Information processing device and system

Abstract

Provided are an information processing device and system capable of easily and quickly calculating the concentration of bacteria in a specimen.　The present invention provides an information processing device comprising a bacteria count calculation unit, a specimen information acquisition unit, and a specimen bacteria concentration analysis unit. The bacteria count calculation unit counts the number of bacteria in a specimen image obtained by imaging at least a portion of a specimen placed on a specimen holding member. The specimen information acquisition unit acquires the specimen image and specimen information including the amount of the specimen placed on the specimen holding member. The specimen bacteria concentration analysis unit calculates the bacteria concentration on the basis of the number of the bacteria and the amount of the specimen.

Description

Information processing device and system

[0001] This invention relates to an information processing device and system.

[0002] Infectious disease diagnosis is performed using a combination of factors, including the patient's symptoms and various tests. A crucial indicator in infectious disease diagnosis is whether or not the specimen obtained from the patient is a bacterial colonization specimen. The concentration of bacteria in the specimen significantly contributes to determining whether or not it is a bacterial colonization specimen.

[0003] Patent Document 1 discloses a method for diagnosing infectious diseases, specifically urinary tract infections, by calculating the concentration of bacteria in a urine sample. In this method, the concentration is calculated based on the number of bacteria in the urine sample counted by photoelectric detection of a fluorescent marker and the amount of filtered urine sample.

[0004] Japanese Patent Application Publication No. 9-119926

[0005] In the current method, it is necessary to label the sample with a marker and filter it in order to calculate the concentration, which incurs costs for laboratory technicians in terms of time and effort. Therefore, there is a need for a technology that can obtain the bacterial concentration in a sample more simply and quickly.

[0006] This invention has been made in view of these circumstances, and provides an information processing device and system that can easily and quickly calculate the concentration of bacteria in a sample.

[0007] An information processing device according to an exemplary embodiment of the present invention has the following configuration: [1] An information processing device comprising: a bacterial count calculation unit; a sample information acquisition unit; and a sample bacterial concentration analysis unit, wherein the bacterial count calculation unit counts the number of bacteria in a sample image taken of at least a portion of a sample placed on a sample holding member; the sample information acquisition unit acquires sample information including the sample image and the amount of the sample placed on the sample holding member; and the sample bacterial concentration analysis unit calculates the bacterial concentration based on the bacterial count and the amount of the sample. [2] The information processing device according to [1], wherein the bacterial count calculation unit counts the number of bacteria in the sample image based on a learning model; and the learning model is configured to output the number of bacteria in the sample image when the sample image is input to the learning model. An information processing device according to [3] [2], wherein the bacterial count calculation unit counts the number of bacteria for each classification of bacteria in the sample image based on the learning model, and the learning model is configured to output the number for each classification of bacteria in the sample image when the sample image is input to the learning model. An information processing device according to any one of [4] [1] to [3], wherein the sample information acquisition unit acquires the shooting position of the sample image in the area placed on the sample holding member, and the sample bacterial concentration analysis unit calculates the bacterial concentration based on the shooting position. An information processing device according to [5] [4], wherein the sample bacterial concentration analysis unit divides the smeared area, which is the area in which the sample is smeared, into a plurality of concentric circular areas with the center as the center point, calculates the partial concentration, which is the bacterial concentration for each divided area, and calculates the total bacterial concentration of the smeared area based on the partial concentration for each divided area. An information processing device according to [6] [4], wherein the sample bacterial concentration analysis unit sets a first region in the smear region, which is the region on which the sample is smeared, where the bacterial concentration is estimated to be high, and a second region which is a region different from the first region, calculates a first partial concentration which is the bacterial concentration in the first region and a second partial concentration which is the bacterial concentration in the second region, and calculates the overall bacterial concentration based on the first partial concentration and the second partial concentration.An information processing device according to any one of [7] [1] to [6], further comprising an enrichment amount calculation unit, wherein the enrichment amount calculation unit calculates the amount of enrichment from the time the sample was collected to the present based on the storage conditions of the sample, and the sample bacterial concentration analysis unit calculates the bacterial concentration at the time the sample was collected based on the enrichment amount. An information processing device according to any one of [8] [1] to [7], further comprising a bacterial colonization sample determination unit, wherein the bacterial colonization sample determination unit determines whether or not the sample is a bacterial colonization sample based on the bacterial concentration. An information processing device according to [9] [8], further comprising a patient information acquisition unit, wherein the patient information acquisition unit acquires patient information of the patient who provided the sample, and the bacterial colonization sample determination unit determines whether or not the sample is a bacterial colonization sample based on the sample information, the patient information and the bacterial concentration. An information processing device according to

[10] and [9], wherein the bacterial colonization specimen determination unit determines whether or not a specimen is a bacterial colonization specimen by comparing the bacterial concentrations obtained from two or more specimen images, and the two or more specimen images are images of specimens taken from spatially different locations of the same patient or specimens taken from the same patient at different time intervals. An information processing device according to

[11] , [9] or

[10] , further comprising an infectious disease possibility determination unit, wherein the infectious disease possibility determination unit determines whether or not the patient is suffering from an infectious disease based on the determination result by the bacterial colonization specimen determination unit and the patient information. An information processing device according to

[12] and

[11] , further comprising an infectious disease progress determination unit, wherein the infectious disease progress determination unit determines the progress of the infectious disease in the patient by comparing the bacterial concentrations obtained from two or more specimen images, and the two or more specimen images are images of specimens of the same type and specimens taken from the same patient at different time intervals. An information processing device according to any one of [1] to

[12] , wherein the sample is urine.

[14] A system comprising: an imaging unit; a bacterial count calculation unit; a sample information acquisition unit; and a sample bacterial concentration analysis unit, wherein the imaging unit is configured to capture a sample image of a sample placed on a sample holding member; the bacterial count calculation unit counts the number of bacteria in the sample image; the sample information acquisition unit acquires the quantity of the sample placed on the sample holding member; and the sample bacterial concentration analysis unit calculates the bacterial concentration based on the bacterial count and the quantity of the sample.

[0008] The information processing device according to the present invention comprises a bacterial count calculation unit configured to count the number of bacteria in a sample image, a sample information acquisition unit configured to acquire the amount of a sample placed on a sample holding member, and a sample bacterial concentration analysis unit configured to calculate the bacterial concentration based on the bacterial count and the amount of the sample. This configuration allows for the simple and rapid calculation of bacterial concentration.

