CN119574844A - Urine test paper collection device and detection method thereof - Google Patents
Urine test paper collection device and detection method thereof Download PDFInfo
- Publication number
- CN119574844A CN119574844A CN202510142568.5A CN202510142568A CN119574844A CN 119574844 A CN119574844 A CN 119574844A CN 202510142568 A CN202510142568 A CN 202510142568A CN 119574844 A CN119574844 A CN 119574844A
- Authority
- CN
- China
- Prior art keywords
- test paper
- module
- sample
- image
- shaft
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/52—Use of compounds or compositions for colorimetric, spectrophotometric or fluorometric investigation, e.g. use of reagent paper and including single- and multilayer analytical elements
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Hematology (AREA)
- Immunology (AREA)
- Engineering & Computer Science (AREA)
- Urology & Nephrology (AREA)
- Molecular Biology (AREA)
- Chemical & Material Sciences (AREA)
- Biomedical Technology (AREA)
- Food Science & Technology (AREA)
- Microbiology (AREA)
- Cell Biology (AREA)
- Biotechnology (AREA)
- Medicinal Chemistry (AREA)
- Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Pathology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
本发明属于试纸检测分析技术领域，公开了一种尿液试纸采集装置及其检测方法，该装置包括：试纸出样模块、试纸运输模块、加热模块、样品移动与加样模块、图像采集模块、支撑装置和底板；试纸出样模块将试纸搓出，经由试纸运输模块将试纸运输至指定位置并由加热模块加热，样品移动与加样模块将样品移动至试纸附近，取样并注射至试纸上，图像采集模块采集试纸上的图像并上传；试纸出样模块、试纸运输模块、加热模块、图像采集模块和支撑装置均设置在底板上。本发明实现全自动尿液试纸分析，具有操作简单、小型化及便于携带等优点。采用新颖的WOA算法优化BP神经网络，通过训练可以直接投入使用，具有自适应性以及很强的鲁棒性。
The present invention belongs to the technical field of test paper detection and analysis, and discloses a urine test paper collection device and a detection method thereof, the device comprising: a test paper sample output module, a test paper transport module, a heating module, a sample movement and sample addition module, an image acquisition module, a support device and a bottom plate; the test paper sample output module rubs out the test paper, transports the test paper to a designated position via the test paper transport module and is heated by the heating module, the sample movement and sample addition module moves the sample to the vicinity of the test paper, takes a sample and injects it onto the test paper, and the image acquisition module collects the image on the test paper and uploads it; the test paper sample output module, the test paper transport module, the heating module, the image acquisition module and the support device are all arranged on the bottom plate. The present invention realizes fully automatic urine test paper analysis, and has the advantages of simple operation, miniaturization and easy carrying. The novel WOA algorithm is used to optimize the BP neural network, which can be directly put into use through training, and has adaptability and strong robustness.
Description
Technical Field
The invention relates to the technical field of test paper detection and analysis, in particular to a urine test paper acquisition device and a detection method thereof.
Background
Urine analyzer is an important tool for detecting urinary system, and is also the first choice for pathological diagnosis, examination and later review. The change of the content of each component in urine produced by human body can reflect the change of health condition. The popular detection method in the current urine detector is to detect the color recognition concentration in the test strip image by adopting a computer vision technology, and has the following problems:
1. the illumination condition, the temperature condition, the pollution and the interference of the collected test paper image environment are not well ensured;
2. the uniformity of the diffusion and reaction of urine on the test strip may affect the uniformity of the color, thereby affecting the reliability of the detection result;
3. The range of color change on the test strip may be very fine, and the method still stays at the present time in using the color value of the test reagent block image and the color value of the standard colorimetric card image to calculate the Euclidean distance, and comparing the values, the minimum value is that the concentration value on the colorimetric card is used as the concentration value of the test reagent, so that a high-precision color classification algorithm is required to accurately distinguish components with different concentrations.
Disclosure of Invention
The invention aims to provide a urine test paper collecting device and a detection method thereof, which are used for solving the problems in the prior art.
To achieve the purpose, the invention adopts the following technical scheme:
The urine test paper collecting device comprises a test paper sample discharging module, a test paper transporting module, a heating module, a sample moving and adding module, an image collecting module, a supporting device and a bottom plate, wherein the test paper sample discharging module rubs out test paper, the test paper is transported to a designated position through the test paper transporting module and heated by the heating module, the sample moving and adding module moves the sample to the vicinity of the test paper, the sample is sampled and injected onto the test paper, the image collecting module collects images on the test paper, the sample moving and adding module is partially arranged on the supporting device, and the test paper sample discharging module, the test paper transporting module, the heating module, the image collecting module and the supporting device are all arranged on the bottom plate.
The test paper sample discharging module comprises a stepping motor, a coupler, a synchronous belt wheel, a paper rubbing wheel, a roller and a test paper box, wherein a motor shaft of the stepping motor is connected with a shaft of the paper rubbing wheel through the coupler and provides power, the synchronous belt wheel transmits power to the shaft of the roller through the synchronous belt to enable the coupler and the synchronous belt wheel to synchronously rotate, and the paper rubbing wheel is arranged in the test paper box.
