Automatic analysis device and automatic analysis method

Abstract

This invention provides a technique for inferring bacterial concentration from a sample containing bacteria and impurities and adjusting the bacterial concentration in the sample to a desired value. The automated analysis device of this invention introduces a substance that destroys the impurities into the sample containing bacteria and impurities, separates the destroyed impurities from the bacteria, and then removes the bacteria using a filter. The concentration of bacteria in the sample is inferred based on the correspondence between the amount of impurities remaining on the filter and the concentration of bacteria in the sample (refer to Figure 5).

Description

Technical Field

[0001] This invention relates to an automated analytical apparatus for analyzing samples containing bacteria and impurities. Background Technology

[0002] Sepsis is a highly fatal infectious disease, making rapid diagnosis and appropriate treatment based on the diagnosis crucial. Blood culture is typically performed to diagnose sepsis. This determines the presence of bacteria in a sterile blood sample. A smear examination is usually performed after the blood culture, followed by differential and susceptibility testing. Differential testing involves isolating and culturing the positive blood culture sample to identify the bacterial species based on the resulting colonies. Susceptibility testing determines the bacteria's susceptibility to antibiotics. Of these tests, blood culture takes one day, isolation culture takes one day, and susceptibility testing takes one day, so the overall testing time is 2-3 days. That is, it currently takes 2-3 days to determine whether the antibiotics being administered in the ongoing treatment are appropriate. Therefore, if ineffective antibiotics are used, the mortality rate of sepsis becomes extremely high.

[0003] In blood culture tests, a very small number of bacteria, typically around 10 CFU / mL (CFU: colony-forming units), present in cases of sepsis, are proliferated in the culture flask. Incubation is usually performed for about 8 hours to one night, allowing the bacteria to multiply to a level where gaseous components produced by bacterial respiration or fermentation can be detected. A known positive blood culture contains approximately 10 CFU / mL. 6 ~10 10 The bacterial concentration at CFU / mL is considered positive for blood cultures. The bacterial concentration varies depending on the patient's blood condition, bacterial species, and the blood culture apparatus, hence this broad range. Besides blood cells and culture medium, the main components in a blood culture bottle include antibiotic-adsorbing resins, beads, and activated carbon. Among these components, red and white blood cells are present in high concentrations, at approximately 10 CFU / mL. 9 cells / mL, 10 7 The concentration is approximately [number] cells / mL, which is the same as or higher than the concentration of bacteria.

[0004] In isolation culture, blood culture-positive samples are plated onto agar medium to allow colonies to develop. Bacterial species are more reliably identified by predicting their characteristics based on colony traits, and bacterial suspensions are prepared using these colonies to obtain bacterial concentrations (typically 10⁻⁶) free of impurities other than bacteria, meeting the sensitivity requirements for testing. 5 ~10 6 The sample contains CFU / mL.

[0005] In susceptibility testing, a specific concentration of antibiotic is typically introduced into a bacterial culture containing bacteria, and the degree of bacterial proliferation corresponding to that antibiotic concentration is determined. Since susceptibility test results can fluctuate, it is important to pre-adjust the culture to ensure a certain bacterial concentration. Research is currently underway to expedite the susceptibility testing process. The current gold standard method uses changes in turbidity to determine bacterial proliferation, requiring a full 24-hour testing period. However, methods are being developed to more rapidly determine turbidity changes using lasers, to quickly assess the proliferation of individual bacteria using microscopy, and to rapidly quantify bacterial proliferation using ATP (adenosine triphosphate) luminescence. These advancements could potentially reduce the time required for susceptibility testing to just a few hours. On the other hand, the pretreatment process for preparing bacterial cultures still relies on a 1-day isolation and culture process to dilute the colonies into liquid.

[0006] To address this, if the following steps can be performed quickly, the time required for sensitivity testing will be reduced by another day: Without performing isolation culture, remove components other than bacteria (such as blood cell components, impurities in the culture medium, etc.) from the positive blood culture sample to prepare a sample with 10... 5 ~10 6 A bacterial solution of a certain concentration of CFU / mL.

[0007] To address this issue, Patent Document 1 discloses a method that selectively destroys blood cell components without affecting bacterial growth by using two different surfactants. Patent Document 2 discloses a method that involves protease-based degradation of blood cells, swelling treatment based on a hypotonic solution, and selective destruction of blood cell components using surfactants.

[0008] Patent document 3 discloses a method for detecting the number and concentration of bacteria by fluorescently labeling bacteria captured on a membrane filter. According to this method, the concentration of bacteria can be measured even in the presence of impurities other than bacteria.

[0009] Existing technical documents

[0010] Patent documents

[0011] Patent Document 1: Japanese Patent Application Publication No. 2014-235076

[0012] Patent Document 2: WO2019 / 097752

[0013] Patent Document 3: Japanese Patent Application Publication No. 2007-006709 Summary of the Invention

[0014] The problem the invention aims to solve

[0015] However, patent documents 1 and 2 do not disclose a method for adjusting the concentration of bacteria to a certain value. For example, when preparing bacterial solutions using colonies, the concentration of the bacterial solution can usually be adjusted based on the turbidity value. However, the absorption wavelengths of blood cell components such as red blood cells, white blood cells, and platelets, or hemoglobin, which is abundant in blood cells, are the same as the wavelengths used in the measurement of scattered light from bacteria, making adjustment based on turbidity difficult.