[0009] This is a block diagram showing the hardware configuration of the information processing device 2, which is a bacterial quantity determination system 1 according to the first embodiment of the present invention. This is a block diagram showing the functional configuration of the information processing device 2 according to the first embodiment of the present invention. This is an explanatory diagram showing the case of calculating bacterial concentration from multiple sample images in the first embodiment of the present invention. This is a schematic diagram of the bacterial quantity determination system 1 according to the second embodiment of the present invention. This is an explanatory diagram showing the case of calculating bacterial concentration from multiple sample images in the third embodiment of the present invention. This is a diagram showing the appearance of the slide glass used in this embodiment. Figures 7A and 7B are schematic diagrams showing the bacterial concentration distribution of sample 820 in this embodiment. Figures 8A and 8B are schematic diagrams showing the bacterial concentration distribution of sample 808 in this embodiment. Figures 9A and 9B are schematic diagrams showing the bacterial concentration distribution of sample 831 in this embodiment. Figures 10A and 10B are schematic diagrams showing the bacterial concentration distribution of sample 806 in this embodiment.

[0010] Embodiments of the present invention will now be described. The various features shown in the embodiments below can be combined with each other. Furthermore, each feature constitutes an independent invention. In addition, any elements in the embodiments below that are not defined in the claims are optional and can be omitted. Any number of zeros (for example, one or two) may be added to the end of the numerical values ​​disclosed in the following description. For example, one or two zeros may be added after "1.4" to make it "1.40" or "1.400".

[0011] 1. The bacterial quantity determination system 1 according to one embodiment of the first embodiment is a system that has the function of calculating the concentration of bacteria in a sample (hereinafter referred to as bacterial concentration). Here, the bacterial concentration is the number of bacteria per 1 ml of sample (volume concentration) or per 1 mm of sample ２ This refers to the number of bacteria per unit area (area concentration). In this embodiment, a system equipped with an information processing device 2 will be described as an example of a bacterial quantity determination system 1.

[0012] 1.1 Information Processing Device 2 1.1.1 Hardware Configuration of Information Processing Device 2 As shown in Figure 1, the information processing device 2 according to the present invention comprises a control unit 10, a storage unit 20, an input unit 30, a communication unit 40, and an output unit 50. These are interconnected by a communication bus 60.

[0013] (1) Control Unit 10 The various functions performed by the control unit 10 may be implemented by software (including so-called applications) or by hardware.

[0014] When implemented through software, various functions can be realized by the processor executing programs that make up the software. For example, processors include CPUs (Central Processing Units), microprocessors, and DSPs (Digital Signal Processors).

[0015] On the other hand, when implemented in hardware, it can be achieved using various circuits such as ASIC (Application Specific Integrated Circuit), SOC (System On a Chip), FPGA (Field Programmable Gate Array), or DRP (Dynamically Reconfigurable Processor).

[0016] (2) Storage Unit 20 A portion of the storage unit 20 is composed of, for example, RAM (Random Access Memory) or DRAM (Dynamic Random Access Memory), and is used as a work area when the control unit 10 executes processing based on various programs.

[0017] Furthermore, a portion of the storage unit 20 is, for example, a non-volatile memory such as ROM (Read Only Memory) or an HDD (Hard Disk Drive), which stores various data and programs used for processing by the control unit 10. The storage unit 20 can also maintain a database including one or more tables for recording various information and processing results.

[0018] (3) Input unit 30 The input unit 30 may include one or more of the following: a keyboard, keypad, mouse, microphone, touchscreen, buttons, etc. The input unit 30 accepts input of various types of information. The information may include, for example, specimen information, patient information, specimen images, etc.

[0019] Here, the specimen image is an image taken of at least a portion of the specimen placed on the specimen holding member. Preferably, the specimen image is an image of a specimen that has been stained in various ways, such as Gram staining, Ziehl-Nielsen staining, India ink staining, Giemsa staining, or Grocott staining. The specimen holding member refers to any member commonly used for observing specimens, such as a glass slide, a prepared slide, or a petri dish. In the following description, the specimen will be described assuming that it is smeared on a glass slide 100.

[0020] (4) Communication Unit 40 The communication unit 40 communicates various signals between the information processing device 2 and any external device, and is configured to acquire sample images. The communication unit 40 may have functions to connect to, for example, a NIC (Network Interface Controller), a wireless LAN (Local Area Network), a wireless WAN (Wide Area Network), or functions to enable short-range wireless communication such as Bluetooth® or infrared communication. In addition, for example, the communication unit 40 may have a wired communication port such as a USB port to enable wired communication.

[0021] (5) Output Unit 50 The output unit 50 is, for example, any display and / or speaker. The output unit 50 can output, for example, the bacterial concentration output from the control unit 10 (described later), the result of determining whether it is a bacterial colonization sample, the result of determining whether it is suffering from an infectious disease, the estimated location of the cause of the infectious disease, the therapeutic effect, etc.

[0022] 1.1.2 As shown in the Functional Configuration Diagram 2 of the Information Processing Device 2, the control unit 10 comprises, as a functional configuration, a sample information acquisition unit 10a, a bacterial count calculation unit 10b, and a sample bacterial concentration analysis unit 10c. The control unit 10 may further comprise a bacterial enrichment amount calculation unit 10d, a patient information acquisition unit 10e, a bacterial colonization sample determination unit 10f, an infectious disease possibility determination unit 10g, an infectious disease progression determination unit 10h, a cause location analysis unit 10i, and a treatment effect determination unit 10j.

[0023] (1) Sample Information Acquisition Unit 10a The sample information acquisition unit 10a can acquire sample information and sample images via the input unit 30 and the communication unit 40 and output them to the bacterial count calculation unit 10b, the sample bacterial concentration analysis unit 10c, the enrichment amount calculation unit 10d, and the bacterial colonization sample determination unit 10f. Specifically, the sample information acquisition unit 10a is configured to acquire the amount of sample smeared on the slide glass 100. Here, the amount of sample refers to, for example, the volume or area of ​​the sample. The sample information acquisition unit 10a is further configured to acquire the shooting position of the sample image in the area where the sample is smeared on the slide glass 100. In addition to these, the sample information acquisition unit 10a may be configured to acquire other sample information such as the smear area on the slide glass 100, the type of sample, storage conditions, pretreatment information, collection conditions, and collection location.