Preferably, the paper rubbing wheel is a cam plastic part, a rubber layer is arranged on the protruding part, a rectangular notch is formed in the tail of the test paper box, a fan-shaped notch is formed in the lower portion of the paper rubbing wheel of the test paper box, and a slope flat plate structure is designed at the front end of the test paper box.
Preferably, the test paper transportation module comprises a motor, a shaft and rubber bands, wherein the shaft is a shaft type three-dimensional printing piece with a concave shaft collar, the rubber bands are stretched and wound at two ends of the shaft to form a platform, and the motor rotates to drive the shaft to rotate so that the rubber bands rotate to drive the test paper to move.
The heating module comprises a heating platform, buckles, bevel gears, spring support columns, cams and rollers, wherein heating resistance wires are distributed in the heating platform, protruding columnar buckles are arranged on the periphery of the heating platform, a motor supplies power for shafts of the cams through the bevel gears, the cams are meshed with the rollers below the heating platform, and the spring support columns arranged on a bottom plate are used for fixedly supporting the heating platform.
The sample moving and adding module comprises a screw rod sliding table module, a sample tube, two axial moving modules and a sample adding needle, wherein the screw rod sliding table module drives the sample tube to move below a supporting device, and the two axial moving modules positioned on the supporting device drive the sample adding needle to move right above the sample tube.
Preferably, the image acquisition module comprises an image acquisition box, a CCD camera and an upper computer, wherein the CCD camera is positioned in the image acquisition box and is used for acquiring images of test paper, and acquired data are transmitted to the upper computer for processing.
A method for detecting a urine test paper collection device, the method comprising:
The test paper strip is placed in the test paper box, a sample tube filled with a sample is placed on the sample moving and adding module, the test paper sample outputting module rubs out a test paper and moves the test paper to the test paper transporting module, and the test paper outputting module moves a designated position and is heated by the heating module;
The test paper transport module drives the test paper to move to the image acquisition module for image acquisition when the test paper strip is completely reacted;
And thirdly, the image acquisition module transmits the acquired image to an upper computer, and the upper computer performs image processing, color value extraction, color correction and concentration calculation to realize detection of the test strip.
Preferably, the image processing in the third step comprises the steps of removing Gaussian noise of the image, removing high light of the image by a weighted superposition method, separating a background area and a foreground area of the image by an Otsu method self-adaptive threshold value calculation, filling up a cavity in the image by morphological opening and closing operation, carrying out image edge extraction by a Canny operator algorithm, determining the center position of a reagent block on test paper through the edge, carrying out rectangular image extraction at the center position, finally carrying out thermodynamic diagram ROI extraction on the reagent block, obtaining a final data image, carrying out color value extraction on the image, carrying out color value extraction on an RGB chromaticity space, finally converting the extracted RGB color value into a CIELab color space, carrying out color correction of a polynomial, carrying out concentration calculation on a WOA algorithm optimization training BP neural network, obtaining an optimal network model, and inputting the RGB chromaticity space color value as an input quantity into the WOA-BP neural network, thus obtaining a concentration result.
Preferably, the training method of the WOA-BP neural network comprises a whale optimization algorithm.
The invention has the beneficial effects that a plurality of modules are designed in a mechanical mode, the modules are mutually matched, the dripping and data acquisition of urine samples are fully and automatically completed, the functions of preventing the interference and pollution of external environments are realized, meanwhile, the invention has a heating function, the stability of the samples in the reaction with the reagents is ensured, and the influence of the color temperature in the process of acquiring images is reduced. The invention can realize full-automatic urine test paper analysis by only putting the test paper strip and the urine sample, and has the advantages of simple operation, miniaturization, portability and the like. Meanwhile, a novel WOA algorithm is adopted to optimize the BP neural network in data analysis, and the BP neural network can be directly put into use after network training.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following description will briefly explain the drawings needed in the description of the embodiments of the present invention, and it should be apparent that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to the contents of the embodiments of the present invention and these drawings without inventive effort for those skilled in the art.
FIG. 1 is an optical-mechanical isometric view of a urine test paper collection device of the present invention;
FIG. 2 is an isometric view of a test strip sample ejection module of a urine test strip collection apparatus of the present invention;
FIG. 3 is a rear view of a test strip transport module and a heating module of a urine test strip collection apparatus of the present invention;
FIG. 4 is a block diagram of an image processing program of a host computer according to the present invention;
FIG. 5 is a diagram of the image processing effect of the present invention;
FIG. 6 is a graph of the color correction effect of the present invention;
FIG. 7 is a graph of the results of semi-quantitative assays performed in accordance with the present invention;
FIG. 8 is a block diagram of a training program for the WOA-BP neural network of the host computer of the invention;
FIG. 9 is a graph of fitness versus WOA-BP neural network training two test items;
FIG. 10 is a graph comparing effects of different algorithms to optimize BP neural network;
Fig. 11 is a graph of real results versus network predicted results.
