Automatic dispenser

Abstract

To enhance the accuracy of determining a type of a specimen container based on an image in an automatic dispenser.SOLUTION: An automatic dispenser includes: a container housing unit (204); an imaging unit (500); a container identification unit (134) for identifying a type of a specimen container from an image of the container housing unit; a container housing information storage unit (131) for storing information on the type of the specimen container to be housed in the container housing unit; a determination unit (136) for inputting the image of the container housing unit into the container identification unit, causing the container identification unit to identify the type of the specimen container, and determining whether the type of the specimen container is same or different with respect to the container housing information; a container posture adjustment unit (137) for adjusting a posture of the specimen container in the container housing unit; and a re-determination unit (138) for causing the container posture adjustment unit to execute operation to adjust the posture of the specimen container when the type of the specimen container is different, inputting the image of the container housing unit into the container identification unit to identify the type of the specimen container, and re-determining whether the type of the specimen container is same or different with reference to the container housing information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] An automatic dispensing device includes a sample rack having multiple container storage sections for storing sample containers, a pipette, and a robot arm capable of holding the pipette. After collecting a liquid sample or reagent with the pipette, the robot arm moves the pipette to the position of a specified sample container stored in the container storage section of the sample rack, and dispenses the collected liquid (see, for example, Patent Document 1).

[0003] An automated pipetting device is used to perform processes such as capturing, recovering, and diluting a target component in a sample. For example, when capturing a target component in a sample, a vial with a column container for capturing the target component attached to its top opening is placed in one container housing of a sample rack, and the sample is dispensed into the column container to capture the target component in the column container. When recovering the target component captured in the column container, a new vial is placed in another container housing, the column container is moved to its top opening, and an eluent is dispensed to recover the target component captured in the column container in the new vial. When diluting the target component, the column container after eluting the target component is removed, and a diluent is dispensed into the vial. In this example, a vial and a column container are placed in a predetermined container housing during the target component capture step, a new vial and the column container, different from those used during the capture step, are placed in another predetermined container housing during the target component recovery step, and only a vial is placed in the container housing during the target component dilution step. In this way, when processing the target component, the position of the container storage section to be used at each stage and the type of sample container, such as a vial or column container, to be stored in that container storage section are predetermined.

[0004] Some automatic dispensing devices are equipped with a camera that photographs the sample rack and a judgment unit that learns the shape and color of the sample containers based on images of the sample rack with the sample containers accommodated in the container storage units, and the judgment unit identifies the type of sample container stored in the container storage units based on the images taken by the camera and determines whether the correct type of sample container is stored in the correct container storage unit in the sample rack. Such automatic dispensing devices make the above judgment at each stage of the processing for the target component, and if the judgment result is negative, they notify the analyst of an error, interrupt the automatic processing, and prompt the analyst to check the position of the container storage unit and the sample container stored therein. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2019-174369 Summary of the Invention [Problem to be solved by the invention]

[0006] To facilitate smooth insertion and removal of sample containers into and removal from the container receptacle, the inner diameter of the container receptacle is slightly larger than the outer diameter of the sample container. This leaves some room within the container receptacle when the sample container is set in the receptacle. Therefore, when a sample container, such as a vial, is placed upright in the receptacle, it may tilt in various directions within the receptacle. If the state of the sample container captured by the camera differs from the state of the sample container captured in the image used for learning, the judgment unit may incorrectly identify the type of sample container contained in the receptacle. Incorrect identification of the sample container in the receptacle results in an erroneous judgment, resulting in an error notification even if the correct type of sample container is contained in the receptacle. When an error is notified, the analyst must check the receptacle each time, placing a heavy burden on the analyst. Furthermore, the automatic processing is interrupted each time an error occurs, reducing processing efficiency.

[0007] The problem that the present invention aims to solve is to provide a technology in an automatic dispensing device that can improve the accuracy of determining whether the type of sample container in a container storage section is correct or not based on an image of the container storage section. [Means for solving the problem]

[0008] In order to solve the above problems, the present invention provides an automatic pipetting device that performs a predetermined process on a sample, comprising: a container accommodating section capable of accommodating a sample container; an imaging unit that acquires an image of the container storage unit; a container identification unit that is constructed by learning an image of the container storage unit in a state where a sample container is stored therein, and that identifies the type of sample container stored in the container storage unit when an image of the container storage unit is input; a container storage information storage unit that stores container storage information, which is information on the type of sample container to be stored in the container storage unit before or after the processing; a determination unit that operates the imaging unit to acquire an image of the container storage unit before or after the processing, inputs the image to the container identification unit to identify the type of sample container stored in the container storage unit, and determines whether the identified type of sample container is the same as the type of sample container to be stored in the container storage unit in light of the container storage information; and a container attitude adjusting unit that performs an operation to adjust the attitude of the sample container accommodated in the container accommodating unit to a predetermined attitude; a re-determination unit that, when the determination unit determines that the type of the sample container is different, causes the container attitude adjustment unit to perform an operation to adjust the attitude of the sample container accommodated in the container accommodating unit to the predetermined attitude, operates the imaging unit to re-acquire an image of the container accommodating unit, inputs the image into the container identification unit to identify the type of sample container accommodated in the container accommodating unit, and re-determines whether the identified type of sample container is the same as the type of sample container to be accommodated in the container accommodating unit in light of the container accommodation information; Equipped with. [Effects of the Invention]

