Method for detecting renal disease of companion animal on basis of protein, creatinine and specific gravity

Abstract

The present invention relates to a method by which a server detects renal disease of a companion animal on the basis of protein, creatinine and specific gravity, and is a technique related to a method for detecting renal disease of a companion animal on the basis of protein, creatinine and specific gravity, the method comprising the steps of: (a) constructing an RGB ratio table, and photographing a urine test strip stained with urine, so as to generate a test image; (b) generating a first RGB value by extracting, from the test image, RGB values for each of a protein pad area and a creatinine pad area corresponding to a renal test item area from among the test item areas of the urine test strip; and (c) generating a first ratio value through the generated first RGB value and a specific gravity RGB value extracted from a specific gravity pad area in the test image, generating a second ratio value, which is the ratio between the specific gravity of the RGB ratio table, and protein and creatinine, correcting the RGB value of the protein pad area and the RGB value of the creatinine pad area by comparing the first ratio value and the second ratio value, and determining the renal health state of the companion animal by comparing the corrected RGB values.

Description

A method for detecting kidney disease in pets based on protein, creatinine, and specific gravity.

[0001] The present invention relates to a method and system for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, and more particularly, to a technology for providing kidney health test results by extracting and analyzing RGB values ​​for color changes in a companion animal urine test strip.

[0002] As the number of people keeping pets has increased recently, animal health management businesses for the health of pets are also increasing.

[0003] Pet health management can be largely carried out through blood collection and urine tests, much like human health management. Urine tests examine the physical properties of urine, such as color and turbidity, and semi-quantitatively detect various types of waste products excreted in urine.

[0004] The conventional urine test method has been based on the color change of the protein area pad, creatinine area pad, and specific gravity pad of the urine test strap when they come into contact with the pet's urine. However, this method has the problem that the color is not clearly distinguished depending on the pet's species, age, and condition.

[0005] Accordingly, there is a need for a testing method that can consistently test the protein, creatinine, and specific gravity contained in the urine of companion animals, regardless of the species, age, and condition of the companion animal.

[0006] The present invention is intended to solve the problems of the above-mentioned prior art, and aims to provide a method and system for detecting kidney disease in companion animals based on protein, creatinine, and specific gravity.

[0007] Through this, the purpose is to perform a urine test based on the color that each sample changes by applying the urine of the pet to a urine test strip equipped with protein, creatinine, and specific gravity samples in a designated area, and to read the color of each test area of ​​the urine test strip based on the color of the standard colorimetric table, thereby collecting samples easily and non-invasively without causing pain to the pet.

[0008] The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned can be clearly understood from the description below.

[0009] As a technical means for achieving the above-described technical task, a method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, performed by a server according to an embodiment of the present invention, may include the steps of: (a) constructing an RGB ratio table and photographing a urine test strip stained with urine to generate a test image; (b) extracting RGB values ​​for each of a protein pad area and a creatinine pad area corresponding to a kidney test item area among test item areas of a urine test strip in the test image to generate a first RGB value; and (c) generating a first ratio value through the generated first RGB value and a specific gravity RGB value extracted from a specific gravity pad area in the test image, generating a second ratio value which is a ratio between the specific gravity of the RGB ratio table and protein and creatinine, comparing the first ratio value and the second ratio value, correcting the RGB value of the protein pad area and the RGB value of the creatinine pad area, and comparing them to determine the kidney health status of the companion animal.

[0010] In addition, the step (a) may be to construct an RGB ratio table by calculating RGB values ​​for combinations between all colors that the protein pad area and the creatinine pad area can express according to the health status of the kidney based on a standard colorimetric table, and calculating a ratio between each RGB value and the RGB values ​​for all colors that the protein pad area and the creatinine pad area can express.

[0011] Additionally, the above standard colorimetric table may define all colors that can be expressed by the pad area corresponding to each inspection item.

[0012] In addition, the step (b) may further include a step of calculating a ratio between the first RGB value of the protein pad region and the first RGB value of the creatinine pad region, and comparing the ratio with the RGB ratio table corresponding to a preset first judgment table classified into normal, borderline, and suspicious, thereby making a first judgment on the kidney health status of the companion animal, and making a first judgment on the kidney health status of the companion animal to be examined as at least one of normal, borderline, and suspicious.

