic analyzer 1 and data calculations within the image processing unit 50, and is configured to include a CPU (Central Processing Unit), FPGA (Field-Programmable Gate Array), and the like.

[0024] The memory 53 is a device that stores programs, parameters, results, etc. related to various processes executed by the control unit 52, and is an SSD, HDD, ROM, RAM, etc. The output unit 54 is an interface that transmits the results of processing by the control unit 52 to the outside, and is connected to, for example, a display device such as a display or touch panel screen (not shown) or a recording device that records the results of processing.

[0025] Figure 3(a) shows an empty reagent container 10, and Figure 3(b) shows the reagent container 10 with a reagent sealed with an aluminum foil seal. The reagent container 10 is a polypropylene resin container consisting of, for example, substantially cylindrical wells 11a and 11b, each with an inner diameter of 6 mm at the opening and a height of 20 mm, arranged side by side with a minimum thickness of about 1 mm between them, and has flanges 14 extending 0.5 mm from each side in the short direction of the top surface. The tips of the flanges 14 are called the tops of the flanges 15.

[0026] The contents of the wells include reagents used in sample measurement, such as calibration reagents and standard reagents for creating calibration curves and calibration curves, sample dilutions for diluting samples, pretreatment reagents for pretreating samples, and enzyme substrates. A two-dimensional code 13 containing various identification information, an item name, and a combination of the item code and lot number, is attached to the aluminum foil seal 12. In the example of Figure 4, well 11a contains magnetic particles with immobilized antibodies that recognize the analyte, and well 11b contains another antibody that recognizes the analyte and is labeled with an enzyme, both in a freeze-dried state and sealed within the aluminum foil seal 12.

[0027] FIG. 4(a) is a plan view of an empty reagent tray 20, and FIG. 4(b) is a plan view of the reagent tray 20 holding reagent containers 10. The reagent tray 20 is made of polypropylene resin and has holders for a total of 32 reagent containers 10 arranged in 4 rows and 8 columns. Each holder in the reagent tray 20 is shaped to fit the shape of the reagent containers 10 and has a large opening 21 on the top surface that allows easy access to the reagent containers 10 from above the reagent tray, and a small circular opening 22 (e.g., 5 mm inner diameter) on the bottom surface that also serves as an air vent. Furthermore, a label 23 bearing identification information such as a tray two-dimensional code, item name, tray ID, expiration date, reagent type, and lot number is affixed to the lower edge of the reagent tray 20.

[0028] Fig. 5 is a flowchart showing an example of a scanning operation for checking the inventory of reagent containers. The processing procedure shown in Fig. 5 is executed by the control unit 52 controlling each component within the automatic analyzer 1 according to a program stored in the memory 53.

[0029] When a scanning operation for checking the inventory of reagent containers is initiated by a predetermined operation, the control unit 52 first operates the reagent tray table drive unit 32 and the gripping mechanism drive unit 42 to move the imaging unit 44 to a two-dimensional code reading position located at the lower edge of the reagent tray 20 (S10). As described above, the reagent tray table 31 can be moved in the Y direction by the reagent tray table drive unit 32, and the gripping mechanism 41 in which the imaging unit 44 is located can be moved in the X and Z directions by the gripping mechanism drive unit 42. Therefore, if predetermined X, Y, and Z positions are stored in memory 53 as parameters for the two-dimensional code reading position, the imaging unit 44 can be quickly moved to the two-dimensional code reading position located at the lower edge of the reagent tray 20.

[0030] Next, the imaging unit 44 reads the two-dimensional code placed on the lower edge of the reagent tray 20, and transmits the information to the control unit 52 (S11).

[0031] Fig. 6 is a diagram showing an example of a reagent tray image acquired for reading a two-dimensional code. As shown in Fig. 6, the reagent tray image includes a two-dimensional code on the bottom edge of the reagent tray 20, a label 23 bearing identification information such as the item name, tray ID, expiration date, reagent type, and lot number, and reagent containers 10. The imaging unit 44 may transmit the acquired reagent tray image to the control unit 52, which may then analyze the two-dimensional code information from the reagent tray image.

