Methods and apparatus suitable for identifying the 3D center position of a sample container using a single image capture device

Abstract

A method for determining the 3D center position of a sample container on a track. The method includes: providing a calibration object on the track; providing an initially calibrated image capture device near the track; moving the calibration object to at least two different longitudinal positions along the track; capturing a first image with the calibration object at a first longitudinal position; capturing a second image with the calibration object at a second longitudinal position; and determining a three-dimensional path trajectory of the center position along the track based at least on the first and second images. This method can be used to determine the 3D center position of a sample container imaged at any location within an observation area. Characterization devices and sample testing devices suitable for performing these methods, among other aspects, are described.

Description

[0001] Cross Reference to Related Applications

[0002] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 148,529, filed February 11, 2021, entitled “METHODS AND APPARATUS ADAPTED TO IDENTIFY 3D CENTER LOCATION OF A SPECIMEN CONTAINER USING A SINGLE IMAGE CAPTURE DEVICE,” the disclosure of which is incorporated by reference herein in its entirety for all purposes. TECHNICAL FIELD

[0003] The present disclosure relates to methods and apparatus for use in biological sample testing, and more particularly to methods and apparatus for characterizing a specimen container in biological sample testing. BACKGROUND

[0004] Automated testing systems can perform immunoassays or clinical chemistry analyses to identify analytes or other constituents in a sample, such as serum, plasma, urine, interstitial fluid, cerebrospinal fluid, and the like. For convenience and safety reasons, these samples are almost always placed in a specimen container (e.g., a blood collection tube), which can be capped with a colored cap. Some of the sample is removed from the specimen container and analyzed by an assay and / or clinical chemistry analysis. Reactions during the assay or clinical chemistry analysis produce various changes, which can be read and / or manipulated to determine concentrations of analytes or other constituents contained in the sample, which in some embodiments can be indicative of a patient’s disease state.

[0005] Advances in automated detection technology have been accompanied by corresponding advances in pre-analytical sample preparation and handling operations, such as centrifuging a specimen container to separate sample constituents, removing a cap (decapping) to facilitate sample access, aliquot preparation, and quality checks, which can be used to identify the dimensions of a specimen container, such as height and width, and / or the presence of interfering substances (such as hemolysis, icterus, or lipemia (HIL)), or the presence of artificial interferences (such as clots, bubbles, or foam). Such pre-analytical devices can be part of a laboratory automation system (LAS). A LAS can automatically transport samples in specimen containers to one or more pre-analytical sample processing stations on a track, enabling various pre-processing operations to be performed on them before analysis.

[0006] The Laboratory Information System (LAS) can handle a variety of different samples in sample containers (e.g., tubes) with barcode labels. The barcode labels may contain a registration number associated with demographic information from the Laboratory Information System (LIS), as well as test orders and other required information. An operator or robot can place the tagged sample containers onto the LAS system, which then automatically transports them along a track for pre-analytical processing. Prior to this, the samples can be tested or subjected to clinical chemistry analysis by one or more analytical instruments coupled to or integrated with the LAS.

[0007] In such testing systems, the sample containers provided for analysis can have various sizes, such as different heights and different widths (e.g., diameters), and need to be identified. Summary of the Invention

[0008] According to a first aspect, this disclosure relates to a method for determining the position of a sample container on a track. The method includes: providing a calibration object on the track; providing an initially calibrated image capture device near the track; moving the calibration object to at least two different longitudinal positions along the track, the at least two different longitudinal positions including a first longitudinal position and a second longitudinal position, the first longitudinal position being different from the second longitudinal position; capturing a first image with the image capture device when the calibration object is located at the first longitudinal position; capturing a second image with the image capture device when the calibration object is located at the second longitudinal position; and determining a three-dimensional path trajectory along the track with a center position based at least on the first and second images.

[0009] According to another aspect, a characterization device is provided. This characterization device includes a calibration object movable on a track, a calibrated image capture device located near the track, and a computer coupled to the calibrated image capture device, the computer being configured and operable such that: the calibration object moves to at least two different longitudinal positions along the track, the at least two different longitudinal positions including a first longitudinal position and a second longitudinal position, wherein the second longitudinal position is different from the first longitudinal position; a first image is captured using the calibrated image capture device when the calibration object is located at the first longitudinal position; a second image is captured using the calibrated image capture device when the calibration object is located at the second longitudinal position; and a three-dimensional path trajectory of a center position along the track is determined based at least on the first and second images. The three-dimensional center position of a sample container stopped anywhere within the imaging area can be determined based on the three-dimensional path trajectory of the center position.

[0010] On the other hand, a sample testing apparatus is provided. The testing apparatus includes: a track; a sample carrier movable on the track, the sample carrier configured to carry a sample container; and one or more characterization devices arranged around the track, each of the one or more characterization devices including: a calibrated image capture device near the track; and a computer coupled to the calibrated image capture device and configured to: determine a three-dimensional path trajectory along a segment of the track based at least on a first image and a second image of a calibrated object captured at an imaging region; move a sample container carried by the carrier on the track to the imaging region; image the sample container within the imaging region to obtain a container image; determine a central plane between the side edges of the sample container; and back-project the central plane to find the intersection point between the central plane and the three-dimensional path trajectory, wherein the intersection point is the three-dimensional center of the sample container at the position of the central plane.

[0011] Other aspects, features, and advantages of this disclosure will readily become apparent from the following description by illustrating multiple exemplary embodiments and implementations, including the best mode contemplated for carrying out this disclosure. This disclosure may also have other different embodiments, and several details thereof may be modified in various respects without departing from the scope of this disclosure. Therefore, the drawings and description are to be considered illustrative rather than restrictive in nature. This disclosure will cover all modifications, equivalents, and substitutions falling within the scope of the claims. Attached Figure Description

[0012] The accompanying drawings described below are for illustrative purposes and are not necessarily drawn to scale. The drawings are not intended to limit the scope of this disclosure in any way. In all the drawings, the same numerals are used to denote the same elements.

[0013] Figure 1 A top view of a sample testing apparatus according to one or more embodiments of the present disclosure is shown. The sample testing apparatus includes one or more characterization devices configured to determine the 3D center position of a sample container and one or more analytical instruments (e.g., clinical chemistry or laboratory instruments).

[0014] Figure 2 A side view of a labeled sample container is shown, the three-dimensional (3D) center position of which can be quantified by a characterization method and characterization apparatus according to one or more embodiments of the present disclosure.

[0015] Figure 3A A perspective view of a characterization device configured to capture multiple images according to one or more embodiments of the present disclosure is shown (where the outline of the housing portion is shown in dashed lines for illustrative purposes), wherein the calibration object is shown in a second position along the track.