[0024] (2) Bacterial count calculation unit 10b The bacterial count calculation unit 10b can count the number of bacteria in a sample image. Specifically, it can count the number of bacteria in a sample image based on the type of sample and the learning model acquired from the sample information acquisition unit 10a. Here, the type of sample is, for example, urine, sputum, cerebrospinal fluid, bile, catheter tip sample, etc. The learning model is a model that has been machine-learned using images of various types of samples as training data. When a sample image is input to the learning model, the learning model is configured to identify bacteria in the sample image and output the number of bacteria. The bacterial count calculation unit 10b can output the counted number of bacteria to the sample bacterial concentration analysis unit 10c.

[0025] Furthermore, the bacterial count calculation unit 10b may be configured to count the number of bacteria for each classification of bacteria in the sample image. Specifically, it can count the number of bacteria for each classification of bacteria in the sample image based on the type of sample and the learning model acquired from the sample information acquisition unit 10a. For example, in the case of a urine sample, the bacterial count calculation unit is configured to count the number of Gram-positive cocci and Gram-negative bacilli. The learning model is configured to output the number of bacteria for each classification of bacteria in the sample image when a sample image is input to the learning model.

[0026] (3) Bacterial enrichment amount calculation unit 10d The bacterial enrichment amount calculation unit 10d can calculate the amount of bacteria added to the sample from the time the sample was collected to the present, based on the storage conditions of the sample. The storage conditions of the sample include, for example, the time the sample was collected, the time the sample was fixed, the time the sample was left to stand, and the temperature at which the sample was left to stand. The bacterial enrichment amount calculation unit can output the calculated bacterial enrichment amount to the sample bacterial concentration analysis unit 10c.

[0027] (4) Sample bacterial concentration analysis unit 10c The sample bacterial concentration analysis unit 10c can calculate bacterial concentration mainly based on the number of bacteria and the amount of sample. If the bacterial count calculation unit 10b counts bacteria for each bacterial classification, the sample bacterial concentration analysis unit 10c can calculate bacterial concentration for each bacterial classification. The sample bacterial concentration analysis unit 10c can output the calculated bacterial concentration to the bacterial colonization sample determination unit 10f and the infectious disease progress determination unit 10h.

[0028] The sample bacterial concentration analysis unit 10c can further calculate the bacterial concentration based on the imaging conditions. The imaging conditions refer to information such as which area of ​​the sample smeared on the glass slide 100 was photographed. In particular, with liquid samples, the peripheral area of ​​the smear region, where the sample is smeared, tends to have a higher bacterial count, while the central area tends to have a lower bacterial count. Therefore, for example, if the sample image was taken near the peripheral area of ​​the smear region, the actual bacterial concentration can be calculated to be lower than the bacterial concentration on the image by correcting the bacterial concentration obtained from the bacterial count calculation unit 10b and the amount of sample (hereinafter referred to as the bacterial concentration on the image). By correcting the bacterial concentration based on the imaging conditions, the sample bacterial concentration analysis unit 10c can accurately calculate the bacterial concentration of the entire sample.

[0029] Further, the specimen bacteria concentration analysis unit 10c can calculate the bacteria concentration based on the number of bacteria in a plurality of specimen images of the same specimen and the amount of the specimen. In this case, the imaging conditions are the number of specimen images, the imaging position of each image, and the smeared area. Specifically, first, the specimen bacteria concentration analysis unit 10c can divide the entire area where the specimen is smeared into areas where the position where each image is taken and the distribution of bacteria in the specimen are approximated (areas considered equal). That is, each image is arranged in each area. Next, the specimen bacteria concentration analysis unit 10c can calculate the area of each area and the imaging area of each specimen image based on the smeared area.

[0030] Here, the specimen images are divided into x images and the smeared area (smeared region) is divided into y regions. At this time, x and y may be the same or different. For example, when x = y and the specimen images are divided into n images and the smeared area is divided into y regions, the specimen bacteria concentration analysis unit 10c can calculate the bacteria concentration of the entire specimen by the following formula. In the following formula, the area concentration is calculated by dividing by the smeared area, but the volume concentration may be calculated by dividing by the volume of the dropped specimen.

[0031] Bacteria concentration [number of bacteria / mm ２ ] = (bacteria concentration [number of bacteria / mm ２ ] derived from the first image × area of the first region [mm ２ ] +... + bacteria concentration [number of bacteria / mm ２ ] derived from the nth image × area of the nth region [mm ２ ]) / smeared area [mm ２ ]

[0032] For easier understanding, here, the case of calculating the bacteria concentration from three specimen images (referred to as the first specimen image 101, the second specimen image 102, and the third specimen image 103) by taking a liquid specimen at three locations will be described more specifically using FIG. 3. FIG. 3 shows a top view of the entire specimen smeared on the slide glass 100.

[0033] As described above, in a liquid specimen, the number of bacteria is relatively small in the central part and tends to be larger in the peripheral part. This is because when the liquid specimen is dropped, it spreads in a concentric circle pattern, and thus the distribution of bacteria tends to change concentrically. Therefore, the specimen bacteria concentration analysis unit 10c divides the smeared area into a plurality of concentric circular areas with the center of the smeared area as the center point, and calculates the partial concentration, which is the concentration of bacteria in each divided area. The specimen bacteria concentration analysis unit 10c calculates the overall bacteria concentration of the smeared area based on the partial concentration for each divided area.

[0034] Specifically, as shown in FIG. 3, the specimen bacteria concentration analysis unit 10c can divide the entire area where the specimen is smeared into concentric circular areas A, B, and C such that the positions where each image is taken are arranged in each area. By dividing the area in this way, the concentration of bacteria within the same area can be regarded as uniform, and the partial concentration can be calculated. The specimen bacteria concentration analysis unit 10c can calculate the area of each of areas A, B, and C and the imaging area of the first, second, and third specimen images 101, 102, and 103 based on the smeared area. The specimen bacteria concentration analysis unit 10c can calculate the bacteria concentration of the entire specimen using the following formula based on the area of each area and the concentration of bacteria (partial concentration) in each specimen image.

