In-container sample state determination device, in-container sample state determination method, and sample analysis system

Abstract

Provided are more reliable in-container sample state determination device, in-container sample state determination method, and sample analysis system. The in-container sample state determination device for determining the state of a sample in a container comprises: an image acquisition unit for acquiring an image obtained by photographing the inside of the container from the outside; a container identification unit for analyzing the acquired image and identifying the type of the container; a sample region detection unit for analyzing the acquired image and detecting a region where the sample is present in the container; a detection accuracy determination unit for analyzing the acquired image and determining the accuracy for detecting the internal state of the container; an object detection unit for analyzing the acquired image and detecting an object in the container; and an output unit for determining the state of the sample in the container on the basis of the object detection result by the object detection unit and the detection accuracy determination result by the detection accuracy determination unit, and outputting the determination result.

Description

Apparatus for determining the state of a sample in a container, method for determining the state of a sample in a container, and sample analysis system

[0001] The present invention relates to an apparatus for determining the state of a sample in a container, a method for determining the state of a sample in a container, and a sample analysis system.

[0002] In systems for analyzing samples such as specimens, there is a growing need to estimate the amount of sample in a container such as a test tube for efficient analysis. Therefore, methods for estimating the amount of sample in a container have been proposed (Patent Documents 1 and 2).

[0003] Japanese Patent Laid-Open No. 9-133687 Japanese Patent Laid-Open No. 2005-265813

[0004] The technology described in Patent Document 1 detects the boundary between serum and other regions based on saturation information of an image. The technology described in Patent Document 2 detects the serum region by moving a transmission infrared sensor up and down the test tube. In this way, the conventional technology detects the upper and lower ends of the serum region and estimates the amount of sample in the test tube from the distance between the upper and lower ends and the thickness (inner diameter) of the test tube.

[0005] However, in recent years, containers with structures such as mechanical separators inside have also come into use, rather than simply shaped containers. They may also contain separating agents, beads, fibrin, etc. Therefore, when inserting a dispensing nozzle into a container to obtain a sample, care must be taken to avoid contact with the structures, separating agents, or other objects. However, if a label is attached to the outside of the container or if the sample inside the container is cloudy and lacks transparency, it can be difficult to accurately detect the presence or location of objects.

[0006] Therefore, the present invention provides a more reliable device for determining the state of a sample in a container, a method for determining the state of a sample in a container, and a sample analysis system.

[0007] In order to solve the above problem, an apparatus for determining the state of a sample in a container according to one aspect of the present invention is an apparatus for determining the state of a sample in a container, and comprises an image acquisition unit that acquires an image of the inside of the container taken from the outside, a container identification unit that analyzes the acquired image and identifies the position and type of the container, a sample area detection unit that analyzes the acquired image and detects the area in the container where the sample is present, a detection accuracy determination unit that analyzes the acquired image and determines the accuracy when determining the internal state of the container, an object detection unit that analyzes the acquired image and detects an object in the container, and an output unit that determines the state of the sample in the container based on the object detection result by the object detection unit and the detection accuracy determination result by the detection accuracy determination unit, and outputs the determination result.

[0008] According to the present invention, the state of the sample in the container can be determined based on the result of detecting an object in the container and the result of determining the accuracy of detecting the internal state of the container.

[0009] 19 is an overall configuration diagram of a sample analysis system including a sample status determination device. FIG. 19 is a functional block diagram of the sample status determination device. FIG. 19 is an example of a sample contained in a container including a structure. FIG. 19 is an example of a cloudy sample contained in a container including a structure. FIG. 19 is an example of a container including a structure with a label affixed to the outside. FIG. 19 is an explanatory diagram showing how a container is identified. FIG. 19 is an explanatory diagram showing how a sample region is detected. FIG. 19 is an explanatory diagram showing a sample acquisition permitted region when there is no object in the container. FIG. 19 is an explanatory diagram showing a sample acquisition permitted region when there is an object in the container. FIG. 19 is an explanatory diagram showing an example of detecting an object in a container. FIG. 19 is an explanatory diagram showing an example of determining detection accuracy. FIG. 19 is an explanatory diagram showing another example of determining detection accuracy. FIG. 19 is a flowchart of a sample status determination process. FIG. 19 is an explanatory diagram showing an example of extracting feature amounts used for detecting or determining a sample status according to Example 2. FIG. 19 is an overall configuration diagram of a sample analysis system according to Example 3. FIG. 19 is an explanatory diagram showing how a sample is acquired by changing the orientation of a container. FIG. 19 is an example of a sample with a tilted separating agent. FIG. 19 is an example of a GUI presented to a user. FIG. 19 is an example of a GUI following FIG. 18. FIG. 19 is an example of a GUI following FIG.