[0013] In addition, the step (c) may mean a ratio between colors of pad areas of the protein pad and creatinine pad corresponding to colors that can be expressed in the specific gravity pad area based on RGB values ​​extracted from the standard colorimetric table, where the second ratio value is used.

[0014] In addition, the first ratio value and the second ratio value may each have a preset effective range, and a numerical value for kidney health may be calculated by determining whether the effective range of the first ratio value is included in the effective range of the second ratio value.

[0015] In addition, the step (c) may further include a step of secondarily judging the kidney health of the companion animal by comparing the second ratio value and the first ratio value and calculating each value corresponding to a preset second judgment table.

[0016] In addition, (d) if the judgment results of the first judgment and the second judgment are the same, the kidney health status of the companion animal is finally judged based on the result, and if the judgment results of the first judgment and the second judgment are different, the step of correcting the first judgment result according to a preset algorithm and making a final judgment may be further included.

[0017] In addition, (e) a step of collecting multiple test images by time period, generating data matching the images with a standard colorimetric table and information on the ratio of each, learning the data as input values ​​according to a supervised learning method, and generating a machine learning model that determines, as an output value, how much time has passed since urine was applied to the strip when a specific test image is input, and, based on the machine learning model, if there is a strip among the strips in the image taken within a preset time period that is determined to have been applied to the strip more than 1 minute ago, providing a guidance message to the user terminal informing that a test on a strip that has passed a specific time period may be inaccurate; may further be included.

[0018] According to one embodiment of the present invention, a server for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity comprises: a memory storing a program for performing a method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity; and

[0019] A processor for executing the above program is included; and the method may include the steps of: (a) constructing an RGB ratio table and photographing a urine test strip stained with urine to generate a test image; (b) extracting RGB values ​​for each of a protein pad area and a creatinine pad area corresponding to a kidney test item area among test item areas of a urine test strip in the test image to generate a first RGB value; and (c) generating a first ratio value through the generated first RGB value and a specific gravity RGB value extracted from a specific gravity pad area in the test image, generating a second ratio value which is a ratio between the specific gravity of the RGB ratio table and protein and creatinine, comparing the first ratio value and the second ratio value, correcting the RGB value of the protein pad area and the RGB value of the creatinine pad area, and comparing them to determine the kidney health status of the companion animal.

[0020] The present invention provides a method and system for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, thereby allowing a urine test to be performed based on the color that each sample changes by applying urine of the companion animal to a urine test strip having protein, creatinine, and specific gravity samples in a predetermined area.

[0021] In addition, since the urine test strip with urine on it is read based on the color of each test area on the standard colorimetric chart, the sample can be collected easily and non-invasively without causing any pain to the pet, so that the pet and its owner are not negatively affected by the health test.

[0022] Figure 1 is a schematic diagram of a system for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, according to one embodiment of the present invention.

[0023] Figure 2 is a block diagram showing the internal configuration of a server for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, according to one embodiment of the present invention.

[0024] Figure 3 is an exemplary diagram of a strip according to one embodiment of the present invention.

[0025] Figure 4 is an exemplary diagram of a standard colorimetric table according to one embodiment of the present invention.

[0026] Figure 5 is an exemplary diagram of a judgment table according to one embodiment of the present invention.

[0027] Figure 6 is an example diagram of an inspection result UI according to one embodiment of the present invention.

[0028] Figure 7 is a flowchart illustrating the execution order of a method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity according to one embodiment of the present invention.

[0029] Below, with reference to the attached drawings, embodiments of the present invention are described in detail so that those skilled in the art can easily implement them. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, irrelevant parts have been omitted for clarity of description, and similar reference numerals have been used throughout the specification to indicate similar elements.

[0030] Throughout the specification, when a part is said to be "connected" to another part, this includes not only the cases where the parts are "directly connected" but also the cases where the parts are "electrically connected" with other elements intervening. Furthermore, when a part is said to "include" a component, this does not exclude other components, but rather includes other components, unless otherwise stated.

[0031] In this specification, the term 'unit' includes a unit realized by hardware, a unit realized by software, and a unit realized using both. In addition, one unit may be realized by using two or more pieces of hardware, and two or more units may be realized by one piece of hardware. Meanwhile, the '~ unit' is not limited to software or hardware, and the '~ unit' may be configured to be in an addressable storage medium or may be configured to reproduce one or more processors. Therefore, as an example, the '~ unit' includes components such as software components, object-oriented software components, class components, and task components, as well as processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functionality provided within the components and '~sub-units' may be combined into a smaller number of components and '~sub-units' or further separated into additional components and '~sub-units'. Furthermore, the components and '~sub-units' may be implemented to activate one or more CPUs within the device or secure multimedia card.