[0035] Bacteria concentration [number of bacteria / mm ２ ] = (concentration of bacteria in the first specimen image 101 [number of bacteria / mm ２ ] × area of area A [mm ２ ] + concentration of bacteria in the second specimen image 102 [number of bacteria / mm ２ ] × area of area B [mm ２ ] + concentration of bacteria in the third specimen image 103 [number of bacteria / mm ２ ] × area of area C [mm ２ ]) / smeared area [mm ２ ]

[0036] Thus, when a plurality of images are taken, by considering the distribution of bacteria according to the specimen type, the bacteria concentration of the entire specimen can be calculated more accurately.

[0037] Furthermore, the sample bacterial concentration analysis unit 10c can also calculate the bacterial concentration based on the enrichment amount. The sample bacterial concentration analysis unit 10c can calculate the number of bacteria at the time the sample was collected based on the enrichment amount and calculate the bacterial concentration. By calculating the bacterial concentration based on the number of bacteria at the time of sample collection, it is possible to contribute to a more accurate determination of the possibility of infection.

[0038] Furthermore, the sample bacterial concentration analysis unit 10c can calculate the bacterial concentration based on pre-treatment information. Pre-treatment may involve changing the viscosity of the sample or diluting the sample. For example, in the case of a diluted sample, the sample bacterial concentration analysis unit 10c is configured to calculate the bacterial concentration of the sample before dilution based on the dilution ratio.

[0039] (5) Patient Information Acquisition Unit 10e The patient information acquisition unit 10e can acquire patient information entered into the input unit 30. Patient information refers to information about the patient who provided the specimen. Specifically, patient information may be any information that a physician would normally use to determine an infectious disease, such as age, sex, symptoms, treatment status, other test results (e.g., microbiological test, urine sediment test, urine culture test), possibility of pregnancy, presence or absence of intravenous fluids, amount of intravenous fluids, presence or absence of a urinary catheter, etc. The patient information acquisition unit 10e outputs the patient information to the bacterial colonization specimen determination unit 10f and the infectious disease possibility determination unit 10g.

[0040] (6) Bacterial colonization specimen determination unit 10f The bacterial colonization specimen determination unit 10f can determine whether a specimen is a bacterial colonization specimen, primarily based on the bacterial concentration. The determination of whether or not a specimen is a bacterial colonization specimen may be the same as the determination criteria used conventionally. For example, when the bacterial colonization specimen determination unit 10f obtains specimen information that the type of specimen is cerebrospinal fluid, and the bacterial concentration is 10 ４ When the concentration is 10 / mL or higher, the sample can be determined to be a bacterial colonization sample. For example, when the bacterial colonization sample determination unit 10f obtains sample information that the type of sample is urine collected by a catheter, and the bacterial concentration is 10 / mL or higher, the sample can be determined to be a bacterial colonization sample. ２ When the value is greater than or equal to / mL, the sample can be determined to be a bacterial colonization sample.

[0041] The bacterial colonization specimen determination unit 10f can further determine whether a specimen is a bacterial colonization specimen based on patient information. For example, the bacterial colonization specimen determination unit 10f can determine that a specimen is a bacterial colonization specimen if any of the following conditions are met: • When specimen information indicating that the type of specimen is urine and patient information indicating that the patient's gender is female are obtained, and the bacterial concentration is 10 for two consecutive times. ５ When the concentration is 10 / mL or higher, and when sample information indicating the sample type is urine and patient information indicating the patient's gender is male, and the bacterial concentration is 10 ５ When it is / mL or more

[0042] Furthermore, the bacterial colonization specimen determination unit 10f can determine whether a specimen is bacterial colonized based solely on the bacterial concentration, even if the specimen is supposed to be sterile. For example, if bacteria are found in a specimen that should be sterile, such as cerebrospinal fluid or bile, the bacterial colonization specimen determination unit 10f can determine that the specimen is bacterial colonized. The bacterial colonization specimen determination unit 10f can output the determination result to the infectious disease possibility determination unit 10g.

[0043] Furthermore, the bacterial colonization specimen determination unit 10f can determine whether a specimen is a bacterial colonization specimen by comparing the bacterial concentrations of two or more specimen images. Here, two or more specimen images can be images of specimens taken from spatially different locations on the same patient, or images of specimens taken from the same patient at different time points in time.

[0044] For example, when determining samples collected from the same patient at different spatial locations, if the bacterial concentration of one or more samples is significantly higher than that of the other samples (specifically, for example, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, or 10.0 times higher, and may be greater than or equal to any of the values ​​exemplified here, or within a range between two of them), the bacterial colonization sample determination unit 10f can determine that each of those samples is a bacterial colonization sample. Also, for example, when determining samples collected from the same patient at different temporal times, if the bacterial concentration of the newer sample is higher than that of the older sample (specifically, for example, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, or 10.0 times higher, and may be greater than or equal to any of the values ​​exemplified here, or within a range between two of them), the bacterial colonization sample determination unit 10f can determine that each of those samples is a bacterial colonization sample.

[0045] Furthermore, the bacterial colonization specimen determination unit 10f can determine whether or not a specimen is a bacterial colonization specimen based on the collection conditions, determine whether or not a specimen is a bacterial colonization specimen, and / or determine the accuracy of whether or not a specimen is a bacterial colonization specimen. Specifically, collection conditions include, for example, in the case of a urine specimen, whether it is the first urine sample or a midstream urine sample, and whether or not disinfection was performed when collecting the urine, and in the case of a sputum specimen, it may include, for example, whether or not it is aspirated sputum.

[0046] For example, in the case of a urine sample, even if the bacterial concentration obtained by the sample bacterial concentration analysis unit 10c shows a high value, if the collection condition is first urine, the bacterial colonization sample determination unit 10f can determine that it is impossible to determine. On the other hand, if the collection condition is mid-stream urine or urine collected after disinfection, it can determine whether or not the sample is a bacterial colonization sample. For example, in the case of a sputum sample, if the collection condition is aspirated sputum, the bacterial colonization sample determination unit 10f can determine whether or not the sample is a bacterial colonization sample and can determine that the accuracy of this determination is high.

