WO2021102741A1

WO2021102741A1 - Image analysis method and system for immunochromatographic detection

Info

Publication number: WO2021102741A1
Application number: PCT/CN2019/121284
Authority: WO
Inventors: 唐奇琛; 朱锋; 曾威雄; 黄健
Original assignee: 深圳加美生物有限公司
Priority date: 2019-11-27
Filing date: 2019-11-27
Publication date: 2021-06-03
Also published as: CN111033563A

Abstract

The present invention relates to an image analysis method for immunochromatographic detection, comprising the following steps: establishing a standard concentration corresponding to a test item; acquiring a chromatographic image of an assay sample on a reagent card, the chromatographic image comprising a quality control strip and at least one test strip; performing image pre-processing on the chromatographic image to obtain a pre-processed chromatographic image; amplifying the pre-processed chromatographic image to obtain a chromatographic image matrix, the chromatographic image matrix comprising a quality control strip matrix and a test strip matrix; on the basis of the greyscale value ratio of the test strip matrix and the quality control strip matrix, determining the test concentration of the assay sample; and comparing the test concentration with the standard concentration to determine the assay result of the assay sample.

Description

Image analysis method and system for immunochromatographic detection

Technical field

This application relates to the technical field of reagent detection, and in particular to an image analysis method and system for immunochromatographic detection.

Background technique

In in vitro diagnostic point-of-care testing (POCT), the reagent card is used to carry the test sample as a whole blood quantitative detection immunochromatographic device, which can effectively filter whole blood and quickly separate red blood cells and plasma.

Existing timely detection, after the test sample on the reagent card fully reacts with the reagent during incubation, the quality control strip and the test strip will be displayed in the detection window of the reagent card, and the result is generally judged by visual judgment to compare quality control The difference and change of the strips and the test strips can be used to obtain the test results.

technical problem

However, this kind of visual inspection result judgment method has a large judgment error, and manual judgment makes the detection process longer, time-consuming and low detection efficiency.

Therefore, the existing immunochromatographic detection and identification technology needs to be improved.

Technical solutions

In view of the above existing technical problems, the present application provides an image recognition reagent card to determine an accurate and efficient image analysis method and system for immunochromatographic detection.

In the first aspect, the technical solution provided by the embodiments of the present application is to provide an image analysis method for immunochromatographic detection, which includes the following steps:

Establish the standard concentration corresponding to the test item;

Acquiring a tomographic image of the test sample on the reagent card, the tomographic image including a quality control strip and at least one test strip;

Perform image preprocessing on the tomographic image to obtain a preprocessed tomographic image;

Enlarging the preprocessed tomographic image to obtain a tomographic image matrix, the tomographic image matrix including a quality control strip matrix and a test strip matrix;

Determine the test concentration of the test sample according to the ratio of the gray value of the test strip matrix to the quality control strip matrix;

Compare the test concentration with the standard concentration to determine the test result of the test sample.

Preferably, in order to determine the edge position information of the graphics, the image analysis method of immunochromatographic detection, while performing the step of establishing the standard concentration corresponding to the test item, further includes an edge positioning step, and the edge positioning step includes:

Establish an edge matrix data model;

The edge matrix data model is used to scan the tomographic image matrix to obtain the edge position coordinates of the detection window image of the reagent card in the tomographic image matrix.

Wherein, the step of performing image preprocessing on the tomographic image includes:

Performing cropping processing on the tomographic image to obtain a cropped tomographic image containing the detection window of the reagent card;

Image edge detection and image edge enhancement processing are performed on the cropped tomographic image based on Sobel operator.

The image analysis method for immunochromatographic detection further includes a background abnormality processing step on the tomographic image matrix, and the background abnormality processing step includes:

Establish a data model for the difference of foreign body shape or color change;

Scan the tomographic image matrix according to the edge position coordinates and the foreign body shape or color change difference data model, and judge the abnormal value in the tomographic image matrix row by row;

Mark or restore the abnormal value.

Further, the foreign body shape or color change difference data model includes an average threshold, a maximum threshold, a minimum threshold, and a standard square deviation threshold.

Wherein, the reagent card includes a main body, a hand-held part, a sample adding window, a detection window and a labeling part arranged on the main body, and an identification code is affixed on the labeling part.

The image analysis method for immunochromatographic detection further includes a reagent card image detection step before the step of obtaining the tomographic image of the test sample on the reagent card, and the reagent card image detection step includes:

Establish a database of image characteristics of reagent cards, including handheld image database and labeling department database;

Acquiring a current reagent card image, the current reagent card image including an image of the hand-held part and an image of the labeling part;

Identify the reagent card information and item detection information based on the image of the labeling part;

Based on the handheld image, the KNN nearest neighbor classification algorithm is used to compare the handheld image and the handheld image database to determine the correctness of the insertion direction and orientation of the reagent card.

Wherein, the step of determining the test concentration of the test sample according to the gray value ratio of the test strip matrix and the quality control strip matrix includes:

Set the first standard formula and the second standard formula;

A fluctuation threshold is set, and the first standard formula or the second standard formula is selected through the fluctuation threshold to calculate the test concentration.