[0032] The "terminal" mentioned below may be implemented as a computer or portable terminal that can access a server or other terminal via a network. Here, the computer may include, for example, a notebook, desktop, laptop equipped with a web browser, VR HMD (e.g., HTC VIVE, Oculus Rift, GearVR, DayDream, PSVR, etc.). Here, the VR HMD includes all of the following: for PC (e.g., HTC VIVE, Oculus Rift, FOVE, Deepon, etc.), for mobile (e.g., GearVR, DayDream, Storm Magic, Google Cardboard, etc.), for console (PSVR), and Stand Alone models (e.g., Deepon, PICO, etc.) that are implemented independently. A portable terminal is, for example, a wireless communication device that ensures portability and mobility, and may include not only a smart phone, a tablet PC, and a wearable device, but also various devices equipped with communication modules such as Bluetooth (BLE, Bluetooth Low Energy), NFC, RFID, ultrasonic, infrared, WiFi, and LiFi. In addition, a "network" refers to a connection structure that enables information exchange between each node, such as terminals and servers, and includes a local area network (LAN), a wide area network (WAN), the Internet (WWW: World Wide Web), wired and wireless data communication networks, telephone networks, and wired and wireless television communication networks.Examples of wireless data communication networks include, but are not limited to, 3G, 4G, 5G, 3GPP (3rd Generation Partnership Project), LTE (Long Term Evolution), WIMAX (World Interoperability for Microwave Access), Wi-Fi, Bluetooth, infrared, ultrasonic, visible light communication (VLC), and LiFi.

[0033] The present invention relates to a method and system for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, and to a technology for providing kidney health test results by extracting and analyzing RGB values ​​for color changes in a companion animal urine test strip.

[0034] To this end, a system for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity according to one embodiment of the present invention may be composed of a server (100), a user terminal (200), and a urine test strip (300).

[0035] Referring to FIG. 2, a server (100) according to one embodiment of the present invention may include a memory storing a program (or application) for performing a method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, a processor for executing the program, and a database (DB, Data Base). Here, the processor may perform various functions depending on the execution of the program stored in the memory. The various functions performed by the processor will be described in detail later.

[0036] Next, the user terminal (200) is capable of communicating with the server (100) via wired or wireless connection, and has a program (or application) installed thereon that performs a method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity prior to the implementation of the present invention, and may include a camera function.

[0037] For example, it can be implemented in the form of a smartphone, laptop, desktop, tablet PC, etc., and the camera function can be performed by the server (100) by granting authority to the camera control signal received by the server (100).

[0038] Next, referring to FIG. 3, a strip (300) according to one embodiment of the present invention includes a region where a reagent for testing each component contained in urine is applied, and each reagent for testing has its own standard color, and when reacting with urine, it may change color to a different color depending on the health condition of the subject.

[0039] Additionally, the strip (300) may include a protein pad region, a creatinine pad region, and a specific gravity pad region arranged in succession.

[0040] Hereinafter, a method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity according to one embodiment of the present invention utilizing the above-described system will be described.

[0041] A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity according to one embodiment of the present invention can be implemented by a processor of a server (100) executing a corresponding application or program.

[0042] First, the server (100) builds an RGB ratio table and photographs a urine test strip (300) stained with urine to create a test image.

[0043] At this time, the RGB ratio table according to one embodiment of the present invention may be one that calculates RGB values ​​for combinations between all colors that the protein pad area and the creatinine pad area can express according to the health status of the kidney based on the standard colorimetric table (310), and calculates the ratio between each calculated RGB value and the RGB values ​​for all colors that the protein pad area and the creatinine pad area on the actual strip (300).

[0044] For example, assuming that the standard colorimetric table (310) includes red (R, Red), green (G, Green), and blue (B, Blue), the RGB ratio table can be formed to include colors for all cases that can be produced by combining the three colors described above by the server (100).

[0045] At this time, each color may be combined in the same ratio, 33.3% for three colors, 50% for two colors, but according to an additional embodiment of the present invention, the colors may be combined in a different ratio depending on the ratio set by the server (100) or the user terminal (200), so that a greater number of combination results may be included than combination results of the same ratio.