[0009] In the automatic dispensing device according to the present invention, when the determination unit determines that the type of sample container is incorrect, the container attitude adjustment unit performs an operation to adjust the attitude of the sample container accommodated in the container housing unit to a predetermined attitude, and then the re-determination unit performs a re-determination. Here, the predetermined attitude refers to a state in which the sample container is slightly tilted, for example, when the sample container is accommodated in the container housing unit in an upright position. Therefore, even if the determination unit incorrectly identifies the type of sample container even though the correct type is accommodated in the container housing unit due to the sample container being accommodated in various tilted positions, and the determination unit determines that the sample container is of a different type, the container attitude adjustment unit performs an operation to adjust the orientation of the sample container accommodated in the container housing unit, and then the re-determination is performed. This improves the accuracy of identifying the type of sample container in the container housing unit from an image of the container housing unit and determining whether the type is correct or not. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a top view showing a schematic configuration of a sample pretreatment device that is an embodiment of a sample dispensing device according to the present invention. [Figure 2] FIG. 2 is a side view of the inside of the sample pretreatment device of the present embodiment. [Figure 3] FIG. 2 is a diagram illustrating the shapes of a spin column and a microtube used in this embodiment. [Figure 4] FIG. 2 is a diagram illustrating the structure of a sample rack according to the present embodiment. [Figure 5] 10 is a flowchart illustrating a procedure for constructing a container identification unit in the present embodiment. [Figure 6] 5A and 5B are diagrams illustrating the state of a sample container accommodated in a sample receptacle portion in this embodiment. [Figure 7] 10A to 10C are diagrams illustrating an operation for adjusting the orientation of a sample container in a sample receptacle in the embodiment. [Figure 8] 10 shows examples of images used for learning of a container identification unit in this embodiment. [Figure 9]10 is a flowchart illustrating a procedure for determining the type of a sample container in this embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] A sample pretreatment device, which is an embodiment of an automatic pipetting device according to the present invention, will be described below with reference to the drawings. Note that the actual scales of the drawings have been changed appropriately to clearly show the respective components.

[0012] Fig. 1 is a top view showing a schematic configuration of the sample pretreatment device of this embodiment, and Fig. 2 is a side view of the inside of the sample pretreatment device (a side view seen from below on the paper surface of Fig. 1).

[0013] The sample pretreatment device of this embodiment includes a housing 100, a centrifuge 110, a micropipette 161, a robot arm 120, and a control unit 130, all of which are arranged in the housing 100. Also provided within the housing 100, surrounding the robot arm 120, are a sample rack arrangement section 220 in which a sample rack 200 having a plurality of container storage sections 204 (see Figure 4) is arranged, a column rack arrangement section 140 in which a column rack 141 storing a plurality of spin columns 300 is arranged, a tube rack arrangement section 150 in which a tube rack 151 storing a plurality of microtubes 400 is arranged, a micropipette storage section 160 in which one or more micropipettes 161 are stored, a tip rack arrangement section 170 in which a tip rack 172 storing a plurality of pipette tips 171 is arranged, a sample liquid / reagent rack arrangement section 180 in which a sample liquid / reagent rack 183 storing reagent containers 181 and sample liquid containers 182 containing sample liquid is arranged, a consumables disposal section 191 in which used spin columns 300 and pipette tips 171 are disposed, and a pretreated sample liquid storage section 192 in which microtubes 400 containing pretreated sample liquid are disposed. Furthermore, a camera 500 is disposed on the top surface inside the housing 100 so as to capture the above-mentioned components in its field of view (shown by dashed lines in FIG. 2).

[0014] The micropipette 161 includes a cylinder, a plunger (or piston), and a tubular nozzle connected to the bottom end of the cylinder, and by moving the plunger up and down within the cylinder with a disposable pipette tip 171 attached to the bottom end of the nozzle, liquid can be aspirated into the pipette tip 171 or discharged from the pipette tip 171. The micropipette 161 is, for example, an electric micropipette that incorporates a motor for driving the plunger and is equipped with operation buttons for instructing the execution of aspirating and discharging operations by driving the motor.

[0015] The robot arm 120 is equipped with a hand unit 121 that can hold the spin column 300, the microtube 400, and the micropipette 161 and can operate the operation buttons provided on the micropipette 161.

[0016] The control unit 130 includes a memory unit 131. The memory unit 131 stores the procedure and content of pre-processing performed on each of one or more samples, whether or not a determination is required by the determination unit 136, information on the type of sample container to be accommodated in the container accommodating unit 204 (see FIG. 4) of the sample rack 200 at each stage of the pre-processing (pre-processing information 1311, including the container accommodating information of the present invention), and information on erroneous determination of the type of sample container accommodated in the container accommodating unit 204 of the sample rack 200 (determination target position information 1312). The memory unit 131 also sequentially stores images captured by the camera 500, determination results by the determination unit 136 and re-determination unit 138, and the like.

[0017] The control unit 130 includes, as functional blocks, a learning execution unit 133, a container identification unit 134, a pre-processing execution unit 135, a determination unit 136, a container attitude adjustment unit 137, and a re-determination unit 138. The actual entity of the control unit 130 is, for example, a personal computer, and the learning execution unit 133, the pre-processing execution unit 135, the determination unit 136, the container attitude adjustment unit 137, and the re-determination unit 138 are realized by executing a dedicated program pre-installed in the computer on a processor. The container identification unit 134 is an algorithm constructed by learning the image pattern of each sample container, which is executed by the learning execution unit 133. Although the control unit 130 is located outside the housing 100 in FIG. 1, the control unit 130 may also be located inside the housing 100. In addition, an input unit 1391 including a keyboard and a mouse and a display unit 1392 including a liquid crystal display or the like are connected to the control unit 130.

[0018] The column rack 141 holds a large number of spin columns 300 as shown in FIG. 3, and the tube rack 151 holds a large number of microtubes 400 as shown in the same figure.