[0016] In Patent Document 3, bacteria must be treated with a fluorescent staining reagent. Exposure to the reagent during treatment can alter the bacteria's morphology, potentially affecting the results of sensitivity testing. Therefore, such a staining method is difficult to apply to sensitivity testing. Furthermore, there are problems such as high reagent costs and the need for a dedicated, expensive optical system for fluorescence excitation.

[0017] The present invention was made in view of the following situation, and provides a technique for inferring the bacterial concentration in a sample from a sample containing impurities such as bacteria and blood cells and adjusting the sample to a desired bacterial concentration.

[0018] Technical means to solve the problem

[0019] The automatic analysis device of the present invention introduces a substance that destroys the impurities into a sample containing bacteria and impurities, separates the destroyed impurities from the bacteria, and then removes the bacteria using a filter. The concentration of bacteria in the sample is inferred based on the correspondence between the amount of impurities remaining on the filter and the concentration of bacteria in the sample.

[0020] The effects of the invention

[0021] The automated analysis apparatus according to the present invention can infer the bacterial concentration in a sample from a sample containing bacteria and impurities and adjust the sample to the desired bacterial concentration. As a result, sensitivity testing can be performed accurately. Issues, configurations, and effects beyond those described above will be clarified through the following description of embodiments. Attached Figure Description

[0022] Figure 1 A flowchart illustrating the general sequence of steps to remove impurities from a blood sample containing bacteria by destroying blood cells.

[0023] Figure 2 This is an example of an image obtained by taking a picture of a filter without staining it.

[0024] Figure 3 A graph showing the relationship between the color information of the filter calculated by processing the filter image and the number of red blood cells in the sample that passed through the filter.

[0025] Figure 4 A graph showing the relationship between the color information of the filter calculated by processing the filter image and the actual concentration of bacteria in the blood.

[0026] Figure 5 A flowchart illustrating the sequence of bacterial concentration adjustments based on blood bacterial concentration inferred from filter images.

[0027] Figure 6 This is a configuration diagram of the automatic analysis device 100 according to Embodiment 2.

[0028] Figure 7 This indicates the results related to the adjusted bacterial concentration from a positive blood culture sample, obtained through turbidity measurement.

[0029] Figure 8 Indicates use Figure 5 The method shown relates results to adjusted bacterial concentrations from positive blood culture samples.

[0030] Figure 9 This indicates the proliferation rate of blood culture-positive samples and samples prepared using bacterial colonies.

[0031] Figure 10 This indicates the proliferation rate of blood culture-positive samples and samples prepared using bacterial colonies.

[0032] Figure 11 This indicates the results of a drug sensitivity test. Detailed Implementation

[0033] Figure 1 This is a flowchart illustrating the general sequence of steps for removing impurities from a blood sample containing bacteria by destroying blood cells. Prior to embodiments of the present invention, according to... Figure 1 The general order of removing impurities from blood cell samples is explained. Subsequently, details of embodiments of the present invention are described.

[0034] In step S10, a surfactant is added to the blood sample to destroy blood cells. The more blood volume processed, the higher the final bacterial count; therefore, it is preferable to pretreat more samples, but this also increases waste liquid. Therefore, it is preferable to pretreat each sample by a few mL to 10 mL. The surfactant is preferably (a) an anionic surfactant having both hydrophilic and hydrophobic portions, with the hydrophobic portion being a chain hydrocarbon, or (b) a surfactant having both hydrophilic and hydrophobic portions, with the hydrophobic portion being a cyclic hydrocarbon, or a combination of (a) and (b). Specifically, examples of the former include sodium dodecyl sulfate, lithium dodecyl sulfate, and sodium N-lauroyl sarcosinate; examples of the latter include saponins, sodium cholate, sodium deoxycholate, 3-[(3-cholamidopropyl)dimethylammonium]-1 propanesulfonate, and 3-[(3-cholamidopropyl)dimethylammonium]-2-hydroxy-1-propanesulfonate. After adding the surfactant, the next step S11 can be carried out immediately, or the mixture can be left to stand for about 5 to 15 minutes to wait for the reaction to complete.

[0035] In step S11, centrifugation is performed to remove components from blood cells that have been damaged by the surfactant, such as hemoglobin. Following this, the supernatant is removed and washed. In this step, centrifugation is preferably performed at 2000G for approximately 5-10 minutes, but the centrifugation speed and time are not limited to these values, as long as the bacteria and blood cell components not damaged by the surfactant are separated from the outflowing hemoglobin. Washing is performed using pure water, physiological saline, etc., and can be performed once or multiple times.

