Biological sample measurement device and biological sample measurement method

Abstract

In order to provide a compact, low-cost, and high-throughput biological sample measurement device and biological sample measurement method capable of detecting the state of a biological sample and the type of a container, the following configuration is adopted.　Provided are: a biological sample measurement device comprising a holder that holds a container capable of accommodating a biological sample, a gripping mechanism that grips and moves the container up and down, a camera that captures an image of the container accommodating the biological sample in a first state in which the container is held by the holder and a second state in which the container is gripped by the gripping mechanism and separated from the holder, and a processor that generates integrated coordinates obtained by integrating vertical coordinates of the images on the basis of the images captured in the first state and the second state, and calculates the liquid height of the biological sample in the container on the basis of the integrated coordinates; and a biological sample measurement method performed by the device.

Description

Biological sample measurement device and biological sample measurement method

[0001] The present invention relates to a biological sample measurement device and a biological sample measurement method.

[0002] For the purpose of improving the efficiency of clinical tests such as blood tests, technologies have been developed to automate the sample pretreatment process that was conventionally performed manually. One such technology is to automate the sample check that was conventionally performed by visual confirmation. The targets of the sample check are the volume and properties (jaundice, hemolysis, turbidity, etc.) of the sample. By confirming these before analysis such as biochemical analysis, it is possible to quickly take measures such as picking up samples that require re-sampling, and improve the test efficiency.

[0003] In addition, in automating the sample check, in addition to detecting the volume and properties of the sample, information on the type of container (shape, color, etc.) is also acquired and used for analyzing the volume and properties, so that the sample check can be performed more accurately and efficiently.

[0004] As such a technology, Patent Document 1 discloses "an imaging unit that images a container containing a biological sample from the side, a container type specifying unit that acquires information for specifying the type of the container, and an image captured by the imaging unit. A control unit having a processing unit that processes based on parameters corresponding to the container type specified by the container type specifying unit and a determination unit that determines the state of the contents in the container based on the processing in the processing unit."

[0005] It is described that this technology can "automatically and accurately check the state and volume of the sample simply by installing various types of blood collection tubes at the input position, and the user does not need to perform laborious operations such as calibration, and can input the blood collection tube into the sample test automation system." (Paragraph

[0011] of Patent Document 1).

[0006] International Publication No. 2015 / 072358

[0007] In automated specimen checking technology that detects the volume and properties of a specimen and acquires information on the type of container, Patent Document 1 has both an imaging unit for specimen checking and an imaging unit for automatic recognition of blood collection tubes. Therefore, space is required for the placement of two imaging units, which may result in a large device. In addition, time is required for the container to move between the two imaging units.

[0008] The present disclosure aims to provide a compact, low-cost, and high-throughput biological sample measurement device and method capable of detecting the state and container type of a biological sample.

[0009] The present invention, in order to achieve the above objectives, has the following configuration: A biological sample measuring device comprising: a holder for holding a container capable of containing a biological sample; a gripping mechanism for gripping the container and raising and lowering it; a camera for imaging the container containing the biological sample in a first state in which the container is held in the holder and a second state in which the container is gripped by the gripping mechanism and separated from the holder; and a processor for generating integrated coordinates by integrating the vertical coordinates of the images based on the images captured in the first and second states, and for calculating the liquid height of the biological sample in the container based on the integrated coordinates.

[0010] A method for measuring a biological sample, comprising: an imaging step of imaging a container containing a biological sample in a first state in which the container capable of containing a biological sample is held in a holder, and a second state in which the container is gripped by a gripping mechanism that grips the container and moves it up and down, and is separated from the holder; an integrated coordinate generation step of generating integrated coordinates by integrating the vertical coordinates of the images based on the images captured in the imaging step; and a liquid height calculation step of calculating the liquid height of the biological sample in the container based on the integrated coordinates generated in the integrated coordinate generation step.

[0011] According to the present invention, it is possible to provide a compact, low-cost, and high-throughput biological sample measuring device and biological sample measuring method that can detect the state of the biological sample and the type of container.

[0012] A diagram showing a schematic configuration example of a biological sample measurement device according to an embodiment. A diagram explaining an example of an imaging method. A flowchart showing an example of an imaging method. A diagram showing an example of an acquired image for the measurement target area. A schematic diagram showing the respective analysis ranges for the sample upper image and sample lower image. A flowchart showing the process of analyzing the sample properties and liquid volume in the processor. An explanatory diagram explaining the integrated coordinates of the sample lower image and sample upper image. A flowchart showing the stopper type discrimination process in the processor. A flowchart showing the process of detecting the diameter and length of the container in the processor. A schematic diagram showing a method for acquiring sample upper and sample lower images of a sample on the transport path. A schematic diagram explaining the camera arrangement when the imaging unit is placed in the sample transport device. A schematic diagram showing a configuration for imaging the entire periphery of a sample using two mirrors. A schematic diagram showing an example of placing the imaging unit using two mirrors in the sample transport device.