[0016] Figure 3B A perspective view of a characterization device according to one or more embodiments of the present disclosure is shown, the characterization device being configured to capture an image of a sample container located at a position along a track within an imaging region and determine its 3D center position.

[0017] Figure 3C A schematic top view of a characterization device according to one or more embodiments of the present disclosure is shown. The characterization device is configured to capture images of a calibration object at first and second positions spaced apart along a track and to determine a three-dimensional path trajectory along the track.

[0018] Figure 3D A side view of a sample container located at a position along a track in an image region according to one or more embodiments of the present disclosure is shown, from which the 3D center position of the sample container at that position can be determined.

[0019] Figure 4A A schematic top view of a characterization device according to one or more embodiments of the present disclosure is shown, the characterization device being configured to determine the 3D center position of a sample container located at a position along a track within an imaging region.

[0020] Figure 4B One or more embodiments according to this disclosure are shown. Figure 4A A schematic side view of the characterization device (in which the outline of the housing portion is shown in dashed lines for illustrative purposes) (the light source 300B is not shown for illustrative purposes).

[0021] Figure 5 A flowchart is shown of functional components of a characterization device suitable for characterizing sample containers and samples, according to one or more embodiments.

[0022] Figure 6 A flowchart is shown of a method for characterizing a 3D trajectory path of a sample container according to one or more embodiments.

[0023] Figure 7 A flowchart is shown of a method for characterizing the 3D center position of a sample container according to one or more embodiments. Detailed Implementation

[0024] Due to tolerance accumulation and variability, the exact location of the sample container's three-dimensional (3D) center position at various points along the track, and particularly its location in front of various pre-analytical operations and analytical instruments, may be unknown. Because of the difficulties encountered in determining the exact location of the sample container in 3D (its center position) and / or the size or type of the sample container, there is an unmet need for methods and apparatus suitable for easily and accurately determining such a center position and such size.

[0025] Specifically, in one or more pre-analytical stages, it may be necessary to obtain the 3D center position and size of various sample containers, as this information helps inform the alignment positions that pre-analytical equipment (e.g., centrifuges, cappers, aspirators, etc.) should follow. Furthermore, this acts as a fail-safe mechanism in cases where unsupported tube geometries are introduced into the sample testing system. On the other hand, knowing the size and 3D center position of the sample containers helps the robotic gripper correctly align them in 3D space when the robot handles them, thus avoiding or minimizing collisions. Additionally, knowing the size and 3D center position of the sample containers helps lower pipettes to the correct position to avoid sample container / pipette collisions and / or aspiration problems.

[0026] In some existing quality inspection modules configured to evaluate the size, sample level, and sample quality (such as determining the HIL level within a container), three cameras are provided, which helps to provide a complete 360-degree view of the sample contained within the sample container. Therefore, in such conventional quality inspection modules, once the cameras are properly calibrated, the geometry of the sample container can be reconstructed in 3D space. Once this reconstruction is complete using input from multiple cameras, the height and width (e.g., diameter) can be determined with fairly high accuracy.

[0027] However, in some systems, using three cameras to perform multi-view imaging is impractical from a cost perspective. Therefore, in some embodiments, this disclosure provides methods, apparatus, and systems capable of measuring the geometry of a sample container using only a single image capture device (e.g., a single camera) and also capable of calculating the 3D center coordinates of the sample container's 3D center position. Furthermore, in some LAS systems, knowing the 3D center position of a sample container within a quality inspection module may not translate into an accurate center position at other locations around the track, because accurate track positioning is challenging due to tolerance stacking and installation variations. By combining a single image capture device with a track that allows the sample container to move around the track in the characterization apparatus, a simple and efficient method and apparatus are achieved for determining the 3D center position and size of a sample container at any desired location along the track in the LAS.

[0028] In some existing testing systems, the geometry of the sample container is measured in a quality inspection module using one of two methods. In the first method, a sample container with a known geometry (e.g., a tube) (such as a cylindrical calibration tool) is moved to a predetermined position on a track within the quality inspection module, and an image is captured. The height (HT) and width (W) are measured in pixels. When a sample container of a different size is encountered, it is moved to the exact same position on the track as before, and its height HT and width W can be derived proportionally based on the previously obtained image measurements (in pixels). While this method can derive the height HT and width W fairly accurately, it does not provide an accurate 3D center position estimate of the sample container.

[0029] Therefore, in a first broad aspect, embodiments of this disclosure provide characterization methods, characterization devices, and sample testing systems configured and operable to determine the 3D center position of a sample container and the physical dimensions of the sample container, such as width W and height HT.

[0030] In one or more embodiments, this characterization method requires neither that the sample container be moved to the exact predetermined position each time, nor that prior computer-aided design (CAD) information about the track geometry be required. Furthermore, this characterization method derives an accurate estimate of the 3D coordinates of the sample container's center position (which can be used for robotic gripper and pipette alignment tasks) without requiring very tight tolerances in the mechanical setup. Finally, this characterization method does not strictly require the track to be parallel to the image capture device (e.g., a camera), and it can even handle slightly tilted or curved tracks, as long as there are no overlapping points on the track along each line of sight from the image capture device.

[0031] Knowing the width W of the sample container can be used to further quantify the various portions of the sample (e.g., its volume), such as quantifying the volume of the serum or plasma portion, the sedimented blood portion, or both. Any robotic system in the testing system can use the height HT of the sample container to determine the initial height of the pipette in the liquid aspiration system, minimizing collisions between the sample container and the pipette when moving the pipette to complete aspiration. Knowing the exact 3D center position of the sample container also allows for the use of a robotic gripper to pick up the sample container while avoiding collisions between the sample container and the robotic gripper. Furthermore, HT and W, along with the exact 3D center position, can be used to position and properly disengage the jaws of any robotic gripper, enabling the gripper to properly grasp the sample container.

[0032] According to this disclosure, the characterization method can utilize a single image-capturing device (e.g., a camera) located at any position along a track, where knowing the 3D center position and / or size of the sample container is useful. For example, the characterization device can be implemented at a location within a quality inspection module, at a centrifuge station (such as at its centrifugation pick-up location), at a dispensing / aspiration location of an aliquoting machine, at any other robotic pick-up and / or placement location on the track, at an analytical instrument location, or at any other suitable location where the robotic pick-up or placement operation is repeated.