[0036] In step S12, the sample is filtered using a filter to further remove blood cell components that could not be destroyed by the surfactant and impurities in the culture medium. By using a pore size (mesh spacing) larger than that of bacteria as the filter pore size, bacteria can pass through and impurities other than bacteria can be captured by the filter. For example, a filter with a pore size of 1 to 40 μm is preferably used. If the amount of impurities is large, multiple filtrations can be performed, such as filtration using a filter with a large pore size followed by filtration using a filter with a small pore size. To inhibit the capture of bacteria by the filter, a filter made of a hydrophobic material is preferably used. Through the above steps S10 to S12, impurities other than bacteria can be removed from the blood sample and bacteria can be extracted.

[0037] <Implementation Method 1>

[0038] In Embodiment 1 of the present invention, a method is shown for obtaining a sample with a bacterial concentration adjusted to a desired value from a blood sample containing bacteria. Furthermore, this embodiment is merely an example and is not limited to this configuration. Pretreatment steps S10 to S12 are performed using a blood sample containing *Escherichia coli* and *Staphylococcus aureus*. The blood sample is prepared in the following order: 10 mL of blood from a healthy volunteer and 0.1 mL of a bacterial culture solution prepared beforehand using bacterial colonies and adjusted to a concentration of approximately 150 CFU / mL are introduced into a blood culture bottle containing pharmaceutically adsorbed beads to prepare blood equivalent to that of an actual sepsis patient. The sample is then cultured in a blood culture apparatus, and after a positive blood culture, it is removed for experimentation. In addition, a sample equivalent to a negative control, with a bacterial concentration of 0 CFU / mL in the blood cultured without the introduction of bacterial culture solution, is prepared, and the bacterial concentration is altered by appropriate dilution.

[0039] Figure 2 This is an example of an image obtained by photographing a filter without staining. The image shown here is an example of an image obtained by changing the concentration of E. coli in a blood sample to 10. 6 ~10 9 The results were obtained from processing pre-prepared blood samples at a concentration of CFU / mL. The filtration area 20, enclosed by the dashed line, is the area of ​​interest. The actual bacterial concentration in the blood was 3.9 × 10⁻⁶. 6 At CFU / mL, the majority of region 22, exhibiting the same color as the filtration zone 21 and free of impurities, is the area without impurities. The actual bacterial concentration in blood is 1×10⁻⁶. 9 At the CFU / mL level, the area 23 with strong redness indicates the presence of impurities, and the redness further increases as the actual blood bacterial concentration rises.

[0040] Figure 3 The results were obtained by filtering a blood sample (a negative control with a bacterial concentration of 0 CFU / mL) with a mixture of surfactant and the amount of red blood cells in the filtered sample, comparing it to the redness of the filter image. Furthermore, the amount of red blood cells was calculated using a hematology counter. To quantitatively evaluate redness, chroma values ​​calculated through the following processing were used. The filter image is a color image, typically represented in the RGB color space. To reduce the influence of ambient brightness during shooting, it was converted from RGB to HSV (hue, chroma, lightness). More specifically, the average chroma value of each pixel within the filtered area 20 was calculated as the redness of the filter image. Figure 3The results showed a positive correlation between the amount of red blood cells in the sample passing through the filter and the chroma value of the filter image. The negative control blood sample contained not only red blood cells but also platelets with attached hemoglobin, fibrin, and impurities from the culture medium; therefore, these impurities were considered to enhance the redness of the filter. In other words, the chroma value of the filter image can be used to detect red blood cells and other impurities.

[0041] Since the redness of the filter element increases depending on the amount of impurities such as red blood cells contained in the sample, it is important to obtain... Figure 2 The reasons for the results can be inferred as follows. Regardless of the actual bacterial concentration in the blood, the surfactant is added at a fixed concentration. When the actual bacterial concentration in the blood is low, most of the blood cells in the sample are destroyed by the surfactant, so most blood cells are removed in step S11, and impurities do not remain on the filter. On the other hand, when the actual bacterial concentration in the blood increases, the concentrations of bacteria and red blood cells become equal, and the destructive effect of the surfactant on the blood cells is hindered by the bacteria themselves. Therefore, the amount of red blood cells that were not completely destroyed in step S11 increases and are captured in the filtration process of step S12. Furthermore, the possibility that bacteria agglutinate with hemoglobin, fibrinogen, and platelets, causing impurities representing redness to be captured in the filtration process of step S12, is also considered. As the actual bacterial concentration in the blood increases, the redness of the filtered area 20 further increases. Therefore, for example, the concentration of bacteria that passed through the filter can be known by calculating the amount of impurities, such as red blood cells, remaining on the filter after filtration using an image of the filter.

[0042] Figure 4 This is a graph illustrating the relationship between the color information of the filter, calculated by processing the filter image, and the actual concentration of bacteria in the blood. To quantify the color information of the filter, the filter image, represented in RGB color space, is converted to HSV (hue, chroma, lightness) using chroma values. Specifically, the average chroma value of each pixel within the filter region 20 is used. Figure 4 The results of repeated experiments using *E. coli* and *Staphylococcus aureus* are also presented. The actual bacterial concentration in blood is 10... 6 ~10 9 Within the CFU / mL range, a positive correlation is observed between the bacterial concentration in the blood and the chroma value of the filter image. By using this blood bacterial concentration and the chroma value of the filter image to obtain a calibration curve, the bacterial count can be inferred from the chroma value information of the filtered filter image. This range of blood bacterial concentration is roughly the same as the bacterial concentration in samples that test positive in normal blood cultures, so it can be applied to various bacterial species and strains.