In the second aspect, the technical solution provided by the embodiments of the present application is to provide an image analysis system for immunochromatographic detection, including a standard storage module, an image acquisition module, an image preprocessing module, an analysis module, and a judgment module,

The standard storage module is used to store the standard concentration corresponding to the test item;

The image acquisition module is used to acquire a tomographic image of the test sample on the reagent card, and the tomographic image includes a quality control strip and at least one test strip;

The image preprocessing module is used to perform image preprocessing on the tomographic image to obtain a preprocessed tomographic image; the image preprocessing module is also used to amplify the preprocessed tomographic image to obtain a tomographic image matrix, the tomographic image matrix Including quality control strip matrix and test strip matrix;

The analysis module is used to determine the test concentration of the test sample according to the gray value ratio of the test strip matrix and the quality control strip matrix; and

The judgment module is used to compare the test concentration with the standard concentration to determine the test result of the test sample.

The image analysis system for immunochromatographic detection further includes an edge positioning module, and the edge positioning module includes an edge data storage module and a position determination module:

The edge data storage module is used to store the edge matrix data model;

The edge position determining module is used to scan the tomographic image matrix using the edge matrix data model to obtain the edge position coordinates of the detection window image of the reagent card in the tomographic image matrix.

Among them, the preprocessing module includes an image cropping module and an edge processing module,

The image cropping module is used to perform cropping processing on the tomographic image to obtain a cropped tomographic image containing the detection window of the reagent card;

The edge processing module is used to perform image edge detection and image edge enhancement processing on the cropped tomographic image based on the Sobel operator.

In order to improve the detection accuracy, the image analysis system for immunochromatographic detection also includes an abnormality processing module for performing background abnormality processing on the tomographic image matrix; the abnormality processing module includes an abnormal model storage module, a scanning judgment module, and an abnormal processing module :

The abnormal model storage module is used to store the foreign body shape or color change difference data model;

The scanning judgment module is used for scanning the tomographic image matrix according to the edge position coordinates and the foreign body shape or color change difference data model, and judging the abnormal values in the tomographic image matrix row by row;

The abnormal processing module is used to mark or restore the abnormal value.

Wherein, the foreign body shape or color change difference data model includes an average threshold, a maximum threshold, a minimum threshold, and a standard square deviation threshold.

The reagent card includes a main body, a hand-held part, a sample adding window, a detection window, and a labeling part arranged on the main body, and an identification code is affixed on the labeling part.

The image analysis system for immunochromatographic detection also includes a reagent card image detection module. The reagent card image detection module includes a reagent card image feature database, a reagent card image acquisition module, a two-dimensional code recognition module, and an orientation judgment module. The card image feature database includes the handheld image database and the labeling department database;

The reagent card image acquisition module is used to acquire the current reagent card image when the reagent card is inserted, and the current reagent card image includes the image of the hand-held part and the image of the labeling part;

The two-dimensional code recognition module is used for recognizing the reagent card information and item detection information based on the image of the labeling part;

The orientation judgment module is used for comparing the handheld image with the handheld image database based on the handheld image and using the KNN nearest neighbor classification algorithm to determine the correctness of the inserting direction and orientation of the reagent card.

In a third aspect, an embodiment of the present application also provides an image analysis system for immunochromatographic detection, including at least one processor, a memory communicatively connected with the at least one processor, and an image acquisition device;

The memory stores instructions that can be executed by the at least one processor, and when the instructions are executed by the at least one processor, the image acquisition device acquires image data and the at least one processor executes the method described above.

In a fourth aspect, the embodiments of the present application also provide a computer program product, the computer program product includes a computer program stored on a non-volatile computer-readable storage medium, the computer program includes program instructions, when the When the program instructions are executed by a computer, the computer executes the method described above.

Beneficial effect

The beneficial effects of the implementation of the present application are: the image analysis method and system for immunochromatographic detection of this embodiment, by pre-establishing the standard concentration corresponding to the test item, based on the tomographic image of the test sample on the reagent card, by calculating the tomographic image The gray-scale ratio of the image quality control band and the detection band is used to determine the test concentration of the test sample, and then by comparing the test concentration with the standard concentration, the test result of the test sample is determined. Only the tomographic image is acquired during the test. It can quickly and accurately analyze the detection results of pattern samples, improving detection efficiency and detection accuracy.

Description of the drawings

One or more embodiments are exemplified by the pictures in the accompanying drawings. These exemplified descriptions do not constitute a limitation on the embodiments. The elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the attached drawings do not constitute a scale limitation.