[0033] The characterization method involves first mapping a track path to a desired region of interest in three-dimensional (3D) space. For example, the characterization method could involve taking multiple images of the same calibration object (e.g., a calibration tool) at various longitudinal positions along the track within the imaging region, and determining a 3D trajectory along the track from these images, centered on the object. This is performed on a calibration object with known geometry, and using this trajectory, the calibration method can map the center of the calibration object in three-dimensional space. This method is applicable to any image capture device (e.g., a camera) and track setup, such as in automated diagnostic equipment.

[0034] Specifically, the track carries samples from the sample container to various locations on the carrier for analysis (e.g., analytical testing or assay), and other locations around the track can utilize geometric dimensions (W and HT) and a 3D center position determined from the quality check module, although these locations may not be perfectly accurate. For more accurate 3D center positioning, one or more characterization devices may be included at other locations around the track. Following pre-screening at the quality check module, chemical analysis or assay can be performed on a suitable analytical instrument. As used herein, the term "analytical instrument" refers to clinical chemistry analysis instruments and / or assay instruments, etc. In one embodiment, the quality check module may be positioned on the track such that the dimensions of the sample container can be characterized while it is stationary on the track, such as on the track's input channel or elsewhere along the track.

[0035] Reference Figures 1 to 7 Further details describe the characterization method, characterization apparatus, and test system including one or more characterization apparatuses of the present invention.

[0036] Figure 1An example embodiment of a sample testing device 100 is shown, which is capable and operable to automatically process multiple sample containers 102 in a sample container 102. These sample containers 102 can be contained in one or more racks 104 located at a loading area 105 prior to analysis by one or more analytical instruments (e.g., a first analytical instrument 106, a second analytical instrument 108, and a third analytical instrument 110, respectively) arranged around a track 121 of the sample testing device 100. It is evident that more or fewer analytical instruments can be used. Analytical instruments 106, 108, and 110 can be one or more clinical chemistry analytical instruments and / or one or more laboratory instruments, or combinations thereof. Sample containers 102 can be any generally transparent or translucent container, such as blood collection tubes (see...). Figure 2 The sample container 102 can move on the carrier 122 about the track 121, as described later herein. More specifically, the sample testing device 100 may include a base 120 (e.g., a frame or other structure) on which the track 121 may be mounted or supported.

[0037] like Figure 2 As shown, a sample 212 to be automatically processed can be provided to a sample testing device 100 in a sample container 102, which can be covered by a lid 214. The lid 214 can have different shapes and / or colors (e.g., red, dark blue, light blue, green, gray, tan, yellow, or other colors). The color and / or shape provide useful information about the test to be performed and / or the additives provided in the sample container 102. Each sample container 102 may include a tube 213, which may be provided with a label 218 containing identification information 215, such as a barcode, letters, numbers, or alphanumeric markings, or a combination thereof, which can be read by the machine at different locations around the sample testing device 100. For example, the identification information 215 may indicate the patient's identity and the test that may be performed on the sample 212. The identification information 215 may also work with a laboratory information system (LIS) 147 to provide additional information about ordered tests, etc. The label 218 is affixed to or otherwise provided on the side of the sample container 102. Label 218 typically does not extend around the entire circumference of sample container 102, or extend the entire length of sample container 102. Therefore, although label 218 may obscure some portions of sample 212, some portions of sample 212 are still visible. In some embodiments, there may be multiple slightly overlapping labels 218. In some embodiments, shelf 104 may have additional identification information available for sample tracking. As should be appreciated, unobstructed areas can be oriented manually or automatically to reside on carrier 122 such that the unobstructed areas face a desired location, such as towards a specific image capture device.

[0038] Refer again Figure 1 A properly calibrated robot 124 can pick up a desired sample container 102 from one or more shelves 104 and place the sample container 102 into a carrier 122 located at a pre-programmed position on track 121 or at an input channel (not shown) of the track, via control commands from computer 123. Computer 123 may include a microprocessor-based central processing unit (CPU), suitable memory, software, and regulating electronics and drivers for operating various test system components. Computer 123 may be mounted as part of or separate from the base 120 of the sample testing equipment 100. Computer 123 is operable to control the movement of carrier 122 back and forth to loading area 105, movement around track 121, movement back and forth to centrifuge station 125, operation of centrifuge station 125, movement back and forth to quality inspection module 130, operation of quality inspection module 130, movement back and forth to equal sample station 131, operation of equal sample station 131, and movement back and forth to each analytical instrument 106, 108, 110. Computer 123 can also interface with one or more characterization devices 101 located around orbit 121 and perform their calculations and operations. In most cases, the operation of each analytical instrument 106, 108, 110 performing various types of tests (e.g., laboratory tests and / or clinical chemistry) is performed by internal software that can interface with computer 123.

[0039] The loading area 105 can serve a dual function, also allowing the sample container 102 to be unloaded from the carrier 122 after processing. The robotic gripper of robot 124 can be configured to grasp sample containers 102 from one or more shelves 104 and move and load them onto carriers 122, typically one sample container 102 per carrier 122. In some embodiments, robot 124 can be configured to remove sample containers 102 from carriers 122 upon completion of testing. Robot 124 may include one or more (e.g., at least two) robotic arms or components capable of X and Z (perpendicular to the XY plane), Y and Z, X, Y and v, or r and θ movements, wherein robot 124 may be equipped with a robotic gripper suitable for picking up and placing sample containers 102 by grasping their sides. However, any suitable type of robot 124 can be used.

[0040] When loaded onto track 121 by robot 124, sample container 102 carried by carrier 122 can proceed to centrifugation station 125 (e.g., an automated centrifuge configured to perform grading separation of sample 212). Characterization device 101 can be positioned near centrifugation station 125 and track 121, at a location where loading / unloading robot 126 can pick up sample container 102 from carrier 122 and place it into the centrifuge at centrifugation station 125. Knowing the exact 3D center position of sample container 102, the location where carrier 122 stops (or at any other location around track) helps avoid collisions between the robot gripper and the container, which could cause sample 212 to spill or break. As will be appreciated, characterization device 101 as described herein can be used at any location where it is desirable to know the 3D center position. For example, characterization device 101 can be positioned at loading area 105, quality inspection module 130 (using one or more of its image capture devices), and aliquot station 131 to avoid pipette-container collisions that could cause sample 212 to spill, sample container 102, or pipette to break. It can also be positioned at one or more of analytical instruments 106, 108, and 110 to avoid pipette-container collisions or robot-container collisions. Characterization device 101 can be positioned at other locations.

