Biological sample measuring device

Abstract

To provide a technique which enables accurate identification of a target portion of a biological sample to be measured regardless of a label orientation without using a large image capturing mechanism when measuring the biological sample that is stored in a labeled container and separated into multiple component regions.SOLUTION: A biological sample measurement device disclosed herein is configured to identify a midpoint of luminance values of a captured image to identify an upper surface boundary and lower surface boundary of a measurement target portion, and applies a correction amounting to a meniscus width to a position of the identified upper surface boundary.SELECTED DRAWING: Figure 1

Description

【Technical Field】 【0001】 The present disclosure relates to an apparatus for measuring a biological sample separated into a plurality of component regions. 【Background Art】 【0002】 For the purpose of improving the efficiency of clinical tests such as blood tests, there is a need for a technology to automate the operation of confirming the volume of a biological specimen before unsealing (before dispensing), which has been conventionally performed by visual confirmation. In particular, biological samples such as blood specimens before dispensing are configured to be separated into multiple layers by centrifugation or the like, and there is a need for a technology to measure only the volume of the sample to be analyzed. In addition, since biological samples before dispensing may have identification barcode labels or pre-labels attached to blood collection tubes and shipped, there is a need for a technology that can measure in a state where these labels are attached. 【0003】 As such a biological sample measuring apparatus, for example, Patent Document 1 discloses a liquid detection apparatus that irradiates a specimen with infrared light, detects transmitted light with a line sensor, obtains the boundary between a label and a serum region (measurement target region) based on the first derivative value thereof, and measures the serum volume of the specimen. 【0004】 In Patent Document 2, by changing the amount of light and acquiring a signal of transmitted light, it is made possible to analyze regardless of the presence or absence of attenuation of transmitted light by a label. Further, Patent Document 2 discloses a technique for detecting the height of a predetermined region of a sample separated into multiple layers by irradiating the sample separated into multiple layers with pulsed light of two wavelengths by switching in a time division manner and measuring the transmitted light while scanning the sample in the vertical direction. 【Prior Art Documents】 【Patent Documents】 【0005】 【Patent Document 1】 Japanese Patent Application Laid-Open No. 2004-037320 【Patent Document 2】 US2012 / 0013889 【Summary of the Invention】 [Problems that the invention aims to solve] 【0006】 The technology described in Patent Document 1 measures the amount of serum in a biological sample composed of multiple components by detecting the vertical position of the label and the height of the upper and lower surfaces of the serum to be measured, using signal processing based on differentiation from the infrared transmitted light signal. However, the signal processing in Patent Document 1 has a problem in that, under conditions where transmitted light is scattered by the label, such as in samples where the label is attached to the detector side, the signal indicating the upper and lower surfaces of the serum becomes blurred due to scattering, resulting in a decrease in detection accuracy. Therefore, a rotation mechanism is required to detect and align the orientation of the label (orientation of the sample), making it difficult to miniaturize the device. 【0007】 The technology disclosed in Patent Document 2 relates to a liquid volume measurement technique for a biological sample composed of multiple components, which accurately identifies the boundary of the measurement target area and measures the liquid volume from the transmitted light signals of two wavelengths with different absorption rates relative to the measurement target. However, Patent Document 2 requires measurement while vertically scanning the sample (blood collection tube) along its long axis, which presents a challenge in miniaturizing the device due to the scanning mechanism. 【0008】 This disclosure has been made in view of the above-mentioned problems, and aims to provide a technology that can accurately identify the target portion to be measured, regardless of the orientation of the label, when measuring a biological sample separated into multiple component regions and stored in a labeled container, without using a large imaging mechanism. [Means for solving the problem] 【0009】 The biological sample measuring device according to this disclosure identifies the upper and lower boundaries of the portion to be measured by identifying the midpoint of the brightness value of the captured image, and applies a meniscus width correction amount to the position of the identified upper boundary. [Effects of the Invention] 【0010】 According to the biological sample measuring device described herein, when measuring a biological sample separated into multiple component regions and stored in a labeled container, it is possible to reduce measurement variability due to the orientation of the label, even under the influence of transmitted light scattering that occurs depending on the orientation of the label. As a result, it is possible to reduce the influence of transmitted light scattering by the label and improve analytical accuracy while achieving miniaturization without installing a rotating mechanism to align the label orientation. [Brief