In the figure, 1, a test paper sample discharging module, 101, a stepping motor, 102, a coupler, 103, a synchronous belt, 104, a synchronous belt pulley, 105, a bearing seat, 106, a paper rubbing wheel, 107, a roller, 108, a test paper box, 2, a test paper conveying module, 201, a shaft, 202, a rubber band, 203, a test paper strip, 3, a heating module, 301, a heating platform, 302, a buckle, 303, a bevel gear, 304, a spring supporting column, 305, a cam, 306, a roller, 4, a sample moving and sample adding module, 401, a sample frame, 402, a sample tube, 403, a sample adding needle, 404, a screw sliding table module, 405, a two-axial moving module, 5, an image collecting module, 501, a CCD camera, 502, an image collecting box, 6, a cleaning tank, 7, an injection pump, 8, a bottom plate, 9 and a supporting device.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.
In the description of the present invention, unless explicitly stated or limited otherwise, the terms "connected," "connected," and "fixed" are to be construed broadly, and may, for example, be fixedly connected, detachably connected, or integrally formed, mechanically connected, electrically connected, directly connected, indirectly connected through an intervening medium, or in communication between two elements or in an interaction relationship between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
In the present invention, unless expressly stated or limited otherwise, a first feature "above" or "below" a second feature may include both the first and second features being in direct contact, as well as the first and second features not being in direct contact but being in contact with each other through additional features therebetween. Moreover, a first feature being "above," "over" and "on" a second feature includes the first feature being directly above and obliquely above the second feature, or simply indicating that the first feature is higher in level than the second feature. The first feature being "under", "below" and "beneath" the second feature includes the first feature being directly under and obliquely below the second feature, or simply means that the first feature is less level than the second feature.
In the description of the present embodiment, the terms "upper", "lower", "left", "right", and the like are orientation or positional relationships based on those shown in the drawings, merely for convenience of description and simplicity of operation, and do not indicate or imply that the apparatus or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the invention. Furthermore, the terms "first," "second," and the like, are used merely for distinguishing between descriptions and not for distinguishing between them.
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative and intended to explain the present invention and should not be construed as limiting the invention.
As shown in FIG. 1, the urine test paper collecting device comprises a test paper sample discharging module 1, a test paper transporting module 2, a heating module 3, a sample moving and adding module 4, an image collecting module 5, a bottom plate 8 and a supporting device 9, wherein the test paper sample discharging module 1 rubs out a test paper strip 203, the test paper strip 203 is transported to a designated position through the test paper transporting module 2 and heated by the heating module 3, the sample moving and adding module 4 moves the sample to the vicinity of the test paper strip 203, the sample is sampled and injected onto the test paper strip 203, the image collecting module 5 collects an image on the test paper strip 203 and uploads the image to an upper computer, the sample moving and adding module 4 is arranged on the supporting device 9, and the test paper sample discharging module 1, the test paper transporting module 2, the heating module 3, the image collecting module 5 and the supporting device 9 are all arranged on the bottom plate 8. All parts are arranged in the shell, so that all works are guaranteed to be completed in a closed environment.
As shown in fig. 2, the test paper sample discharging module 1 comprises a stepping motor 101, a coupler 102, a synchronous belt 103, a synchronous belt pulley 104, a bearing seat 105, a paper rubbing wheel 106, a roller 107 and a test paper box 108, wherein the stepping motor 101 is matched with the synchronous belt pulley 104 and the paper rubbing wheel 106 through the coupler 102 and the bearing seat 105, meanwhile, the synchronous belt pulley 104 is matched with the roller 107 through the synchronous belt 103 and the bearing seat 105, a motor shaft of the stepping motor 101 is connected with a shaft of the paper rubbing wheel 106 through the coupler 102 and provides power, the synchronous belt pulley 104 transmits power to the shaft of the roller 107 through the synchronous belt 103 so that the coupler 102 and the synchronous belt pulley 104 synchronously rotate, and the paper rubbing wheel 106 is arranged in the test paper box 108.
The paper rubbing wheel 106 is a cam plastic part, a rubber layer is arranged on the protruding part, a rectangular notch is formed in the tail of the test paper box 108, a fan-shaped notch is formed in the lower portion of the paper rubbing wheel 106 of the test paper box 108, and a slope flat plate structure is designed at the front end of the test paper box.
As shown in fig. 3, the test paper transportation module 2 comprises a motor, a shaft 201 and a rubber band 202, wherein the shaft 201 is a shaft type three-dimensional printing piece with a concave shaft collar integrated, the rubber band 202 is stretched and wound at two ends of the shaft 201 to form a platform, and the motor rotates to drive the shaft 201 to rotate, so that the rubber band 202 rotates to drive the test paper strip 203 to move.
The heating module 3 comprises a heating platform 301, a buckle 302, a bevel gear 303, a spring support column 304, a cam 305 and a roller 306, wherein heating resistance wires are distributed in the heating platform 301, and the temperature stability in the environment can be ensured. The periphery of the heating platform 301 is provided with a convex columnar buckle 302, a motor supplies power to the shaft of the cam 305 through a bevel gear 303, the cam 305 is meshed with a roller 306 below the heating platform 301, and a spring support column 304 arranged on the bottom plate 8 is used for fixedly supporting the heating platform 301.