[0041] Figures 3A to 3D An example embodiment of the characterization device 101 is shown. The characterization device 101 includes a calibration object 325 that can move on a track 121. Figures 3A to 3C (Only a portion of track 121 is shown in the image). Track 121 may be a collection of rails (e.g., monorails or multirails), conveyors, chains, mobile platforms, or other suitable transport mechanisms. Track 121 may have a circular, serpentine, or other shape, and in some embodiments may be a closed (i.e., endless) track. In some embodiments, track 121 may transport a single sample container 102 of sample containers 102 via carrier 122, or multiple sample containers 102 on each carrier 122. In some embodiments, sample containers 102 are configured to be housed within containers of carrier 122 that are vertically movable on track 121.

[0042] In the depicted embodiment, the carrier 122 may be carried on the track 121, for example, by a trolley 324. The trolley 324 may be programmed, commanded, or otherwise forced to stop at a desired position along the track 121. The carrier 122 may be removed from the trolley 324 and may include any suitable means for alignment with the trolley 324, such as a plurality of pins registered into holes. This positions the carrier 122 on the trolley 324 in a fixed orientation. In some embodiments, the trolley 324 may include an onboard drive motor, such as a linear motor, configured to move the sample container 102 about the track 121 while stopping at a desired position along the track 121 according to programmed instructions. Each carrier 122 may include a retainer 122H suitable for holding and securing the sample container 102 in a defined vertical position. Figure 3B The retainer 122H may include three or more leaf springs or fingers that provide a common center when the sample container 102 is inserted therein. The use of leaf springs allows the carrier 122 to accommodate sample containers 102 of varying widths W while still positioning the sample containers 102 at the common center position on the carrier 122. In some embodiments, the cart 324 and the carrier 122 may be integral.

[0043] like Figure 3A and 3CAs best illustrated, the calibration object 325 may include a V-shaped marking tool having at least a first plane 325A and a second plane 325B set at an angle thereto. For example, the two planes 325A, 325B may be set at an angle relative to each other of approximately 90 degrees to approximately 150 degrees. In some embodiments, a third plane is provided, which may have approximately 120 degrees between all the planes. In particular, the calibration object 325 includes a three-dimensional tool having a known geometry and one or more calibrated patterns 325P disposed on each of the planes 325A, 325B, the positions of which on the three-dimensional tool are known. Any suitable pattern, such as a checkerboard pattern, or one or more geometric objects with recognizable edges, may be used. The calibration object 325 may include a center position indicated by axis 329, which may be positioned on the base 331 at the same center position as the center of the holder 122H of the carrier 122, i.e., the center position 329 of the sample container 102 when held in the holder 122H. Therefore, the center 329 of the calibration object 325 is the same as the center 329 of the holder 122H of the carrier 122, that is, when held in the holder 122H, it is the same as the center position 329 of the sample container 102. The base 331 is positioned on the carriage 324 in a fixed orientation, so the calibration object 325 and the base 331 move together with the carriage 324. The size and position of the calibration pattern 325P are known, as are their spatial relationships with the axis 329 and the base 331. In some embodiments, the V-shaped marking tool includes a Huffman mark disposed thereon, the Huffman mark having a known position and geometry relative to the center 329 and the base 331, and therefore relative to the carriage 324.

[0044] The characterization device 101 also includes a calibrated image capture device 328 located near the track 121, such as along the side of the track 121. The calibrated image capture device 328 can be calibrated by any suitable means to obtain its inherent characteristics (e.g., focal length, image center, skewness, and lens distortion coefficients), such as by using standard, automated calibration techniques (e.g., camera calibration techniques). These calibration techniques typically involve using a printed planar target of known size (e.g., a Huffman-marked grid or checkerboard pattern) and applying iterative refinement techniques to determine the inherent parameters of the image capture device 328. Knowing the inherent characteristics of the image capture device 328 is a prerequisite for any 3D imaging task, as it enables the estimation of the scene's structure in Euclidean space while at least removing some inaccuracies caused by any lens distortion (e.g., possibly stemming from imperfect lens manufacturing).

[0045] The calibrated image capture device 328 can be any combination of a focusing lens system and one or more sensors. For example, the calibrated image capture device 328 can be a conventional digital camera (e.g., a color or monochrome camera), or a charge-coupled device (CCD), photodetector array, one or more CMOS sensors, etc., coupled to any suitable focusing lens system. For example, the calibrated image capture device 328 can be configured to capture images at multiple different imaging positions (including position A and a second position B) along track 121. The calibrated image capture device 328 can be a device capable of capturing digital images (i.e., pixelated images) at multiple different imaging positions. The image resolution of each image can be approximately 0.5 MP or higher, such as from 0.5 MP to 10 MP. Other pixel resolutions can also be used. The calibrated image capture device 328 can be a high-speed image capture device, and although it is desirable to stop the calibration object 325 at the first position A and the second position B and to stop the carrier 122 at the imaging position, if the speed is high enough, images can be captured while the carrier 122 or base 331 and the calibration object 325 are still moving.

[0046] The characterization device 101 also includes a computer 123, which is coupled to the calibrated image capture device 328, such as via a USB cable. The computer 123 is configured and operable to cause the calibrated image capture device 328 to capture side images of the calibrated object 325 at multiple imaging positions (A and B) along track 121. The calibrated object 325 can be illuminated during imaging. For example, illumination of the calibrated object 325 can be achieved by providing illumination with one or more light sources 330A, 330B, such as a light panel described in US201 / 0041318. The light sources can be positioned relative to the calibrated object 325 such that surfaces 325A, 325B of the calibrated object 325 are illuminated. For example, the light panel can provide front illumination and can be positioned in front of the calibrated object 325, and can include multiple light sources 330A, 330B, for example, positioned on either side of the calibrated image capture device 328. Other positioning and forms of the light sources can be used.

[0047] Specifically, the computer 123, through drive signals to the trolley 324, can move the calibration object 325 within the imaging region 335 (e.g., the wide-angle observation region) to at least two different longitudinal positions along the track 121, including at least the following: Figure 3CThe diagram shows a first longitudinal position A and a second longitudinal position B (indicated by dashed lines), where the second longitudinal position B differs from the first longitudinal position A. Computer 123 can trigger a calibrated image capture device 328 via a appropriately timed trigger signal to capture a first image of the calibration object 325 when it is located at the first longitudinal position A. Computer 123 can also command a cart 324 to move the calibration object 325 to the second longitudinal position B and capture a second image of the calibration object 325 using the calibrated image capture device 328. The second longitudinal position B should be sufficiently spaced from the first longitudinal position A so that when the carrier 122 transports the sample container 102 to the location where the 3D center position 350 is to be determined later, a representative and accurate travel path between positions A and B can be obtained in front of said position. Furthermore, for higher accuracy or in cases where the track segment is not straight, images can be captured at multiple longitudinal positions between the first position A and the second position B. For example, a wide-angle lens (35 mm or less) with a short focal length and a wide field of view (e.g., 50 degrees or more) can be used. Other wide-angle lenses can be used, as long as the calibrated object 325 is within the field of view (viewing window) of the calibrated image capture device 328.