explanation of the drawing] 【0011】 [Figure 1] This is a diagram showing the configuration of the biological sample measuring device 1 according to Embodiment 1. [Figure 2A] This diagram explains the principle for identifying the measurement target area. [Figure 2B] This diagram explains the principle for identifying the measurement target area. [Figure 3A] This shows an example of a container for biological sample 2 with a label attached. [Figure 3B] This shows an example of a container for biological sample 2 with a label attached. [Figure 4A] This is an example of a transparent image. [Figure 4B] The following shows the results of extracting the slope of the boundary on the serum surface by imaging the same simulated sample with the label orientation (sample orientation) actually changed, using illumination at a wavelength (1550 nm) that is absorbed in the measurement target area. [Figure 4C] The following shows the results of extracting the slope of the boundary on the serum surface by imaging the same simulated sample with the label orientation (sample orientation) actually changed, using illumination at a wavelength (1550 nm) that is absorbed in the measurement target area. [Figure 5A] This is an example of a transparent image. [Figure 5B] The results of extracting the signal from the serum surface when imaging with illumination at a wavelength of 970 nm are shown. [Figure 6] This graph shows the results obtained by attaching two labels 301 to one side of the sample, with Figures 4B and 4C superimposed. [Figure 7A] This document describes a method for detecting the center point of brightness changes in a one-dimensional signal. [Figure 7B] Disclosed is a method for detecting the center point of the luminance change of a one-dimensional signal. [Figure 8] It is a flowchart for explaining the procedure in which the image processing unit 14 measures the liquid volume (serum liquid volume) of the measurement target area 21. [Figure 9] It is a flowchart for explaining the details of S2. [Figure 10] It is a flowchart for explaining the details of S23. [Figure 11] It is a flowchart for explaining the details of S24. [Figure 12] It is a flowchart for explaining the details of S3. [Figure 13A] Disclosed is a simulation model in which the meniscus is modeled. [Figure 13B] It is the result of performing a simulation of transmitted light (one-dimensional signal near the serum surface) by ray tracing simulation using the model of FIG. 13A. 【Mode for Carrying Out the Invention】 【0012】 <Embodiment 1> In an embodiment of the present disclosure, the biological sample to be measured is a specimen with a label attached before unsealing (before analysis), and it is assumed that the specimen is separated into a plurality of component layers (typically 1 to 3 layers) depending on whether or not centrifugation is performed. The measurement target is a part of the specimen separated into a plurality of layers, such as plasma or serum, which is an analysis target (dispensing target) by a biochemical analyzer or the like. <00001​​​​​​Figure 1 is a diagram showing the configuration of a biological sample measuring device 1 according to Embodiment 1 of the present disclosure. The biological sample measuring device 1 is a device for measuring a biological sample 2. The biological sample 2 is a sample configured as described above. The biological sample measuring device 1 identifies a measurement target area 21 of the biological sample 2 and measures its liquid volume. The biological sample measuring device 1 includes a surface illumination light source 11, an area camera 12, a time-division control driver 13, and an image processing unit 14. 【0015】 The surface illumination light source 102 is configured to switch the wavelength of the emitted light between two wavelengths, and uses this light to illuminate the biological sample 2. The light emitted by the surface illumination light source 11 can simultaneously illuminate two or more component layers constituting the biological sample 2 (i.e., across two or more component layers). In addition, it can illuminate the upper surface of the topmost layer (the boundary between the sample and the air layer) regardless of the component layer. Switching the wavelength of light does not necessarily require emitting light with only a single wavelength; it is sufficient to switch the wavelength component with the strongest intensity (e.g., wavelength λ1 = 1550 ± 100 nm, wavelength λ2 = 970 ± 100 nm). In a biological sample 2 with one or three component layers, if the number of component layers is known before measurement, it is sufficient to use either of the two wavelengths, and there is no need to switch wavelengths. The reason and method of wavelength selection will be explained later in the measurement principle section. 【0016】 The area camera 12 generates a two-dimensional image of the biological sample 2 by capturing the surface illumination light that has passed through the biological sample 2. The area camera 12 has sensitivity characteristics that allow it to detect light in the wavelength range emitted by the surface illumination light source 11. The area camera 12 can be constructed using, for example, an InGaAs camera. 【0017】 The time-division control driver 13 switches the wavelength of light emitted by the surface illumination light source 11 in a time-division manner. The time-division control driver 13 adjusts the exposure time (or gain) of the area camera 12 to a time suitable for that wavelength in synchronization with the wavelength switching. The imaging timing by the area camera 12 is controlled by the time-division control driver 13 in synchronization with the wavelength emitted by the surface illumination light source 11. The time-division control driver 13 may also receive processing results from the image processing unit 14 and control re-imaging according to those results. 