[0046] The standard colorimetric table (310) utilized here defines all colors that can be expressed by the pad area corresponding to each inspection item, and as illustrated in FIG. 4, may be implemented in the form of a palette that includes multiple colors on the left and right sides centered on the strip (300).

[0047] In addition, when a strip (300) is overlapped on a standard colorimetric table (310), each pad region included in the strip (300) can be formed to be perpendicular to one of the color types of one row of the palette, and the rows of the palette can represent different disease states depending on the color type.

[0048] After the user applies the pet's urine to the strip (300), waits for the pad area of ​​the strip (300) to be sufficiently discolored by the pet's urine, and then transmits the captured image to the server (100), the server (100) creates the image as an inspection image and stores it.

[0049] Thereafter, the server (100) extracts RGB values ​​for each of the protein pad area and the creatinine pad area corresponding to the kidney test item area among the test item areas of the urine test strip (300) in the test image to generate a first RGB value.

[0050] At this time, the server (100) according to one embodiment of the present invention calculates a ratio value between the first RGB value of the protein pad area and the first RGB value of the creatinine pad area among the plurality of pads included in the strip (300) in order to determine the kidney health status.

[0051] In addition, the server (100) compares the ratio value between the two calculated areas with the previously constructed RGB ratio table, and makes a primary judgment on the pet's kidney health status by corresponding the comparison value to a preset primary judgment table (320) that is divided into normal, borderline, and suspicious.

[0052] Referring to FIG. 5, the judgment table (320) includes information on protein and creatinine content and specific gravity (protein:creatinine ratio) that can determine the kidney health status of the pet to be examined as at least one of normal, borderline, and suspicious, and a normal category, a borderline category, and a suspicious category may be set for each section.

[0053] Meanwhile, the protein and creatinine contents shown in Figure 5 may be formed with significant differences depending on the variety of companion animal species, so the drawing shows arbitrary numbers and ratios.

[0054] After the above first judgment is performed, the server (100) generates a first ratio value using the generated first RGB value and the specific gravity RGB value extracted from the specific gravity pad area within the inspection image.

[0055] The specific gravity pad area is formed separately from the protein pad and creatinine pad in the strip (300), and may have different colors depending on the specific gravity, which represents a numerical value indicating how much of a substance is contained in the urine.

[0056] Therefore, assuming that a protein pad, a creatinine pad, and a specific gravity pad exist in the strip (300), the content of protein and creatinine in the urine of a companion animal can be intuitively estimated in such a way that, based on the color change of the specific gravity pad, if the specific gravity is measured to be high, the content of other components other than protein and creatinine is high, and thus the content of protein and creatinine is estimated to be somewhat low; conversely, if the specific gravity is measured to be low, the content of other components other than protein and creatinine is low, and thus the content of protein and creatinine is estimated to be somewhat high.

[0057] This is a key indicator for assessing kidney health. If there is a lot of protein in the urine, it is estimated that the subject of the test is very likely to suffer from diseases such as proteinuria and glomerular dysfunction, so it can be used as a clear indicator for assessing kidney health.

[0058] In addition, the server (100) can generate a second ratio value, which is the ratio between the specific gravity of the constructed RGB ratio table and the protein and creatinine, and correct the RGB value of the protein pad area and the RGB value of the creatinine pad area by comparing the first ratio value and the second ratio value.

[0059] Here, the second ratio value may mean a ratio between the colors of the pad area of ​​the protein pad and the creatinine pad corresponding to the colors that can be expressed in the specific gravity pad area based on the RGB values ​​extracted from the standard colorimetric table (310).

[0060] That is, as with the first ratio value explained above, when the specific gravity is measured high, since other components other than protein and creatinine are contained in large quantities, the protein and creatinine content is estimated to be somewhat low, and conversely, when the specific gravity is measured low, since other components other than protein and creatinine are contained in small quantities, the protein and creatinine content is estimated to be somewhat high, so different colors are displayed for cases where the specific gravity value is high and low, respectively.

[0061] Reflecting this, the server (100) knows in advance the RGB ratio values ​​of protein and creatinine according to the specific gravity value when constructing the RGB ratio table described above, and can therefore use it to compare with the first ratio value, which is the ratio value shown on the actual urine test strip (300).