[0048] According to this method, the three-dimensional path trajectory of the center position 340 along the track 121 is determined based at least on the first and second images. The center position 340 can be at any predetermined height on the calibration object 325 and can be determined relative to the imaging position of two or more calibration patterns 325P. In particular, if the path is not straight, such as along the curved track 121, one or more additional images can be captured. Using calculated inherent camera parameters, such as focal length, image center, skew, and possible lens distortion coefficients (for greater accuracy), the method can calculate the relative external pose of the 3D center position 350 of the calibration object 325 relative to the calibrated image capture device 328, at least for the first and second images. For each image, such as the first and second images and any other captured images, the relative external pose of the 3D center position 350 of the calibration object 325 relative to the calibrated image capture device 328 can be calculated using an algorithm such as Perspective-n-Point. Perspective-n-Point is the problem of estimating the pose of a calibrated image capture device (e.g., a camera) given a set of n 3D points in the world and their corresponding 2D projections in an image. The pose of the image capture device 328 comprises six degrees of freedom, consisting of rotations (roll, pitch, and yaw) and 3D translations (X, Y, Z) of the image capture device 328 relative to the world. Given a set of n 3D points in a world reference frame, their corresponding 2D image projections, and calibrated intrinsic parameters, the six-DOF pose of the image capture device 328 in the form of rotations and translations relative to the world can be determined as follows:

[0049] sp c K[R|T]p w

[0050] Where, p w =[xyz 1] T It is a homogeneous world point, p c =[uv 1] T Let fx and fy be the corresponding homogeneous image points, K be a matrix of intrinsic parameters of the image capture device 328, where fx and fy are the scaled focal lengths, γ is the skew parameter (sometimes assumed to be 0), and (u0, v0) are the principal points, S is the scaling factor of the image points, and R and T are the desired 3D rotation and 3D translation (extrinsic parameters) of the image capture device being calculated. This yields the following equations for the model:

[0051]

[0052] Optionally, P3P can be used when there are n = 3 points, or EPnP can be used for n ≥ 4 points. If outliers exist, RANSAC can be used.

[0053] Now that the three-dimensional path trajectory of the center position 340 along track 121 has been determined, in the next stage, the exact position of the 3D center position 350 of any sample container 102 that stops within the imaging area 335 of the characterization device 101 can be obtained. A particular advantage is that the stopping position of the carrier 122 does not need to be exact within the imaging area 335, because the method can determine the 3D center position 350 at any location within the imaging area 335, as long as the sides and top of the sample container 102 can be observed / imaged from said location. The imaging area 335 is an area that can be imaged by a calibrated image capture device 328. The image area 335 can be at least as high as the intended sample container 102 and can be wide-angle, as disclosed herein.

[0054] Once the three-dimensional path trajectory of the center position 340 along the track 121 is determined, the 3D center position 350 of the sample container 102 at the imaging position within the imaging area 335 for imaging the sample container 102 is determined. For example... Figure 3D As shown, with the carrier 122 positioned at imaging position 333 within imaging region 335, a container image 336 of the sample container 102 is captured using a calibrated image capture device 328. As shown in the magnified view, the carrier 122 does not need to be positioned at the exact center of imaging region 335 to find the 3D center position 350, because any point on the trajectory can be mapped back to 2D image space. Specifically, since the method has previously calculated the extrinsic relationship between the trajectory and the calibrated image capture device 328, we now also have a corresponding 2D trajectory in image space. Using this method, it is even possible to estimate the 3D center position 350 of the sample container 102 along a slightly curved trajectory, such as by fitting a polynomial function (where more than two imaging positions are used) instead of a straight line.

[0055] Calculate the center position of the sample container

[0056] When the sample container 102 in the carrier 122 appears at the imaging position 333 of the characterization device 101, the method can capture a container image 336 and estimate the 3D center position 350 of the sample container 102 in the first two dimensions (2D) from the captured container image 336. In image space, the method first calculates the center of the sample container 102 in the X dimension by determining the positions of the first edge 341 and the second edge 342 in pixel space (e.g., at the same height as the center position 340), where the trajectory of the center point is marked by line 344. Edges 341, 342 can be detected by any conventional edge detection method, such as detecting sudden changes in light intensity above a preset threshold by raster scans across the trajectory path 344. When an intensity abrupt change is found at the same X position in the pixel (e.g., in the space above and / or below), one or more raster scans above and / or below the trajectory path 344 can be used to confirm that edges 341, 342 are indeed edges. Once the positions of vertical edges 341 and 342 are determined, the center point in 2D space (in the XY plane) along the trajectory path 344 can be found by adding these two dimensions and dividing by 2. The determined 2D centerline is shown as the center plane 346. The intersection of the trajectory path 344 and the center plane 346 includes the 2D center point. The 2D center point can then be mapped to 3D space, i.e., mapped to the nearest point on the 3D trajectory (in the Z dimension, i.e., the dimension through the paper), to determine the 3D center position 350. The 3D center position 350 is calculated by using the 2D image coordinates from the sample container 102 and projecting these 2D image coordinates into the Z dimension using the inherent parameters of the calibrated image capture device 328. One implementation involves drawing a line from the center of image capture device 328 to 350 (in 3D Euclidean space) extending to infinity, and finding the nearest point on the 3D trajectory 340 (i.e., 344 in 3D Euclidean space) that intersects this line. If there is no intersection, we can choose a point on 344 (in 3D Euclidean space) that minimizes the distance to the Z-projection of the 3D center position 350. This point on the trajectory path 344 is the 3D center position 350 of sample container 102. Therefore, robot 126 can accurately know the 3D center position 350 of sample container 102, which it will use to pick up sample container 102, place it in the centrifuge of centrifuge station 125, and return it to carrier 122 after grading.