[0019] The microtube 400 has a cylindrical portion with a closed bottom and an open top, and a generally conical portion whose diameter decreases downward. However, the microtube 400 in this embodiment is merely an example of a sample container, and is not limited to this shape. Any shape may be used as long as it can contain a liquid and can accommodate a spin column 300 (described later).

[0020] The spin column 300 has a cylindrical portion 301 with a circular upper opening and a lower opening with a smaller diameter than the upper opening, and a membrane filter 302 held inside the cylindrical portion 301. As shown on the right side of Figure 3, the spin column 300 can be inserted into the upper opening of a microtube 400. The membrane filter 302 may be any filter used in solid-phase extraction, and may be, for example, a porous carrier such as a silica monolith, a silica membrane, or glass fiber. In this specification, the spin column 300 alone, the microtube 400 alone, and the spin column 300 attached to the microtube 400 are all referred to as sample containers.

[0021] As shown in FIG. 4 , the sample rack 200 includes a main body 201 and a heat transfer block 203 placed on the main body 201. The main body 201 has a recess for accommodating the heat transfer block 203, and a heater 202 for heating the heat transfer block 203 and a temperature sensor (not shown) for measuring the temperature of the heat transfer block 203 are provided below the inner bottom surface of the recess. In addition to the heating function provided by the heater 202 as described above, the sample rack 200 may also be provided with a cooling function, for example, using a fan or a Peltier element. The heat transfer block 203 is made of a material with high thermal conductivity, such as aluminum, and its upper surface is provided with a container receiving portion 204, which is a recess for accommodating a microtube 400. A cylindrical cavity is provided below each container receiving portion 204, and a spring plunger 205 is fitted into the cavity. The cavity may be a through-hole whose upper end opens to the inner bottom surface of the container accommodating section 204 and whose lower end opens to the lower surface of the heat transfer block 203, or may be a concave shape whose upper end is open only. Here, the sample rack 200 has a temperature control function, but it is also possible to use one that does not have a temperature control function (one that has only multiple container accommodating sections 204 and spring plungers 205). The spring plunger 205 is also a preferred component but is not essential.

[0022] Spring plunger 205 includes hollow cylindrical holder 206, compression coil spring 210 housed in holder 206, and pin member 207, a portion of which can be freely extended and retracted from holder 206. Pin member 207 includes protrusion 208 that protrudes from holder 206 to the outside through a through-hole (not shown) provided in the upper surface of holder 206, and flange-shaped retaining portion 209 that is provided at the lower end of protrusion 208 and can slide up and down within holder 206. The lower surface of retaining portion 209 abuts against the upper end of compression coil spring 210, and pin member 207 is always biased upward by the biasing force of compression coil spring 210. Instead of spring plunger 205 including pin member 207 as described above, a so-called ball plunger or the like may be used as the biasing member, which includes a ball, a portion of which can be freely extended and retracted from holder 206, and a spring that biases the ball.

[0023] Next, the operation of the sample pretreatment device of this embodiment will be described.

[0024] In the sample pretreatment device of this embodiment, before pretreatment of a sample is performed, the container identification unit 134 is constructed by the learning execution unit 133. FIG.

[0025] When the analyst instructs the start of learning, the learning execution unit 133 has the analyst set the type of sample rack 200 and the type of sample container to be set in the sample rack 200 (step 1). This setting can be performed, for example, by displaying pre-registered types of sample rack 200 and types of sample container (in this example, spin column 300, microtube 400, spin column 300 + microtube 400) in a pull-down format and having the analyst select one. Note that if only one type of sample rack 200 is used, it is not necessary to set the type of sample rack 200.

[0026] The analyst selects the type of sample container, sets the sample rack 200 containing the selected type of sample container in the container storage section 204 in the sample rack placement section 220 (step 2), and when the analyst issues an instruction to start processing by pressing the complete button or the like, the container attitude adjustment section 137 moves the robot arm 120 to a position above the sample rack 200. Next, the hand section 121 is lowered from the position above the sample containers stored in the container storage sections 204 of the sample rack 200, and the hand section 121 grips and releases the sample container for each sample container (step 3).

[0027] Here, the reason for performing the above operation will be explained.

[0028] The inner diameter of the container receptacle 204 in the sample rack 200 is larger than the outer diameter of the sample container so that the sample container can be smoothly stored in and removed from the container receptacle 204. Furthermore, during the manufacturing of the sample containers, individual differences in the size and shape of the sample containers occur within the tolerance range.

[0029] The left side of FIG. 6 shows a state (standard state) in which a sample container (here, a microtube 400 with a spin column 300 inserted into the top opening) is accommodated in the center of the container accommodating section 204. However, the sample container can be accommodated in various positions within the container accommodating section 204. Specifically, for example, as shown in the center of FIG. 6, the spin column 300 and the microtube 400 may be accommodated in a position offset toward the edge of the container accommodating section 204, or as shown on the right side of FIG. 6, the sample container may be accommodated in a tilted position within the container accommodating section 204. When such a state is photographed from above with the camera 500 and an image is acquired, the positions of the lines indicating the outer edges of the microtube 400 and the spin column 300 are significantly different, as shown in the lower part of FIG. 6. For example, if the container identification unit 134 is constructed by learning images of extreme states such as those shown in the center and right sides of FIG. 6, erroneous determinations are likely to occur when determining the type of sample container accommodated in the container accommodating section 204 at each stage of actual pretreatment. Note that the spring plunger 205 is not shown in FIGS. 6 and 7.