[0047] (7) Infection possibility determination unit 10g The infection possibility determination unit 10g can determine whether or not a patient has an infectious disease based on the determination result from the bacterial colonization specimen determination unit 10f and patient information. Alternatively, the infectious disease may be determined from the bacterial concentration output by the specimen bacterial concentration analysis unit 10c without going through the bacterial colonization specimen determination unit 10f.

[0048] For example, if a urine sample is determined to be a bacterial colonization sample by the bacterial colonization sample detection unit 10f and there are no symptoms of infection, the infection possibility detection unit 10g determines that the urine sample is asymptomatic bacteriuria. Alternatively, if a urine sample is determined to be a bacterial colonization sample by the bacterial colonization sample detection unit 10f and there are symptoms of infection, the infection possibility detection unit 10g determines that the patient has a urinary tract infection.

[0049] The infection possibility determination unit 10g may be further configured to recommend treatment according to the determination result. Specifically, if the infection possibility determination unit 10g determines that the patient has an infection or is highly likely to have one, it may be configured to recommend medication in accordance with known guidelines, etc. More specifically, for example, in the case of a urine sample, if the infection possibility determination unit 10g determines that the patient has a urinary tract infection (or is highly likely to have one), it may be configured to select an antimicrobial agent based on guidelines or the facility's antibiogram.

[0050] (8) Infectious Disease Progression Determination Unit 10h The infectious disease progression determination unit 10h can determine the progression of a patient's infectious disease by comparing the bacterial concentrations obtained from two or more sample images obtained from the sample bacterial concentration analysis unit 10c. Here, the two or more sample images may be images of the same type of sample, taken from the same patient at different time points.

[0051] Specifically, the infectious disease progression determination unit 10h can determine that the progression is good if the bacterial concentration in the new sample is lower than that of the old sample, and that it is worsening if the bacterial concentration is higher. Furthermore, if bacteria are counted according to their classification, the progression of the infectious disease can be determined by the increase or decrease in the concentration of a specific bacterium.

[0052] (9) Cause Location Analysis Unit 10i The cause location analysis unit 10i can estimate the location of the infectious disease based on the sample information from the sample information acquisition unit 10a and the bacterial concentration from the sample bacterial concentration analysis unit 10c. Specifically, the cause location analysis unit 10i compares the bacterial concentrations of multiple samples collected from different locations and estimates the area near the collection site of the sample with the highest bacterial concentration as the cause location.

[0053] (10) Treatment effect determination unit 10j The treatment effect determination unit 10j can determine the effectiveness of the treatment based on the bacterial concentration from the sample bacterial concentration analysis unit 10c. Specifically, the treatment effect determination unit 10j compares the bacterial concentration before and after treatment and determines that the treatment is effective if the bacterial concentration after treatment is lower than the bacterial concentration before treatment, and determines that the treatment is ineffective if the bacterial concentration after treatment is the same as the bacterial concentration before treatment or if the bacterial concentration has increased. Furthermore, depending on whether or not there is a treatment effect and the magnitude of the effect, it can recommend additional tests or other treatments.

[0054] 1.2 Effects The bacterial load determination system 1 according to one embodiment of the present invention can calculate bacterial concentration from a sample image and sample volume through information processing. For example, in order to obtain bacterial concentration for a definitive diagnosis of a urinary tract infection, conventional methods such as urine culture tests required a separate procedure by a medical technologist. However, with the configuration of this system, bacterial concentration can be obtained without the need for a medical technologist to perform a separate procedure. Therefore, infectious diseases can be diagnosed more quickly.

[0055] 2. Second Embodiment This embodiment is similar to the first embodiment, and the contents described in the first embodiment are applicable to this embodiment insofar as they do not contradict its intent. The main difference between the bacterial load determination system 1 of this embodiment and the first embodiment is that it includes a sample image acquisition device 3. The differences will be explained below.

[0056] The specimen image acquisition device 3 is a device for acquiring specimen images and includes an imaging unit 70. The imaging unit 70 can have any configuration as long as it can acquire specimen images, and may be, for example, a camera or a microscope with imaging capabilities.

[0057] Furthermore, the specimen image acquisition device 3 may include a communication unit 80. The communication unit 80 is configured to transmit and receive data and has the function of connecting to a communication line. The specimen image acquisition device 3 is configured to communicate with the communication unit 40 of the information processing device 2 via the communication unit 80. Therefore, in this embodiment, the information processing device 2 can acquire specimen images from the specimen image acquisition device 3 via the communication unit 40.

[0058] 3. Third Embodiment This embodiment is similar to the first embodiment, and the contents described in the first embodiment are applicable to this embodiment insofar as they do not contradict its intent. The main difference in the bacterial quantity determination system 1 according to this embodiment is the configuration of the sample bacterial concentration analysis unit 10c. The differences will be explained below.

[0059] The sample bacterial concentration analysis unit 10c sets a first region in the smear area where the bacterial concentration is estimated to be high, and a second region which is different from the first region. It calculates the first partial concentration, which is the bacterial concentration in the first region, and the second partial concentration, which is the bacterial concentration in the second region, and calculates the overall bacterial concentration based on the first partial concentration and the second partial concentration.

[0060] Figure 5 is a schematic diagram of the smear region according to this embodiment. The sample bacterial concentration analysis unit 10c divides this smear region into a region D near the periphery and a region E located inside region D. Here, region D corresponds to a first region (hereinafter referred to as a hot spot) where the bacterial concentration is presumed to be high, and region E corresponds to a second region different from the first region.

[0061] As described in the first embodiment, in liquid samples, the bacterial count tends to be high at the periphery of the smear area and low in the center. On the other hand, in the periphery itself and the area very close to the periphery (area F in Figure 5), the liquid thickness is less than in other areas, so bacteria are often almost nonexistent. Therefore, in this embodiment, the area that extends a predetermined distance inward from the periphery of the smear area toward the center is defined as area D, which is presumed to be a hotspot.

[0062] Therefore, the smear area can be divided into the area inside the hotspot (area E), the hotspot itself (area D), and the area outside the hotspot (area F). However, as mentioned above, bacteria are not normally distributed in area F, and it is highly likely that this area does not adequately reflect the overall bacterial concentration. Therefore, the existence of sample images located in area F can be disregarded. In one embodiment, it is expected that excluding the number and concentration of bacteria in area F from the calculation of the overall bacterial concentration will result in a value closer to the actual situation.