The sample moving and sample adding module 4 comprises a screw rod sliding table module 404, a sample tube 402, two axial moving modules 405 and a sample adding needle 403, wherein the screw rod sliding table module 404 drives the sample tube 402 to move below the supporting device 9, and the two axial moving modules 405 positioned on the supporting device 9 drive the sample adding needle 403 to move to the position right above the sample tube 402, so that the position of the sample adding needle 403 is rapidly and accurately controlled.
The image acquisition module comprises an image acquisition box 502, a CCD camera 501 and an upper computer, wherein the CCD camera 501 is positioned in the image acquisition box 502 and is used for acquiring images of the test strip 203, and acquired data are transmitted to the upper computer for processing.
A method for detecting a urine test paper collection device, the method comprising:
The test paper 203 is put into a test paper box 108, a sample tube 402 filled with a sample is put onto a sample moving and adding module 4, a paper rubbing wheel 106 is driven by a stepping motor 101 to rotate, a convex part of the paper rubbing wheel 106 rotates until a sector notch of the test paper box 108 is tightly pressed against all test papers, one test paper 203 is continuously rotated and taken out, the test paper 203 is pushed to a flat plate with a slope and a rubber layer of the test paper box 108 by the paper rubbing wheel 106 to be separated from the test paper below the flat plate, meanwhile, a synchronous belt 103 drives a synchronous belt wheel 104 to rotate, a roller 107 is driven to rotate, the test paper 203 is provided with subsequent power, the sample outputting of the test paper 203 is completed, the test paper is moved onto a test paper transporting module 2, the test paper is placed onto a platform composed of a plurality of rubber strips 202, a specified position is heated by a heating module 3, the motor and a bevel gear 303 are matched and are installed, the two bevel gears are meshed with each other, the rotating force from the motor to the cam 305 is transmitted, the cam 305 is contacted with a roller 306 under the heating platform 301, the cam 305 rotates, the heating platform 301 is blocked by the movement of the test paper 203 by the buckle 302 on the heating platform 301, thus the test paper 203 is positioned, the test paper 203 is continuously heated, and the test paper 203 is continuously heated, and the test paper is heated by the heating platform 203.
And secondly, the screw rod sliding table module 404 of the sample moving and adding module 4 drives the screw rod to rotate through the rotation of the motor, so that the sliding table moves, the sample frame 401 is driven to move, and the central axis of the sample tube 402 positioned on the sample frame 401 and the central axis of the sample adding needle 403 are positioned on the same straight line, so that the sample adding needle 403 is convenient for sampling. The sample adding needle 403 is matched with the injection pump 7 to complete sampling and discharging, in this embodiment, three sections of gas-liquid columns are formed in the sample adding needle 403, the uppermost column is a system liquid (normal saline) column, the middle column is an air column, the lowermost column is a urine sample column, and the discharging is to control the amount of the discharged system liquid through the injection pump 7, so as to control the discharge of the urine sample amount. During sample discharging, the two axial moving modules 405 drive the sample adding needle 403 to a position set by the system, and control the sample adding needle 403 to descend to a certain height to discharge samples, and the two axial moving modules 405 start stepping operation according to program setting to drop samples onto the reagent blocks of the test strip 203 one by one. After the sample addition is completed, the sample addition needle 403 is moved to the position of the cleaning pool 6, system liquid is pumped into the cleaning pool 6, the head of the sample addition needle is inserted into the cleaning pool 6 during cleaning, the infiltration cleaning is repeatedly moved, the outer wall of the sample addition needle 403 is completed, then the system liquid is discharged from the inner wall of the sample addition needle 403, the inner wall of the sample addition needle 403 is completed to clean, finally, the waste liquid in the cleaning pool 6 is discharged, new system liquid is pumped, and the next cleaning is waited. When the test paper 203 is completely reacted, the test paper transportation module 2 drives the test paper 203 to move to the image acquisition module 5 for image acquisition.
And thirdly, the CCD camera 501 in the image acquisition module 5 acquires images in the image acquisition box 502, acquired data are transmitted to an upper computer for processing, and image processing, color value extraction and concentration calculation are carried out in the upper computer to realize detection of the test strip 203.
The image processing in the second step comprises the following steps:
As shown in fig. 4, consider that in the image acquisition process, the camera introduces certain noise due to factors such as devices, and the image filtering work is completed by using a gaussian smoothing filter based on gaussian noise. The weighted overlap-add method removes image highlights, and the highlight region in the image usually causes local pixel values to be too high, thereby causing interference to binarization, segmentation and other operations. By adjusting the weights of the original image and the smoothed image, the influence of the highlight region can be weakened, and the brightness of the image is more uniform. Details in the original image are enhanced, and the brightness of the highlight region is suppressed, thereby improving the overall contrast. The difference between the original image and the blurred image can effectively keep edge details, and meanwhile, the brightness is balanced, so that the method is suitable for subsequent processing. The specific mathematical formula is as follows:
Wherein I output (x, y) is the value of the pixel (x, y) in the output image, I image (x, y) and I blurred (x, y) are the values of the corresponding pixels of the original image and the blurred image, respectively, α and β are weights for controlling the contribution of the two images to the result, and γ is a constant offset for overall brightness adjustment. In this embodiment, α is 1.5, the weight for enhancing the original image brightness, β is-0.5, the weight for suppressing the blurred image brightness, and the highlight region is reduced, γ is 0, and the overall brightness is not adjusted.