[0062] For example, when comparing a first ratio value, which is a ratio between the color of the protein pad area and the color of the creatinine pad area shown in an actual urine test strip (300), and a second ratio value, which is a ratio between the protein pad area and the creatinine pad area (i.e., a second ratio value according to specific gravity) in a pre-built RGB ratio table, in an ideal case, the two ratios should be completely identical, but it is difficult for the two ratios to be completely identical due to various variables during a urine test (the pet's condition, urine volume, and individual differences).

[0063] Accordingly, the server (100) of the present invention can compare two ratios and, if the two ratio values ​​are less than or equal to a preset difference, correct each color actually measured with the specific gravity, protein, and creatinine color of the RGB ratio table.

[0064] For example, if a dark red color appears with black mixed in the protein pad area of ​​the tested urine test strip (300), the server (100) can correct the color of the pad to a vivid red color, which is the ideal color of the RGB ratio table.

[0065] According to an additional embodiment of the present invention, after the server (100) performs such correction for all pad areas in the inspection image received from the user terminal (200), the server (100) may display and provide the correction-performed inspection image together with the inspection results when providing the inspection results to the user terminal (200).

[0066] Returning to the above, the first ratio value and the second ratio value according to one embodiment of the present invention may each have a preset effective range, and a numerical value for kidney health may be calculated by determining whether the effective range of the first ratio value is included in the effective range of the second ratio value.

[0067] The server (100) can make a second judgment on the kidney health of the companion animal by comparing the second ratio value and the first ratio value and corresponding each value calculated to a preset second judgment table (not shown). The second judgment table is written in the same format as the first judgment table (320), but may be created by a machine learning model to be described later, or may receive each value to compose the table from the server (100) or the user terminal (200).

[0068] Next, if the judgment results of the first judgment and the second judgment are the same, the server (100) can make a final judgment on the pet's kidney health status based on the result, and if the judgment results of the first judgment and the second judgment are different, the server can correct the first judgment result according to a preset algorithm to make a final judgment.

[0069] At this time, according to an additional embodiment of the present invention, correction of the first judgment result may be performed by color correction based on the RGB ratio table described above, and may be repeated until each judgment result is the same.

[0070] Referring to Fig. 6, the judgment result as described above may be provided through the test result UI (210). The test result UI (210) according to one embodiment of the present invention may include, as illustrated, disease information, disease grade information, a bar for visualizing and displaying the disease grade information, and an additional identifier (251) displayed on the bar to indicate a detailed disease grade.

[0071] Here, the disease grade information is divided into normal, significant, and suspicious, the disease information includes one disease among multiple diseases preset for each test type, and the test result may further include at least one of the test round and disease solution.

[0072] In addition, the bar may be displayed by dividing it into multiple sections, with a separate color displayed in the normal range sections, an additional identifier (251) displayed within each section, and the detailed disease grade may be displayed based on the color or position of the additional identifier (251).

[0073] At this time, the additional identifier (251) is composed of a geometric image, and the closer it is to the normal range within the corresponding cell, the lower the risk of the detailed disease grade.

[0074] In addition, when an additional identifier (251) is placed in the cell immediately following a normal range cell, and when the cell is divided into N equal parts, the cell closest to the normal range among the N equal parts sub-cells, the detailed grade indicates suspected disease grade information close to normal, and when an additional identifier (251) is placed in the remaining cells among the N equal parts sub-cells, the detailed grades can all indicate suspected disease grade information at the same level.

[0075] Here, the additional identifier (251) may be used to determine the severity of the symptom as the degree of agreement is higher depending on how closely the first and second ratio values ​​calculated previously match.

[0076] Accordingly, by utilizing the test result UI (210) of the present invention, even if the strip (300) test result contains suspicion of a specific disease, the degree can be intuitively determined through the additional identifier (251) as to whether it is severe or mild, and since it is possible to suspect a case in which the symptom temporarily appears even though the health condition of the pet is normal depending on the pet's condition, it is possible to prevent in advance a pet that can be determined to be normal from suffering damage due to excessive treatment.

[0077] In addition, as illustrated, in an additional embodiment of the present invention, the examination date on which the examination was performed, the veterinarian's opinion on the examination results, and the component determined to be abnormal, for example, glucose, as illustrated, may be displayed together on the examination result UI (210), and the types of diseases that the component may cause or contain and that have a high probability of developing may be provided in the form of thumbnails at the bottom.