【0018】 The image processing unit 14 extracts the measurement target area (image area of ​​the sample) from the image captured by the area camera 12. The extraction principle will be described later. The image processing unit 14 includes a measurement target area identification unit 141, a meniscus correction unit 142, and a liquid volume calculation unit 143. The operation of these units will also be described later. 【0019】 Figures 2A and 2B illustrate the principle of identifying the measurement target area. In the configuration shown in Figure 1, the area camera 12 captures a two-dimensional transmission image of the biological sample 2, as shown in Figures 2A and 2B. 【0020】 Figure 2A shows an example where the biological sample 2 is separated into three layers. Figure 2B shows an example where the biological sample 2 is separated into two layers. The blood clot 23 is formed in the lower layer when the biological sample 2 is centrifuged. The separation agent 22 is added to separate the blood clot 23 from the measurement target area 21. 【0021】 Comparing the first wavelength (wavelength 1) and the second wavelength (wavelength 2) emitted by the surface illumination light source 102, the transmittance when wavelength 1 passes through the blood clot 23 is approximately the same as the transmittance when wavelength 2 passes through the blood clot 23. Therefore, the difference between the image of the blood clot 23 acquired using wavelength 1 and the image of the blood clot 23 acquired using wavelength 2 is very small. Similarly, for the separating material 22, the transmittance of wavelength 1 and the transmittance of wavelength 2 are approximately the same, so the difference between the two images is very small. 【0022】 In contrast, the transmittance when wavelength 1 passes through the measurement target region 21 is significantly different from the transmittance when wavelength 2 passes through the measurement target region 21. Therefore, the difference between the image of the measurement target region 21 acquired using wavelength 1 and the image of the measurement target region 21 acquired using wavelength 2 is significant. By identifying this difference, the measurement target region 21 can be extracted from the image. 【0023】 The wavelength bands to be used as wavelength 1 and wavelength 2 must be selected in advance so that, as illustrated in Figures 2A and 2B, a significant difference occurs between the wavelengths in the measurement target region 21, but almost no difference occurs in other areas. As long as this condition is met, the specific numerical values ​​of the wavelengths can be arbitrary. That is, at a minimum, the difference between the transmittance when wavelength 1 passes through the measurement target region 21 and the transmittance when wavelength 2 passes through the measurement target region 21 must be greater than the difference between the transmittance when wavelength 1 passes through areas other than the measurement target region 21 and the transmittance when wavelength 2 passes through areas other than the measurement target region 21. 【0024】 By using the above measurement principle, the measurement target region 21 can be identified even when the biological sample 2 is separated into multiple component layers. The image processing unit 14 identifies the measurement target region 21 according to this principle. The number of component layers does not matter, but typical biological samples such as plasma and serum are separated into 1 to 3 layers. In either case, the measurement target region 21 can be identified with high accuracy. 【0025】 For biological samples 2 with one component layer and biological samples 2 with three component layers, if the number of component layers is known in advance before measurement, measurement is possible using a transmission image of only one wavelength. In the case of three layers as shown in Figure 2A, a wavelength should be selected that has a difference in absorption rate between the separation material 22 or air layer located above and below the measurement target area 21 and the measurement target area 21. In the case of one layer, a wavelength should be selected that has a difference in absorption rate between the air layer above the measurement target area 21 and the measurement target area 21 itself. 【0026】 <Embodiment 1: Issues with transmitted light scattering by labels> Figures 3A and 3B show examples where a label is attached to the container of the biological sample 2. Figure 3A is a side view of the container, and Figure 3B is a top view. The container of the biological sample 2 (e.g., a blood collection tube) may have a barcode label or pre-label (label 301) attached to it. Even in such cases, it is required to accurately identify the measurement target area 21. Furthermore, if there is no mechanism to control the orientation of the biological sample 2, the positional relationship between the label 301, the surface illumination light source 11, and the area camera 12 will vary for each biological sample 2. For example, the label 301 may be on the illumination side (Figure 3B(1)) or on the opposite side from the illumination (Figure 3B(2)). In both cases, it is required to identify the measurement target area 21 with the same accuracy. 【0027】 As shown in Figure 3, when the orientation of label 301 is changed (the orientation of biological sample 2 is changed) and a transmission image is acquired, it has been found that the serum boundary becomes blurred when label 301 is on the detector (area camera 12) side. This is because the light transmitted through the sample is scattered by label 301 on the surface of the sample. 