[0043] The specific concentration of the surfactant used in the treatment can be specified within the following range. Typically, the concentration of red blood cells in blood is 10... 9 The number of red blood cells is on the order of magnitude of cells / mL. If a 1 mL sample is pretreated, the number of red blood cells in the sample is 10. 9 One. Here, Figure 2 The range shown, which can be inferred from the color of the filter element, is for impurities such as red blood cells, and is approximately 10. 6 ~10 8 In other words, to infer the actual bacterial concentration based on the color of the filter element, it is only necessary to destroy the red blood cells to 1 / 1000 to 1 / 10. That is, a surfactant capable of destroying 90 to 99.9% of the red blood cells in blood containing bacteria is needed.

[0044] Furthermore, as an example, the surfactant concentration range shown here is for the case of pretreatment of 1 mL of sample. When the treatment volume increases, it is preferable to increase the surfactant concentration in a way that correspondingly increases the red blood cell destruction rate. For example, when pretreating 10 mL of sample, a surfactant concentration capable of destroying 99–99.99% of red blood cells is required, and this can be varied depending on the amount of sample treated.

[0045] Specifically, regarding the concentration of a surfactant capable of destroying 99-99.99% of red blood cells in blood containing bacteria, taking sodium dodecyl sulfate as an example—an anionic surfactant having both hydrophilic and hydrophobic portions, with the hydrophobic portion being a chain hydrocarbon—the concentration is in the range of 0.05% by weight to 0.5% by weight. For other surfactants such as lithium dodecyl sulfate and sodium N-lauroyl sarcosinate, a concentration of a surfactant capable of destroying the desired red blood cells is also preferred.

[0046] In addition, other surfactants, such as saponins, sodium cholate, sodium deoxycholate, 3-[(3-cholamidopropyl)dimethylammonium]-1 propanesulfonate, and 3-[(3-cholamidopropyl)dimethylammonium]-2-hydroxy-1-propanesulfonate, which have both hydrophilic and hydrophobic portions and the hydrophobic portion has a cyclic hydrocarbon, can also be mixed.

[0047] In Embodiment 1, a 2.7 mL blood sample was treated with a surfactant such as sodium dodecyl sulfate, which is capable of destroying 99 to 99.99% of red blood cells in blood containing bacteria at a final concentration ranging from 0.05% to 0.5% by weight.

[0048] Figure 5 This is a flowchart illustrating the sequence of steps for adjusting bacterial concentration using blood bacterial concentration inferred from a filter image. Steps S10-S12 are... Figure 1 The procedures shown are the same.

[0049] In step S50, the filter element is photographed to obtain the color of the impurities remaining on it. In this flowchart, since the amount of impurities is inferred from the color of the red blood cells remaining on the filter element, there is no need to stain the sample and impurities.

[0050] In step S51, correspondence data describing the relationship between the color of impurities remaining on the filter and the actual blood bacterial concentration is read. The color obtained from the filter image is used as a reference to this correspondence data to calculate the inferred blood bacterial concentration. Specifically, a calibration curve, for example, expressed as a single logarithm, is obtained for the relationship between the color of impurities remaining on the filter and the actual blood bacterial concentration. The inferred blood bacterial concentration is calculated based on the data from the calibration curve. The correspondence data is... Figure 4 The data illustrated herein are pre-prepared and stored in a storage device. Furthermore, regardless of the bacterial species and type of surfactant, the method of the present invention can be applied as long as there is at least one corresponding data. In the case of more accurately determining the inferred bacterial concentration in the blood, corresponding data can also be maintained based on information about each bacterial species or drug-resistant strain, such as methicillin-resistant Staphylococcus aureus, and the type of surfactant.

[0051] In step S52, the sample is diluted based on the blood bacterial concentration deduced in S51 to obtain the desired bacterial concentration. For example, the blood bacterial concentration deduced in S51 is 5 × 10⁻⁶. 8 When the concentration is CFU / mL, if the desired bacterial concentration is 5 × 10⁻⁶ 5 If the concentration of bacteria in the blood is CFU / mL, then it is diluted 1000 times. If the bacterial concentration in the blood inferred in S51 does not reach the expected bacterial concentration, it is preferable to proceed to step S53 and determine it as a defective specimen. In the case of a defective specimen, it is difficult to prepare a specimen suitable for sensitivity testing, so further culturing is performed in the blood culture bottle to allow bacterial proliferation before returning to step S10. Alternatively, colonies obtained by performing isolation culture can be used to perform identification and sensitivity testing.