[0063] Since liquid samples tend to spread in concentric circles, the shape of region D is preferably annular. In the example in Figure 5, the smear area may be circular, and the shape of region D is annular, but depending on the shape of the smear area, it can be set to other geometric shapes.

[0064] Region D is specifically a region located a predetermined distance inward from the periphery of the smeared area. The predetermined distance is, for example, 100 to 300 μm. Specifically, the predetermined distance is, for example, 100, 150, 200, 250, or 300 μm, and may be within the range of any two of the values ​​exemplified here. The width of region D is, for example, 5 to 20 μm. Specifically, the width of region D is, for example, 5, 10, 15, or 20 μm, and may be within the range of any two of the values ​​exemplified here.

[0065] The position of region D may be set in advance, or it may be set during analysis (by the "sample bacterial concentration analysis unit 10c" or a separately incorporated "region setting unit," etc.). If the position of region D is set in advance, it may be a fixed value, or it may be a value that fluctuates based on the slide glass used, the shape of the smear area, the smear conditions, the type and amount of the sample, etc. If the position of region D is set during analysis, for example, the position of region D may be determined based on the identified position after identifying the location where hotspots are formed (for example, where many hotspots are formed) by analyzing each sample image.

[0066] The following describes the details of the process performed by the sample bacterial concentration analysis unit 10c.

[0067] The sample bacterial concentration analysis unit 10c first divides the smear area into areas D, E, and F. For example, the sample bacterial concentration analysis unit 10c sets area D by obtaining the diameter and center coordinates of the smear area from the input unit 30. Based on the setting of area D, the inner area E and the outer area F are set. Specifically, for example, if the diameter of the smear area is 10 mm, the sample bacterial concentration analysis unit 10c sets the area with a width of 200 μm from the outer edge of the smear area toward the center as "area D". Then, the sample bacterial concentration analysis unit 10c sets the entire circular area inside area D as "area E", and the entire area outside area D as "area F".

[0068] Next, the sample bacterial concentration analysis unit 10c receives the shooting location information for each acquired sample image from the sample information acquisition unit 10a. The sample bacterial concentration analysis unit 10c compares the received shooting location information with the information for regions D and E that were set earlier. Through this process, the sample bacterial concentration analysis unit 10c determines where in regions D, E, and F each sample image was taken.

[0069] For example, in Figure 5, if the coordinates of the first sample image 104 are within the range of region D, the sample bacterial concentration analysis unit 10c classifies the sample image taken at that location as an "image belonging to region D". Similarly, if the coordinates of the second sample image 105 and the third sample image 106 are within the range of region E, the sample bacterial concentration analysis unit 10c classifies the sample images taken at those locations as "images belonging to region E". Since region E has a larger area than region D, it is preferable to acquire multiple sample images. Therefore, it is desirable for the sample bacterial concentration analysis unit 10c to randomly select many shooting locations within region E, or to select multiple locations from a crosshair passing through the center of the smear area. This makes it possible to acquire data with less bias.

[0070] Next, the sample bacterial concentration analysis unit 10c calculates the number of bacteria per unit area for each region and integrates them to calculate the overall bacterial concentration of the sample. Here, even within each of regions D and E, there may be variations in the density of bacteria distribution. Therefore, in order to calculate the representative bacterial concentration (first partial concentration, second partial concentration) for each region, it is preferable to acquire multiple sample images from each region and use the average of their measured values. The number of sample images to be acquired is set to a range of 1 to 200 for each of regions D and E, and more preferably 50 to 200.

[0071] The sample bacterial concentration analysis unit 10c obtains the bacterial count from the bacterial count calculation unit 10b and obtains area information in the sample images from the sample information acquisition unit 10a. Then, it calculates the average bacterial count for the image group belonging to region D and the image group belonging to region E, respectively.

[0072] Next, the sample bacterial concentration analysis unit 10c estimates the total number of bacteria in the entire sample using the bacterial concentrations and area of ​​each region. Furthermore, it obtains the amount of sample dropped from the sample information acquisition unit 10a and calculates the bacterial concentration of the entire sample based on the following formula. Note that in the formula below, the volume concentration is calculated by dividing by the volume of the sample, but the area concentration may also be calculated by dividing by the smear area. Bacterial concentration [number of bacteria / ml] = (Bacterial concentration in region D [number of bacteria / mm²] ２ ] × Area of ​​region D [mm² ２ ] + bacterial concentration in area E [bacterial count / mm³ ２ ] × Area of ​​region E [mm² ２ ]) ÷ Volume of the sample dropped [ml]

[0073] Specifically, as shown in Figure 5, the bacterial concentration when three sample images are obtained is calculated using the following formula: Bacterial concentration [bacteria / ml] = {First sample image 104 [bacteria / mm²] ２ ] × Area of ​​region D [mm² ２ ] + (Concentration of the second sample image 105 [bacterial count / mm³) ２ ] + Concentration of the third sample image 106 [bacterial count / mm³] ２ ]) ÷ 2 × Area of ​​region E [mm²] ２ ]} ÷ Volume of the sample dropped [ml]

[0074] According to this embodiment, the bacterial concentration of the entire sample is estimated by a weighted average that takes into account the area of ​​regions where bacteria are likely to be concentrated and other regions, thus enabling a more accurate determination that reflects the actual situation. Furthermore, by acquiring multiple sample images within each region and using their average value, the influence of local variations in density can be reduced, and the bacterial concentration can be calculated with greater accuracy.

[0075] 4. Other Embodiments The present invention can also be implemented in the following embodiments.

[0076] 4.1 Bacterial Count Calculation Unit 10b In the above embodiment, the bacterial count calculation unit 10b was configured to count the number of bacteria in the sample image based on the type of sample acquired by the sample information acquisition unit 10a and a learning model. However, it may also be configured to count the number of bacteria in the sample image based only on the learning model. In this case, the bacterial count calculation unit 10b can also identify the type of sample when counting the number of bacteria.