[0057] After fractionation and separation using a centrifuge, such as Figure 2As best illustrated herein, sample 212 may include a serum or plasma portion 212SP and a sedimentation blood portion 212SB contained within tube 213. Air 217 may be provided above the serum and plasma portions 212SP, and the line or boundary between air 217 and the serum or plasma portion 212SP is defined herein as the liquid-air interface (LA). The boundary between the serum or plasma portion 212SP and the sedimentation blood portion 212SB is defined herein as the serum-blood interface (SB). The interface between air 217 and cap 214 is referred herein as the tube-cap interface (TC). The tube height (HT) is defined as the height from the actual bottom of tube 213 to the bottom of cap 214. The height of serum or plasma portion 212SP (HSP) is defined as the height from the top of sedimentation blood portion 212SB to the top of serum or plasma portion 212SP, i.e., from SB to LA. The height of sedimentation blood portion 212SB (HSB) is defined as the height from the bottom of sedimentation blood portion 212SB to the top of sedimentation blood portion 212SB. In embodiments using a gel separator, an interface exists between the serum or plasma portion 212SP and the gel separator. Similarly, an interface exists between the sedimented blood portion 212SB and the gel separator. HTOT is HSB plus HSP. W is the width of tube 213. In some embodiments, the size of sample container 102 may be represented by a combination of width W and height HT.

[0058] As described above, carrier 122 can be moved to quality inspection module 130. Optionally, centrifugation can be performed prior to this, and sample 212 contained in sample container 102 can be directly loaded into quality inspection module 130 located in loading area 105 (such as as part of input channel). Quality inspection module 130 is configured and adapted to automatically determine / characterize the physical properties of sample container 102 containing sample 212 to be processed by sample testing device 100. Characterization may include characterizing tube size, cap type and / or cap color. Once characterized, sample 212 can be further characterized to determine the depth and / or volume of sample 212, for screening against hemolysis, jaundice or lipemia (HIL) and / or the presence of one or more artificial interfering substances (such as clots, bubbles or foam). If no HIL and / or one or more man-made interferences are found, sample 212 can continue on track 121 and then be analyzed in one or more analytical instruments (e.g., first, second and third analytical instruments 106, 108 and / or 110), and then each sample container 102 is returned to loading area 105 for unloading.

[0059] In some embodiments, the quantification of the physical properties of the sample container 102 can be performed at the quality inspection module 130 (i.e., determining the height HT, width W, cap color, cap type, and / or tube type). In some embodiments, the quantification of the sample 212 can also be performed at the quality inspection module 130, and may involve determining the HSB, HSP, HTOT, and the vertical positions of SB and LA.

[0060] The sample testing device 100 may include multiple sensors 116 at one or more locations around the track 121. The sensors 116 can be used to read identification information 215 placed on the tag 218. Figure 2 Alternatively, similar information (not shown) (such as barcodes) on each carrier 122 can be provided to detect the position of sample container 102 along track 121. Other means for tracking the position of carrier 122 can be used. All sensors 116 interface with computer 123, so that the position of each sample container 102 and sample 212 is known at all times. Computer 123 can interface and communicate with laboratory information system (LIS) 147 in a known manner to provide test results and status information to requesters.

[0061] Embodiments of this disclosure may be implemented using a computer interface module (CIM) 145, which allows a user to easily and quickly access various control and status displays. These control and status displays may describe some or all aspects of multiple interrelated automated devices used for the preparation and analysis of sample 212. CIM 145 may be used to provide information about the operational status of multiple interrelated automated devices, as well as information describing the location of any sample 212 and the status of screening or testing to be performed or currently being performed on sample 212. CIM 145 may be adapted to facilitate interaction between the operator and sample testing equipment 100. CIM 145 may include a display screen suitable for displaying menus, including icons, scroll bars, boxes, and buttons through which the operator can interact with sample testing equipment 100. The menu may include multiple function buttons programmed to display various functional aspects of sample testing equipment 100.

[0062] refer to Figures 4A to 4BAn embodiment of the quality inspection module 430 is shown and described. The quality inspection module 430, as shown, can be configured and adapted to automatically characterize the physical structure (e.g., size) of the sample container 102. The characterization method can be performed by the quality inspection module 430 prior to automated processing by one or more of the analytical instruments 106, 108, 110. In this way, the size of the sample container 102 (e.g., width W and height HT) is known for any subsequent processing. The quality inspection module 430 can also be used to quantify the sample container 102, i.e., quantify certain physical dimensional characteristics of the sample container 102, such as the position of TC, HT and / or W of the sample container 102, and / or the color and / or type of the lid 214.

[0063] In addition to sample container quantification, other detection methods can be performed on the sample 212 contained in the sample container 102 at the quality inspection module 430. For example, the quality inspection module 430 can be used to quantify the sample 212, that is, to determine certain physical dimensional characteristics of the sample 212 (e.g., the physical location of LA, SB and / or the determination of HSP, HSB and / or HTOT, and / or the volume of the serum or plasma fraction and / or the volume of the sedimented blood fraction).

[0064] Refer again Figure 4A and 4B The inexpensive quality inspection module 430 may include a single (one and only one) calibrated image capture device 328 (e.g., a single conventional digital camera, such as a color or monochrome camera), or a lens system coupled to a charge-coupled device (CCD), photodetector array, CMOS sensor, etc. For example, the single calibrated image capture device 328 may be configured to capture images of sample container 102 and sample 212 from a single viewpoint at imaging position 333. In this embodiment, sample container 102 may be positioned in a rotatable orientation, making a clear image of sample 212 possible, such as by a user or robot determining an unobstructed orientation (not obstructed by label 218), and then inserting the sample container into carrier 122 in that orientation.

[0065] In addition to determining the geometric properties of the sample container 102 (e.g., width W and height HT), this embodiment, including the quality inspection module 430 of a single image capture device 328, can be used for pre-screening against HIL, as described in US10,816,538, entitled "Methods and Apparatus for Detecting an Interferent in a Specimen" by Kluckner et al., and / or for pre-screening against the presence of artificial interference, as described in US10,746,665, entitled "Methods and Apparatus for Classifying an Artifact in a Specimen" by Kluckner et al. For example, backlighting using a backlight source 400C (such as a panel light source) can be used to perform HIL pre-screening.

[0066] In one or more embodiments, the characterization method for determining the 3D center position 350 can be performed using a characterization device 101, which is a sub-component of the quality inspection module 430. The characterization device 101 includes one or more light sources 300A, 300B, a calibrated image capture device 328, and a calibration object 325, as described above. Figure 3B The characterization device 101 and characterization method described herein can be performed within the quality inspection module 430. Knowing the 3D center position 350 at imaging position 333 can be used as a rough estimate of the 3D center position at other locations along track 121 (at least on any straight segment of it). If a more accurate 3D center position determination is desired at another location, the characterization device 101 and characterization method can be performed at that location.

[0067] In operation, each of the front light and backlight images captured by the quality inspection module 430 can be triggered and captured in response to a trigger signal. The trigger signal can be generated by computer 123 and provided in a communication line coupled to computer 123. Each captured image can be processed according to one or more embodiments of the characterization methods provided herein. In particular, image processing can be used to determine the width W and height HT. Furthermore, the color and type of the cover can be determined using known methods. Additionally, the presence of pre-screening and / or artificial interference for HIL can be determined, such as backlight images provided by a backlight using a light source 400C.