And the threshold segmentation is to automatically acquire an image threshold by adopting an Ojin method, and the background area and the foreground area are separated by using the Ojin method to perform self-adaptive threshold calculation.
Morphological processing, namely firstly performing open operation on an image to eliminate some interference color blocks with smaller pixels, and then filling a target color block area by using closed operation, wherein the main purpose of the morphological processing is to divide independent elements of the image and reduce the interference of small areas in the image. After the image operation, the edge of the image of the urine dry chemical analysis test paper is obvious and easy to position, so that the edge detection of a Canny operator is directly adopted to carry out edge extraction on the image, after the edge is obtained, the center of the image is determined according to the edge, a rectangular image with a certain pixel size is extracted at the center position, and then the image of each detection item can be obtained.
Because the reaction process cannot ensure that each region of the test strip 203 completely reacts with the sample, and there are different colors of pixel points in a plurality of regions caused by various factors, the region with the deepest color can be found out through a thermodynamic diagram algorithm, the region is the region with the best reaction, and a rectangular region of the part is extracted through an ROI algorithm to obtain a color block diagram of each detection item.
And finally extracting color values, traversing each detection item color block diagram, and calculating the average RGB value of pixels in the region, as shown in figure 5.
As shown in fig. 6, for the extracted color patch, corresponding color correction is required, where the color correction belongs to device color correction, and the color deviation shot by the device is corrected by fitting a mapping relationship between the color value shot by the device and the known color value of the standard color card, and creating a polynomial regression model. First, it is necessary to capture a standard color chart in a fixed illumination environment, where the standard color chart is an international standard color chart, and convert all the obtained data from the RGB color space to the CIELab color space. The device collects Lab values of color patches, known Lab values of standard color patches, and corresponding color patch images before and after polynomial nonlinear correction. A polynomial fitting model is built for each color channel (L, a, b):
Where x s is the standard value for the color channel, a 0、a1、...、an is the fitting polynomial coefficient, x m is the measurement of the color channel, n is the polynomial order, and given the standard color card dataset, the polynomial coefficient for each channel, a i, is fitted using the least squares method:
Where x i is the device measurement and m is the standard color chip sample number. And finally, calculating to obtain corresponding fitting coefficients, obtaining a final polynomial model result, carrying out model fitting on all three color channels to obtain three color correction models, bringing the L, a and b three channels of the color block to be measured into the three correction models to obtain new L, a and b three values, and finally converting back to RGB color space to finish correction.
Before converting RGB color space, semi-quantitative detection can be carried out on the color block to be detected in CIELab color space and the test paper standard colorimetric card, namely Euclidean distance color difference discrimination is carried out, and the smallest color difference item between the color block and the colorimetric card is the corresponding colorimetric result. Here, a negative urine quality control liquid sample is prepared according to a standard formula of the negative urine quality control liquid, mainly 10g of urea, 5g of sodium chloride, 0.1g of creatinine, 1g of potassium chloride and 20mmol of Phosphate Buffer (PBS), the pH value is adjusted to about 6.0, and purified water is added to a constant volume value of 1000ml. For testing the reliability of the instrument, a total of 360 sets of data were tested by testing 30 negative samples, and the experimental results obtained by running are shown in fig. 7. In the figure, the vertical axis is the CIELab color difference value of the instrument test result, the horizontal axis is each detection item, the detection threshold is set to be 10 through experiments, and when the CIELab color difference value of the detection result and the standard result is greater than 10, the experimental result is determined to be inaccurate. 360 data in the graph, which are matched with the standard result, have the accuracy rate up to 99.4%.
The concentration calculation is to optimally train a BP (Back propagation) neural network through a WOA algorithm to obtain an optimal network model, and input RGB (red, green and blue) chromaticity space color values into the WOA-BP neural network as input quantities to obtain a concentration result.
The training method of the WOA-BP neural network is a whale optimization algorithm. The whale optimization algorithm (Whale Optimization Algorithm, WOA) is an optimization algorithm based on whale behaviors, and the bubble net feeding method of the whale in the head is a core inspiration source of the whale optimization algorithm. The whale forms a net by surrounding the prey and spitting out air bubbles, and then gathers the prey together for easier capture. WOA searches and optimizes the solution space by simulating such predation behavior, with each whale's location representing a potential solution, and the whale's location is updated continuously in the solution space to obtain a globally optimal solution. The predation behavior is classified into three, surrounding prey, bubble net predation, searching prey. The surrounding behavior is by calculating the distance between the whale and the game (the current best position) and adjusting the whale's position based on this distance. The specific flow of the WOA hybrid algorithm optimization BP neural network is shown in figure 8, wherein 1) data normalization preprocessing, 2) calculation is carried out by an exhaustion method to obtain an optimal network topological structure, 3) BP neural network initialization is carried out to determine an input and output structure, initial connection weight and threshold value of the network, 4) Latin hypercube (Latin Hypercube Sampling, LHS) is adopted to initialize the population of WOA, latin hypercube sampling can ensure uniform sample distribution in each dimension, the diversity of the population is enhanced, LHS can more effectively utilize search space compared with random initialization, unnecessary calculation cost is reduced, convergence speed is accelerated, 5) WOA optimization is carried out by taking training errors of the BP neural network as fitness values and training weights and threshold values as population individuals, phase optimal population positions and phase optimal fitness 6) iterative updating is carried out by repeating the optimization and updating processes until the main cycle is ended, 7) optimal network weight and threshold value are obtained, and 9) network training is carried out. The model obtained by training can be directly used as a mathematical model, each examination item needs to train a corresponding mathematical model, and when the model is used, only the data to be measured is input into the model corresponding to the data, so that a corresponding concentration value can be obtained.