[0013] Embodiments of the present invention will be described below with reference to the drawings.

[0014] Figure 1 shows a schematic configuration of a biological sample measuring device according to Example 1. The biological sample measuring device detects the type of container 101 (stopper type, stopper color, shape) and the properties of the biological sample (normal, jaundice, chyle, hemolysis) and liquid volume.

[0015] Container 101 is for containing biological samples, and is, for example, a cylindrical blood collection tube for containing blood samples as biological samples. A label 102 is affixed along the shape of the outer surface of container 101. Examples of labels 102 include barcodes for identifying biological samples, as well as pre-labels that are attached to blood collection tubes. In addition, container 101 has an exposed surface 109 on at least a portion of its circumference where the label 102 is not affixed, allowing the biological sample to be seen from the outside. Note that some containers do not have a label 102 affixed.

[0016] The biological sample measuring device acquires information such as the diameter and length (height) of the container 101, the color of the container 101, the color and type of the stopper of the container 101, and the type of container 101 identified from this information. Since there are various types of containers 101 depending on the analysis method of the biological sample, acquiring this information allows for more accurate detection of the liquid volume and properties of the biological sample. Furthermore, by transmitting the acquired information to a sample transport device that transports and pre-processes the sample, it becomes possible to perform sample handling and pre-processing with higher accuracy and quality.

[0017] Specifically, information about the shape of the container 101, such as its diameter and the shape of its bottom, can be used when detecting the liquid volume of a biological sample to measure the liquid volume more accurately. In addition, the color of the container 101 can be used to determine the properties of the measurement target 103 without being affected by the color of the container 101, by performing color correction on the container 101.

[0018] The biological sample stored in container 101 is a blood sample or a urine sample. An example of a blood sample is described below. Before opening (analysis), the blood sample is separated into multiple component layers by centrifugation or the like, for example, three layers consisting of a blood clot, a separating agent, and serum or plasma, or two layers consisting of a blood clot and serum or plasma. Of these, the serum or plasma becomes the measurement target 103 of the biological sample measuring device according to this embodiment. However, the single layer (whole blood) without separation may also be the measurement target 103. The blood clot is formed in the lower layer when the biological sample is centrifuged, and the separating agent is added to separate the blood clot from the measurement target 103.

[0019] The biological sample measuring device detects the volume and color of the sample to be measured 103 and determines its properties. As mentioned above, by utilizing the container information obtained by container 101, the volume detection and property determination can be performed with high accuracy.

[0020] In determining the characteristics of the sample, the characteristics (normal, jaundice, hemolysis, chyle, etc.) are determined based on the color of the sample 103. Based on the determination results, normal samples and abnormal samples (jaundice, hemolysis, chyle, etc.) can be separated. This allows for the quick identification of abnormal samples and arrangements for re-collection, thereby improving testing efficiency.

[0021] As shown in Figure 1, the biological sample measurement device according to this embodiment consists of an imaging unit 108 and a processor 107. The imaging unit 108 consists of a camera 104, a gripping mechanism 105, and a holder 106. Hereinafter, the container 101 containing the biological sample will be referred to as the sample.

[0022] Camera 104 captures color images of the container 101 and the object to be measured 103, and is an imaging device such as a CCD camera or a CMOS camera. Camera 104 transmits the captured color images to the processor 107. Camera 104 can also be used to capture images of the label 102 attached to the container 101 and read information (such as a barcode).

[0023] Camera 104 acquires both images of the sample container 101 and the measurement target 103 in the state where they are held by the gripping mechanism 105 (referred to as the "second state") (also referred to as the "lower sample image" because it captures the area below the sample container), and images of the sample container 101 and the measurement target 103 in the state where they are mounted on the holder 106 (referred to as the "first state") (also referred to as the "upper sample image" because it captures the area above the sample container). Camera 104 may capture a two-dimensional image with the sample stationary, or it may capture a video of the sample in motion. In the case of a video, the processor 107 acquires the image to be analyzed and performs the image processing described below.