[0030] Therefore, in this embodiment, as shown in FIG. 7 , after the analyst sets the sample rack 200, the sample containers (spin columns 300 + microtubes 400) contained in the container storage units 204 are grasped (arrows in the left diagram of FIG. 7 ) and released (arrows in the center diagram of FIG. 7 ) by the hand unit 121 of the robot arm 120. The robot arm 120 is taught with high positional accuracy at the time of shipment or installation of the device, with an error of, for example, approximately ±0.01 mm. Therefore, by grasping and releasing each sample container with the robot arm 120, each sample container is accommodated in the center of the container storage unit 204 in an upright state (right diagram of FIG. 7 ). Even after the hand unit 121 of the robot arm 120 grasps and releases the sample containers, some variation in the sample container position may occur from a completely upright state (standard state). However, the variation in the sample container position is significantly smaller than before the above process.

[0031] When the robot arm 120 has adjusted its posture by gripping and releasing all of the sample containers, the learning execution unit 133 operates the camera 500 to photograph the sample rack 200 (step 4). The image of the sample rack 200 photographed by the camera 500 is stored in the memory unit 131.

[0032] When an image of the sample rack 200 is acquired, the learning execution unit 133 divides the image at a position according to the type of sample rack 200 set by the analyst (cuts out an image for each container storage unit 204 of the sample rack 200), acquires information (learning image information) that associates the position of the container storage unit 204 with the image of the sample container stored in that position (step 5), and stores this information in the memory unit 131. In this embodiment, the camera 500 captures the sample rack 500 from almost directly above, but depending on the positional relationship between the camera 500 and the sample rack placement unit 220, the sample rack 200 may be captured from an oblique direction, and the angle at which the sample containers in the container storage units 204 are captured may vary slightly for each container storage unit 204. In addition, the amount of light hitting the sample containers may vary depending on the position of the container storage unit 204. Therefore, in this embodiment, an image of the state in which the sample containers are stored in each container storage unit 204 is acquired for each container storage unit 204.

[0033] Once the learning image information is saved, the learning execution unit 133 causes a previously prepared learning model to learn predetermined features contained in the reference image (step 6). Examples of predetermined features include the outline of the sample container and the proportion of RGB values ​​in the image, and the learning model is made to learn features corresponding to the characteristics (shape and color) of the sample container.

[0034] Next, the learning execution unit 133 checks whether learning has been completed for the state in which all set types of sample containers are accommodated and the state in which no samples are accommodated (step 7). If there is an unlearned state (NO in step 7), a screen prompting the analyst to set the sample rack 200 in that state is displayed on the display unit 1392, and a series of processes similar to those described above are performed to learn the features contained in the reference image.

[0035] When learning is complete for all types of sample containers (YES in step 7), a container identification unit 134 for identifying sample containers is constructed (step 8). Here, we have described a case where one image (see FIG. 8) is acquired for each state in which each type of sample container is contained in each container housing unit 204 and for a state in which no sample container is contained. However, taking into account variations in the orientation of the sample containers contained in each container housing unit 204, multiple images may be acquired for each state, and the container identification unit 134 may be constructed by learning predetermined features (such as the outline of the sample container and the proportion of RGB values ​​in the image) from the multiple images. The construction of the container identification unit 134 may be performed at the time of shipment or installation of the device. Furthermore, when starting to use a new type of sample container, the analyst may operate the learning execution unit 133 at that time to have the container identification unit 134 perform additional learning for that sample container.

[0036] Next, as an example of sample pretreatment, a case where a series of pretreatments for cleaving sugar chains contained in a sample solution with an enzyme is automatically performed will be described.

[0037] When the analyst instructs the start of sample pretreatment, the pretreatment execution unit 135 prompts the analyst to set the type of sample to be treated and the details of the pretreatment. This setting can be performed, for example, by displaying in a pull-down format items stored in the storage unit 131 in advance and prompting the analyst to select an option.

[0038] When the analyst sets the type of sample and the details of pre-processing, the pre-processing execution unit 135 operates the robot arm 120 to move unused microtubes 400 arranged in the tube rack arrangement unit 150 to each container receptacle 204 of the sample rack 200. Note that here, microtubes 400 are set in only half (or less) of the container receptacles 204 provided in the sample rack 200. The container receptacles 204 into which the microtubes 400 have been set are then subject to sample container determination, which will be described later.

[0039] The pre-processing execution unit 135 also operates the robot arm 120 to insert an unused spin column 300 placed in the column rack 141 into the upper opening of the microtube 400 housed in each container housing portion 204 of the sample rack 200. As a result, each container housing portion 204 is set with a spin column 300 and a microtube 400.

[0040] Here, it is determined whether the type of sample container that should be accommodated at this stage is correctly accommodated in the container accommodation portion 204 of the sample rack 200. Hereinafter, this determination will be referred to as sample container determination.

[0041] The flow of sample container determination will be described with reference to the flowchart in FIG.

[0042] First, the determination unit 136 operates the camera 500 to take an image of the sample rack 200 (step 11). The taken image is stored in the storage unit 131.

[0043] When the image of the sample rack 200 is saved, the container identification unit 134 cuts out an image of each container holding section 204 included in the image of the sample rack 200, and for each container holding section 204, identifies the type of sample container held in that container holding section 204.

[0044] The determination unit 136 then reads the type of sample container (spin column 300 + microtube 400) that should be accommodated in the container receptacle 204 at this stage (before sample liquid is introduced) from the pretreatment information 1311 stored in the memory unit 131, compares it with the identified type of sample container for each container receptacle 204, and determines whether the two match (different or the same) (step 12). This determination result is stored in the memory unit 131. Next, it is confirmed whether the types of sample containers have been determined for all container receptacles 204 (step 13). If an unidentified container receptacle 204 remains (NO in step 13), the container identification unit 134 identifies the type of sample container for that container receptacle 204 in the same manner as above. Once the types of sample containers have been determined for all container receptacles 204 (YES in step 13), it is confirmed whether the types of sample containers match for all container receptacles 204 (step 14). If the types of sample containers match for all container receptacles 204 (YES in step 14), the sample container determination is completed.