The WOA-BP neural network is trained, meanwhile, the particle swarm algorithm PSO and the genetic algorithm GA are adopted to optimize the BP neural network, the training effect of the particle swarm algorithm PSO and the genetic algorithm GA is compared with that of the WOA-BP neural network, and as shown in figure 9, the fitness calculation iteration curve of the detection protein item and the leucocyte item shows that all three optimization algorithms can obviously optimize fitness, the performance of a model is improved, and particularly the WOA-BP algorithm shows the fastest convergence speed and the lowest final fitness value in the iteration process. The BP neural network optimized by the three algorithms is obviously reduced in Mae and Rmse on the test set, and compared with the GA and PSO algorithms, the WOA algorithm is best in three indexes of Mae, rmse and fitness, so that the convergence speed is highest in the optimization process, the prediction precision is highest, and the prediction effect is best.
As can be seen from the results of predicting the concentration of protein and white blood cells in urine by each algorithm in fig. 10, the predicted values of the four algorithms are very close to the true values, and in particular, the three optimization algorithms, i.e., PSO-BP, GA-BP and WOA-BP, have their predicted values almost coincident with the true values, indicating that they are excellent in prediction accuracy. The predicted value of the standard BP algorithm is closer to the true value, but has a certain deviation on part of samples. PSO-BP prediction error is small and stable, with relative error mostly between-10% and 10%. The prediction error of the GA-BP is also small for most samples, but the error is large on individual samples, approaching-60%. The overall prediction error of WOA-BP is minimal and stable, with relative errors mostly concentrated between-5% and 5%.
The results of the predicted proteins were validated and compared with the results of manual testing as shown in FIG. 11. To prepare a 500mg/dL solution of Bovine Serum Albumin (BSA), 0.5g of BSA powder was first weighed using a precision electronic balance and placed into a clean, sterile 50mL centrifuge tube. 30mL of sterile 1 XPBS buffer was slowly added thereto and gently shaken to initially dissolve the powder, after which the tube was slowly inverted upside down to mix the solution until the powder was completely dissolved, and no air bubbles were generated during the whole process. The fully dissolved solution was transferred to a 100mL volumetric flask and sterile 1 x PBS buffer was added to the scale line to volume. Finally, the volumetric flask was slowly inverted several times to ensure uniform mixing of the solutions. The prepared solution was stored in a 4 ℃ refrigerator. The remaining concentration gradients were prepared as described above. The detection of the protein in the urine is realized based on the principle of a protein error method of a PH indicator, and a reagent block for detecting the protein contains an indicator tetrabromophenol blue and a buffer solution, and positive charges of the protein can be combined with anions of the tetrabromophenol blue, so that the PH change generates a color change from yellow green to blue green. At low concentration, the test paper is yellow or light green, and when the concentration is increased, the color is gradually deepened to be blue-green. And 20 mu L of each gradient is dripped on the reagent block, after 30s of reaction, the reagent block is placed in a data acquisition system, corresponding RGB values are read, data acquired by each gradient are acquired by ten times of acquisition and averaging, the RGB values are corrected and then taken as input features to enter a neural network, and corresponding concentration results are obtained through prediction.
It is to be understood that the above examples of the present invention are provided for clarity of illustration only and are not limiting of the embodiments of the present invention. Various obvious changes, rearrangements and substitutions can be made by those skilled in the art without departing from the scope of the invention. It is not necessary here nor is it exhaustive of all embodiments. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are desired to be protected by the following claims.
Claims (10)
1. A urine test paper collecting device is characterized by comprising a test paper sample discharging module, a test paper transporting module, a heating module, a sample moving and adding module, an image collecting module, a supporting device and a bottom plate, wherein the test paper sample discharging module rubs out test paper, the test paper is transported to a designated position through the test paper transporting module and heated by the heating module, the sample moving and adding module moves the sample to the vicinity of the test paper, the sample is sampled and injected onto the test paper, the image collecting module collects images on the test paper, the sample moving and adding module is partially arranged on the supporting device, and the test paper sample discharging module, the test paper transporting module, the heating module, the image collecting module and the supporting device are all arranged on the bottom plate.