[0024] The gripping mechanism 105 has the function of gripping the specimen from above in the vertical direction and moving it in the horizontal and vertical directions. The gripping mechanism 105 grips the specimen stored in the biological specimen storage area (not shown) from above, moves it to the imaging unit 108, and lowers the specimen into the field of view of the camera 104. The camera 104 captures a lower image of the specimen while it is being gripped by the gripping mechanism 105 (second state). After imaging, the gripping mechanism 105 lowers the specimen further and places the container 101 on the holder 106, and the camera 104 captures a higher image of the specimen while the container 101 is placed on the holder 106 (first state).

[0025] Note that the order in which the upper and lower images of the specimen are captured can be reversed. If the upper image of the specimen is captured first while it is placed on the holder 106 (first state), the upper image is captured first, and then the gripping mechanism 105 raises the specimen placed on the holder 106 while gripping it from above, moves it within the field of view of the camera 104, and captures the lower image of the specimen. In this case, after capturing, the gripping mechanism 105 may lower the specimen to the height of the holder 106 and place it back on the holder 106, or it may transfer the gripped specimen to a specimen transport path or biological sample storage area (not shown).

[0026] The holder 106 fixes and holds the specimen. The camera 104 takes images in both the state when it is grasped by the grasping mechanism 105 (second state) and the state when it is mounted on the holder 106 (first state). By controlling the operating positions of the holder 106 and the grasping mechanism 105 so that the distance from the camera 104 to the container 101 is the same in both states, the scale of the upper and lower images of the specimen can be made the same. Note that the annotations for the first and second states will be omitted below.

[0027] The processor 107 analyzes the sample top image and sample bottom image generated by the camera 104 and outputs the container information and the determination results of the liquid volume and properties of the biological sample.

[0028] Figure 2 is a schematic diagram showing the method for acquiring the upper image 202 and lower image 201 of the specimen. First, the container 101 (specimen) containing the biological sample is moved to the imaging unit 108 while being gripped from above by the gripping mechanism 105 (Figure 2(a)). Next, the lifting mechanism of the gripping mechanism 105 is used to lower the specimen so that the lower side (the area exposed from the gripping mechanism) is within the field of view of the camera 104 (Figure 2(b)). In this way, the lower image 201 of the specimen is acquired. Subsequently, the gripping mechanism 105 is lowered further to place the specimen on the holder 106, and the upper image 202 of the specimen is acquired in the mounted state (Figure 2(c)).

[0029] As described above, after first acquiring the upper image 202 of the sample with the holder 106 installed, the container 101 may be grasped and raised with the gripping mechanism 105 to acquire the lower image 201 of the sample.

[0030] The processor 107 uses both the upper sample image 202 and the lower sample image 201 to detect the type of container 101 (stopper type, stopper color, shape) and the properties of the biological sample (normal, jaundice, chyle, hemolysis) and the volume of liquid in the sample image captured across both (Figure 2(d)).

[0031] Figure 3 is a flowchart illustrating the method for acquiring the upper sample image 202 and lower sample image 201 as described in Figure 2. Figure 3(a) shows the process of acquiring the upper sample image after acquiring the lower sample image 201, while Figure 3(b) shows the process of acquiring the lower sample image 201 after acquiring the upper sample image 202.

[0032] In Figure 3(a), the gripping mechanism 105 first grasps the sample and moves it to the imaging unit (step S1), then the gripping mechanism is lowered (step S2) to capture a lower image 201 of the sample (step S3). Next, the gripping mechanism 105 is lowered toward the holder 106 (step S4) and mounted on the holder 106 (step S5). An upper image of the sample is captured (step S6), and after imaging, the sample is moved (step S7). At this time, the sample may be grasped again with the gripping mechanism 105 and moved, or it may be moved manually or by other methods while mounted on the holder 106.

[0033] In Figure 3(b), the sample is placed in the holder 106 (step S11), and an image of the top of the sample 202 is captured while the sample is in place (step S12). Then, the gripping mechanism 105 is lowered to the position of the sample (step S13), and the sample is grasped (step S14). While the sample is grasped, the sample is raised to the field of view of the camera 104 (step S15), and an image of the bottom of the sample 201 is captured (step S16). After imaging, the sample may be transported to another area (step S17), or it may be returned to the holder 106 and placed back into the holder 106. Furthermore, the method of loading the sample into the holder 106 in step S11 may be to transport it from a sample storage area (not shown) using the gripping mechanism 105, or to load it manually.

[0034] Furthermore, the destination of the sample after capturing the upper and lower images may be determined based on the volume and properties of the biological sample detected by the processor 107. For example, if there are no problems with the volume or properties, the sample may be moved to a space where opening and dispensing are performed for biochemical analysis, etc. If there are problems with the volume or properties, the sample may be moved to an abnormal sample storage area. This enables automated sorting of samples, leading to improved testing efficiency and increased reliability of test results.