【0028】 Figure 4A is an example of a transmission image. Figures 4B to 4C show the results of extracting the slope of the boundary of the serum surface by actually changing the orientation of the label (orientation of the sample) and imaging the same simulated sample using illumination at a wavelength (1550 nm) that is absorbed in the measurement target area. The image processing unit 14 obtained a one-dimensional signal by extracting a 3 mm width from the center of the sample from the image acquired at a wavelength of 1550 nm and averaging it in the x direction. The "label illumination side" corresponds to Figure 3B(1), and the "label camera side" corresponds to Figure 3B(2). Figure 4B shows the measurement results for a sample with one label 301 attached to the camera (area camera 12) side, a sample with two labels 301 attached to the camera side, and a sample with labels 301 attached all around (2 in total). Figure 4C shows the measurement results for a sample with one label 301 attached to the camera side, a sample with two labels 301 attached to the camera side, and a sample without a label 301. 【0029】 As shown in Figures 4B to 4C, the slope of the signal near the serum boundary differs significantly depending on whether the label 301 is on the camera side or not. Therefore, in liquid volume measurement, it is necessary to detect the boundary position of the measurement target area 21 from the measurement signal which has a different slope depending on the orientation of the label 301, that is, to reduce the variation in detection coordinates due to the orientation of the label 301. 【0030】 Figures 5A and 5B show the results of extracting the signal from the serum surface when imaging with illumination at a wavelength of 970 nm. The image processing unit 14 obtained a one-dimensional signal by extracting a 3 mm width from the center of the sample from the image acquired at a wavelength of 970 nm, similar to Figures 4B and 4C, and averaging it in the x direction. Figure 5B shows the results when two labels 301 are attached to the illumination (surface illumination light source 11) side and when two labels 301 are attached to the camera side. 【0031】 Illumination at a wavelength of 970 nm shows less attenuation in the serum region compared to illumination at 1550 nm. However, for samples where label 301 is on the illumination side, a characteristic attenuation of transmitted light is observed in the meniscus region on the upper surface of the serum. This is because the transmitted light is scattered and refracted by the curved meniscus. On the other hand, when label 301 is on the camera side, the shape of the dip is unclear. This is because, as in Figures 4B to 4C, the transmitted light is scattered by the label on the sample surface. In other words, when the label is not on the camera side, the underside of the meniscus can be accurately measured by detecting the dip in the transmitted light at a wavelength of 970 nm. However, when the label is on the camera side, the dip becomes blurred due to scattering, making it difficult to detect the underside of the meniscus. 【0032】 As described above, the challenges are to detect the lower surface of the meniscus under conditions where scattering of transmitted light occurs due to the label 301, and to reduce the variation in detection position due to the orientation of the label 301. 【0033】 <Embodiment 1: Principle of serum boundary signal processing> Figure 6 is a graph showing the results obtained by attaching two labels 301 to one side of the sample, with Figures 4B and 4C superimposed. As shown in Figure 6, the slope of the signal differs greatly depending on the orientation of the label, but both exhibit point symmetry at a point around 50% brightness. Therefore, in this disclosure, we aim to accurately detect the boundary of the measurement target area 21 regardless of the label orientation by utilizing this symmetry and detecting the center point of the brightness change. 【0034】 In the image region encompassing the boundary of the measurement target area 21 (either the upper boundary or the lower boundary), the luminance value of the one-dimensional signal takes on a maximum value (the left half of Figure 6) and a minimum value (the right half of Figure 6). Here, 50% luminance refers to the luminance value corresponding to the midpoint between these maximum and minimum values. 【0035】 Figures 7A and 7B show a method for detecting the center point of the brightness change in a one-dimensional signal. The image processing unit 14 can detect the inflection point of the one-dimensional signal by detecting the peak of the first derivative of the one-dimensional signal (e.g., detecting the zero-crossing point of the second derivative of the one-dimensional signal), and can identify this as the center point of the brightness change (i.e., the boundary of the measurement target area 21). Figure 7A shows the profile of the one-dimensional signal and its first derivative acquired with light of wavelength 1550 nm when the label 301 is on the camera side. Figure 7B shows the profile of the one-dimensional signal and its first derivative acquired with light of wavelength 1550 nm when the label 301 is on the illumination side. It can be seen that the peak coordinate of the first derivative corresponds to the coordinate around 50% of the brightness. Therefore, by identifying the peak position of the first derivative of the one-dimensional signal, the boundary of the measurement target area 21 can be identified. The same applies to both the upper and lower boundaries. 