[0032] Next, the presence or absence of a reagent tray is determined using the results of reading the two-dimensional code (S12). If the control unit 52 successfully reads the two-dimensional code located on the lower edge of the reagent tray 20 and receives information on the type of reagent tray 20, it determines that a "reagent tray is present" at the corresponding position. On the other hand, if the control unit 52 cannot read the two-dimensional code, it determines that a "reagent tray is not present" at the corresponding position.

[0033] If it is determined in S12 that a reagent tray is present, the control unit 52 operates the reagent tray table driving unit 32 and the gripping mechanism driving unit 42 to move the imaging unit 44 to the reagent container number confirmation position in order to take an image of the reagent tray 20 (S13).

[0034] Next, the control unit 52 controls the imaging unit 44 to capture images of the reagent tray (S14). Specifically, the images are captured separately for a first reagent tray image including 16 positions in 4 rows and 4 columns on the right side as shown by dotted line A in Fig. 4(b) and a second reagent tray image including 16 positions in 4 rows and 4 columns on the left side as shown by dotted line B in Fig. 4(b).

[0035] Next, the control unit 52 performs image processing on each of the first reagent tray image and the second reagent tray image to determine the presence or absence of the reagent container 10 at each position (16 positions in 4 rows and 4 columns) on one side of each reagent tray 20 (S15). As a result, it becomes possible to determine the presence or absence of the reagent container 10 at each position (32 positions in 4 rows and 8 columns) on the entire reagent tray 20.

[0036] Next, the control unit 52 determines whether scanning of all reagent trays 20 has been completed (S16), and if scanning has not been completed, repeats S10 and subsequent steps for the positions of the other reagent trays 20. On the other hand, if scanning of all reagent trays 20 has been completed, the scanning operation for checking the inventory of reagent containers ends. In the automatic analyzer 1 shown in FIG. 1, the storage unit 30 is provided with a three-tiered reagent tray table 31 that can accommodate four reagent trays 20, so scanning is performed at 12 positions.

[0037] As described above, the automated analyzer 1 according to this embodiment is provided with a three-tiered reagent tray table 31 capable of mounting up to four reagent trays 20 in the storage unit 30 (a total of 12 reagent trays 20). In the example of the scanning operation shown in FIG. 5, the reagent trays 20 are scanned one by one. However, this is not limited to this, and various modifications are possible. For example, the two-dimensional codes of all four reagent trays 20 on each tier may be read, and then the number of reagent containers 10 may be confirmed. Furthermore, from the perspective of shortening the scanning time, it is desirable to shorten the moving distance of the reagent trays 20 and the imaging unit 44 to read the two-dimensional codes of the reagent trays 20 and confirm the number of reagent containers 10.

[0038] Fig. 7 is a flowchart showing an example of the image processing procedure. The procedure shown in Fig. 7 corresponds to S15 in Fig. 5, and is executed by the control unit 52 controlling each component in the automatic analyzer 1 using a program stored in the memory 53. By executing the image processing operation shown in Fig. 7, the presence or absence of reagent containers 10 at each position on one side of the reagent tray 20 (16 positions in 4 rows and 4 columns) is determined. Below, we will explain the image processing related to the first reagent tray image including the 16 positions in 4 rows and 4 columns on the right side as shown by dotted line A in Fig. 4(b).

[0039] The control unit 52 performs image distortion correction on the first reagent tray image captured in S14 of Fig. 5 (S20). This is because image distortion may occur if the imaging unit is not installed so that the imaging target is facing forward, or due to assembly errors, etc. Specifically, the device references projective transformation coordinates of the four corners of the reagent tray 20 that are preset, and performs keystone correction on the captured reagent tray image 100. By performing keystone correction, even when capturing an image at an angle, it is possible to convert the image into an image similar to an image captured from the front, making it possible to equalize the number of pixels of the reagent containers 10 at each position and align the rows and columns of the reagent containers 10.