[0052] In step S54, the prepared bacterial culture is used to perform identification and sensitivity tests. Any method can be used as the testing method. Examples include identification tests using automated devices, gene testing, sensitivity tests using micro-liquid dilution methods, sensitivity tests using the paper disc method, and rapid sensitivity tests based on microscopic images or laser scattering measurements.

[0053] <Implementation Method 1: Summary>

[0054] In this embodiment 1, the sample after the blood cells have been destroyed is filtered using a filter element. The inferred blood bacterial concentration is calculated by referring to the correspondence data between the color components of the image of the impurities remaining on the filter element and the actual blood bacterial concentration. Therefore, a sample with the desired bacterial concentration can be prepared without performing isolation and culture. Consequently, the isolation and culture process, which typically requires about one day, can be shortened to, for example, about 30 minutes.

[0055] <Implementation Method 2>

[0056] In Embodiment 2, an automated analysis apparatus is shown for obtaining a sample from a blood sample containing bacteria, with a bacterial concentration adjusted to a desired value. Furthermore, this embodiment is merely an example and is not limited to this configuration.

[0057] Figure 6 This is a configuration diagram of the automatic analysis device 100 according to Embodiment 2 of the present invention. The automatic analysis device 100 is automatically implemented. Figure 5 The device describes the pretreatment sequence. Typically, blood culture bottles are fitted with rubber stoppers to prevent contamination, and the inside of the bottle is under vacuum. An operator uses a syringe to remove a blood sample from the blood culture bottle and dispenses the blood sample into a container containing a surfactant. This is how S10 is performed. Alternatively, an automatic surfactant introduction device 101 can be provided as part of the automatic analysis device 100 to replace the operator's introduction of the surfactant.

[0058] The sample containing the surfactant is introduced into centrifuge 102. The blood cells in the sample are destroyed by the surfactant. Centrifuge 102 separates the dissolved hemoglobin and other substances from the bacteria and the undestroyed blood cells. This performs the separation process in S11.

[0059] After centrifugation, the sample is introduced into the washing section 103. The washing section 103 removes the supernatant from the sample and washes it with a washing solution such as approximately 1 mL of physiological saline, pure water, or culture medium. The supernatant is aspirated using a pipette 104. The supernatant can be treated up to a certain height from the bottom of the sample container. This is used to process the remaining portion in S11. More precisely, it is preferable to install a level sensor or the like to detect the interface between the particulate portion formed by bacterial and blood cell coagulation and the liquid portion, and treat the portion up to the vicinity of the interface as supernatant.

[0060] After washing, the sample is introduced into the filter section 105. The filter section 105 is composed, for example, a filter for processing and a syringe for capturing impurities onto the filter. S12 is performed by the filter section 105. In some cases, a centrifuge 102 may also be used to filter the sample.

[0061] The camera 106 (equivalent to a sensor for detecting the amount of impurities) photographs the impurities remaining on the filter element section 105. The storage section 107 stores... Figure 4 The correspondence data described earlier. The computer 110 (processing unit) converts the RGB image captured by the camera 106 into an HSV color space image, then detects the color components (e.g., chroma values) of the impurity region (S50), and uses this color to refer to the correspondence data to calculate the inferred concentration of bacteria in the blood (S51).

[0062] The filtered sample is introduced into the dilution unit 108. The dilution unit 108 adjusts the dilution ratio to achieve the desired bacterial concentration based on the bacterial concentration predicted by the computer 110. The dilution pipette 109 then introduces the diluent according to this dilution ratio. This completes step S52.

[0063] The computer 110 automatically performs the above procedures by controlling the various components equipped in the automatic analysis device 100. Preferably, the computer 110 is equipped with input / output devices, allowing operators to instruct the computer 110 on bacterial types and desired bacterial concentrations. The computer 110 can also change the reference correlation data based on the input bacterial types and adjust the dilution ratio of the dilution unit 108 based on the desired bacterial concentration. The bacterial concentration value in the blood estimated by the computer 110 is output to an output device such as a display. If the bacterial concentration is below the desired level, a mark indicating a defective sample may be displayed (S53).

[0064] In addition to the above-described configuration, the automatic analysis device 100 may also be equipped with a sensitivity testing device 111. The computer 110 automatically performs sensitivity testing by controlling the sensitivity testing device 111. Thus, all steps from S10 to S54 can be performed automatically. In addition to the contents described in Embodiment 1, the sensitivity testing may also measure the minimum growth inhibition concentration, as described in the embodiments below.

[0065] <Implementation Method 3>

[0066] It is believed that the method for adjusting the concentration of bacteria in blood samples as described in Embodiments 1 and 2 is affected to some extent by the amount of red blood cells in the original blood. The concentration of red blood cells in a person also varies depending on gender and health status, and is approximately 3 × 10⁻⁶. 9 ~6×10 9 Approximately [number] cells / mL, within a bacterial concentration range of 10. 6 ~10 10Compared to the value per mL, the variation is minimal. Therefore, it is considered to have little impact on the inferred blood bacterial concentration. However, if the amount of impurities has been determined beforehand through other hematology analyses such as erythrocyte concentration and hematocrit, this value can be used to correct the result of step S51. This allows for a more accurate calculation of the inferred blood bacterial concentration. For example, the following correction order can be considered.