[0010] An embodiment of the present invention will be described below with reference to the drawings. The in-container sample state determination device of this embodiment acquires and analyzes a specified amount of sample, such as serum, stored in a container such as a test tube. Analysis is performed multiple times, either for different purposes or for the same purpose, and the remaining sample volume decreases each time a sample is acquired from the container. To determine whether the required amount of sample for analysis can be acquired from inside the test tube, it is necessary to accurately determine the range within which the sample can be acquired (the permitted sample acquisition area). As mentioned above, samples may be stored in containers containing structures such as mechanical separators or beads. Furthermore, in containers containing separating agents, the interface between the sample and the separating agent may be tilted or uneven depending on the centrifugation conditions. To avoid the risk of damage or clogging of the dispensing nozzle, it is desirable to calculate the permitted sample acquisition area as the area within which the dispensing nozzle does not come into contact with the mechanical separator, beads, separating agent, etc. However, it may be difficult to accurately determine the state of the bottom of the sample due to reasons such as the sample being suspended or being obscured by a label.

[0011] Therefore, the in-container sample state determination device determines whether an object, such as a mechanical separator, separating agent, beads, or fibrin, is present at the bottom of the sample in the container, and also determines the possibility (accuracy) of detecting the object by observing it from outside the container. The in-container sample state determination device then calculates the position at which the sample can be acquired based on the state of the sample's bottom in the container and the level of accuracy, and controls the device so that the sample can be acquired up to the calculated position. For example, if there is no object at the bottom of the sample and the object detection accuracy is high, the limit position at which the sample can be acquired from the container (the vertical lower limit of the container) expands downward. In other words, the in-container sample state determination device determines that the sample can be acquired up to the bottom of the container. On the other hand, even if it is determined that no object is present in the container, if the object detection accuracy is low, there is a possibility that an object is present in the container, and the sample acquisition device for acquiring the sample may interfere with the object, separating agent, or the like. Therefore, the in-container sample state determination device provides a margin at the limit position at which the sample can be acquired from the container, thereby preventing the sample acquisition device from interfering with the object in the container.

[0012] The device for determining the state of a sample in a container according to this embodiment makes it possible to estimate the limit position at which the sample can be safely acquired from the container, even when an object that interferes with sample acquisition is present inside the container and the position or shape of the object is unclear. This reduces the risk of damage or clogging of the sample acquisition device, and enables the sample in the container to be acquired without waste.

[0013] In the following, a specimen such as serum will be taken as an example of a "sample." A sample may be called serum or a specimen. The specimen state determination device 101 will be taken as an example of an "in-container specimen state determination device." The specimen analysis system 102 will be taken as an example of a "sample analysis system."

[0014] One embodiment will be described below using Figures 1 to 13 and 17 to 20. Figure 1 is a diagram showing the overall configuration of a sample analysis system 102 as an example of a sample analysis system. The sample analysis system 102 includes, for example, a sample status determination device 101, a camera 120, a control device 121, a sample acquisition device 122, an analyzer 123, and a display device 124. In addition to these, the sample analysis system 102 also includes a transport mechanism (not shown) for transporting containers 130 such as test tubes.

[0015] The specimen state determination device 101 is a device that determines the state of a specimen (such as serum as a sample) in a container 130. The state of the specimen includes, for example, the transparency (turbidity) of the specimen, the state of separation of the specimen such as serum and blood clots, the shape and various dimensions of the container 130 that stores the specimen, the presence or absence of an object in the container 130, the presence or absence and shape of a label affixed to the container 130, and the like.

[0016] The specimen status determination device 101 comprises, for example, an interface unit 110, a calculation unit 111, and a memory 112, and these circuits 110, 111, and 112 are interconnected via a bus 113. A predetermined computer program is stored in the memory 112. The calculation unit 111 reads and executes the predetermined computer program to realize each function of the specimen status determination device 101, which will be described later with reference to FIG.

[0017] The interface unit 110 is, for example, a circuit connected to the camera 120 and the control device 121 to send and receive information. The interface unit 110 may include a user interface in addition to an input / output interface, a communication interface, etc. A user of the specimen status determination device 101 can obtain information from the specimen status determination device 101 and input setting values ​​and the like into the specimen status determination device 101 via the user interface (not shown).

[0018] The memory 112 may include a main storage device and an auxiliary storage device. A predetermined computer program is transferred from the auxiliary storage device to the main storage device and executed by the processor 111.

[0019] The specimen status determination device 101 is connectable to a storage medium MM. The storage medium MM is configured as, for example, a memory device, a hard disk device, an optical disk device, a magneto-optical disk device, or a magnetic tape device, and non-temporarily stores computer programs and data. The storage medium MM can transfer and store computer programs and data to the memory 112 of the specimen status determination device 101. Computer programs and data can also be transferred and stored from the memory 112 to the storage medium MM. By storing predetermined computer programs that realize the functions of the specimen status determination device 101 in the storage medium MM, connecting the storage medium MM to another computer, and installing the predetermined computer programs stored in the storage medium MM on the other computer, the other computer can function as the specimen status determination device 101.

[0020] The calculation unit 111 is not limited to a CPU (Central Processing Unit), and may include a GPU (Graphics Processing Unit) or an ASIC (Application Specific Integrated Circuit).

[0021] The sample status determination device 101 can be constructed as a single physical computer, or can be constructed by linking one or more computers and external storage devices.

[0022] The camera 120 photographs the container 130 from the side and transmits the image data to the specimen state determination device 101. The camera 120 may be configured as, for example, a visible light camera, an infrared camera, or a camera that detects visible light and infrared light, by providing an optical filter or the like as necessary.