【0036】 Another method for detecting the center point of the brightness change in a one-dimensional signal is to identify the maximum and minimum values ​​of the brightness change near the boundary of the measurement target area 21, and then determine the center point between them to identify the point where the brightness value is at 50% of the maximum value. The same method applies to both the upper and lower boundaries. This method is preferable when the first derivative peak of the one-dimensional signal does not coincide with the center point. 【0037】 For example, depending on the individual and species variations of the biological sample 2, the boundary of the measurement target area 21 may not be the midpoint between the maximum and minimum brightness near the boundary (it may not be at the 50% brightness level). In such cases, the image processing unit 14 can also treat a brightness level that minimizes such variations as the boundary of the measurement target area 21. For example, a brightness level slightly closer to the maximum brightness than the midpoint between the maximum and minimum brightness near the boundary of the measurement target area 21 (e.g., the 60% brightness level) may be considered as the boundary. In this case, instead of the midpoint between the maximum and minimum brightness, the midpoint between them (in this example, the 60% brightness level) is used. 【0038】 Similarly, when using the first derivative peak, the boundary position obtained by the inflection point of the brightness change may be further corrected and considered as the final boundary. For example, if the boundary position obtained using the first derivative peak and the boundary position obtained using the brightness center point are different, one possible method is to consider the midpoint between the two as the boundary. 【0039】 Such boundary position correction can be performed by first determining the correction amount through experiments or other means, and then applying that correction amount to the boundary position determined by the first derivative peak or luminance center point. The process of determining the correction amount can be performed, for example, in a calibration process before conducting measurements. 【0040】 Figure 8 is a flowchart illustrating the procedure by which the image processing unit 14 measures the volume of fluid (serum volume) in the measurement target area 21. The image processing unit 14 acquires transmission images of two wavelengths captured by the area camera 12 (S1). The measurement target area identification unit 141 identifies the measurement target area 21 from the images of each wavelength (S2: details will be described later). The meniscus correction unit 142 applies a meniscus width correction to the boundary of the identified measurement target area 21 (S3: details will be described later). The fluid volume calculation unit 143 calculates the volume of fluid in the measurement target area 21 (S4). 【0041】 Figure 9 is a flowchart illustrating the details of S2. The following describes each step in Figure 9. 【0042】 (Figure 9: Step S21) The measurement target area identification unit 141 acquires a one-dimensional signal from the acquired spectral image. The one-dimensional signal is obtained by extracting a predetermined width of the area near the center along the horizontal direction from the image of the biological sample 2 and using its pixel values. The width of the extracted central area is between 1 pixel and the width of the biological sample 2. The average or median of the pixel values ​​in the central area is used as the one-dimensional signal. 【0043】 (Figure 9: Step S22) The measurement target area identification unit 141 acquires the difference between the one-dimensional signal of the transmitted image acquired using illumination at a wavelength of 970 nm and the one-dimensional signal of the transmitted image acquired using illumination at a wavelength of 1550 nm. Based on this difference, the measurement target area identification unit 141 provisionally identifies the measurement target area 21. Specific examples are explained in Figures 2A and 2B. 【0044】 (Figure 9: Step S23) The measurement target area identification unit 141 acquires a partial image of the area near the upper boundary of the measurement target area 21 that was provisionally identified in S22, and uses this partial image to identify the upper boundary of the measurement target area 21. For example, it is sufficient to acquire a partial image of the area within approximately ±5 mm of the boundary of the measurement target area 21 that was provisionally identified in S22. S23 improves noise immunity in subsequent analysis. If there is little noise, this step can be omitted. The same applies to S24. Details of this step will be described later. 【0045】 (Figure 9: Step S24) The measurement target area identification unit 141 acquires a partial image of the area near the lower boundary of the measurement target area 21 that was provisionally identified in S22, and uses this partial image to identify the lower boundary of the measurement target area 21. The range of the partial image is, for example, approximately ±5 mm of the boundary of the measurement target area 21 that was provisionally identified in S22. Details of this step will be described later. 【0046】 Figure 10 is a flowchart illustrating the details of S23. This flowchart implements the procedure described in Figures 6 to 7B. The serum surface is analyzed using a signal at a wavelength absorbed by the serum, such as 1550 nm. This flowchart shows an example using the peak of the first derivative of a one-dimensional signal, but the maximum and minimum brightness values ​​near the boundary may also be used, or a combination of these may be used. 