[0035] Figure 4 shows an example of integrating the upper sample image 202 and the lower sample image 201. The upper sample image 202 is acquired when the sample is placed in the holder 106, and the lower sample image 201 is acquired when the sample is grasped by the gripping mechanism 105. By using both the upper sample image 202 and the lower sample image 201, information from the bottom of the sample to the top of the stopper can be obtained without the sample being obscured by the gripping mechanism 105 or the holder 106.

[0036] In the case of blood samples, the volume and properties of the serum or plasma, which is the supernatant after standing or centrifugation of a whole blood sample, become the measurement target 103. However, in addition to the volume of the measurement target 103, the volume of the blood clot 401 and the separation material 402 differ from sample to sample. Therefore, whether the measurement target 103 is imaged in the upper sample image 202 or the lower sample image 201 differs depending on the sample. Figure 4(a) shows an example in which the measurement target 103 is imaged across the upper sample image 202 and the lower sample image 201, while Figures 4(b) and 4(c) show examples in which it is imaged only in the lower sample image 201 or the upper sample image 202.

[0037] The processor 107 uses the integrated image information of the acquired sample upper image 202 and sample lower image 201 to identify the position of the measurement target 103, determine the liquid volume from the liquid height, and determine the properties from the color. Note that analyzing the measurement target 103 requires imaging the area where the measurement target 103 is exposed through the gap in the label 102 (exposed surface 109 in Figure 1).

[0038] Methods for imaging the exposed surface of the object to be measured 103 with the camera 104 include aligning the orientation of the object using a mechanism or by hand, rotating the object in the circumferential direction while it is mounted on the holder 106 or held by the gripping mechanism 105 to acquire an image of the entire circumference, and arranging multiple cameras 104 to image the entire circumference of the object.

[0039] Figure 5 is a schematic diagram showing the analysis ranges in the upper sample image 202 and the lower sample image 201, respectively. As described above, the position of the measurement target 103 in the image differs depending on the sample, so in order to accurately detect the liquid height, it is necessary to image the entire area from the bottom of the sample to below the stopper without any omissions. For this reason, it is necessary to set the gripping position of the gripping mechanism 105 and the height of the holder 106 so that the entire sample can be imaged, including the lower part of the sample exposed from the gripping mechanism 105 and the upper part of the sample exposed from the holder 106.

[0040] Although the length (height) of the container 101 varies depending on the type of container, by keeping the exposed length from the gripping mechanism 105 (the length of the part of the container 101 that is not hidden by the gripping mechanism) constant, it is possible to accommodate various types of containers 101 without changing the settings according to the container 101.

[0041] Next, the property detection and liquid volume detection processes performed by the processor 107 will be explained using Figure 6. Figure 6 is a flowchart showing the process by which the processor 107 detects the properties and liquid volume of the sample to be measured 103 from the upper sample image 202 and the lower sample image 201.

[0042] First, the processor 107 acquires the sample image 202 and the sample image 201 generated by the camera 104 (step S21). Next, the processor 107 identifies the position of the label 102 from the sample image 202 and the sample image 201 (step S22). The identification of the label 102 can be done using the color features of the label 102, edge detection, barcode detection, etc.

[0043] Next, the liquid surface of the measurement target 103 is searched for in the sample lower image 201 (step S23). The liquid surface is indicated by the lower and upper surfaces of the measurement target area in the image. The liquid surface is detected by gradient or edge detection of color features at the boundary between the measurement target area (e.g., serum area) and other areas (blood clot area, air layer area, etc.). Depending on the sample, both the upper and lower surfaces of the liquid surface may be present in the sample lower image 201, only the lower surface may be present, or neither upper nor lower surface may be present. Next, the liquid surface of the measurement target 103 is searched for in the sample upper image 202 (step S24). The method for detecting the liquid surface is the same as in step S24, but if both the upper and lower surfaces of the liquid surface are detected in the sample lower image 201 in step S24, step S24 may be omitted. Also, if only the lower surface is detected in the sample lower image 201, only the upper surface is detected in step S24.

[0044] Next, the horizontal region of the measurement target 103 (the range of the measurement target 103 in the image) is identified from the detection result of the label 102 in step S22, and the vertical region of the measurement target 103 is identified from the liquid level detection results in steps S23 and S24 (step S25).