[0023] The control device 121 is a device that controls the overall operation of the sample analysis system 102. When the control device 121 receives data indicating the results determined by the sample status determination device 101 from the sample status determination device 101, it generates a control signal for aspirating the sample from the container 130 and transmits it to the sample acquisition device 122. The control device 121 can also detect whether the container 130 has been transported to a predetermined position and stopped. In the figure, the area where the sample is present is denoted by the symbol 301. For convenience, the sample may also be referred to as sample 301.

[0024] Based on a control signal from the control device 121, the sample acquisition device 122 (which can also be called the specimen acquisition device 122) lowers the dispensing nozzle 1221 a predetermined amount toward the container 130, or stops it by detecting the liquid level based on capacitance, etc., and aspirates and acquires a specified amount of sample from the container 130.

[0025] When the analysis device 123 receives a specified amount of sample from the sample acquisition device 122, it analyzes the sample by adding a predetermined reagent to the sample, etc. The analysis results are displayed on the display device 124. The results may also be printed out on a printer (not shown) along with the display device 124.

[0026] For example, when specimen amount estimation, sample acquisition, and analysis are repeated multiple times, the remaining specimen amount at a certain analysis stage can be calculated because a specified amount of sample 301 is acquired at each analysis stage from the remaining specimen amount initially estimated by the specimen status determination device 101. Nevertheless, the container 130 may be photographed again using the same camera as camera 120 or a different camera, and the latest remaining specimen amount may be estimated by the specimen status determination device 101.

[0027] 2, the functional blocks of the specimen state determination device 101 will be described. The specimen state determination device 101 includes, for example, an image acquisition unit 201, a container identification unit 202, a specimen region detection unit 203, a detection accuracy determination unit 204, an object detection unit 205, and an output unit 206.

[0028] The image acquisition unit 201 has a function of receiving image data of the container 130 from the camera 120. The container identification unit 202 has a function of identifying the shape of the container 130, the dimensions of each part (inner diameter, height), etc., based on the image data of the container 130. The container identification unit 202 may previously store a classification table (not shown) for the container 130, and may determine which type of container the container corresponds to based on the captured image data. Alternatively, the classification table may record dimensional information for each part, and after determining only the type of container through image processing, the dimensional information may be referenced from the classification table.

[0029] The sample region detection unit 203 has a function of detecting the region of the sample 301 contained in the container 130 from image data obtained by photographing the container 130. In other words, the sample region detection unit 203 identifies the region of the sample (serum, plasma, urine, etc.) within the entire region of the container 130.

[0030] The detection accuracy determination unit 204 determines the possibility (accuracy) of accurately detecting the object 303 (see FIG. 3) in the container 130 based on image data obtained by photographing the container 130. This is because, as will be described later, the detection accuracy of the object 303 decreases if the sample in the container 130 is cloudy or if a label 304 (see FIG. 5) is attached to the outside of the container 130.

[0031] The object detection unit 205 detects an object 303 in the container 130 based on image data of the container 130. The object 303 is a structure such as a mechanical separator provided in the container 130. A foreign object other than a structure may also be detected as the object 303.

[0032] The output unit 206 outputs a determination result including items such as the identification result of the container 130, the range of the sample area, the presence or absence of the object 303, and the detection accuracy. This determination result is sent to the control device 121 and serves as the basis for generating a control signal. Note that the output unit 206 does not need to send all of the above items to the control device 121. The output unit 206 only needs to send to the control device 121, as the determination result, only the items necessary for the control device 121 to generate a control signal.

[0033] Furthermore, the output unit 206 may present the above-mentioned determination results to the user via the display device 124. Examples of GUIs (Graphical User Interfaces) presented to the user are shown in Figures 18, 19, and 20. A GUI 1800 represents the entire screen displayed on the display device 124. A processing status display unit 1801 displays the status of processing by the inspection device. Furthermore, the GUI 1800 may display information other than the processing status in an other information display unit 1802.

[0034] 19 shows an example of the processing status display section 1801. The processing status display section 1801 includes, for example, a progress status display section 1803, a detection result display section 1804, and a detection result display button 1805. The progress status display section 1803 displays the number of specimens scheduled for processing (Ordered), the number of specimens currently being processed (Processing), the number of specimens for which processing has been completed successfully (Complete), and the number of specimens for which an abnormality has been detected (Incomplete) as progress status. The detection result display section 1804 displays overall information on the determination results by the specimen status determination device 101. For example, this includes the number of specimens with insufficient liquid amount (Insufficient amount), the number of specimens in which an object has been detected (Object), and the number of specimens for which the detection accuracy has been determined to be low (Low confidence).

[0035] Furthermore, for example, by pressing the detection result display button 1805, a determination result display screen 1810 for each sample may be displayed as shown in FIG. 20 . The determination result display screen 1810 for each sample includes, for example, a detection result drawing section 1811, a sample information display section 1812, and an operation section 1813. The detection result drawing section 1811 displays the determination result by the sample state determination device 101 superimposed on an image of the sample captured by the camera 120. As an example, FIG. 20 displays the upper end 501 and lower end 502 of the sample area and the area 602 from which the sample can be acquired. Alternatively, if an object or label is detected, the detected position of the object or label may be displayed, or the detection accuracy for each area may be displayed superimposed as shown in FIG. 12 . Furthermore, instead of displaying the detection results as a rectangle as in the example of FIG. 20 , for example, the sample acquireable area may be displayed by the upper and lower ends, or the sample acquireable area may be displayed by pixel or small area.