[0078] Meanwhile, the color of the strip (300) changes over time after urine is applied to it. Normally, the test result is most accurate when measured after 1 minute, and the color after 2 minutes, 5 minutes, and 1 hour are all slightly different. If the strip (300) is photographed with the color at that time and then tested, there is a problem that the test result is inaccurate.

[0079] In order to solve such a problem, in addition, in another embodiment of the present invention, if the disease grade (significant, suspicious, borderline, etc.) changes rapidly within a preset short period of time (about 20 minutes), or the displacement of the additional identifier (251) indicating the detailed disease grade changes significantly, and the user is performing multiple tests, the server (100) can estimate that the user is performing multiple tests over time using one strip (300) that has been soaked in urine once.

[0080] Alternatively, a user may intentionally or accidentally use a strip (300) that was previously urinated on during the current test. In such a case, the litmus paper color of the strip (300) may change, and the image captured by the camera that captured the strip (300) may also change. In such a case, a serious illness may suddenly be diagnosed, as in an error phenomenon, or the displacement of the additional identifier (251) of the detailed identification information may significantly change.

[0081] Accordingly, in the embodiment, if the server (100) detects that the displacement of the additional identifier (251) is outside the preset range or that the disease grade changes rapidly, it can perform a past analysis on the cycles and patterns in which the user performed the test (e.g., a past analysis on how many times or how many cycles the strip (300) test was performed in the past).

[0082] Afterwards, if a peculiarity is found by comparing with the recent inspection pattern and cycle, it can be assumed that the same strip (300) is being inspected too frequently, and if no peculiarity is found, it can be assumed that a past strip (300) or a strip (300) of another animal was inspected incorrectly, and the estimated result value can be provided as a message to the user terminal (200).

[0083] Next, the server (100) can go through a feedback process of collecting responses to the message (responses to whether the estimate made by the server (100) is correct or incorrect), and then store the feedback.

[0084] If the server (100) receives a lot of feedback of “O” through the above processes, it can collect all of the user’s feedback information, such as the user’s examination cycle, number of examinations, change period of disease grade, degree of change, degree of displacement of additional identifier (251) within the detailed disease grade, etc., and input them into an unsupervised learning model to perform learning, thereby classifying the patterns of users performing normal examinations and patterns in which errors or abnormal patterns occur.

[0085] Through the machine learning model learned as above, the server (100) inputs the user's examination information (examination cycle, number of examinations, change period of disease grade, degree of change, degree of displacement of additional identifier (251) within detailed disease grade, etc.) into the model, and can determine in advance the possibility of an error occurring or an abnormal examination result being provided to the user and provide a guidance message including guidance thereon.

[0086] In addition, a machine learning model according to an embodiment of the present invention may, in addition to the above-described functions, collect multiple inspection images by time period, generate data matching the images with a standard colorimetric table (310) and information on each ratio, learn the data as input values ​​according to a supervised learning method, and, when a specific inspection image is input as an output value, determine how much time has passed since the strip (300) was soaked with urine.

[0087] In the embodiment, if the server (100) determines, based on a machine learning model, that a strip (300) among the strips (300) of images taken within a preset time period has been exposed to urine for more than 1 minute, the server (100) may provide a guidance message to the user terminal (200) informing that testing on a strip (300) that has been exposed to urine for a certain period of time may be inaccurate.

[0088] Hereinafter, with reference to FIG. 7, a method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity according to one embodiment of the present invention will be described again.

[0089] First, the user terminal (200) photographs a urine test strip (300) stained with urine and provides it to the server (100), and the server (100) generates a test image (S101).

[0090] Next, RGB values ​​are extracted for each of the protein pad area and the creatinine pad area among the test item areas of the urine test strip (300) in the test image (S102).

[0091] Afterwards, the server (100) calculates and corrects the RGB values ​​of the protein and creatinine pad areas using specific gravity to determine the kidney health status (S103).

[0092] An embodiment of the present invention may also be implemented in the form of a recording medium containing computer-executable instructions, such as program modules, executed by a computer. Computer-readable media may be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Furthermore, computer-readable media may include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data.

[0093] Although the methods and systems of the present invention have been described with respect to specific embodiments, some or all of their components or operations may be implemented using a computer system having a general-purpose hardware architecture.

[0094] The foregoing description of the present invention is for illustrative purposes only, and those skilled in the art will readily appreciate that the present invention can be readily modified into other specific forms without altering the technical spirit or essential characteristics of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single entity may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined manner.