[0068] To improve discrimination, a spectrum of more than one wavelength can be used. Image capture device 328 can then capture multispectral images. Each color spectral image (represented by a nominal wavelength and some relatively narrow wavelength bands) is captured one after another in one or more exposures (e.g., 4 to 8 or more exposures). The duration of each exposure can vary. Spectral images can be captured in any order, such as multiple exposures of red, green, and blue. For the detection method, transmittance images can be calculated, where each transmittance image (for each of the R, G, and B illuminations) can be calculated from the optimal exposure images. The optimal exposure images can be normalized according to their respective pixel-by-pixel intensities.

[0069] In one or more embodiments, the characterization device 101 and the quality inspection module 430 may include a housing 345 that may at least partially surround or cover the track 121 and provide a closed or semi-closed environment for image capture, such as minimizing the influence of external light. During each image capture, the sample container 102 may be located inside the housing 345. The housing 345 may include one or more doors to allow the carrier 122 to enter and / or exit the housing 345. In some embodiments, the canopy may include an opening to allow the sample container 102 to be loaded from above into the carrier 122 located inside the housing 345 by a robot (e.g., robot 124) including a movable robotic gripper, such as when the characterization device 101 and / or the quality inspection module 430 are located in the loading area 105. In the case of using front light without backlighting (e.g., Figures 3A to 3C The characterization device 101 may include a rear baffle in the housing 345 to provide improved image contrast.

[0070] Figure 5 A functional diagram 500 illustrating the characterization apparatus and method is shown, wherein the characterization of the sample container 102 containing sample 212 is just one of many items that can be characterized or classified using a broader method employing the quality inspection module 430. According to one or more embodiments of the method, images are captured, for example, by a calibrated image capture device 328 (e.g., a calibrated monochrome camera). As described above, the images captured by the image capture device 328 can be multispectral and / or multi-exposure images. In particular, multiple exposures (e.g., 4 to 8 or more exposures) can be performed for each wavelength of light used for illumination (e.g., R, G, and B). Front-lit images can be captured (obtained) using front light sources 300A, 300B, and backlit images can be obtained using a backlight source 400C, such as... Figures 4A to 4B As described in [the text]. Alternatively, a white light source and a color camera can be used to obtain multi-exposure images with frontal illumination.

[0071] Then, the image can be further processed to determine segmentation 550, as described in US10,816538, entitled "Methods and Apparatus For Detecting An Interferent In A Specimen," by Kluckner et al., and US2019 / 0041318, entitled "Methods and Apparatus For Imaging A Specimen Container Using Multiple Exposures," by Wissmann et al. Other suitable segmentation methods based on artificial intelligence, such as convolutional neural networks (CNNs), can be used. In some embodiments, images from front lighting can be optimally used for segmentation 550. Similarly, images captured using backlighting can be optimally used for HILN classification 552 and / or artificial interference detection 556 using the methods described above.

[0072] Liquid quantization 554 can also be performed after segmentation 550. Liquid quantization may involve determining certain physical dimensional characteristics of sample 212, such as the physical locations of LA and SB, and / or determining the volumes of HSP, HSB, and / or HTOT, and / or the volumes of serum or plasma portions and / or sedimented blood portions. Identification can be accomplished by selecting pixels at these boundary regions and averaging their positional values ​​in pixel space to obtain the values ​​of LA and SB. Based on this information, the volume of serum or plasma portion 212SP can be determined using the width W and cross-sectional shape of sample container 102. The correlation from pixel space to mechanical measurement results can be achieved by using any suitable calibration method to calibrate the pixel space in pixels to the mechanical space in millimeters.

[0073] Further characterization of the sample container 102 can be performed using characterization methods, such as determining the 3D center position 350. As described above, the 3D path trajectory is first determined using 3D path trajectory determination 551, and then the 3D center position 350 is determined in the 3D center position determination block 553. Tube type detection 558, cap type detection 560, and cap color detection 562 can be implemented based on processing images from the image capture device 328 using conventional methods.

[0074] Figure 6A flowchart is shown of a characterization method 600 for determining the position (3D center position 350) of a sample container (e.g., sample container 102) at an imaging location on a track (e.g., track 121) according to one or more embodiments. Method 600 includes, in block 602, providing a calibration object (e.g., calibration object 235) on the track, and in block 604, providing an initially calibrated image capture device (e.g., calibrated image capture device 328) near the track. The calibration of the initially calibrated image capture device 328 can be performed by any suitable method, such as by using a marker grid (e.g., a checkerboard or Huffman marker) of known size and using nonlinear thinning techniques to optimize inherent parameters such as focal length, image center, skewness, and distortion coefficients.

[0075] Method 600 further includes, in block 606, moving the calibration object to at least two distinct longitudinal positions along track 121, including a first longitudinal position (e.g., Figure 3B The longitudinal position A) and the second longitudinal position (e.g., Figure 3B The first longitudinal position is different from the second longitudinal position, and in block 608, a first image is captured using a calibrated image capture device when the calibration object 325 is located at the first longitudinal position A, and in block 610, a second image is captured using an image capture device when the calibration object 325 is located at the second longitudinal position B. Once the images are obtained, method 600 includes, in block 612, determining a three-dimensional path trajectory 344 of the center position along the track (the segment of track 121 within the imaging region 335) based at least on the first and second images.

[0076] Once the three-dimensional path trajectory (three-dimensional path trajectory 344) within the imaging region 335 is known, it can be used to determine the 3D center position (e.g., 3D center position 350) of any sample container 102 brought into the imaging region on the carrier 122.

[0077] like Figure 7According to one or more embodiments, a flowchart of a method 700 for determining the 3D center position (e.g., 3D center position 350) of a sample container (e.g., sample container 102) on a track (e.g., track 121) is provided. Method 700 includes, at block 702, moving a sample container (e.g., sample container 102) carried by a carrier (e.g., carrier 122) on the track (e.g., track 121) to an imaging region (e.g., imaging region 335). An exact position for imaging within imaging region 335 is not required. Preferably, sample container 102 may stop in imaging region 335 for imaging, but sample container 102 may not need to stop there if the image capture speed is sufficient. Next, at block 704, the method includes imaging the sample container (e.g., sample container 102) within the imaging region (e.g., imaging region 335) to obtain a container image (e.g., container image 336). Method 700 further includes discovering a central plane (e.g., central plane 346) by means such that, in box 706, the side edges (e.g., edges 341, 342) of the sample container (e.g., sample container 102) are found in the container image (e.g., container image 336), and in box 708, a central plane (e.g., central plane 346) is determined between these edges. Finally, method 700 operates to back-project the central plane to find the intersection between central plane 346 and a three-dimensional path trajectory 344 (e.g., from the previously described method 600), wherein the intersection is the 3D center position 350 of the sample container 102 at the location of central plane 346.