[0036] The specimen information display section 1812 displays, for example, the sample ID, information on whether the processing was normal (Complete), the container type identification result (Container) by the container identification section 202, the detection accuracy determination result (Detection confidence) by the detection accuracy determination section 204, the object detection result (Object) by the object detection section 205, the obtainable liquid volume (Amount), etc. In addition, for example, the position of the sample region detected by the specimen region detection section 203, such as the position of the upper or lower end of the sample, may be displayed numerically (not shown). The operation section 1813 displays buttons for accepting user operations, making it possible to switch the sample displayed in the detection result drawing section 1811 and the specimen information display section 1812.

[0037] The detection result display button 1805 and the operation unit 1813 may be of a touch panel type, or physical buttons may be provided. Furthermore, the display methods of items in the progress status display unit 1803, the detection result display unit 1804, and the specimen information display unit 1812 are merely examples, and items may be displayed using icons or the like instead of text. Furthermore, the detection result display button 1805 does not necessarily have to be in the processing status display unit 1801, but may be located anywhere in the GUI 1800. Furthermore, the determination result display screen 1810 may be displayed in a manner other than that described above. For example, it may be displayed as one of the items in the other information display unit 1802, or may be displayed by switching between them using tabs.

[0038] If the detection accuracy is below a predetermined threshold, the judgment result may be accompanied by a message describing the situation at the time of judgment, such as "Structure not detected. However, since the detection accuracy is below a predetermined value, the amount that can be aspirated is judged to be 100 μl."

[0039] In the specimen state determination device 101, when it is determined that a specified amount of specimen cannot be obtained from the container 130, the container 130 may be transported to a storage unit that collects only specimens with an insufficient amount of specimen.

[0040] 3, 4, and 5, an example in which the container 130 includes an object 303 will be described. As described above, the container 130 is provided with the object 303 as an internal structure such as a mechanical separator.

[0041] Below the serum or plasma serving as sample 301, a blood clot 302 that does not constitute a sample has settled. For example, blood can be separated into serum or plasma and the blood clot by centrifuging the blood. Since only sample 301 is the subject of analysis (test), dispensing nozzle 1221 of sample acquiring device 122 is slightly immersed in the area where sample 301 exists, and a specified amount of sample is sucked and acquired.

[0042] If the dispensing nozzle 1221 is advanced into the container 130 up to the area where the object 303 is present, there is a risk that the tip of the dispensing nozzle 1221 may come into contact with the object 303. Therefore, the area above the object 303 is the limit position at which the sample can be obtained.

[0043] 3 shows an example in which the transparency is relatively high and an object 303 such as a mechanical separator is easily visible. In the example of FIG. 3, simply detecting the object 303 can reduce the possibility of the dispensing nozzle 1221 coming into contact with the object 303.

[0044] Figure 4 shows a case where the sample 301 is more turbid and has lower visibility than the example in Figure 3. In the case of Figure 4, the object 303 is difficult to see, and the overall shape of the object 303 is unclear, making it difficult to detect the object 303. If the dispensing nozzle 1221 is inserted into the container 130 in a state where the object 303 cannot be correctly detected, there is a high possibility that the tip of the dispensing nozzle 1221 will come into contact with the object 303.

[0045] 5 shows an example in which a label 304 is attached to the outside of the container 130. Because part of the object 303 is covered by the label 304, the overall shape of the object 303 is unclear, making it difficult to detect the object 303, and there is a possibility that the dispensing nozzle 1221 may come into contact with the object 303.

[0046] As will be described later, a layer of separating agent may be formed between the sample 301 and the blood clot 302. The separating agent is not shown in Figures 3 to 5.

[0047] FIG. 6 shows how the container 130 is identified. The container identification unit 202 determines the type of container 130. At this time, the position of the container 130 may also be detected. In FIG. 6, only the head (upper portion) of the container 130 is detected and identified as the container identification result 401. By detecting and identifying only the head of the container 130, the insertion start position of the dispensing nozzle 1221, the type of container 130, the inner diameter of the container 130, and the like can be detected. Container information such as the inner diameter and length for each type of container may be stored in advance as a classification table in the memory 112 and referenced from the container type identified by image processing. Instead of an image of the head of the container 130, the container 130 may also be identified from an entire image. The container identification unit 202 may be equipped with a trained model that has been machine-learned to learn the type, etc., from partial or entire images of various containers. Alternatively, the container identification unit 202 may identify the container 130 using manually designed features such as edges obtained by image processing of the container.

[0048] 7 is an explanatory diagram showing the detection of the sample region 301. The upper end 501 of the sample region 301 is the interface between the sample and air. The lower end 502 of the sample region 301 is the interface between the sample and the blood clot 302.

[0049] The sample region detection unit 203 detects and outputs the upper end 501 and lower end 502 of the sample region. Alternatively, the sample region detection unit 203 can detect the entire sample region 301. The sample region detection unit 203 may perform the detection using a trained model that has been machine-learned to detect sample regions in various containers, or may perform the detection from manually designed feature quantities. The upper end 501 and lower end 502 of the sample region are not limited to being displayed as rectangles as in the example of Figure 7, but may also be displayed as straight lines, inclined straight lines, curves, etc.