[0067] (Revision order 1) Implemented by operators or computer 110 Figure 5 If the blood bacterial concentration inferred from the sequence of events is an outlier (beyond a predetermined tolerance range), then an alternative test result is used instead of the impurity level inferred from the filter image. That is, the blood bacterial concentration is inferred by referring to corresponding data using an alternative test result.

[0068] (Revision order 2) Implemented by operators or computer 110 Figure 5 The impurity levels on the filter element are inferred more than once in order of sequence, and additional test results are also obtained. The impurities, after removing outliers, are averaged to obtain the final impurity level. This impurity level is then used as a reference to corresponding data to calculate the inferred blood bacterial concentration.

[0069] In the above correction sequence, additional detection results obtained from measuring the amount of impurities are used as a reference for the correspondence data. Therefore, the correspondence data must accurately describe the relationship between the amount of impurities and the actual concentration of bacteria in the blood. For example, the correspondence data can simultaneously record the chroma value of the filter image and its corresponding amount of impurities, or a conversion formula between chroma values ​​and impurities can be predefined and used to convert the additional detection results into chroma values. In other words, the correspondence data only needs to describe the relationship between the value of the amount of impurities remaining on the filter and the actual concentration of bacteria in the blood sample.

[0070] In embodiments 1 and 2, the chroma value of the filter image is used to detect red blood cells remaining on the filter. Alternatively, a light sensor that detects absorption and reflection of a specific wavelength corresponding to red can be used instead of the camera 106. That is, by using an optical sensor capable of detecting at least the largest component of the RGB components of impurities remaining on the filter, the same information as the chroma value of the image captured by the camera 106 can be obtained. In this case, the correspondence data must also record the value measured by the optical sensor instead of the chroma value. Alternatively, the amount of impurities can be visually measured by a person using a color chart, and the correspondence data can be referenced based on the measurement result. Furthermore, the color chart itself can also record the correspondence between the color of the impurities and the concentration of bacteria in the blood.

[0071] <Example 1>

[0072] In Example 1 of the present invention, the advantages of the pretreatment method of the present invention are described together with a comparative example. In this Example 1, concentration adjustments were made within the recommended bacterial concentration range for sensitivity testing using turbidity measurement and the concentration adjustment of the present invention, and the results were compared. In the case of turbidity measurement, absorbance measurement at a wavelength of 600 nm was used. In the case of concentration adjustment of the present invention, absorbance measurement at a wavelength of 600 nm was used. Figure 1 The pretreatment method shown is the same as in Embodiment 1, using the same types and concentrations of bacteria and surfactants.

[0073] Figure 7 Demonstration and utilization through Figure 1 The bacterial suspension obtained by the pretreatment method shown was adjusted for bacterial concentration using turbidity measurements previously used in bacterial testing pretreatment. The shaded area is a recommended bacterial concentration range for sensitivity testing established by the American Institute for Clinical Laboratory Standards. The shaded area is 5 × 10⁻⁶. 5 The range of CFU / mL (±60%) was adjusted to the central value of 5 × 10⁻⁶. 5 The method was adjusted using CFU / mL. Specifically, the bacterial count concentration was adjusted to the equivalent of 1.5 × 10⁻⁶ using the McFarland turbidimetric method. 8 The bacterial suspension with CFU / mL and a McFarland turbidity of 0.5 was diluted 300 times.

[0074] The bacterial concentration in the blood sample was 10. 8 At concentrations above CFU / mL, the concentration can be adjusted to near the desired range, but the bacterial concentration in the blood sample is 10. 6 ~10 7 At a CFU / mL level, the adjusted bacterial count will decrease by 1-2 places. This is because hemoglobin not removed in steps S11 and S12, as well as microparticles in the culture medium, contribute to increased light scattering, resulting in an increased turbidity value despite a low bacterial concentration. Consequently, the bacterial concentration estimate is overestimated, making it difficult to adjust the bacterial concentration using turbidity measurements. Figure 7 The fact that the plotting did not fall within the shadow area representing the recommended bacterial concentration range also confirms this fact.

[0075] Figure 8 Display based on usage Figure 5 The method shown is used to adjust the concentration based on the inferred bacterial concentration in the blood. Figure 7 In comparison, even if the actual blood bacterial concentration in the blood sample is 10... 6 ~10 7 At the CFU / mL level, the adjusted bacterial concentration did not decrease and remained roughly within the range indicated by the shadow. Furthermore, Figure 8 This presents the results of pretreatment of *Escherichia coli* and *Staphylococcus aureus* based on the same correspondence data. Figure 8 This indicates that even bacteria with vastly different characteristics, such as Gram-negative and Gram-positive bacteria, can have their adjusted bacterial concentrations fall within a certain range based on a single correspondence data point. For more accurate concentration adjustments, it is preferable to maintain correspondence data for each bacterial species and refer to the corresponding data for each species.