2. The urine test paper collecting device according to claim 1, wherein the test paper sampling module comprises a stepping motor, a coupler, a synchronous belt pulley, a paper rubbing wheel, a roller and a test paper box, a motor shaft of the stepping motor is connected with a shaft of the paper rubbing wheel through the coupler and provides power, the synchronous belt pulley transmits power to the shaft of the roller through the synchronous belt to enable the coupler and the synchronous belt pulley to synchronously rotate, and the paper rubbing wheel is arranged in the test paper box.
3. The urine test paper collecting device according to claim 2, wherein the paper rubbing wheel is a cam plastic part, a rubber layer is arranged on the protruding portion, a rectangular notch is formed in the tail portion of the test paper box, a fan-shaped notch is formed in the lower portion of the paper rubbing wheel of the test paper box, and a slope flat plate structure is designed at the front end of the test paper box.
4. The urine test paper collecting device according to claim 1, wherein the test paper transporting module comprises a motor, a shaft and a rubber band, the shaft is a shaft type three-dimensional printing piece with a concave shaft collar, the rubber band is stretched and wound at two ends of the shaft to form a platform, and the motor rotates to drive the shaft to rotate, so that the rubber band rotates to drive the test paper to move.
5. The urine test paper collecting device is characterized in that the heating module comprises a heating platform, buckles, bevel gears, spring support columns, cams and rollers, heating resistance wires are distributed in the heating platform, protruding columnar buckles are arranged on the periphery of the heating platform, a motor supplies power for a shaft of the cams through the bevel gears, the cams are meshed with the rollers below the heating platform, and the spring support columns arranged on a bottom plate are used for fixedly supporting the heating platform.
6. The urine test paper collecting device according to claim 1, wherein the sample moving and adding module comprises a screw rod sliding table module, a sample tube, two axial moving modules and a sample adding needle, wherein the screw rod sliding table module drives the sample tube to move below a supporting device, and the two axial moving modules located on the supporting device drive the sample adding needle to move right above the sample tube.
7. The urine test paper collecting device according to claim 1, wherein the image collecting module comprises an image collecting box, a CCD camera and an upper computer, wherein the CCD camera is located in the image collecting box and used for collecting images of test papers, and data obtained through collection are transmitted to the upper computer for processing.
8. A method for detecting a urine test paper collection device according to any one of claims 1 to 7, comprising:
The test paper strip is placed in the test paper box, a sample tube filled with a sample is placed on the sample moving and adding module, the test paper sample outputting module rubs out a test paper and moves the test paper to the test paper transporting module, and the test paper outputting module moves a designated position and is heated by the heating module;
The test paper transport module drives the test paper to move to the image acquisition module for image acquisition when the test paper strip is completely reacted;
And thirdly, the image acquisition module transmits the acquired image to an upper computer, and the upper computer performs image processing, color value extraction, color correction and concentration calculation to realize detection of the test strip.
9. The detection method according to claim 8, wherein the step of image processing includes the steps of removing Gaussian noise of the image, removing high light of the image by a weighted superposition method, separating a background area and a foreground area of the image by an Otsu self-adaptive threshold calculation, filling a cavity in the image by morphological opening and closing operation, performing image edge extraction by a Canny operator algorithm, determining a center position of a reagent block on test paper through the edge, performing rectangular image extraction at the center position, and finally performing thermodynamic ROI extraction on an optimal reaction area to obtain a final data image, wherein the color value extraction is performed on the image by performing color value extraction of RGB chromaticity space, and finally converting the extracted RGB color value into CIELab color space to perform polynomial color correction, and the concentration calculation is performed by optimizing a training BP neural network by a WOA algorithm to obtain an optimal network model, and inputting the RGB color value as an input quantity into the WOA-BP neural network to obtain a concentration result.