[0050] 8 is an explanatory diagram showing the sample acquisition permitted area when there is no object in the container. When there is no object 303 in the container 130, the lower limit (limit position) 601 of the area where a sample can be acquired from the container 130 coincides with the lower end 502 of the sample area 301. Therefore, the area 602 where a sample can be acquired extends from the upper end 501 to the lower end 503 (= symbol 601) of the sample area 301.

[0051] 9 shows the sample acquisition permitted area 602 when the object 303 is present in the container 130. The lower end of the sample acquisition permitted area 602 must be changed to the upper end 601 of the object 303.

[0052] 10 shows examples of detecting an object 303 in containers 130(1) to 130(3) under various conditions. The area indicated by the reference numeral 701 shows the detection result of the object 303. The lower limit 601 of the area in which the sample can be obtained does not change depending on the transparency of the sample or the presence or absence of a label in each of the containers 130(1) to 130(3).

[0053] In the case of the container 130(1) shown on the left side of the figure, the sample has good visibility, so the object 303 can be accurately detected. In the case of the container 130(2) shown in the center of the figure, the sample is cloudy and has low visibility, so it is difficult to accurately detect the position and shape of the object 303 inside the container 130. In the case of the container 130(3) shown on the right side of the figure, at least a portion of the object 303 is hidden by the label 304, so the overall shape of the object 303 cannot be detected.

[0054] 10 , simply knowing that an object 303 is present in the container 130 does not allow for accurate determination of the lower limit 601 of the area in which the sample can be obtained, and does not improve the reliability of the specimen status determination device 101. Therefore, the specimen status determination device 101 of this embodiment determines not only the presence or absence of the object 303, but also the detection accuracy of the object 303, and outputs both pieces of information as determination results.

[0055] Figure 17 shows an example of a sample with an inclined separating agent. The separating agent 1701 separates the serum 301 and the blood clot 302. Depending on the centrifugation conditions, the boundary between the separating agent 1701 and the serum 301 may be inclined relative to the horizontal. The left side of Figure 17 shows an example of an inclination from the front to the back. The right side of Figure 17 shows an example of an inclination from left to right. The lower limit 601 of the sample acquisition area should be detected at the position shown in Figure 17 or at the highest point of inclination. In the case of the right side of Figure 17, the lower limit 601 is detectable. However, if the sample is inclined toward the back, as in the left side of Figure 17, and the sample visibility is low, the inclination of the separating agent cannot be determined. Therefore, for example, by detecting the lower end 502 as the lower end of the detectable sample area and correcting it upward by a certain value, it is possible to prevent the dispensing nozzle from contacting the separating agent.

[0056] 11 shows an example of determining detection accuracy. Image data of a container 130 shown on the left side of the figure is input to a detection accuracy determination unit 204. The detection accuracy determination unit 204 inputs image data 801 corresponding to a sample region 301 from the input image data to a detection accuracy determiner 802, which then outputs a detection accuracy 803.

[0057] 11, the detection accuracy is determined for the entire sample region 301. The detection accuracy 802 may be a binary value of high or low, or may be a multi-value of three or more values ​​such as high, medium, or low. The detection accuracy 802 can also be displayed as a percentage. If it is difficult to accurately determine the overall shape of the object 303, the detection accuracy will be low.

[0058] 11, the entire sample region 301 is input as image data 801 to the detection accuracy determiner 802. Alternatively, image data of a portion of the sample region 301 may be extracted and input to the detection accuracy determiner 802. Furthermore, preprocessing may be performed on the image data, such as by calculating edge information within the sample region 301. Image data used to determine the detection accuracy may be selected depending on the results of preprocessing the image data of the entire sample region 301.

[0059] The detection accuracy determiner 802 may use machine learning, or may make a determination based on edge information, saturation, hue, brightness information, or the degree of change therein.

[0060] Fig. 12 shows another example of determining the detection accuracy. In the modification shown in Fig. 12, the detection accuracy is determined and output for each local region of the sample region 301, which is a predetermined region in the image of the container 130.

[0061] As shown in FIG. 12 , the detection accuracy determination unit 204 inputs each local region (region of interest) 804 from an image 801 of the sample region 301 to a detection accuracy determiner 802, and determines the detection accuracy for each local region 804. The detection accuracy determination result 803 by the detection accuracy determiner 802 is output in the form of a map. For example, the determination result 803 in FIG. 12 illustrates an example of a map display in which the higher the detection accuracy of the local region, the darker the color, and the lower the detection accuracy, the lighter the color of the local region. If a label 304 is affixed to the container 130, the color of the region 804 to which the label 304 is affixed will be lighter. The color shading may be reversed. In other words, the determination result may be expressed so that the higher the detection accuracy, the lighter the color of the local region, and the lower the detection accuracy, the darker the color of the local region. Furthermore, the detection accuracy may be expressed by different colors, such as by displaying local regions with high detection accuracy in green and local regions with low detection accuracy in red.