[0045] Subsequently, the processor 107 acquires the color of the measurement target 103 and classifies the properties based on the color (step S26). The method for determining the properties will be described. The correspondence between the classification of the properties of the measurement target 103 and the range (threshold) of the color feature amount corresponding to each classification is stored in the storage unit in advance. Therefore, the processor 107 can determine the properties of the measurement target 103 by comparing the acquired information regarding the color (color feature amount) with the correspondence stored in the storage unit. For example, when the measurement target 103 is serum, since the serum color is characteristic depending on the serum state (normal, hemolysis, milkiness, jaundice, etc.), if the corresponding relationship is stored in advance, it is possible to determine the serum state.

[0046] Next, in order to detect the liquid height and calculate the liquid volume from the specimen image captured across the upper-specimen image 202 and the lower-specimen image 201, the vertical coordinates of the upper-specimen image 202 and the lower-specimen image 201 are integrated (step S27). FIG. 7 is an explanatory diagram for explaining the integrated coordinates with respect to each of the lower-specimen image 201 and the upper-specimen image 202. Here, the method for determining the integration position of the lower-specimen image 201 and the upper-specimen image 202 may be determined based on the exposed height of the specimen in the gripping state and the holder 106 mounted state. That is, it can be determined based on the height of the portion of the container 101 that is not hidden by the gripping mechanism 105 and the holder 106 (specimen exposed height).

[0047] Also, the integration position may be set for each specimen, or when the specimen exposed height from the gripping mechanism 105 and the specimen exposed height in the holder 106 mounted state do not change for each specimen, a common one can be used among the specimens. When setting for each specimen, for example, the bottom position may be detected from the lower-specimen image 201 for each specimen, and the upper part of the holder 106 or the position of the stopper may be detected from the upper-specimen image 202, and these may be set as a reference. Through the above coordinate integration process, it becomes possible to detect the liquid height and liquid volume of the measurement target 103 from the specimen image captured across the upper-specimen image 202 and the lower-specimen image 201.

[0048] Subsequently, the processor 107 detects the liquid height using the liquid surface position of the measurement target 103 detected in steps S23 and S24 and the integrated coordinates of the upper and lower specimen images (step S28). The liquid height is the height between the liquid surfaces of the measurement target 103. After obtaining it in terms of the number of pixels, the number of pixels is converted into height [mm] using the liquid height conversion coefficient. The liquid height conversion coefficient is stored in advance in a storage unit (not shown).

[0049] After that, the processor 107 calculates the liquid volume from the information on the liquid height (step S29). To calculate the liquid volume, in addition to the liquid height, information on the inner diameter of the container 101 is required. The information on the inner diameter is obtained by accessing the type of the container 101 specified by the method described later and the inner diameter data stored in a storage unit (not shown) associated with the type. Alternatively, if the type of the container 101 is limited, even without using the type result of the container 101, information input by an input device (not shown) or the inner diameter information stored in a storage unit in advance may be used.

[0050] Furthermore, before step S28, the inclination of the specimen is detected using edge detection or line detection from the upper specimen image 202 and the lower specimen image 201, and after performing inclination correction, the liquid height is detected, so that the liquid height and the liquid volume can be detected with higher accuracy.

[0051] Also, from the lower specimen image 201, the shape of the bottom of the container 101 is detected, and by calculating the liquid volume based on the shape of the bottom in addition to the inner diameter of the container 101, the liquid volume can be detected with higher accuracy. As a method for detecting the shape of the bottom, there are edge detection and line detection of the container 101, and it may be a process for detecting the contour of the specimen in combination with the above-mentioned process for detecting the inclination of the specimen.

[0052] According to the above, it is possible to detect the properties and the liquid volume from the specimen images captured across the upper specimen image 202 and the lower specimen image 201. In FIG. 6, after classifying the properties in step S26, the liquid volume calculation process is performed in step S29, but the property classification process may be executed after the liquid volume calculation process. Also, the detection results of the liquid volume and the properties detected by the processor 107 can be transmitted to a control mechanism (not shown) of the gripping mechanism 105 and the holder 106, and the conveyance path of the specimen after measurement can also be determined.

[0053] Next, the method for analyzing the type of container 101 (stopper type, stopper color, shape) performed by the processor 107 will be described. The processor 107 acquires information about the container 101 from the upper sample image 202 and lower sample image 201 captured by the camera 104. The analysis targets are the stopper type, stopper color, and the diameter and length of the container 101, and the type of container 101 is identified from this information.

[0054] First, an example of a method for identifying the type of valve stopper will be explained using Figure 8. Figure 8 is a flowchart showing an example of identifying the type of valve stopper by template matching using a template of valve stopper images registered during calibration.