[0045] If the type of sample container to be stored in any of the container storage sections 204 does not match the type of sample container identified (NO in step 14), the container posture adjustment section 137 operates the robot arm 120, and uses the hand section 121 to grasp and open the sample container stored in the container storage section 204 whose type did not match as determined above (step 15).

[0046] Next, the re-evaluation unit 138 operates the camera 500 to photograph the sample rack 200 again (step 16), and stores the image in the memory unit 131. The container identification unit 134 again identifies the type of sample container accommodated in each container housing unit 204. The re-evaluation unit 138 then compares the type of sample container to be accommodated in the container housing unit 204 with the identified type of sample container for that container housing unit 204, and re-evaluates whether the two match (step 17). If the re-evaluation determines that the two match, and there are no more mismatched container housing units 204 (YES in step 18), the sample container evaluation is terminated.

[0047] On the other hand, if the re-evaluation in any of the container storage units 204 determines that the sample container type does not match (NO in step 18), it is determined whether the number of re-evaluations has reached a predetermined number (step 19). In this embodiment, the predetermined number is three. If the sample container type does not match even after three re-evaluations, it is highly likely that the sample container actually stored in the container storage unit 204 is incorrect, and performing re-evaluation four or more times is unlikely to change the result. Furthermore, the more re-evaluations are performed, the longer the time required. Therefore, the predetermined number may be set to three or less. Here, if the number of re-evaluations has reached the predetermined number (YES in step 19), the pre-processing execution unit 135 concludes that the wrong type of sample container is stored in that container storage unit 204, and displays an error on the display unit 1392 together with information about the location of that container storage unit 204, the identified type of sample container, and the type of sample container that should be stored in that container storage unit 204 (step 20), prompting the analyst to confirm and ending the sample container determination. In this case, the automatic pre-processing is interrupted. Thereafter, when an instruction to resume is received from the analyst, the process resumes by returning to step 11. If the number of re-evaluations has not reached the predetermined number (NO in step 19), the process returns to step 15 to adjust the attitude of the sample container, acquire an image, identify the sample container, and perform re-evaluation.

[0048] When it is determined that the types of sample containers match for all container storage sections 204 (YES in step 14 or step 18), the pre-processing execution section 135 resumes the pre-processing of the sample.

[0049] The pre-processing execution unit 135 operates the robot arm 120, causes the hand unit 121 to grasp the micropipette 161 housed in the micropipette housing unit 160, and attaches an unused pipette tip 171 arranged in the tip rack 172 to the tip of the micropipette. Then, a predetermined amount of sample liquid is collected from the sample liquid container 182 and dispensed into a sample container (a spin column 300 arranged on a microtube 400) housed in the container housing unit 204. As a result, the sugar chains (target components) contained in the sample liquid are captured by the membrane filter 302, and the rest is discarded into the microtube 400.

[0050] After dispensing the sample liquid into all the sample containers, the preprocessing execution unit 135 then discards the used pipette tip 171 attached to the tip of the micropipette 161 into the consumables disposal unit 191 and attaches a new pipette tip 171. Then, a predetermined amount of reagent (enzyme) stored in the sample liquid / reagent rack 183 is collected and dispensed into the sample container (spin column 300 placed on the microtube 400) stored in the container storage unit 204. When dispensing the enzyme, the tip of the pipette tip 171 is lightly pressed against the membrane filter 302 of the spin column 300. As described above, the spring plunger 205 is disposed below the container storage unit 204, which biases the sample container in the container storage unit 204 upward. When the tip of the pipette tip 171 is pressed against the membrane filter 302, the force pushes the sample container downward. Therefore, excessive force is not applied to the membrane filter 302 from the tip of the pipette tip 171 .

[0051] Because enzymes often have high viscosity and only a small amount is dispensed, they are difficult to remove from the tip of the pipette tip 171. In this embodiment, the enzyme is dispensed by lightly pressing the tip of the pipette tip 171 against the membrane filter 302 and pushing the sample container downward. Because the membrane filter 302 is porous, when the tip of the pipette tip 171 is brought into contact with the membrane filter 302 and the enzyme is discharged, the enzyme at the tip of the pipette tip 171 is taken up into the pores of the membrane filter 302 and removed from the pipette tip. Note that, since the tip of the pipette tip 171 is brought into contact with the membrane filter 302 that has captured the target component during enzyme dispensing, it is recommended to discard the used pipette tip 171 after each enzyme dispensing into one sample container and attach a new pipette tip 171 to prevent contamination.

[0052] After dispensing the enzyme into all sample containers, the pre-processing execution unit 135 subsequently operates the robot arm 120 to remove the spin column 300 on which the cleaved glycans have been captured from the microtube 400, and move (remove) it to a container storage section 204 in the sample rack 200 that was not used in the above process. The remaining microtubes 400 are then discarded in the consumables disposal unit 191. Furthermore, unused microtubes 400 that are placed in the tube rack placement section 150 are moved to each container storage section 204 in the sample rack 200. Thereafter, the retracted spin column 300 is set in a microtube 204 that is stored in the sample storage section 204 in the same position as before the retraction.

[0053] After the above process is completed, the sample container is identified again. The flow of the sample container identification is the same as above, and the type of sample container to be set in the container storage section 204 to be identified is also the spin column 300+microtube 400.