【0047】 The measurement target area identification unit 141 acquires the first derivative of the one-dimensional signal (S231) and detects the peak of that first derivative (S232). A smoothing filter, such as a Gaussian filter, may be applied to the one-dimensional signal before acquiring the first derivative or to the first derivative before peak detection (both may be applied). This has the effect of making it easier to separate from noise. The measurement target area identification unit 141 selects from the detected peaks that have an absolute value greater than or equal to a preset threshold (S233). For example, from the detected peaks, the unit selects the peak that is closer to the serum side, using the upper boundary of the measurement target area provisionally identified in S22 of Figure 9 as a reference. Alternatively, the system may check whether the serum boundary has been correctly detected by checking whether the brightness value of the one-dimensional signal at coordinates before and after the candidate peak (a few millimeters) increases or decreases before and after the peak. This allows for separation from noise and improves accuracy. As a result, the upper boundary of the measurement target area 21 can be identified (S234). 【0048】 Figure 11 is a flowchart illustrating the details of S24. As explained in Figures 2A and 2B, the wavelength at which contrast is produced at the lower boundary of the measurement target region 21 varies depending on the substance present below the measurement target region 21. In the examples of Figures 2A and 2B, the wavelength of the transmission image used differs depending on the number of layers constituting the biological sample 2. For example, in Figure 2A, wavelength 2 is used to identify the lower boundary, and in Figure 2B, wavelength 1 is used to identify the lower boundary. Therefore, the measurement target region identification unit 141 switches which wavelength to use according to the number of layers constituting the biological sample 2. 【0049】 The measurement target area identification unit 141 confirms the number of layers constituting the biological sample 2 (S241). Specifically, the difference in brightness values ​​(contrast) of a partial region spanning the lower boundary provisionally identified in S2 is obtained based on the image or one-dimensional signal acquired at wavelengths of 970 nm and 1550 nm. If the contrast at 970 nm is greater, it is a two-layer sample, and the lower boundary is detected by analyzing the 970 nm transmission image. If the contrast at 1550 nm is greater, it is a three-layer sample, and the lower boundary is detected by analyzing the 1550 nm transmission image. If the contrast at both wavelengths is about the same (the contrast across the lower boundary is below the threshold), it is a one-layer sample, and the lower boundary is detected by analyzing the 1550 nm transmission image. S242 to S245 are the same as S231 to S234. 【0050】 Figure 12 is a flowchart illustrating the details of S3. The meniscus correction unit 142 acquires the upper boundary coordinates identified in S2 (S31). The meniscus correction unit 142 acquires the meniscus width (S32). The meniscus correction unit 142 has previously acquired the expected meniscus width from the blood collection tube to be used and experimental data acquired in advance. Alternatively, while continuously measuring samples, the meniscus width may be determined from the 970nm transmitted light signal (at which time a characteristic dip can be seen in the meniscus region) when the label 301 is on the illumination side. The meniscus correction unit 142 corrects the position of the upper boundary downward (towards the lower surface of the meniscus) by the width obtained by multiplying the meniscus width by a correction coefficient (S33). An example of the correction coefficient will be described later. The meniscus correction unit 142 determines the position of the upper boundary after correcting the meniscus width as the final upper boundary position (S34). 【0051】 Figure 13A shows a schematic simulation model of the meniscus. The upper boundary of the measurement target area 21 is the lower surface of the meniscus. Here, the lower surface of the meniscus is assumed to be at a Y coordinate of 15 mm. The meniscus width is 1.7 mm. 【0052】 Figure 13B shows the results of a ray tracing simulation of transmitted light (one-dimensional signal near the upper surface of the serum) using the model in Figure 13A. The upper surface boundary identified in S2 (50% brightness level in Figure 13B) is shifted upward from the lower surface of the meniscus due to the influence of the meniscus. The meniscus correction unit 142 corrects this shift, bringing the estimated position of the upper surface boundary closer to the lower surface of the meniscus. The correction coefficient in Figure 12 is defined by the ratio between this shift amount and the meniscus width. In Figure 13B, the simulation model in Figure 13A shows that the coordinate of the 50% brightness level is above the lower surface of the meniscus by meniscus width × 0.94. Therefore, the meniscus correction unit 142 calculates the correction amount by multiplying the previously acquired meniscus width by this correction coefficient (the correction coefficient is 0.94 in the result of Figure 13B), and applies this correction amount to correct the position of the upper surface boundary downward toward the lower surface of the meniscus. 【0053】 Considering the mechanism by which the estimated position of the upper boundary shifts due to the meniscus, the correction coefficient should preferably be a value between 0.5 and 1.0. The correction coefficient can be determined in advance by simulation or calibration. The correction coefficient may be stored in advance in the memory of the biological sample measurement device 1, or it may be obtained from a memory device of an external device. A correction coefficient may be determined in advance for each type of biological sample 2, and the correction coefficient used may be selected according to the type of biological sample 2 at the time of measurement. 