[0055] First, let's explain the calibration shown in Figure 8(a). Calibration is performed in advance before measuring the sample. The container 101 to be used is imaged by the imaging unit 108 (step S31). The images to be acquired may be both an image of the top and bottom of the container, similar to the sample imaging method shown in Figure 3, or, if only the stopper type, diameter and length are to be analyzed, only an image of the top of the container may be acquired. The stopper region is detected from the acquired image (step S32), and preprocessing such as cropping, edge detection, and contrast adjustment is performed (step S33).

[0056] The plug region can be extracted based on the color and shape of the plug, or by setting the coordinates of the plug in the image as fixed values. Alternatively, it can be specified using an input device (not shown) while checking the captured image. Preprocessing, such as cropping or reducing the resolution, can reduce the computational load during template matching and improve the throughput during measurement. Furthermore, preprocessing such as vertical or horizontal edge detection or contour enhancement to emphasize the features of the plug of container 101, background color removal, or image color adjustment can improve the accuracy of template matching.

[0057] In step S34, the pre-processed image and data on the type of container 101 are stored as template data in a storage unit (not shown) or similar. The image stored may be the image captured before pre-processing. Furthermore, one template image may be used for each type, or multiple images with different pre-processing methods may be prepared.

[0058] Furthermore, as template data, information such as the type of inspection associated with the container 101, the stopper color, label color, and stopper characteristics can also be saved as text data. The calibration image can also be saved along with the template data after performing the diameter and length detection process for the container 101, which will be performed in the processing described later. By registering this information, it becomes possible to refer to this information together when analyzing the stopper type during measurement.

[0059] Next, a method for determining the type of plug will be explained. Figure 8(b) shows an example of a process for determining the plug type by analyzing a sample image captured by the imaging unit 108. First, a sample image 202 is acquired (step S41), the plug region is extracted from the sample image 202 (step S42), and preprocessing is performed (step S43). The method for extracting the plug region and the preprocessing method are the same as in step S32 of Figure 8(a). The preprocessing method is the same as that used for the template image acquired in calibration. Subsequently, template data is acquired from a storage unit (not shown) or the like (step S44), and template matching is performed with the template data (step S45). One method of template matching is to calculate the cross-correlation coefficient between the plug data of the sample and the template. In this way, the plug type can be determined by selecting the most similar template.

[0060] Next, an example of a method for detecting the diameter and length of the container 101 will be explained using Figure 9. First, an image 202 of the sample is acquired (step S51), and edge-enhanced images are obtained in both the horizontal and vertical directions by edge detection (step S52). Next, the height (top position) of the container 101 is detected by searching for edge positions in the vertical direction of the horizontal edge-enhanced image, and the length is determined according to the height (step S53). Furthermore, the positions of both ends of the container width are identified by searching for edge positions in the horizontal direction from the vertical edge-enhanced image, and the diameter is detected (step S54). By doing so, information on the length and diameter of the container 101 can be obtained.

[0061] Furthermore, the color of container 101 can also be obtained as information about container 101. The material of container 101 varies depending on the type, and can be transparent or opaque. If information on the stopper area and the width of container 101 can be obtained using the method described above, the area in the image where container 101 exists can be identified. After identifying the area of ​​the container, the color of the area without the biological sample (such as the area below the stopper area) can be obtained to identify container 101. Note that even without performing the color detection process, it is also possible to obtain this information as information linked to the stopper type and container type by registering it in the template data saved in the calibration described above or in the database for identifying the container type described later.

[0062] The color of the container 101 obtained in this way can be used for color correction of the measurement target 103 and adjustment of the color threshold for property classification when classifying the properties of the measurement target 103, thereby enabling more accurate discrimination.

[0063] Next, we will explain how to identify the type of container 101 based on the detected stopper type, stopper color, and container 101 shape. The type is determined by referring to a pre-prepared database that records information about containers 101. The information about container 101 includes the color, stopper type, and shape of the stopper being measured. The measurement results and the database are referred to to select the appropriate container 101 type.

[0064] Furthermore, by recording the color, bottom shape, and inspection type of the container 101 in the database, this information can be obtained as container type-specific data once the container type is determined, even without detecting it during measurement. This information can be used for analysis within the biological sample measuring device, as well as for pre-processing in transport devices, etc. For example, it becomes possible to change the transport route depending on the type of container 101, or to change the holder 106 mounted on the container depending on its shape.