[0062] The specimen state determination device 101 can also output only the area where the detection accuracy is equal to or greater than a predetermined value as the sample acquisition permitted area to the control device 121. The control device 121 can generate and output a control signal so that the tip of the dispensing nozzle 1221 of the sample acquisition device 122 is immersed only in the sample acquisition permitted area.

[0063] Fig. 13 is a flowchart of the specimen state determination process. Steps S10, S11, S12, S13, S15, and S19 in Fig. 13 correspond to the image acquisition unit 201, container identification unit 202, sample region detection unit 203, detection accuracy determination unit 204, object detection unit 205, and output unit 206 in Fig. 2.

[0064] The specimen state determination device 101 acquires image data from the camera 120 using the image acquisition unit 201 (S10). The container identification unit 202 identifies the type of container 130 based on the image data of the container 130 and acquires the dimensions of each part (S11). The sample region detection unit 203 detects the sample region based on the image data of the container 130 (S12).

[0065] The detection accuracy determination unit 204 determines the detection accuracy based on the image data of the container 130 and the sample region determined as the identification result of the container 130 (S13).

[0066] If the detection accuracy determined by the detection accuracy determination unit 204 is equal to or less than a predetermined threshold Th (S14: YES), the specimen status determination device 101 determines the position obtained by subtracting a predetermined height α from the lower end of the specimen region 301 as the lower limit (limit position) at which the specimen can be acquired (S17). In other words, the specimen status determination device 101 sets the lower limit α above the lower end detected as the specimen region.

[0067] On the other hand, if the detection accuracy determined by the detection accuracy determination unit 204 exceeds the predetermined threshold value Th (S14: NO), the object detection unit 205 detects the presence or absence of the object 303 in the container 130 (S15).

[0068] When the object detection unit 205 detects the object 303 (S16: YES), the specimen state determination device 101 sets the lower end of the specimen acquireable area to be higher than the upper end of the object 303 by α (S17). When the object detection unit 205 does not detect the presence of the object 303 (S16: NO), the specimen state determination device 101 sets the lower end of the specimen acquisition area to the lower end of the specimen area (S18).

[0069] The output unit 206 outputs information specifying the area where sample acquisition is possible as a determination result to the control device 121 (S19). In addition to the information on the area where sample acquisition is possible, the output unit 206 can also include, for example, the type of container 130, the transparency (turbidity) of the sample, the presence or absence of an object 303, the presence or absence of a label 304, and the detection accuracy in the determination result.

[0070] According to this embodiment configured as described above, even if an object is present in the container 130 and its position and shape are unclear, it is possible to estimate the limit position at which a sample can be safely acquired from the container 130. Therefore, it is possible to estimate the amount of sample that can be acquired from the container while reducing the risk of damage or clogging of the sample acquisition device, thereby improving analysis efficiency.

[0071] According to this embodiment, whether or not to execute the object detection unit 205 and / or the detection accuracy determination unit 204 is selected based on the container identification result by the container identification unit 202 and / or the sample region detection result by the sample region detection unit 203, so it is possible to prevent the execution of unnecessary processing and to effectively utilize computer resources. This also makes it possible to simultaneously process the estimation of the sample amounts in multiple containers in parallel.

[0072] Because the detection accuracy determination unit 204 determines the detection accuracy for each predetermined local region of the container 130, the control device 121 can obtain the sample by causing the dispensing nozzle 1221 to enter a region of the container 130 that has been determined to be able to safely obtain the sample. Therefore, compared to determining one detection accuracy for the entire sample region, the sample can be obtained more efficiently, improving analysis efficiency.

[0073] Example 2 will be described with reference to Fig. 14. In the following examples, including this example, differences from Example 1 will be mainly described. This example shows an example of extracting feature amounts used for detecting or determining the state of a sample.

[0074] A plurality of regions of interest 1002 are set in image data 1001 captured by a camera 120. Each region of interest 1002 is multiplied by a weight 1006, and the sum of the weighted values ​​is taken as a feature, generating a feature extraction result 1007.

[0075] A region of interest 1003, which is an enlarged view of one region of interest 1002, can be divided into a DC component 1004 and an AC component 1005. The DC component 1004 of the region of interest can be calculated, for example, by averaging the values ​​within the region of interest 1003. Each value of the AC component 1005 of the region of interest can be calculated, for example, by subtracting the DC component 1004 from each value of the region of interest 1003.

[0076] By doing this, in the sample status determination device 101 of this embodiment, even if the brightness in the image changes uniformly (the DC component changes) due to changes in lighting when capturing an image with the camera 120, the effect of this can be reduced and stable feature values ​​can be obtained.

[0077] Note that the magnitude of the signal of the region of interest 1003 or the AC component 1005 may be normalized to correct the magnitude of the AC component so as to maintain it constant. For example, the region of interest 1003 or the AC component 1005 may be divided by the norm (the square root of the sum of the squares of each element) or maximum value of the signal of the region of interest 1003 or the AC component 1005. This makes it possible to extract stable feature amounts even if the brightness scale of the image changes due to gain adjustment of the camera 120, for example.

[0078] This embodiment configured in this manner also achieves the same effects as those of Example 1. Furthermore, in this embodiment, even if the brightness or color of the lighting fluctuates, the feature quantities of the container 130 can be stably obtained by the camera 120, thereby improving the reliability and robustness of the specimen state determination device 101.