[0095] The scope of the present invention is indicated by the claims described below rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention.

Claims

1. A method for detecting kidney disease in a companion animal based on protein, creatinine and specific gravity, performed by a server. (a) a step of constructing an RGB ratio table and photographing a urine test strip stained with urine to generate a test image; (b) a step of extracting RGB values ​​for each of the protein pad area and the creatinine pad area corresponding to the kidney test item area among the test item areas of the urine test strip in the test image, thereby generating a first RGB value; and (c) A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, comprising: generating a first ratio value using the generated first RGB value and the specific gravity RGB value extracted from the specific gravity pad area within the test image; generating a second ratio value which is the ratio between the specific gravity of the RGB ratio table and protein and creatinine; comparing the first ratio value and the second ratio value, correcting the RGB value of the protein pad area and the RGB value of the creatinine pad area, and comparing them to determine the kidney health status of the companion animal; 2. In paragraph 1, Step (a) above, A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, wherein RGB values ​​for combinations of all colors that can be expressed by protein pad areas and creatinine pad areas according to the health status of the kidney are calculated based on a standard colorimetric table, and an RGB ratio table is constructed by calculating the ratio between each RGB value and the RGB values ​​for all colors that can be expressed by protein pad areas and creatinine pad areas.

3. In paragraph 2, The above standard colorimetric table is, A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, which defines all colors that can be expressed in the pad area corresponding to each test item.

4. In paragraph 1, Step (b) above, A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, further comprising: calculating a ratio between a first RGB value of the protein pad area and a first RGB value of the creatinine pad area, and comparing the ratio with the RGB ratio table corresponding to a preset first judgment table classified into normal, borderline, and suspicious, thereby making a first judgment on the kidney health status of the companion animal, wherein the kidney health status of the companion animal to be examined is made as at least one of normal, borderline, and suspicious.

5. In paragraph 1, Step (c) above, A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, wherein the second ratio value means a ratio between colors of pad areas of a protein pad and a creatinine pad corresponding to colors that can be expressed in a specific gravity pad area based on RGB values ​​extracted from a standard colorimetric table.

6. In paragraph 5, The above first ratio value and the second ratio value are, A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, wherein each has a preset effective range, and a numerical value for kidney health is calculated by determining whether the effective range of the first ratio value is included in the effective range of the second ratio value.

7. In paragraph 1, Step (c) above, A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, further comprising a step of making a second judgment on the kidney health of the companion animal by comparing the second ratio value and the first ratio value and corresponding each value calculated to a preset second judgment table.

8. In paragraph 1, (d) A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, further comprising: a step of making a final judgment on the kidney health status of the companion animal based on the results of the first and second judgments if the results of the first and second judgments are the same; and making a final judgment by correcting the results of the first judgment according to a preset algorithm if the results of the first and second judgments are different.

9. In paragraph 1, (e) A method for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, further comprising: a step of collecting multiple test images by time period, generating data matching the images with a standard colorimetric table and information on each ratio, learning the data as input values ​​according to a supervised learning method, and generating a machine learning model that determines, when a specific test image is input as an output value, how much time has passed since the strip was dipped in urine, and, based on the machine learning model, if there is a strip among the strips in the images taken within a preset time period that is determined to have been dipped in urine for more than 1 minute, providing a guidance message to a user terminal informing that testing on a strip that has passed a specific time period may be inaccurate; 10. In a server that detects kidney disease in pets based on protein, creatinine, and specific gravity, A memory storing a program for performing a method for detecting kidney disease in a pet based on protein, creatinine and specific gravity; and A processor for executing the above program; The above method, (a) a step of constructing an RGB ratio table and photographing a urine test strip stained with urine to generate a test image; (b) a step of extracting RGB values ​​for each of the protein pad area and the creatinine pad area corresponding to the kidney test item area among the test item areas of the urine test strip in the test image, thereby generating a first RGB value; and (c) A server for detecting kidney disease in a companion animal based on protein, creatinine, and specific gravity, comprising: a step of generating a first ratio value using the generated first RGB value and the specific gravity RGB value extracted from the specific gravity pad area within the test image, generating a second ratio value which is the ratio between the specific gravity of the RGB ratio table and protein and creatinine, comparing the first ratio value and the second ratio value to correct the RGB value of the protein pad area and the RGB value of the creatinine pad area, and comparing them to determine the kidney health status of the companion animal;

Images