[0078] As part of the edge detection box 706, method 700 may include identifying the width W of sample container 102. Based on the calibration of image capture device 328, pixel width can be simply converted to a distance in millimeters. The height HT of sample container 102 can be determined using similar conventional edge detection methods, where the top of tube 213 at TC is determined. Edge detection can be performed by segmenting or otherwise searching for light intensity transitions above a threshold within the area of ​​imaging region 335 where the tube cap interface TC is likely to be located.

[0079] In some embodiments, once a size characterization, such as in terms of width W and height HT, has been assigned to sample container 102, the volume of sample 212 can be obtained. The inner width can be determined, for example, by using a lookup table based on the size of sample container 102. For example, the inner width can be used to accurately calculate the volume of serum or plasma fraction 212SP and / or the volume of sedimented blood fraction 212SB based on the positions of the serum-blood interface SB and liquid-air interface LA obtained from the segmentation.

[0080] Therefore, based on the foregoing, it should be clear that the characterization methods 600 and 700 performed by the characterization device 101 (which may be included in the quality inspection modules 130 and 430 or may be a standalone characterization device 101) can rapidly characterize the 3D trajectory path 344 and the 3D center position 350 of the sample container 102. The characterization device 101 can also be used to obtain the physical properties of the sample container 102, such as tube size (W and HT), cap type, and cap color. In some embodiments including backlighting, such as Figures 4A to 4B Those shown can also achieve HIL detection and / or detection of man-made interference.

[0081] While this disclosure may have various modifications and alternatives, specific device embodiments and methods thereof are illustrated by way of example in the accompanying drawings and described in detail herein. However, it should be understood that this disclosure is not intended to limit it to the specific device or method disclosed, but rather to cover all modifications, equivalents, and alternatives falling within the scope of the claims and their equivalents.

Claims

1. A method of determining a position of a sample container on a track, comprising: providing a calibration object on the track; providing an initially calibrated image capture device in proximity to the track; moving the calibration object to at least two different longitudinal positions along the track, the at least two different longitudinal positions comprising a first longitudinal position and a second longitudinal position, the first longitudinal position being different from the second longitudinal position; capturing a first image with the initially calibrated image capture device with the calibration object positioned at the first longitudinal position; capturing a second image with the initially calibrated image capture device with the calibration object positioned at the second longitudinal position; determining a three-dimensional path trajectory of a center position along the track based on at least the first image and the second image; moving a sample container carried by a carrier on the track to an imaging region; imaging the sample container within the imaging region to obtain a container image; finding lateral edges of the sample container in the container image; determining a center plane between the lateral edges; and back-projecting the center plane to find an intersection between the center plane and the three-dimensional path trajectory, wherein the intersection is a three-dimensional center of the sample container at the position of the center plane.

2. The method of claim 1, further comprising stopping the sample container on the track within the imaging region at the time of imaging.

3. The method of claim 1, comprising determining a width W of the sample container.

4. The method of claim 1, comprising determining a height HT of the sample container.

5. The method of claim 1, wherein, The calibration object comprises a three-dimensional tool having a known geometry, and one or more calibration patterns are provided thereon.

6. The method of claim 1, wherein, The calibration object comprises a V-shaped marker tool comprising at least two planar surfaces.

7. The method of claim 6, wherein, The V-shaped marker tool comprises a Hofmann marker thereon.

8. The method of claim 1, comprising computing a relative extrinsic pose of a three-dimensional center of the calibration object with respect to the initially calibrated image capture device for at least the first image and the second image.

9. The method of claim 8, wherein, The computation of the relative extrinsic pose is done using Perspective-n-Point.

10. The method of claim 1, comprising capturing one or more additional images with the initially calibrated image capture device with the sample container positioned at one or more additional longitudinal positions along the track.

11. A characterization apparatus, comprising: a calibration object movable on a track; an initially calibrated image capture device in proximity to the track; and a computer coupled to the initially calibrated image capture device, the computer configured and operable to cause: the calibration object to move to at least two different longitudinal positions along the track, the at least two different longitudinal positions comprising a first longitudinal position and a second longitudinal position, wherein the second longitudinal position is different from the first longitudinal position, the initially calibrated image capture device to capture a first image with the calibration object positioned at the first longitudinal position, the initially calibrated image capture device to capture a second image with the calibration object positioned at the second longitudinal position, a three-dimensional path trajectory of a center position along the track to be determined based on at least the first image and the second image, a sample container carried by a carrier on the track to be moved to an imaging region, ​ imaging the sample container within the imaging area to obtain a container image, finding lateral edges of the sample container in the container image, determining a center plane between the lateral edges, and back-projecting the center plane to find an intersection between the center plane and the three-dimensional path trajectory, wherein the intersection is a three-dimensional center of the sample container at the location of the center plane.

12. The characterization device of claim 11, located near one or more of an analysis instrument, a loading station, a centrifugation station, a quality control module, and an aliquot machine station.

13. The characterization device of claim 11, comprising one or more light sources configured to backlight the sample container during imaging.

14. The characterization apparatus of claim 11, wherein, The calibration object comprises a three-dimensional tool having a known geometry, and one or more calibration patterns are provided thereon.

15. The characterization apparatus of claim 14, wherein, The calibration object comprises a V-shaped marker tool comprising at least two planar surfaces.

16. The characterization apparatus of claim 15, wherein, The V-shaped marker tool comprises a Huffman marker thereon.

17. The characterization apparatus of claim 11, wherein, The initially calibrated image capture device is an RGB camera in a quality check module.

18. A sample testing device, comprising: a track; a sample carrier movable on the track, the sample carrier configured to carry a sample container; and one or more characterization devices arranged around the track, each of the one or more characterization devices comprising: a calibrated image capture device near the track; and a computer coupled to the calibrated image capture device and configured to: determine a three-dimensional path trajectory of a center position along a section of the track based on at least a first image and a second image of a calibration object taken at an imaging area, cause a sample container carried by the carrier on the track to move to the imaging area, cause the sample container to be imaged within the imaging area to obtain a container image, determine a center plane between lateral edges of the sample container, and back-project the center plane to find an intersection between the center plane and the three-dimensional path trajectory, wherein the intersection is a three-dimensional center of the sample container at the location of the center plane.

Images