[0079] Example 3 will be described with reference to Figures 15 and 16. Figure 15 is a diagram showing the overall configuration of a sample analysis system 102A. Figure 16 is an explanatory diagram showing how a sample is obtained by changing the orientation of a container 130.

[0080] The sample analysis system 102A includes a sample posture control device 1101 and a sample holding mechanism 1102. The sample posture control device 1101 operates the sample holding mechanism 1102 in response to a control signal from the control device 121, and controls the posture of a container 130 (sample) containing a sample.

[0081] 16 , when an object 303 is detected in the container 130, the specimen position control device 1101 rotates the specimen holding mechanism 1102 to tilt the container 130. This shifts the sample to one side, increasing the amount that can be obtained. The specimen acquisition device 122 controls the insertion direction and insertion position of the dispensing nozzle 1221 in accordance with the degree of tilt of the container 130 by the specimen position control device 1101. After tilting the container 130, the specimen state determination device 101 may re-estimate the remaining amount of sample and acquire the sample.

[0082] This embodiment configured in this manner also achieves the same effects as those of Embodiment 1. Furthermore, in this embodiment, the container 130 can be tilted, and by tilting the container 130 depending on the presence or absence, shape, etc. of the object 303, the sample that has been moved to one side can be aspirated, allowing for even more efficient use of the sample.

[0083] The present invention is not limited to the above-described embodiments, and includes various modifications. The above-described embodiments have been described in detail to clearly explain the present invention, and are not necessarily limited to those including all of the described configurations. Furthermore, part of the configuration of one embodiment can be replaced with the configuration of another embodiment. Furthermore, the configuration of another embodiment can be added to the configuration of one embodiment. Furthermore, part of the configuration of each embodiment can be added to, deleted from, or replaced with other configurations.

[0084] The above-described configurations, functions, processing units, processing means, etc. may be partly or entirely implemented in hardware, for example, by designing them as integrated circuits. Furthermore, the above-described configurations, functions, etc. may be implemented in software, with a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function can be stored in a recording device, such as a memory, a hard disk, or an SSD (Solid State Drive), or in a recording medium, such as an IC card, an SD card, or a DVD.

[0085] The service provision method adjustment system 1000 may be configured from a plurality of computers, or may be configured from a single computer.

[0086] The technical features included in the above-described embodiments can be combined as appropriate, not limited to the combinations explicitly stated in the claims.

[0087] For example, in the above embodiment, the configurations expressed as follows are described so that those skilled in the art can implement them.

[0088] (Representation 1) A device for determining the state of a sample in a container, comprising: an image acquisition unit that acquires an image of the inside of the container taken from the outside; a container identification unit that analyzes the acquired image and identifies the type of container; a sample area detection unit that analyzes the acquired image and detects an area in the container where a sample is present; a detection accuracy determination unit that analyzes the acquired image and determines the accuracy when detecting the internal state of the container; an object detection unit that analyzes the acquired image and detects an object in the container; and an output unit that determines the state of the sample in the container based on the object detection result by the object detection unit and the detection accuracy determination result by the detection accuracy determination unit, and outputs the determination result.

[0089] (Representation 2) A sample state determination device in a container described in Representation 1, in which the execution of at least one of the object detection unit and the detection accuracy determination unit is selected based on at least one of the container identification result by the container identification unit and the sample area detection result by the sample area detection unit.

[0090] (Representation 3) A sample state determination device in a container described in either Representation 1 or Representation 2, in which the detection accuracy determination unit determines the accuracy when detecting the internal state of the container based on at least one of the transparency of the sample and a label attached to the outer surface of the container.

[0091] (Representation 4) An apparatus for determining the state of a sample in a container described in any one of Representations 1 to 3, wherein the detection accuracy determination unit determines the accuracy for each predetermined local area of ​​the container.

[0092] (Representation 5) A device for determining the state of a sample in a container described in any one of Representations 1 to 4, which extracts features to be used when analyzing the acquired image based on the AC components of the color information and brightness information of the acquired image.

[0093] (Representation 6) An apparatus for determining the state of a sample in a container according to any one of Representations 1 to 5, in which an image of a predetermined region among the images acquired by the image acquisition unit is analyzed.

[0094] (Representation 7) A sample analysis system for analyzing a sample, comprising: a container sample state determination device that determines the state of a sample in a container; a control device that generates a first control signal regarding a sample acquisition operation based on the determination result output from the container sample state determination device; a sample acquisition device that acquires the sample in the container based on the first control signal; and an analysis device that analyzes the sample acquired by the sample acquisition device, wherein the container sample state determination device comprises: an image acquisition unit that acquires an image of the inside of the container taken from the outside; a container identification unit that analyzes the acquired image and identifies the type of the container; a sample area detection unit that analyzes the acquired image and detects an area where a sample is present in the container; a detection accuracy determination unit that analyzes the acquired image and determines the accuracy when detecting the internal state of the container; an object detection unit that analyzes the acquired image and detects an object in the container; and an output unit that determines the state of the sample in the container based on the object detection result by the object detection unit and the detection accuracy determination result by the detection accuracy determination unit, and outputs the determination result.