[0040] Fig. 8(a) is a diagram showing an example of a jig 110 for acquiring projective transformation coordinates, Fig. 8(b) is a diagram showing an example of a captured image 120 of the jig 110 captured by the imaging unit 44, Fig. 8(c) is a diagram showing projective transformation coordinates in the captured image 120, Fig. 8(d) is a diagram showing an example of a captured first reagent tray image 100, Fig. 8(e) is a diagram showing projective transformation coordinates in the captured first reagent tray image 100, and Fig. 8(f) is a diagram showing an example of a first reagent tray image 100' after distortion correction. Note that the first reagent tray image 100 in Figs. 8(d) and (e) is an example of an image rotated 90 degrees clockwise from the position indicated by dotted line A in Fig. 4(b).

[0041] As shown in FIG. 8(a), a jig 110 is prepared with black circles P1 to P4 arranged at its four corners. The black circles P1 to P4 on the jig 110 have the same distance between the top left and bottom left, the same distance between the top right and bottom right, and the same distance between the top left and top right, and the same distance between the bottom left and bottom right. That is, the black circles P1 to P4 are arranged at the four corners of a rectangle. In the example of FIG. 8(a), four black circles are used, but different colors or marks may be used. Furthermore, the number of marks on the jig 110 does not have to be four, and the marks do not have to be spaced equally apart. It is desirable that the color of the marks on the jig 110 have high contrast with the color of the jig material when an image is captured.

[0042] The control unit 52 places the jig 110 on the reagent tray table 31 in the same manner as the reagent tray 20, and performs imaging using the imaging unit 44 to obtain a captured image 120 in advance. In the captured image 120 shown in Fig. 8(b), the black circles P1 to P4 of the jig 110 shown in Fig. 8(a) are imaged as black circles P1' to P4'. Fig. 8(c) shows the black circles P1' to P4' of the jig 110 and the projective transformation coordinates P1" to P4"

[0043] A method for calculating the projection transformation coordinates will be described with reference to FIG. 8(c). First, the control unit 52 detects the center coordinates of the black circles P1′ to P4′. Then, the control unit 52 calculates the projection transformation coordinates P1″ to P4″ by adding or subtracting correction values ​​P1# to P4# that are pre-stored in the memory 53. For example, if the center coordinate of the black circle P1′ is (X1, Y1), then P1# (correction value) = (-X2, -Y2) is added, resulting in the projection transformation coordinates P1″ = (X1-X2, Y1-Y2). The projection transformation coordinates P2″ to P4″ are also calculated using the correction values ​​P2# to P4#, just like the projection transformation coordinate P1′. Note that if the jig is designed so that the center coordinates (X1, Y1) of the black circles are set to the projection transformation coordinates, the correction values ​​can be eliminated. Fig. 8(f) shows a corrected first reagent tray image 100' obtained by performing the distortion correction described above on the photographed first reagent tray image 100 (see Fig. 8(d)). Fig. 8(e) shows an example of the photographed first reagent tray image 100 shown in Fig. 8(d) with projective transformation coordinates P1" to P4" and figures connecting the respective coordinates with dotted lines drawn on the image.

[0044] The center coordinates of each of the black circles P1' to P4' at the four corners were detected using a Hough transform, but other detection methods such as contour detection, template matching, and edge detection may also be used. Keystone correction was performed for distortion correction, but image correction methods such as horizontal movement, vertical movement, enlargement / reduction, rotation, keystone correction, and projective transformation, or a combination of these, or image analysis may also be used. The preset projective transformation coordinates for the four corners of the reagent tray may also be set for each stage and the reagent tray position on each stage.

[0045] Next, the control unit 52 extracts a plurality of specific regions from the first reagent tray image 100' after the image distortion correction (S21).