[0076] <Example 2>

[0077] In Example 2 of the present invention, an example is shown of adjusting bacterial concentration using *E. coli* and a positive blood culture sample via the pretreatment method of the present invention. As a comparative example, the proliferation rate was measured when bacterial culture was prepared using colonies after 1 day and night of isolation and culture. Regarding the bacterial concentration in the positive blood culture sample, 10... 7 ~10 9 Three different concentrations of CFU / mL were used to achieve a final bacterial concentration of 5 × 10⁻⁶. 5 Adjustments were made in the form of CFU / mL. When using colonies to prepare the bacterial culture, the final bacterial concentration was also adjusted to 5 × 10⁻⁶. 5 The CFU / mL method was adjusted using turbidity measurement. 50 μL of the sample was aliquoted into 96-well plates, along with 50 μL of Miller-Hinton medium adjusted to twice its original concentration. The 96-well plates were then incubated at 35–37°C, and proliferation was observed using a bright-field microscope. The degree of proliferation was calculated over time using the area identified as bacteria in the microscopic image as an indicator of proliferation.

[0078] Figure 9 The proliferation rates of blood culture-positive samples and samples prepared from bacterial colonies are shown. Between blood culture-positive_1 and isolated culture_1, there was a difference of approximately 0.5 hours in the time it took for bacterial proliferation to increase, but the final proliferation rates were roughly the same. The same was true for blood culture-positive_2 and blood culture-positive_3. This indicates that even with pretreatment of blood culture-positive samples, the bacterial concentration can be adjusted to the same level as that adjusted from the colonies, without affecting bacterial development during sensitivity testing.

[0079] <Example 3>

[0080] In Example 3 of the present invention, the results of replacing Escherichia coli in Example 2 with Staphylococcus aureus will be described.

[0081] Figure 10The proliferation rates of blood culture-positive samples and samples prepared from bacterial colonies are shown. Similar to Example 2, the final proliferation rates were consistent between blood culture-positive and isolated cultures. This demonstrates that, as in Example 2, even with pretreatment of blood culture-positive samples, the bacterial concentration can be adjusted to the same level as that adjusted from the colonies, without affecting bacterial development during sensitivity testing.

[0082] <Example 4>

[0083] In Example 4 of this invention, the results of drug susceptibility testing are demonstrated using both blood culture-positive samples and samples prepared from bacterial colonies. In the susceptibility test, the lowest concentration of the drug that exhibits antibacterial activity against bacteria (Minimum Inhibitory Concentration, MIC) is determined. Here, the susceptibility test uses the microdilution method. The bacterial culture and different concentrations of the drug are mixed, and the turbidity of each well in a 96-well plate is visually assessed 18 hours after incubation to determine the MIC.

[0084] Figure 11 Present the results of the drug sensitivity test. Figure 11 In one case, cefepime (CFPM), cefotaxime (CTX), gentamicin (GM), and levofloxacin (LVFX) were used to treat Escherichia coli, while erythromycin (EM), oxacillin (MPIPC), penicillin G (PCG), and vancomycin (VCM) were used to treat Staphylococcus aureus.

[0085] In all cases, the MIC for blood culture samples pretreated fell within ±1 tube (1 times or half) of the MIC for samples prepared using colonies, indicating that samples obtained by pretreatment of blood culture samples can also be correctly used for sensitivity testing.

[0086] Figure 11 The text demonstrates an example of using micro-liquid dilution to determine MIC, but it can also be done through methods such as... Figures 8-9 The MIC can be determined by rapid sensitivity testing using microscopic images, as obtained in the previous method. Alternatively, rapid sensitivity testing using lasers can also be performed.

[0087] <Regarding variations of the present invention>

[0088] In the above embodiments, the case where the average chroma value of the filter region 20 is used as a reference to the corresponding data has been described, but the maximum value or mode may also be used instead of the average value. Alternatively, the amount of impurities remaining on the filter may be represented by a characteristic quantity expressed in at least two of hue / brightness / chroma instead of the chroma value.

[0089] In the above embodiments, an example of detecting blood cells contained in a blood sample as impurities was described, but the present invention can also be used in other bacterial samples. That is, the present invention can be used for any sample in which there is a correspondence between the image information obtained by photographing the impurities remaining on the filter and the bacterial concentration in the sample. The substances added to destroy the impurities can be changed according to the type of impurity.

[0090] Symbol Explanation

[0091] 100…Automatic Analysis Device

[0092] 101…Import Device

[0093] 102… Centrifugal separator

[0094] 103… Cleaning Department

[0095] 104… Cleaning pipette

[0096] 105… Filter Components Section

[0097] 106… camera

[0098] 107… Storage Department

[0099] 108…Dilution Department

[0100] 109… Dilution pipette

[0101] 110…computer.