[0077] 4.2 Multiple Information Processing Devices In the above embodiment, each information processing was performed by a single information processing device 2, but it may also be performed by multiple information processing devices that can communicate with each other. For example, the processing can be configured to be performed by an information processing device comprising a sample information acquisition unit 10a, a bacterial count calculation unit 10b, a bacterial enrichment amount calculation unit 10d, and a sample bacterial concentration analysis unit 10c, and an information processing device comprising a patient information acquisition unit 10e, a bacterial colonization sample determination unit 10f, an infection possibility determination unit, a cause occurrence location analysis unit 10i, and a treatment effect determination unit 10j.

[0078] 4.3 The smear area calculation unit information processing device 2 may further include a smear area calculation unit. In the above embodiment, the smear area was described as a value input by the input unit 30, but the smear area calculation unit may be configured to calculate the smear area. Specifically, the smear area calculation unit can acquire an image of the entire sample that has been smeared and calculate the smear area from the image.

[0079] 4.4 The catheter-related bloodstream infection possibility determination unit information processing device 2 may further include a catheter-related bloodstream infection possibility determination unit. The catheter-related bloodstream infection possibility determination unit can determine whether or not a patient has a catheter-related bloodstream infection. The catheter-related bloodstream infection possibility determination unit makes the determination based on the bacterial concentration obtained from the sample bacterial concentration analysis unit 10c. The sample used in this case is a catheter tip sample.

[0080] 4.5 Specimen bacterial concentration analysis unit 10c (1) Arrangement of multiple specimen images in each region In the first embodiment, an example in which one specimen image corresponds to each divided region has been mainly described, but the present invention is not limited thereto. Depending on the imaging settings of the device and the division size of the region, multiple specimen images may be arranged within a single region.

[0081] In this case, the sample bacterial concentration analysis unit 10c first identifies all sample images belonging to a specific region (for example, region A when the smear region is divided into concentric circles). Next, the sample bacterial concentration analysis unit 10c obtains the number of bacteria counted by the bacterial count calculation unit 10b for each of the identified sample images. Then, the sample bacterial concentration analysis unit 10c calculates the average value of the obtained bacterial counts.

[0082] The sample bacterial concentration analysis unit 10c treats this calculated average value as the representative bacterial count for that specific region (region A) and uses it in subsequent bacterial concentration calculation processing. This process equalizes the influence of local bacterial biases within the region, allowing for more reliable analysis results.

[0083] (2) Subdivision of the hotspot region In the third embodiment, an example was described in which the smear region is divided into two areas, a hotspot (region D) and other regions (region E), for analysis, but the present invention is not limited thereto. In actual clinical specimens, even within the hotspot, there may be further variations in the concentration of bacteria. Therefore, in this modified example, the specimen bacterial concentration analysis unit 10c may be configured to further subdivide the hotspot (region D) into multiple regions for analysis.

[0084] For example, the sample bacterial concentration analysis unit 10c may divide region D in a direction from the outer edge toward the center. Specifically, region D with a width of 20 μm, as set in the third embodiment, is divided into multiple ring-shaped regions: an outer region D1 (width 10 μm) and an inner region D2 (width 10 μm).

[0085] Alternatively, for example, the sample bacterial concentration analysis unit 10c may divide region D in the periphery direction. In this case, for example, the smear region is divided into multiple fan-shaped regions by multiple radiation lines extending from the center point outwards. For example, by drawing radiation lines at 45-degree intervals from the center point, the entire smear region is divided into eight regions.

[0086] The sample bacterial concentration analysis unit 10c can calculate the bacterial concentration by calculating the bacterial concentration for each of the divided regions, summing them up, and dividing by the smear area or the volume of the dropped sample.

[0087] (3) Selection of bacterial concentration calculation method The bacterial quantity determination system 1 may be configured to be able to execute both the calculation method described in the first embodiment (a method of dividing the area into concentric circles and calculating the bacterial concentration) and the calculation method described in the third embodiment (a method of dividing the area into a hotspot and other areas and calculating), so that it can be selected according to the situation.

[0088] For example, a healthcare professional, acting as the user, can pre-select which calculation method to use via the input unit 30. In this configuration, the healthcare professional can instruct the system on the optimal analysis method based on their own experience and visual impression of the specimen.

[0089] Alternatively, the sample bacterial concentration analysis unit 10c may be configured to automatically select the optimal calculation method based on sample information or an image of the entire smear area. When based on sample information, for example, in the case of a urine sample, bacteria tend to concentrate at the edges when dried, so the sample bacterial concentration analysis unit 10c can be configured to automatically select the calculation method of the third embodiment. Furthermore, when based on an image of the entire smear area, the sample bacterial concentration analysis unit 10c can be configured to analyze the image and select the calculation method of the third embodiment if bacteria are concentrated near the edges, and the calculation method of the first embodiment otherwise. This automatic selection function makes it possible to provide highly accurate test results by automatically performing the optimal analysis tailored to the characteristics of each sample, even if the user does not have specialized knowledge.

[0090] This section describes an experiment that confirmed the distribution of bacteria within a smear area using actual clinical specimens.

[0091] (1) Experimental Objectives: The purpose of this experiment was to confirm how bacteria are actually distributed using urine samples and to identify bacterial distribution patterns.

[0092] (2) Experimental Method Sample: Urine samples collected in a clinical setting were used. Smear Method: 5 to 10 μl of urine sample was dropped into a circular frame on a glass slide (sample holding member 100), and dried to prepare a smear. Figure 6 shows the glass slide actually used in this experiment, and as shown in the figure, the sample was smeared into eight circular areas. Observation Method: After Gram staining, the smear area was observed using a microscope. Then, in order to evaluate the distribution of bacteria, sample images of multiple fields within the smear area were acquired. Data Collection: The observation results were recorded on a schematic diagram in which the smear area was divided into concentric circles and radially, with the bacterial concentration in five stages (0 to 10, 11 to 25, 26 to 50, 51 to 80, and 81 or more, depending on the number of bacteria per field), color-coded so that the color becomes darker as the number of bacteria increases. In addition, representative sample images for each concentration stage were saved.

[0093] (3) Experimental Results and Discussion The experimental results are shown in Figures 7 to 10. From these experimental results, it was observed that the distribution of bacteria tended to spread concentrically from the center to the periphery. For example, in sample 808B shown in Figure 8A, although the overall bacterial count was relatively high, a tendency similar to the distribution assumed in the first embodiment was confirmed, where the concentration increased concentrically from the center to the outer edge.