[0046] FIG. 9 illustrates an example of a method for extracting a specific region and calculating the total number of pixels below the threshold in the specific region. The first reagent tray image 100′ after image distortion correction in FIG. 9 is the same image as FIG. 8(f), but corresponds to an image rotated 90 degrees clockwise from the position indicated by dotted line A in FIG. 4(b), and shows an image of 16 positions in 4 rows and 4 columns. Reference numeral 130 in FIG. 9 illustrates multiple specific regions 105 extracted from the first reagent tray image 100′ after image distortion correction, corresponding to a total of 16 positions in 4 rows and 4 columns. Reference numeral 140 in FIG. 9 illustrates a pixel count distribution diagram showing the number of pixels for each gradation level for all pixels included in the specific region 105. Reference numeral 150 in FIG. 9 illustrates the total number of pixels below the threshold in each specific region 105 illustrated in reference numeral 130 in FIG. 9.

[0047] The specific area 105 is selected as an area that can maintain high gradation when the reagent container 10 is not present (an area that is not in the shadow of the reagent container 10), regardless of the two-dimensional code, item name, etc. printed on the aluminum foil seal of the reagent container 10.

[0048] In this embodiment, the reference coordinates (Xn, Yn) of the specific area 105 were determined by referring to the top position 15 of the fan on the aluminum foil seal of the reagent container 10 (see the first reagent tray image 100' after image distortion correction in Figure 9). The specific area 105 was set to a width of +50 pixels from the X coordinate and a height of ±25 pixels from the Y coordinate, and the X and Y coordinates were set to be the same in the row and column of the reagent tray 20. However, the specific area 105 in the fourth column from the front of the reagent tray 20 is shifted +15 pixels forward from the Y coordinate of the reference coordinate that should be set. This is to prevent an increase in the number of pixels caused by the reagent container 10 in the fifth column being reflected in the extracted area when there is no reagent container 10 in the fourth column position but there is a reagent container 10 in the fifth column position at the back. In other words, this is to prevent erroneous detection, such as determining that a reagent container is present due to the reagent container 10 being reflected in the fifth column position even though there is no reagent container 10 in the fourth column position. In this way, the specific area 105 is rectangular and has the same number of pixels at each position, but the shape and number of pixels of the area at each position may be different, and the reference coordinates may also be determined for each position.

[0049] Next, the control unit 52 calculates the sum of the number of pixels equal to or less than the threshold in each specific region 105 (S22).

[0050] Graph 140 in Fig. 9 is a distribution diagram showing all pixels included in specific region 105 as the number of pixels for each gradation, and 150 in Fig. 9 is the calculated total number of pixels below a threshold (number of gradations = 120 in graph 140) in each specific region 105 corresponding to 130 in Fig. 9. The greater the number of pixels below a predetermined threshold, the higher the probability that a reagent container 10 will be present at the position indicated by specific region 105. Note that, here, when the gradation is low, the pixel density is high (close to black), and when the gradation is high, the pixel density is low (close to white).

[0051] Next, the control unit 52 determines the presence or absence of reagent containers 10 at each position (16 positions in 4 rows and 4 columns) on one side of the reagent tray 20 (see dotted line A in Fig. 4(b)) from the total number of pixels calculated in S22 (S23), and ends the series of image processing operations. In reality, image processing is then also performed on the second reagent tray image, which includes the 16 positions in 4 rows and 4 columns on the left side shown as dotted line B in Fig. 4(b), but since this is the same as the image processing for the first reagent tray image, a description thereof will be omitted.

[0052] 10 is a diagram showing the determination result of whether or not a reagent container is present from the total number of pixels. In this embodiment, a pixel count of 450 or more indicates "reagent container present," and a pixel count of less than 450 indicates "reagent container absent." The pixel count (threshold) for determining whether or not a reagent container is present may be determined based on the lowest possible pixel count, or a statistical determination method using, for example, the standard deviation of the pixel count when a reagent container is present may be used.

[0053] As described above, by repeating the image processing procedure described with reference to Fig. 7, the automatic analyzer 1 can determine the presence or absence of the reagent container 10 at each position on the reagent tray 20. Based on the determination result of the presence or absence of the reagent container 10, the automatic analyzer 1 can grasp the inventory of the reagent containers 10 stored in the storage unit 30. In this way, by determining the presence or absence of multiple reagent containers based on image data captured by a single imaging unit, or by combining the function of determining the presence or absence of multiple reagent containers based on image processing with the function of reading two-dimensional codes, the automatic analyzer 1 can save space.