【0054】 By following the above procedure, the biological sample measuring device 1 can accurately identify the upper and lower boundaries of the measurement target area 21 without adjusting the position of the label 301 attached to the biological sample 2 (position relative to the surface illumination light source 11 or position relative to the area camera 12). 【0055】 <Embodiment 2> In Embodiment 1, the biological sample measuring device 1 can be installed within an existing device as an additional option, or it can be used as a standalone device. 【0056】 In Embodiment 1, a specific example of the wavelength used to obtain a transmission image was described, assuming a specific composition of the biological sample 2 (serum sample). This disclosure is not limited to this and can be applied to other types of samples. That is, by appropriately selecting the wavelength to obtain a transmission image as exemplified in Figures 2A to 2B based on the absorbance characteristics of the sample, it can be applied to other samples. 【0057】 The image processing unit 14 (and each functional unit provided within the image processing unit 14) can be configured by hardware such as circuit devices that implement the function, or by a computing device such as a CPU (Central Processing Unit) executing software that implements the function. 【0058】 As a method for acquiring a one-dimensional signal, one may extract a predetermined width of the area near the center along the horizontal direction from the image of the biological sample 2, and calculate the average or median of the pixel values ​​from that area to obtain a one-dimensional signal. Alternatively, each pixel row of the two-dimensional image of the biological sample 2 may be treated as a one-dimensional signal, and the analysis process shown in Figure 9 may be performed on a pixel row basis. When performing the analysis on a pixel row basis, the upper and lower surfaces are determined by calculating the average or median of the boundary coordinates determined for each pixel row. [Explanation of symbols] 【0059】 1: Biological sample measuring device 11: Area lighting light source 12: Area Camera 13: Time-division control driver 14: Image Processing Unit 2: Biological samples 21: Measurement target area

Claims

[Claim 1] A biological sample measuring device for measuring a biological sample separated into multiple component regions, A light source that irradiates the biological sample with first light having a first wavelength component and second light having a second wavelength component, An imaging device that generates an image of the biological sample using the first and second light transmitted through the biological sample. The system includes an image processing unit that identifies the target portion to be measured from the component region in the captured image, The image processing unit obtains the brightness value of the first light along the height direction of the target portion from the captured image obtained using the first light, and identifies the position of the upper boundary of the target portion along the height direction by identifying the midpoint of the change in the brightness value of the first light in the first partial region of the captured image. The image processing unit obtains the brightness value of the first light or the second light along the height direction from the captured image, and identifies the midpoint of the change in the brightness value of the first light or the second light in the second partial region of the captured image, thereby determining the position of the lower boundary of the target portion along the height direction. The image processing unit corrects the position of the identified upper boundary by applying a correction amount to correct the meniscus width of the target portion to the identified upper boundary position, The image processing unit measures the liquid volume of the target portion based on the identified upper boundary position and the identified lower boundary position, The image processing unit calculates the correction amount by multiplying the meniscus width by a correction coefficient, The image processing unit obtains the correction coefficient from a storage unit that stores the correction coefficient corresponding to the type of biological sample, The light source is a surface illumination light source, and the imager is an area camera. The transmittance when the first wavelength component passes through the target portion and the transmittance when the second wavelength component passes through the target portion are different from each other. A biological sample measuring device characterized by the following features. [Claim 2] The image processing unit provisionally determines the positions of the upper boundary and the lower boundary along the height direction from the captured image acquired using the first light and the second light, The image processing unit identifies the position of the upper boundary along the height direction by identifying the midpoint of the change in the brightness value of the first light in the first partial region encompassing the provisionally identified upper boundary, The image processing unit identifies the position of the lower boundary along the height direction by identifying the midpoint of the change in the brightness value of the first or second light in the second partial region encompassing the provisionally identified lower boundary. The biological sample measuring device according to claim 1, characterized in that it is a