[0065] To achieve efficient and high-quality specimen testing, there is a specimen transport device that automates the specimen preprocessing process and transport to the analyzer. Figure 10 is a schematic diagram showing an example in which the imaging unit 108 is arranged inside the specimen transport device. The specimen transport device described in Figure 10 has a gripping mechanism 105 for gripping and moving specimens, a holder 106 for mounting specimens, and a transport path 1001 for moving the holder 106. By making the transport path 1001 line belt-shaped, the holder 106 on the transport path 1001 can be transported along the transport path 1001.

[0066] Figure 10 also shows an example of a method for acquiring an upper sample image 202 and a lower sample image 201 of a sample on the transport path 1001. The imaging unit 108 uses the gripping mechanism 105 and holder 106 of the sample transport device to capture the lower sample image 201 and the upper sample image 202 in the gripped state and in the holder 106-mounted state. The gripping mechanism 105 moves the sample from a sample storage area (not shown) or a pre-processing area to the imaging unit 108 while gripping it, captures the lower sample image 201, and then lowers the sample to the holder 106. Alternatively, the sample may be gripped while mounted on the holder 106 and moved to the imaging unit 108, or the upper sample image 202 may be captured first while the sample is mounted on the holder 106, and then the sample is gripped to acquire the upper sample image 202.

[0067] After imaging by the imaging unit 108, the sample is transported along the transport path 1001 while mounted on the holder 106 and moved to the next pre-processing step. If the analysis results from the biological sample measuring device indicate that a sample has a small liquid volume or abnormal properties, the transport destination can be changed based on the results to pick up the abnormal sample.

[0068] Figure 11 is a schematic diagram illustrating the placement of the camera 104 when the imaging unit 108 is placed in the specimen transport device. The camera 104 is positioned so that its height does not overlap with the transport path 1001, and specimen images are acquired with the specimen placed in the holder 106 located on the transport path 1001. By placing a stopper for the holder 106 on the transport path 1001 and turning on the stopper only during imaging to keep the holder 106 stationary, specimen images can be acquired with greater accuracy.

[0069] With the above configuration, by utilizing the gripping mechanism 105 and holder 106 of the sample transport device to capture the upper sample image 202 and lower sample image 201 on the transport path 1001, it is possible to miniaturize and reduce costs without installing additional mechanisms for the biological sample measurement device.

[0070] Furthermore, by imaging the sample on the sample transport path 1001, the sample movement time between the imaging unit 108 and the transport path 1001 can be shortened, enabling efficient measurement without reducing the throughput of the sample transport device.

[0071] A label 102 is usually attached to the sample. Label types include barcode labels for sample identification and pre-labels that are initially attached to the container 101. Therefore, in order to analyze the properties and volume of the object to be measured 103, it is necessary to image the area where there is no label and the entire height of the object to be measured 103 is exposed (label opening).

[0072] To image the label opening, one method is to orient the label opening towards the camera, or to image the entire perimeter of the sample. Figure 12 is a schematic diagram showing a configuration that uses two mirrors to image the back of the camera 104 and image the entire perimeter of the sample.

[0073] In addition to the gripping mechanism 105, holder 106, camera 104, and processor 107, the system also includes mirrors 1201 and 1202, which are positioned opposite each other at the back of the specimen. The camera 104 simultaneously images both the specimen positioned in front and the specimen reflected in the mirrors 1201 and 1202.

[0074] Furthermore, while the camera 104 is held by the gripping mechanism 105, it captures an upper image 202 of the specimen using the front of the camera 104 and the mirrors 1201 and 1202, and while the specimen is mounted on the holder 106, it captures a lower image 201 of the specimen using the front of the camera 104 and the mirrors 1201 and 1202. This makes it possible to image the entire periphery of the specimen and the entire top and bottom surface of the specimen.

[0075] Figure 13 is a schematic diagram showing an example of arranging the configuration of Figure 12 in a specimen transport device. By positioning the camera 104 and mirrors 1203 and 1204 so that their height positions do not overlap with the transport path 1001, specimens on the transport path 1001 can be imaged without obstructing the transport of the holder 106 and specimens. Furthermore, by positioning the mirrors 1203 and 1204 within the field of view of the camera 104, an image of the entire surroundings of the specimen can be captured with a single image captured by the camera 104, thereby reducing the imaging time.

[0076] With the above configuration, imaging of the entire periphery and top and bottom of the specimen becomes possible by capturing two images: the upper image 202 and the lower image 201 of the specimen. This improves throughput compared to methods that capture the entire periphery while rotating the specimen. Furthermore, by placing mirrors 1203 and 1204 within the field of view of camera 104, it becomes possible to capture the entire periphery using only camera 104, thus achieving miniaturization and low cost.