[0054] If the sample container type is determined to match in the sample container determination (YES in step 14 or step 18), the preprocessing execution unit 135 discards the used pipette tip 171 attached to the tip of the micropipette 161 into the consumables disposal unit 191 and attaches a new pipette tip 171. Then, a predetermined amount of reagent (eluent) stored in another sample liquid / reagent rack 183 is collected and dispensed into the sample container (spin column 300 placed on microtube 400) stored in the container storage unit 204. As a result, the target component (glycan cleaved by the enzyme) captured on the membrane filter 302 is collected in the microtube 400.

[0055] When the collection of the target components into the microtubes 400 in all of the container storage units 204 is completed, the preprocessing execution unit 135 subsequently returns the micropipettes 161 held by the robot arm 120 to the micropipette storage unit 160. Then, the spin columns 300 held in each container storage unit 204 are removed from the microtubes 400 by gripping and lifting them with the hand unit 121, and discarded in the consumable disposal unit 191.

[0056] After all the spin columns 300 have been discarded, the sample container is again determined. The flow of the sample container determination is the same as above, but here, the only type of sample container that should be set in the container storage section 204 to be determined is the microtube 400.

[0057] If the sample container type is determined to match in the sample container determination (YES in step 14 or step 18), the pre-processing execution unit 135 operates the robot arm 120 to move the microtube 400 containing the target component (cleaved glycan) contained in each container storage unit 204 to the pre-processed sample liquid storage unit 192.

[0058] Conventionally, when dispensing highly viscous liquids such as enzymes or minute amounts of liquid samples, analysts first contact the tip of a pipette tip with a membrane filter on which glycans are captured, and then operate the micropipette to inject the enzyme. However, the membrane filter 302 is made of glass fiber or other materials, and can be damaged if the tip of the pipette tip is pressed against it with too much force. On the other hand, if the tip of the pipette tip is separated from the membrane filter, droplets cannot be removed from the tip of the pipette tip. Therefore, analysts must carefully operate the micropipette while checking the position of the tip of the pipette tip, which is time-consuming and places a heavy burden on the analyst.

[0059] In contrast, in this embodiment, as described above, it is possible to automatically dispense minute amounts of highly viscous liquid, such as an enzyme, into the spin column 300 attached to the upper opening of the microtube 400. In this embodiment, a spring plunger 205 is provided below the container storage section 204, and while the sample container is biased upward, the tip of the pipette tip 171 is lightly pressed against the membrane filter 302 to dispense the reagent (enzyme). This prevents the tip of the pipette tip 171 from being pressed against the membrane filter 302 with a strong force and being damaged. Furthermore, because the tip of the pipette tip 171 is in contact with the membrane filter 302, droplets can be reliably released from the tip of the pipette tip.

[0060] Furthermore, in this embodiment, at the end of each stage of sample pretreatment that includes an operation of moving a sample container, the determination unit 136 photographs the sample rack 200 with the camera 500 to obtain an image and identifies the type of sample container held in each container holding unit 204. Then, the determination unit 136 reads the type of sample container that should be held in each container holding unit 204 at that stage from the pretreatment information 1311 stored in the memory unit 131, compares it with the identified type of sample container for each container holding unit 204, and determines whether the two match. Therefore, there is no risk of subsequent processing being performed without the correct sample container being placed due to some kind of trouble, resulting in wasted processing.

[0061] During the initial determination, the sample container is often placed in the container storage unit 204 in an unbalanced position, which can cause the determination unit 136 to mistakenly identify the type of sample container, resulting in an erroneous determination. If an error is output each time this occurs, unnecessary confirmation work must be performed before the analysis can be performed, placing a heavy burden on the analyst. Furthermore, the automatic processing is interrupted each time an error occurs, which reduces processing efficiency.

[0062] Therefore, in this embodiment, if the initial judgment result is a mismatch, the container attitude adjustment unit 137 corrects the variation in the attitude of the container storage unit 204, photographs the sample rack 200 again to obtain an image, and then a re-judgment is performed. Therefore, even if the initial judgment is an incorrect judgment, a correct re-judgment is performed, eliminating the need for the analyst to perform unnecessary confirmation work. Furthermore, if the sample container types match in the re-judgment, the process proceeds to the next step, so automatic processing continues and processing efficiency does not decrease. Furthermore, the number of re-judgments is limited to three, so there is no need to waste time by repeating re-judgments unnecessarily.

[0063] In the above example, the robot arm 120 was used to store and retrieve the sample containers in the container storage units 204, but depending on the contents of the pre-processing, a step may be included in which the analyst manually sets the sample containers in each container storage unit 204 and then sets the sample rack 200. In this case, if a step in which the analyst manually sets the sample containers is included, it is advisable to subsequently perform sample container determination in the same manner as described above.

[0064] In the above example, the type of sample container stored in each of the container storage units 204 used for pretreatment among those provided in the sample rack 200 is identified, and it is determined whether the type matches the type of sample container listed in the pretreatment information 1311. If the determination result indicates a mismatch, the attitude of the sample container is adjusted and re-determination is performed. However, these processes may be performed only for container storage units 204 at specific positions. Specifically, for example, the above processes may be performed only for container storage units 204 at locations where the angle between the optical axis and the vertical axis during image capture by the camera 500 is greater than a predetermined angle, or where the probability of an erroneous determination in a previous determination by the determination unit 136 or re-determination unit 138 is equal to or greater than a predetermined rate. When performing these processes, the container storage unit 204 to be subjected to sample container determination may be determined based on the determination target position information 1312 stored in the memory unit 131. It is also preferable to update the determination target position information 1312 each time an erroneous determination occurs in the sample container determination.