[0094] Other results also confirmed a tendency for bacteria to be distributed randomly. For example, in sample 820B shown in Figure 7B and sample 831B shown in Figure 9B, no clear concentric pattern was observed in the distribution of bacteria in the internal region; instead, a mixture of light and dark areas was present. In such cases, it is considered effective to take images from multiple locations and calculate the average value in order to accurately evaluate the bacterial concentration in the internal region (region E in the third embodiment).

[0095] Furthermore, the experimental results confirmed that the bacterial concentration tends to be generally higher near the periphery of the smear area. For example, in sample 808A shown in Figure 8A and sample 806A shown in Figure 10A, it can be seen that bacteria are distributed at a higher concentration near the periphery compared to the central area. Therefore, the approach of the third embodiment, which clearly divides the smear area into regions where the bacterial concentration is presumed to be high near the periphery and other internal regions for analysis, has been demonstrated to be rational.

[0096] From these results, it was found that the concentric circle approach according to the first embodiment and the approach according to the third embodiment that distinguishes between hotspots and other areas are superior methods for accurately calculating bacterial concentration.

[0097] 1: Bacterial load determination system, 2: Information processing device, 3: Sample image acquisition device, 10: Control unit, 10a: Sample information acquisition unit, 10b: Bacterial count calculation unit, 10c: Sample bacterial concentration analysis unit, 10d: Bacterial enrichment amount calculation unit, 10e: Patient information acquisition unit, 10f: Bacterial colonization sample determination unit, 10g: Infection possibility determination unit, 10h: Infection course determination unit, 10i: Cause and occurrence location analysis unit, 10j: Treatment effect determination unit, 20: Storage unit, 30: Input unit, 40: Communication unit, 50: Output unit, 60: Communication bus, 70: Imaging unit, 80: Communication unit, 100: Slide glass, 101: First sample image, 102: Second sample image, 103: Third sample image, 104: First sample image, 105: Second sample image, 106: Third sample image

Claims

1. An information processing device comprising: a bacterial count calculation unit; a sample information acquisition unit; and a sample bacterial concentration analysis unit, wherein the bacterial count calculation unit counts the number of bacteria in a sample image taken of at least a portion of a sample placed on a sample holding member; the sample information acquisition unit acquires sample information including the sample image and the amount of the sample placed on the sample holding member; and the sample bacterial concentration analysis unit calculates the bacterial concentration based on the bacterial count and the amount of the sample.

2. An information processing device according to claim 1, wherein the bacterial count calculation unit counts the number of bacteria in the sample image based on a learning model, and the learning model is configured to output the number of bacteria in the sample image when the sample image is input to the learning model.

3. An information processing device according to claim 2, wherein the bacterial count calculation unit counts the number of bacteria for each classification of bacteria in the sample image based on the learning model, and the learning model is configured to output the number for each classification of bacteria in the sample image when the sample image is input to the learning model.

4. An information processing device according to claim 1, wherein the sample information acquisition unit acquires the shooting position of the sample image in the area placed on the sample holding member, and the sample bacterial concentration analysis unit calculates the bacterial concentration based on the shooting position.

5. An information processing device according to claim 4, wherein the sample bacterial concentration analysis unit divides the smear region, which is the area on which the sample is smeared, into a plurality of concentric circular regions with the center of the smear region as the central point, calculates a partial concentration which is the bacterial concentration for each divided region, and calculates the total bacterial concentration of the smear region based on the partial concentration for each divided region.

6. An information processing apparatus according to claim 4, wherein the sample bacterial concentration analysis unit sets a first region in a smear region, which is a region on which the sample is smeared, where the bacterial concentration is estimated to be high, and a second region, which is a region different from the first region, calculates a first partial concentration, which is the bacterial concentration in the first region, and a second partial concentration, which is the bacterial concentration in the second region, and calculates the overall bacterial concentration based on the first partial concentration and the second partial concentration.

7. An information processing device according to claim 1, further comprising an enrichment amount calculation unit, wherein the enrichment amount calculation unit calculates the amount of bacteria added to the sample from the time the sample was collected to the present based on the storage conditions of the sample, and the sample bacterial concentration analysis unit further calculates the bacterial concentration at the time the sample was collected based on the enrichment amount.

8. An information processing apparatus according to claim 1, further comprising a bacterial colonization specimen determination unit, wherein the bacterial colonization specimen determination unit determines whether or not the specimen is a bacterial colonization specimen based on the bacterial concentration.

9. An information processing device according to claim 8, further comprising a patient information acquisition unit, wherein the patient information acquisition unit acquires patient information of the patient who provided the sample, and the bacterial colonization sample determination unit determines whether or not the sample is a bacterial colonization sample based on the sample information, the patient information, and the bacterial concentration.

10. An information processing device according to claim 9, wherein the bacterial colonization specimen determination unit determines whether or not a specimen is a bacterial colonization specimen by comparing the bacterial concentrations obtained from two or more specimen images, and the two or more specimen images are images of specimens taken from spatially different locations of the same patient, or images of specimens taken from the same patient at different time points in time.

11. An information processing device according to claim 9 or claim 10, further comprising an infectious disease possibility determination unit, wherein the infectious disease possibility determination unit determines whether or not the patient is suffering from an infectious disease based on the determination result from the bacterial colonization specimen determination unit and the patient information.

12. An information processing device according to claim 11, further comprising an infectious disease progression determination unit, wherein the infectious disease progression determination unit determines the progression of the infectious disease of the patient by comparing the bacterial concentrations obtained from two or more of the sample images, and the two or more sample images are images of the same type of sample and of the same patient taken at different times in time.

13. An information processing apparatus according to claim 1, wherein the sample is urine.

14. A system comprising: an imaging unit; a bacterial count calculation unit; a sample information acquisition unit; and a sample bacterial concentration analysis unit, wherein the imaging unit is configured to capture a sample image of a sample placed on a sample holding member; the bacterial count calculation unit counts the number of bacteria in the sample image; the sample information acquisition unit acquires the quantity of the sample placed on the sample holding member; and the sample bacterial concentration analysis unit calculates the bacterial concentration based on the bacterial count and the quantity of the sample.

Images