Claims

1. An automated analytical method for non-diagnostic purposes, for analyzing samples containing bacteria and impurities, characterized in that, It includes the following steps: Introduce a substance that destroys the impurities into the sample; The impurities and bacteria in the sample into which the substance was introduced were separated by centrifugation. The bacteria are removed from the sample by using a filter element from which the bacteria are separated from the impurities; The corresponding relationship data is read from the storage unit that stores the corresponding relationship data, which records the correspondence between the amount of the impurities remaining on the filter and the concentration of the bacteria in the sample; as well as The concentration of bacteria in the sample is inferred by referring to the corresponding data using a numerical value representing the amount of impurities remaining on the filter after the bacteria are removed from the sample. The sample was a blood sample. The substance is a surfactant.

2. The automatic analysis method according to claim 1, characterized in that, The sample contains blood cells as the impurity. The surfactant comprises at least one of the following: an anionic surfactant having a hydrophilic portion and a hydrophobic portion, wherein the hydrophobic portion is a chain hydrocarbon, and a surfactant having a hydrophilic portion and a hydrophobic portion, wherein the hydrophobic portion is a cyclic hydrocarbon.

3. The automatic analysis method according to claim 1, characterized in that, The filter element is a filter element used to separate the impurities from the bacteria by filtering the sample. In the step of inferring the concentration of the bacteria, the amount of the impurities remaining on the filter is detected using an image obtained by photographing the impurities on the filter without staining. In the step of inferring the concentration of the bacteria, the amount of the impurities detected using the image is used as a reference to the correspondence data.

4. The automatic analysis method according to claim 3, characterized in that, The automatic analysis method further includes the step of capturing an RGB image as the image. In the step of inferring the concentration of the bacteria, the RGB image is converted into an HSV color space image. In the step of inferring the concentration of the bacteria, the chroma value on the HSV color space image of the impurities remaining on the filter is used as a numerical value representing the amount of the impurities remaining on the filter.

5. The automatic analysis method according to claim 3, characterized in that, The automatic analysis method further includes the step of capturing an RGB image as the image. In the step of inferring the concentration of the bacteria, the RGB image is converted into an HSV color space image. In the step of inferring the concentration of the bacteria, feature values ​​represented by the hue, chroma, and lightness values ​​on the HSV color space image of the impurities remaining on the filter are used as numerical values ​​representing the amount of the impurities remaining on the filter.

6. The automatic analysis method according to claim 1, characterized in that, In the step of inferring the concentration of the bacteria, the result of detecting the impurity is obtained from an optical sensor that detects at least the largest component of the RGB color components possessed by the impurity. In the step of inferring the concentration of the bacteria, the result of detecting the impurities remaining on the filter element using the optical sensor is used as a numerical value representing the amount of the impurities remaining on the filter element.

7. The automatic analysis method according to claim 1, characterized in that, The filter element is a filter element used to filter the impurities and bacteria by filtering the sample. The automated analysis method further includes the step of detecting the amount of the impurities remaining on the filter element. In the step of inferring the concentration of the bacteria, a separate detection result is obtained, distinct from the step of detecting the amount of impurities remaining on the filter element, to determine the amount of impurities within the sample. In the step of inferring the concentration of the bacteria, the amount of the impurities detected in the step of detecting the amount of the impurities remaining on the filter element is corrected using the additional detection results, and the corrected amount of the impurities is used to refer to the correspondence data.

8. The automatic analysis method according to claim 1, characterized in that, The automated analysis method further includes a step of performing a drug sensitivity test on the bacteria. In the step of performing the drug susceptibility test, after the concentration of the bacteria in the sample is inferred, the drug susceptibility test against the bacteria in the sample is performed without culturing the bacteria in the sample.

9. The automatic analysis method according to claim 8, characterized in that, The automated analysis method further includes a step of diluting the sample. In the dilution step, after determining the concentration of the bacteria in the sample, the sample is diluted to prepare a test sample with the concentration of the bacteria required for performing the drug susceptibility test. In the step of performing the drug sensitivity test, the drug sensitivity test is performed on the test sample prepared in the dilution step.

10. The automatic analysis method according to claim 8, characterized in that, In the step of performing the drug sensitivity test, images of the sample placed in an incubator maintained at 35–37°C are captured using a camera device. In the step of performing the drug susceptibility test, the image of the sample captured by the imaging device is used to determine the bacterial proliferation rate, thereby determining the minimum growth inhibition concentration of the drug.

11. The automatic analysis method according to claim 1, characterized in that, The filter element has a pore size of 1 to 40 μm and is made of a hydrophobic material.

12. The automatic analysis method according to claim 1, characterized in that, The impurities include at least red blood cells from the blood.

13. An automated analytical apparatus for analyzing samples containing bacteria and impurities, characterized in that, have: A separator that separates the impurities from the bacteria in a sample containing a substance designed to destroy the impurities by centrifugation. A filter element that removes the bacteria from the sample after the impurities and bacteria have been separated. The storage unit stores correspondence data, which records the correspondence between the amount of impurities remaining on the filter and the concentration of bacteria in the sample. as well as The calculation unit uses a numerical value representing the amount of impurities remaining on the filter after the bacteria are removed from the sample, referencing the corresponding relationship data, to infer the concentration of the bacteria in the sample. The sample was a blood sample. The substance is a surfactant.

Images