[0065] The above embodiment is merely an example and can be modified as appropriate in accordance with the spirit of the present invention.

[0066] In the above embodiment, the camera 500 is disposed approximately in the center of the top surface inside the housing 100. However, depending on the arrangement of the robot arm 120, etc., the sample rack 200 may be shaded by the robot arm 120 and unable to be photographed. In such cases, the camera 500 may be attached to the periphery of the top surface inside the housing 100 or to a side surface inside the housing 100. In these cases, the optical axis of the camera 500 photographing the sample rack 200 is tilted relative to the vertical direction, and the angle at which the sample containers in the container housing units 204 are photographed often differs for each container housing unit 204. Therefore, as in the above embodiment, it is necessary to obtain an image of each container housing unit 204 with the sample containers accommodated therein.

[0067] In the above embodiment, the container attitude adjustment unit 137 is configured to eliminate variations in the attitude of the sample containers by gripping and releasing the sample containers with the robot arm 120, but the attitude of the sample containers may be adjusted by other methods. For example, by using the hand unit 121 of the robot arm 120 to push the sample containers in the container storage unit 204 in a specific direction, all of the sample containers can be adjusted to an attitude in which they stand upright at a specific position in the container storage unit 204.

[0068] In the above embodiment, the process of identifying and determining the sample container is performed before the start of all stages of pre-processing in which the type of sample container to be accommodated in the container accommodation section 204 is different. However, the process of identifying and determining the sample container may also be performed at a specific timing that is pre-associated with the pre-processing procedure and contents stored in the memory section 131.

[0069] In the above embodiment, spin columns 300 and microtubes 400 are used as sample containers, but the present invention is not limited to these, and various types of containers such as column containers and vials can be used. Furthermore, the pretreatment of the sample described in the above embodiment is merely an example, and various types of pretreatment can be performed.

[0070] [Aspect] It will be apparent to those skilled in the art that the above-described exemplary embodiments are examples of the following aspects.

[0071] (Section 1) An automatic pipetting device according to one aspect of the present invention is an automatic pipetting device that performs a predetermined process on a sample, a container accommodating section capable of accommodating a sample container; an imaging unit that acquires an image of the container storage unit; a container identification unit that is constructed by learning an image of the container storage unit in a state where a sample container is stored therein, and that identifies the type of sample container stored in the container storage unit when an image of the container storage unit is input; a container storage information storage unit that stores container storage information, which is information on the type of sample container to be stored in the container storage unit before or after the processing; a determination unit that operates the imaging unit to acquire an image of the container storage unit before or after the processing, inputs the image to the container identification unit to identify the type of sample container stored in the container storage unit, and determines whether the identified type of sample container is the same as the type of sample container to be stored in the container storage unit in light of the container storage information; and a container attitude adjusting unit that performs an operation to adjust the attitude of the sample container accommodated in the container accommodating unit to a predetermined attitude; a re-determination unit that, when the determination unit determines that the type of the sample container is different, causes the container attitude adjustment unit to perform an operation to adjust the attitude of the sample container accommodated in the container accommodating unit to the predetermined attitude, operates the imaging unit to re-acquire an image of the container accommodating unit, inputs the image into the container identification unit to identify the type of sample container accommodated in the container accommodating unit, and re-determines whether the identified type of sample container is the same as the type of sample container to be accommodated in the container accommodating unit in light of the container accommodation information; Equipped with.

[0072] In the automatic dispensing device according to paragraph 1, container accommodation information, which is information on the type of sample container to be accommodated in the container accommodation unit before or after a predetermined sample processing, is stored in advance. Furthermore, the information is constructed by learning images of multiple types of sample containers individually accommodated in the container accommodation unit, and a container identification unit is provided in advance that identifies the type of sample container accommodated in the container accommodation unit when an image of the container accommodation unit is input. Then, before or after performing a predetermined sample processing, an image of the container accommodation unit is acquired and input to the container identification unit to identify the type of sample container accommodated in the container accommodation unit. Next, a determination unit determines whether the type of sample container identified in light of the container accommodation information is the same as the type of sample container to be accommodated in the container accommodation unit.

[0073] In the automatic dispensing device according to paragraph 1, if the determination unit determines that the type of sample container is incorrect, the container attitude adjustment unit performs an operation to adjust the attitude of the sample container accommodated in the container storage unit to a predetermined attitude, and then the re-determination unit performs a re-determination. Here, the predetermined attitude refers to a state in which the sample container is slightly tilted, for example, when the sample container is stored upright in the container storage unit. Therefore, even if the determination unit incorrectly identifies the type of sample container even though the correct type is stored in the container storage unit due to the sample container being stored in a tilted state in various directions, and the determination unit determines that the type of sample container is incorrect, the container attitude adjustment unit performs an operation to adjust the sample container accommodated in the container storage unit to a predetermined attitude, and then the re-determination is performed. This improves the accuracy of identifying the type of sample container in the container storage unit from an image of the container storage unit and determining whether the type is correct or not.

[0074] (Section 2) The automatic dispensing device according to paragraph 2 is the automatic dispensing device according to paragraph 1, When the re-determination unit determines that the types of the sample containers are different, the re-determination is performed up to a predetermined upper limit.

[0075] (Section 3) The automatic dispensing device according to paragraph 3 is the automatic dispensing device according to paragraph 2, The predetermined number of times is three or less.