[0054] In the above description, image processing was performed on two images: a first reagent tray image including 16 positions in 4 rows and 4 columns on the right side indicated by dotted line A in Fig. 4(b), and a second reagent tray image including 16 positions in 4 rows and 4 columns on the left side indicated by dotted line B in Fig. 4(b). However, it is also possible to capture a single reagent tray image including the entire reagent tray 20 and perform the image processing shown in Fig. 7 only once to determine the presence or absence of reagent containers 10 at each position on the reagent tray 20 at once. Furthermore, the method of dividing the reagent tray 20 into images may be a method other than that shown in Fig. 4(b), and the number of divisions may be three or more.

[0055] Furthermore, after capturing the first reagent tray image and the second reagent tray image, distortion image correction may be performed on each to generate a corrected first reagent tray image and a corrected second reagent tray image.The corrected first reagent tray image and the corrected second reagent tray image may be combined to generate a single corrected reagent tray image, and the presence or absence of reagent containers 10 at each position on the reagent tray 20 may be determined at once based on the single corrected reagent tray image.

[0056] The threshold value of the gradation value and the threshold value of the number of pixels in the image processing procedure will be considered below.

[0057] FIG. 11 is a distribution diagram showing the number of pixels in each gradation for all pixels included in a specific region. In this embodiment, the threshold value is set to 120 gradations. However, a distribution diagram of the number of pixels as shown in FIG. 11 may be created and the threshold value may be determined using mathematical analysis. For example, the median value of the peaks of the distribution diagram of the number of pixels representing the presence or absence of a reagent container may be used as the threshold. In S22, the sum of pixels below the threshold (number of gradations = 120) is calculated. However, the sum of pixels below the threshold, the sum of pixels above or equal to the threshold, or the sum of pixels within a predetermined range may also be calculated. Furthermore, the sum of pixels may be calculated after applying various binarization methods, such as adaptive threshold processing or Otsu's algorithm, which can automatically set a threshold value. Furthermore, instead of the sum of the total number of pixels after binarization, the ratio of white pixels to black pixels may be calculated, and the presence or absence of a reagent container may be determined based on the ratio of white pixels to black pixels.

[0058] FIG. 12 is a diagram showing an example of the illuminance of light at each position on the reagent tray. Reference numeral 170 in FIG. 12 is an example of a reagent tray image, with region 171 representing a high-illuminance region and region 172 representing a low-illuminance region. Reference numeral 180 in FIG. 12 indicates the gradation number with the largest number of pixels in each specific region 105 corresponding to 170. The gradation number with the largest number of pixels was determined from a pixel count distribution map created (see 140 in FIG. 9 ) in which all pixels included in each specific region 105 are represented as the number of pixels for each gradation. In FIG. 12, region 181 corresponds to region 171 as a high-illuminance region, and region 182 corresponds to region 172 as a low-illuminance region. In this way, when high-illuminance and low-illuminance regions coexist in the reagent tray image, it is preferable to set a threshold value for each region. In the case of FIG. 12, if the threshold value in the area 171 is set to 180 and the threshold value in the area 172 is set to 120, the presence or absence of the reagent container can be determined satisfactorily.

[0059] FIG. 13 is a diagram showing an example of a specific region 105. In FIG. 13, 16 (4 rows x 4 columns) specific regions 105 are arranged in an example of a reagent tray image. In reality, the position of the reagent container 10 may be shifted on the reagent tray 20, so it is necessary to set a pixel count threshold so that the presence or absence of the reagent container can be determined appropriately even when the displacement of the reagent container 10 is at its maximum. In the example of FIG. 13, the reagent container 10 is tilted by hand at the position indicated by dotted line C, causing the displacement of the reagent container to be at its maximum, but the pixel count threshold is set so that the presence or absence of the reagent container can still be determined.

[0060] The optimization of imaging conditions in the image processing procedure will be considered below.