biological sample measuring device. [Claim 3] The imaging device generates a first image of the biological sample using the first wavelength component of the first light. The imaging device generates a second image of the biological sample using the second wavelength component of the second light. The image processing unit calculates the difference between the portion generated using the first wavelength component in the first captured image and the portion generated using the second wavelength component in the second captured image. The image processing unit identifies the portion where the difference exceeds a threshold, thereby provisionally determining the positions of the upper boundary and the lower boundary. The biological sample measuring device according to claim 2, characterized in that it is a biological sample measuring device. [Claim 4] The imaging device is an area camera that generates a two-dimensional image of the biological sample as the captured image using the first or second light transmitted through the biological sample. The image processing unit acquires a one-dimensional signal of the first or second light by statistically processing the pixel values ​​of the two-dimensional image along the horizontal direction in the central region of the biological sample along the horizontal direction perpendicular to the height direction. The biological sample measuring device according to claim 1, characterized in that it is a biological sample measuring device. [Claim 5] The image processing unit obtains the first derivative of the luminance value with respect to a position along the height direction and identifies the peak of the first derivative to determine the position of the upper boundary or the lower boundary. The biological sample measuring device according to claim 1, characterized in that it is a biological sample measuring device. [Claim 6] The image processing unit determines the maximum and minimum values ​​of the luminance value in the region encompassing the upper boundary identified in the first partial region or the lower boundary identified in the second partial region, respectively. The image processing unit identifies the position in the height direction corresponding to the brightness value obtained by adding a predetermined percentage of the difference between the maximum value and the minimum value to the minimum value as the position of the upper boundary or the position of the lower boundary. The aforementioned predetermined percentage is 50% or 60%. The biological sample measuring device according to claim 1, characterized in that it is a biological sample measuring device. [Claim 7] The image processing unit corrects the position of the upper boundary by applying the correction amount to the position of the upper boundary that was identified. The biological sample measuring device according to claim 1, characterized in that it is a biological sample measuring device. [Claim 8] The image processing unit corrects the position of the upper boundary using the correction coefficient obtained from the storage unit. The biological sample measuring device according to claim 7, characterized in that it is a biological sample measuring device. [Claim 9] The correction coefficient is configured such that the correction amount is 0.5 to 1.0 times the meniscus width. The biological sample measuring device according to claim 7, characterized in that it is a biological sample measuring device. [Claim 10] The image processing unit compares the first contrast of the lower boundary provisionally identified using the first captured image with the second contrast of the lower boundary provisionally identified using the second captured image. If the first contrast is greater than the second contrast by a predetermined value or more, the image processing unit obtains the luminance value of the first light along the height direction from the captured image obtained using the first light, and identifies the position of the lower boundary along the height direction by identifying the inflection point of the change in the luminance value of the first light in the region encompassing the provisionally identified lower boundary. If the second contrast is greater than the first contrast by a predetermined value or more, the image processing unit obtains the luminance value of the second light along the height direction from the captured image obtained using the second light, and identifies the position of the lower boundary along the height direction by identifying the inflection point of the change in the luminance value of the second light in the region encompassing the provisionally identified lower boundary. The biological sample measuring device according to claim 3. [Claim 11] If the difference between the first contrast and the second contrast is less than the predetermined value, the image processing unit obtains the luminance value of the first light along the height direction from the captured image obtained using the first light, and identifies the position of the lower boundary along the height direction by identifying the inflection point of the change in the luminance value of the first light in the region encompassing the provisionally identified lower boundary. The biological sample measuring device according to claim 10, characterized in that it is a biological sample measuring device. [Claim 12] A label is affixed to the side of the container that holds the biological sample. The biological sample measuring device acquires the captured image using the imager without adjusting the positional relationship between the imager and the label. The biological sample measuring device according to claim 1, characterized in that it is a biological sample measuring device.

Images