10. The method of claim 9, wherein the training method of the WOA-BP neural network comprises a whale optimization algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510142568.5A CN119574844A (en) | 2025-02-10 | 2025-02-10 | Urine test paper collection device and detection method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510142568.5A CN119574844A (en) | 2025-02-10 | 2025-02-10 | Urine test paper collection device and detection method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN119574844A true CN119574844A (en) | 2025-03-07 |
Family
ID=94796676
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510142568.5A Pending CN119574844A (en) | 2025-02-10 | 2025-02-10 | Urine test paper collection device and detection method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119574844A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201434869Y (en) * | 2009-03-10 | 2010-03-31 | 杭州健宝医疗器械有限公司 | Body-fluid leucorrhea dry-chemistry auto-analyzer |
| CN111069091A (en) * | 2019-12-14 | 2020-04-28 | 西安交通大学 | An intelligent test strip quality tester based on computer vision detection |
| CN212111456U (en) * | 2020-03-06 | 2020-12-08 | 深圳市瑞图生物技术有限公司 | Full-automatic urine analysis system |
| WO2022052364A1 (en) * | 2020-09-09 | 2022-03-17 | 广州万孚生物技术股份有限公司 | Biochemical analyzer |
| CN216484762U (en) * | 2021-11-30 | 2022-05-10 | 成都斯马特科技有限公司 | Test card top heating device for blood gas analyzer |
| CN114486895A (en) * | 2022-01-21 | 2022-05-13 | 长沙协大生物科技有限公司 | Method for detecting sample substance concentration based on pushing of urine drying test strip to identification |
| CN114755441A (en) * | 2022-05-26 | 2022-07-15 | 深圳市凯特生物医疗电子科技有限公司 | Test paper pusher and applied device's automatic urine analysis appearance |
| CN219245553U (en) * | 2022-12-07 | 2023-06-23 | 四川新健康成生物股份有限公司 | Full-automatic dry-type fluorescence immunoassay appearance |
| CN117849043A (en) * | 2024-03-08 | 2024-04-09 | 长春理工大学 | Urine test paper analysis device and detection method thereof |
-
2025
- 2025-02-10 CN CN202510142568.5A patent/CN119574844A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201434869Y (en) * | 2009-03-10 | 2010-03-31 | 杭州健宝医疗器械有限公司 | Body-fluid leucorrhea dry-chemistry auto-analyzer |
| CN111069091A (en) * | 2019-12-14 | 2020-04-28 | 西安交通大学 | An intelligent test strip quality tester based on computer vision detection |
| CN212111456U (en) * | 2020-03-06 | 2020-12-08 | 深圳市瑞图生物技术有限公司 | Full-automatic urine analysis system |
| WO2022052364A1 (en) * | 2020-09-09 | 2022-03-17 | 广州万孚生物技术股份有限公司 | Biochemical analyzer |
| CN216484762U (en) * | 2021-11-30 | 2022-05-10 | 成都斯马特科技有限公司 | Test card top heating device for blood gas analyzer |
| CN114486895A (en) * | 2022-01-21 | 2022-05-13 | 长沙协大生物科技有限公司 | Method for detecting sample substance concentration based on pushing of urine drying test strip to identification |
| CN114755441A (en) * | 2022-05-26 | 2022-07-15 | 深圳市凯特生物医疗电子科技有限公司 | Test paper pusher and applied device's automatic urine analysis appearance |
| CN219245553U (en) * | 2022-12-07 | 2023-06-23 | 四川新健康成生物股份有限公司 | Full-automatic dry-type fluorescence immunoassay appearance |
| CN117849043A (en) * | 2024-03-08 | 2024-04-09 | 长春理工大学 | Urine test paper analysis device and detection method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 黄建邦等: "基于鲸鱼算法优化的 BP 神经网络的尿液成分浓度预测方法", 《长春理工大学学报（自然科学版）》, vol. 47, no. 5, 31 October 2024 (2024-10-31), pages 89 - 98 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112074725B (en) | Detection test paper reader system based on accurate colorimetry | |
| US8922761B2 (en) | Measuring volume and constituents of cells | |
| CN103827658B (en) | Identify and measure desmacyte | |
| WO2015022318A1 (en) | Test method for determining biomarkers | |
| CN109061131A (en) | Dye picture processing method and processing device | |
| CN208441190U (en) | The intelligent closestool of automatic urine examination | |
| US20240371141A1 (en) | Methods and apparatus providing training updates in automated diagnostic systems | |
| CN105223196A (en) | A kind of urinalysis method based on intelligent terminal | |
| Zhao et al. | A customizable automated container-free multi-strip detection and line recognition system for colorimetric analysis with lateral flow immunoassay for lean meat powder based on machine vision and smartphone | |
| US12579432B2 (en) | Methods and apparatus for automated specimen characterization using diagnostic analysis system with continuous performance based training | |
| CN114486895A (en) | Method for detecting sample substance concentration based on pushing of urine drying test strip to identification | |
| JP7273631B2 (en) | Specimen property identification device, specimen property identification method, and specimen transport system | |
| JP7724356B2 (en) | Site-specific adaptation of automated diagnostic analysis systems. | |
| CN118130397A (en) | Detection method of tea pigment types and contents in instant tea powder based on computer vision | |
| CN117288751A (en) | Household novel intelligent urine detection method and device | |
| CN112200803B (en) | Sperm nucleoprotein maturity detection method and device | |
| Weitz et al. | Prediction of Ki67 scores from H&E stained breast cancer sections using convolutional neural networks | |
| Yudhana et al. | Optimizing urine protein detection accuracy using the K-nearest neighbors algorithm and advanced image segmentation techniques | |
| CN213337356U (en) | Water body hexavalent chromium detection device based on machine vision | |
| CN121186026A (en) | A method for detecting residual protein in lactose excipients based on machine learning-assisted visualization chips | |
| JP7458481B2 (en) | Method and apparatus for hashing and retrieving training images used for HILN determination of specimens in automated diagnostic analysis systems | |
| Rocha et al. | An Edge-Based Computer Vision | |
| CN120177328A (en) | Microfluidic chip, intelligent staining detection device and intelligent staining method for detecting urine sediment | |
| HK40105569A (en) | Methods and apparatus providing training updates in automated diagnostic systems | |
| CN117470785A (en) | Multi-index rapid detection method of Ganoderma lucidum spore powder based on hyperspectral technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20250307 |
|
| WD01 | Invention patent application deemed withdrawn after publication |