[0076] The automatic dispensing device according to paragraph 2 can eliminate erroneous determinations that cannot be resolved by a single re-evaluation. If the type of sample container does not match even after three re-evaluations, it is highly likely that the sample container actually stored in the container storage unit 204 is incorrect, and performing re-evaluations four or more times is unlikely to change the result, and the more re-evaluations are performed, the more time it takes. In the automatic dispensing device according to paragraph 3, by limiting the number of re-evaluations to three or less, it is possible to more reliably eliminate erroneous determinations while avoiding the time required for re-evaluation from increasing.

[0077] (Section 4) The automatic dispensing device according to paragraph 4 is the automatic dispensing device according to any one of paragraphs 1 to 3, further comprising: a pre-processing information storage section in which pre-processing information relating to the contents of the multiple stages of processing to be performed on the sample and whether or not a judgment by the judgment section is necessary is recorded; Equipped with.

[0078] In the automatic dispensing device according to paragraph 4, it is possible to effectively check at a specific timing whether the correct type of sample container is contained in the container container portion based on the information stored in the memory portion.

[0079] (Section 5) The automatic pipetting device according to paragraph 5 is the automatic pipetting device according to any one of paragraphs 1 to 4, further comprising: a robot arm having a hand portion for gripping and moving the sample container; Equipped with The operation for adjusting the attitude of the sample container to a predetermined attitude is the operation of gripping and releasing the sample container with the hand portion.

[0080] Many automatic pipetting devices are equipped with a robot arm having a hand unit for grasping and moving sample containers. The automatic pipetting device according to paragraph 5 can effectively use this hand unit to adjust the position of the sample container without adding any additional components.

[0081] (Section 6) The automatic pipetting device according to paragraph 6 is the automatic pipetting device according to any one of paragraphs 1 to 5, further comprising: a sample rack placement section on which a sample rack having a plurality of the container storage sections is placed; a determination target position information storage unit in which determination target position information relating to the position of a container storage unit that is the target of the determination among the plurality of container storage units is stored; Equipped with The determination unit executes the determination only for a specific container storage unit based on the determination target position information storage unit.

[0082] In the automatic dispensing device according to paragraph 6, by limiting the position of the sample container at which the type of sample container is determined, the type of sample container in the container storage section at a position where erroneous determination is likely to occur can be determined effectively and in a short time. [Explanation of symbols]

[0083] 100…Case 120...Robot arm 121...Hand part 130...Control unit 131...Storage section 1311...Preprocessing information 1312...Determination target location information 133…Learning Execution Department 134…Container identification department 135...Preprocessing execution unit 136…Judgment section 137...Container posture adjustment section 138...Rejudgment Department 1391...input section 1392...Display section 140...Column rack placement section 141...Column rack 150...Tube rack placement area 151...Tube Black 160...Micropipette storage section 161...Micropipette 170...Tip rack placement section 171...Pipette tip 172...Tip rack 180...Sample solution / reagent rack placement area 181...Reagent container 182...Sample liquid container 183...Sample solution / reagent rack 191...Consumables Disposal Department 192...Pretreated sample liquid storage section 200...Sample rack 201...Main body 202...Heater 203...Heat transfer block 204...Container storage section 205...Spring plunger 220...Sample rack placement area 300...Spin column 301...Cylinder part 302...Membrane filter 400...microtube

Claims

1. An automatic dispensing device that performs a predetermined process on a sample, a container accommodating section capable of accommodating a sample container; an imaging unit that acquires an image of the container storage unit; a container identification unit that is constructed by learning an image of the container storage unit in a state where a sample container is stored therein, and that identifies the type of sample container stored in the container storage unit when an image of the container storage unit is input; a container storage information storage unit that stores container storage information, which is information on the type of sample container to be stored in the container storage unit before or after the processing; a determination unit that operates the imaging unit to acquire an image of the container storage unit before or after the processing, inputs the image to the container identification unit to identify the type of sample container stored in the container storage unit, and determines whether the identified type of sample container is the same as the type of sample container to be stored in the container storage unit in light of the container storage information; and a container attitude adjusting unit that performs an operation to adjust the attitude of the sample container accommodated in the container accommodating unit to a predetermined attitude; a re-determination unit that, when the determination unit determines that the type of the sample container is different, causes the container attitude adjustment unit to perform an operation to adjust the attitude of the sample container accommodated in the container accommodating unit to the predetermined attitude, operates the imaging unit to re-acquire an image of the container accommodating unit, inputs the image into the container identification unit to identify the type of sample container accommodated in the container accommodating unit, and re-determines whether the identified type of sample container is the same as the type of sample container to be accommodated in the container accommodating unit in light of the container accommodation information; An automatic dispensing device comprising:

2. 2. The automatic dispensing device according to claim 1, wherein when the re-determination unit determines that the types of the sample containers are different, the re-determination is performed up to a predetermined upper limit number of times.

3. 2. The automatic dispensing device according to claim 1, wherein the predetermined number of times is three or less.

4. moreover, a pre-processing information storage section in which pre-processing information relating to the contents of the multiple stages of processing to be performed on the sample and whether or not a judgment by the judgment section is necessary is recorded; The automated pipetting device of claim 1 , comprising:

5. moreover, a robot arm having a hand portion for gripping and moving the sample container; Equipped with 2. The automatic dispensing device according to claim 1, wherein the operation for adjusting the attitude of the sample container to a predetermined attitude is an operation of gripping and releasing the sample container with the hand unit.

6. moreover, a sample rack placement section on which a sample rack having a plurality of the container storage sections is placed; a determination target position information storage unit in which determination target position information relating to the position of a container storage unit that is the target of the determination among the plurality of container storage units is stored; Equipped with The automatic dispensing device according to claim 1 , wherein the determining unit performs the determination only for a specific container storage unit based on the determination target position information storage unit.

Images