[0077] In Examples 1, 2, and 3 described above, the present invention was applied to cases where two images, an upper image 202 of the sample and a lower image 201 of the sample, were captured and analyzed. However, in order to detect the type of container 101 (stopper type, stopper color, shape) and the properties of the biological sample (normal, jaundice, chyle, hemolysis) and liquid volume, it is sufficient to obtain information on the entire upper and lower region of the sample, or information on the entire upper and lower region and the entire surrounding area, based on the gripping state by the gripping mechanism 105 and the mounting state in the holder 106. Therefore, the present invention can also be applied to cases where three or more sample images are captured and analyzed, or where multiple images extracted from a video are captured and analyzed.

[0078] 101...Container, 102...Label, 103...Measurement target, 104...Camera, 105...Gripping mechanism, 106...Holder, 107...Processor, 108...Imaging unit, 201...Bottom image of sample, 202...Top image of sample, 401...Blood clot, 402...Separation material, 1001...Transport path, 1201...Mirror

Claims

1. A biological sample measuring device comprising: a holder for holding a container capable of containing a biological sample; a gripping mechanism for gripping the container and raising and lowering it; a camera for imaging the container containing the biological sample in a first state in which the container is held in the holder, and in a second state in which the container is gripped by the gripping mechanism and separated from the holder; and a processor for generating integrated coordinates by integrating the vertical coordinates of the images based on the images captured in the first and second states, and for calculating the liquid height of the biological sample in the container based on the integrated coordinates.

2. A biological sample measuring device according to claim 1, wherein the processor analyzes at least one of the shape data of the container and the color data of the stopper of the container based on the image captured in the first state.

3. A biological sample measuring device according to claim 1, wherein the processor identifies a measurement target area of ​​the biological sample contained in the container and analyzes the color data of the measurement target area based on the images captured in the first state and the second state.

4. A biological sample measuring device according to claim 3, wherein the processor searches for the position of the interface of the measurement target area based on the images captured in the first state and the second state, and identifies the measurement target area based on the image of the interface.

5. A biological sample measuring device according to claim 1, characterized in that the processor identifies the position of a label attached to the container based on the images captured in the first state and the second state.

6. A biological sample measuring device according to claim 1, wherein the processor detects the tilt of the container based on the images captured in the first state and the second state, corrects the integrated coordinates from the detected tilt, and calculates the liquid height.

7. A biological sample measuring device according to claim 1, characterized in that the processor analyzes the diameter and length of the container based on the images captured in the first state and the second state.

8. A biological sample measuring device according to claim 1, wherein the processor identifies at least one of the type of stopper and the color of the stopper of the container based on the image captured in the first state, and identifies the type of container based on at least one of the identified type of stopper and the color of the stopper.

9. A biological sample measuring device according to claim 3, wherein the processor detects the shape of the bottom of the container based on the image captured in the second state, and calculates the liquid volume of the measurement target area based on the detected bottom shape.

10. A biological sample measuring device according to claim 8, wherein the processor identifies the type of stopper by template matching using a template of the stopper of the container and information on the type of stopper of the container that has been acquired in advance.

11. A biological sample measuring device according to claim 1, characterized in that the processor determines the gripping position of the gripping mechanism and the height of the holder such that the images of the biological sample captured in the first state and the second state on the integrated coordinate system are continuous.

12. A biological sample measuring device according to claim 1, wherein the camera has a mirror that enables imaging of the back side of the container, and the processor simultaneously images the biological sample in front of the camera and the biological sample projected onto the mirror.

13. A biological sample measuring device according to claim 12, characterized in that the angle of view of the camera is determined based on the height of the sample exposed from the holder and the gripping mechanism, and the coordinates of the mirror.

14. A method for measuring a biological sample, comprising: an imaging step of imaging a container containing a biological sample in a first state in which the container capable of containing a biological sample is held in a holder, and a second state in which the container is gripped by a gripping mechanism that grips the container and moves it up and down, and is separated from the holder; an integrated coordinate generation step of generating integrated coordinates by integrating the vertical coordinates of the images based on the images captured in the imaging step; and a liquid height calculation step of calculating the liquid height of the biological sample in the container based on the integrated coordinates generated in the integrated coordinate generation step.

15. A method for measuring a biological sample according to claim 14, characterized in that it includes a container data analysis step of analyzing at least one of the shape data of the container and the color data of the stopper of the container based on the image captured in the first state by the imaging step.

16. A method for measuring a biological sample according to claim 14, characterized in that it includes a measurement target analysis step of identifying a measurement target area of ​​the biological sample contained in the container and analyzing the color data of the measurement target area based on the images captured in the imaging step in the first state and the second state.

Images