[0061] To determine the presence or absence of reagent containers based on the number of images below a set threshold, it is necessary to increase the contrast between the presence and absence of reagent containers in the reagent tray images. To increase this contrast, it is necessary to select an appropriate color temperature for the lighting unit. Reagent types include reaction reagents, calibration reagents, sample dilution solutions, and pretreatment reagents, and the aluminum foil seal colors are black, green, blue, and purple, respectively. An LED with a color temperature of 1800K and a wavelength near red, which is the complementary color to the aluminum foil seal colors, was selected for the lighting unit 45.

[0062] Figure 14(a) is an image of a sample dilution reagent container captured using a white LED, and Figure 14(b) is an image of a sample dilution reagent container captured using an LED with a color temperature of 1800 K. As can be seen from Figures 14(a) and (b), the color of the aluminum foil seal appears gray in the image using the white LED, while it appears black in the image using the LED with a color temperature of 1800 K.

[0063] It is also possible to increase the contrast by increasing the light intensity of the illumination unit 45 and optimizing parameters such as the gain and exposure time in the settings of the imaging unit 44. Generally, the contrast is high in areas of high illumination in the reagent tray 20, and low in areas of low illumination. It is desirable to optimize the light intensity of the illumination unit 45 and the gain and exposure time of the imaging unit 44 in each area so that the gradation does not become saturated when a reagent container is present, and so that the gradation does not become too low when a reagent container is absent.

[0064] Figure 15(a) shows an example of a reagent tray image under direct lighting, and Figure 15(b) shows an example of a reagent tray image under indirect lighting. As can be seen from Figures 15(a) and 15(b), the reagent tray image using indirect lighting is illuminated evenly from the front to the back, maintaining a constant contrast. Therefore, in order to illuminate the reagent containers 10 in each position on the reagent tray 20 evenly (32 positions in total, 4 rows and 8 columns), the illumination unit 45 may be configured for indirect lighting.

[0065] The image transfer time was the most time-consuming part of image processing. To shorten this time, it is preferable to reduce the resolution of the reagent tray image. When the resolution was set to 640 x 480 pixels, 320 x 240 pixels, and 160 x 120 pixels, the image transfer time was 3.2, 2.2, and 2 seconds, respectively.

[0066] It should be understood by those skilled in the art that various changes, substitutions, and alterations can be made to the present invention without departing from the scope of the present invention. For example, the above-described embodiments and modifications may be implemented in appropriate combination within the scope of the present invention. [Explanation of symbols]

[0067] 1 Automatic analyzer 10 Reagent containers 13 QR Code 20 Reagent Trays 23 Labels with printed identification information 30 Storage area 31 Reagent tray mounting area 40 Transfer section 41 Gripping mechanism 44 Imaging unit 50 Image processing section 52 Control section

Claims

1. a reagent tray capable of holding a plurality of reagent containers; a gripping mechanism movable on the reagent tray for gripping the reagent container; an imaging unit capable of simultaneously reading a plurality of reagent containers held on the reagent tray; a control unit capable of acquiring position information of a plurality of reagent containers held in the reagent tray based on image information read by the imaging unit; An automatic analyzer comprising:

2. further comprising an illumination unit including a light source that illuminates the reagent container; The automated analyzer according to claim 1 , wherein the imaging unit and the illumination unit are arranged to be movable together with the gripping mechanism.

3. 2. The automated analyzer according to claim 1, wherein the control unit extracts images of a plurality of specific regions in which the reagent containers are arranged on the reagent tray from the image information, and determines the presence or absence of the reagent containers in each of the plurality of specific regions based on a predetermined number of pixels included in the plurality of specific regions.

4. The automatic analyzer according to claim 3 , wherein the control unit performs image distortion correction on the image information before determining whether the reagent container is present.

5. The automatic analyzer according to claim 3 , wherein the control unit changes a threshold value for determining whether or not the reagent container is present, depending on the position on the reagent tray.

6. The automatic analyzer according to any one of claims 1 to 5, wherein the imaging unit has a function of reading reagent identification information for identifying the contents attached to the reagent container and / or a code placed on the reagent tray.

Images