[0095] (Representation 8) The control device can also output a second control signal regarding the attitude of the container, and further includes an attitude control device that controls the attitude of the container based on the second control signal.

[0096] (Representation 9) A method for determining the state of a sample in a container using an apparatus for determining the state of a sample in a container, the method comprising the steps of: acquiring an image of the inside of the container taken from the outside; analyzing the acquired image and identifying the type of the container; analyzing the acquired image and detecting an area in the container where the sample is present; analyzing the acquired image and determining the accuracy of detecting the internal state of the container; analyzing the acquired image and detecting an object in the container; and determining the state of the sample in the container based on the object detection result from the object detection step and the detection accuracy determination result from the step of determining the accuracy of detecting the internal state of the container, and outputting the determination result.

[0097] 101: specimen state determination device, 102, 102A: specimen analysis system, 120: camera, 121: control device, 122: specimen acquisition device, 123: analysis device, 124: display device, 130: container, 201: image acquisition unit, 202: container identification unit, 203: specimen region detection unit, 204: detection accuracy determination unit, 205: object detection unit, 206: output unit, 301: specimen region, 302: blood clot, 303: object, 304: label, 1101: specimen position control device, 1102: specimen holding mechanism 1102

Claims

1. An apparatus for determining the state of a sample in a container, comprising: an image acquisition unit that acquires an image of the inside of the container taken from the outside; a container identification unit that analyzes the acquired image and identifies the type of container; a sample area detection unit that analyzes the acquired image and detects an area in the container where a sample is present; a detection accuracy determination unit that analyzes the acquired image and determines the accuracy of detecting the internal state of the container; an object detection unit that analyzes the acquired image and detects an object in the container; and an output unit that determines the state of the sample in the container based on the object detection result by the object detection unit and the detection accuracy determination result by the detection accuracy determination unit, and outputs the determination result.

2. A sample state determination device in a container as described in claim 1, in which the execution of at least one of the object detection unit and the detection accuracy determination unit is selected based on at least one of the container identification result by the container identification unit and the sample area detection result by the sample area detection unit.

3. The device for determining the state of a sample in a container as described in claim 1, wherein the detection accuracy determination unit determines the accuracy of detecting the internal state of the container based on at least one of the transparency of the sample and a label attached to the outer surface of the container.

4. The device for determining the state of a sample in a container according to claim 1, wherein the detection accuracy determination unit determines the accuracy for each predetermined local area of ​​the container.

5. The device for determining the state of a sample in a container according to claim 1, wherein feature quantities used in analyzing the acquired image are extracted based on the AC components of color information and brightness information of the acquired image.

6. The device for determining the state of a sample in a container according to claim 1, wherein an image of a predetermined region among the images acquired by the image acquisition unit is analyzed.

7. A sample analysis system for analyzing a sample, comprising: an in-container sample state determination device that determines the state of a sample in a container; a control device that generates a first control signal related to a sample acquisition operation based on the determination result output from the in-container sample state determination device; a sample acquisition device that acquires the sample in the container based on the first control signal; and an analysis device that analyzes the sample acquired by the sample acquisition device, wherein the in-container sample state determination device comprises: an image acquisition unit that acquires an image of the inside of the container taken from the outside; a container identification unit that analyzes the acquired image and identifies the type of the container; a sample area detection unit that analyzes the acquired image and detects an area in the container where a sample is present; a detection accuracy determination unit that analyzes the acquired image and determines the accuracy of detecting the internal state of the container; an object detection unit that analyzes the acquired image and detects an object in the container; and an output unit that determines the state of the sample in the container based on the object detection result by the object detection unit and the detection accuracy determination result by the detection accuracy determination unit, and outputs the determination result.

8. The sample analysis system according to claim 7, wherein the control device is also capable of outputting a second control signal related to the attitude of the container, and further comprising an attitude control device that controls the attitude of the container based on the second control signal.

9. A method for determining the state of a sample in a container by an apparatus for determining the state of a sample in a container, the method comprising the steps of: acquiring an image of the inside of the container taken from the outside; analyzing the acquired image and identifying the type of the container; analyzing the acquired image and detecting an area in the container where the sample is present; analyzing the acquired image and determining the accuracy of detecting the internal state of the container; analyzing the acquired image and detecting an object in the container; and determining the state of the sample in the container based on the object detection result from the object detection step and the detection accuracy determination result from the step of determining the accuracy of detecting the internal state of the container, and outputting the determination result.

10. A method for determining the state of a sample in a container as described in claim 9, wherein whether or not to perform at least one of the steps of detecting the object and determining the accuracy of detecting the internal state of the container is selected based on at least one of the results of identifying the type of container and the results of detecting the area where the sample is present.

11. A method for determining the state of a sample in a container as described in claim 9, wherein the step of determining the accuracy of detecting the internal state of the container determines the accuracy of detecting the internal state of the container based on at least one of the transparency of the sample and a label attached to the outer surface of the container.

12. A method for determining the state of a sample in a container according to claim 9, wherein feature quantities used in analyzing the acquired image are extracted based on the AC components of color information and luminance information of the acquired image.

13. The method for determining the state of a sample in a container according to claim 9, wherein an image of a predetermined region among the images acquired in the image acquisition step is analyzed.

Images