FIG. 1 is a schematic diagram of a three-dimensional structure of an immunochromatographic detection device according to an embodiment of the present application;

FIG. 2 is a structural diagram of a reagent card of an image analysis method for immunochromatographic detection according to an embodiment of the present application;

3 is a schematic diagram of a tomographic image of a detection window of an image analysis method for immunochromatographic detection according to an embodiment of the present application;

4 is a main processing flowchart of the image analysis method for immunochromatographic detection according to an embodiment of the present application;

5 is a schematic diagram of the overall flow of an image analysis method for immunochromatographic detection according to an embodiment of the present application;

6 is a flowchart of edge positioning of an image analysis method for immunochromatographic detection according to an embodiment of the present application;

FIG. 7 is a flowchart of reagent card image detection in an image analysis method for immunochromatographic detection according to an embodiment of the present application;

FIG. 8 is a background abnormality processing flowchart of the image analysis method for immunochromatographic detection according to an embodiment of the present application;

9 is a schematic diagram of the module structure of an image analysis system for immunochromatographic detection according to an embodiment of the present application;

FIG. 10 is a schematic diagram of the device hardware architecture of the image analysis system for immunochromatographic detection according to an embodiment of the present application;

11 is a schematic diagram of image processing of the labeling part of the reagent card of the image analysis method for immunochromatographic detection according to an embodiment of the present application; and

FIG. 12 is a schematic diagram of image processing of the reagent card holding part of the image analysis method for immunochromatographic detection according to an embodiment of the present application.

Embodiments of the present invention

In order to make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following further describes the embodiments of the present application in detail with reference to the accompanying drawings. Here, the exemplary embodiments of the present application and the description thereof are used to explain the present application, but are not intended to limit the present application.

As shown in Fig. 1 and Fig. 4, the present application relates to an image analysis method and system for immunochromatographic detection.

From the perspective of software design, the image analysis method for immunochromatographic detection in this embodiment mainly includes the following steps:

Establish a standard concentration corresponding to the test item; obtain a tomographic image of the test sample on the reagent card, the tomographic image includes a quality control strip and at least one test strip; perform image preprocessing on the tomographic image to obtain a preprocessing layer Analyze the image; enlarge the preprocessed tomographic image to obtain a tomographic image matrix, the tomographic image matrix includes a quality control strip matrix and a test strip matrix; according to the gray value of the test strip matrix and the quality control strip matrix The ratio determines the test concentration of the test sample; compares the test concentration with the standard concentration to determine the test result of the test sample.

Correspondingly, from the perspective of the hardware system, the image analysis system for immunochromatographic detection involved in this application includes at least one processor, a memory that is communicatively connected with the at least one processor, and an image acquisition device; the memory stores data that can be used by the at least one processor. An instruction executed by a processor, when the instruction is executed by the at least one processor, causes the image acquisition device to acquire image data and causes the at least one processor to execute the aforementioned method.

From the perspective of the software modules involved in the foregoing method, the image analysis system for immunochromatographic detection includes a standard storage module, an image acquisition module, an image preprocessing module, an analysis module, and a judgment module.

The standard storage module is used to store the standard concentration corresponding to the test item. The image acquisition module is used to acquire a tomographic image of the test sample on the reagent card, and the tomographic image includes a quality control strip C and at least one test strip T. The image preprocessing module is used to perform image preprocessing on the tomographic image to obtain a preprocessed tomographic image; the image preprocessing module is also used to amplify the preprocessed tomographic image to obtain a tomographic image matrix, the tomographic image matrix Including quality control strip matrix and test strip matrix. The analysis module is used to determine the test concentration of the test sample according to the gray value ratio of the test strip matrix and the quality control strip matrix. The judgment module is used to compare the test concentration with the standard concentration to determine the test result of the test sample.

The image analysis method and system for immunochromatographic detection of this embodiment establishes a standard concentration corresponding to the test item in advance, and calculates the quality control strip C and the test strip of the tomographic image based on the tomographic image of the sample The gray scale ratio with T indirectly infers the test concentration of the test sample, and then compares the difference between the test concentration and the standard concentration, and uses the standard concentration as the threshold to determine the test result of the test sample. Only by acquiring the tomographic image during detection, the detection result of the pattern sample can be quickly and accurately analyzed, thereby improving detection efficiency and detection accuracy.

Example 1

Please refer to FIG. 9, which shows the software module design diagram of the image analysis system for immunochromatographic detection of the present application. The image analysis method of immunochromatographic detection is executed by the system.

The image analysis system 1 for immunochromatographic detection includes a standard storage module 2, an image acquisition module 3, an image preprocessing module 4, an analysis module 5, a judgment module 6, an abnormality processing module 7, a reagent card image detection module 8 and an edge positioning module 9.

The standard storage module 2 is arranged on the memory, and stores the standard concentration corresponding to the test item.

The image acquisition module 3 is connected to the output of the second camera to acquire the tomographic image of the test sample on the reagent card.

Please refer to FIG. 2, the reagent card includes a main body 10, a handle 11, a sample adding window 12, a detection window 13, and a labeling portion 14 provided on the main body 10. An identification code is affixed to the labeling part 14. In this embodiment, the identification code is a two-dimensional code. The two-dimensional code includes the correction curve parameters of the reagent card, project name, reagent batch number and expiration date and other related information.

Please refer to FIG. 3. In this embodiment, the tomographic image is an image displayed on the detection window 13 of the reagent card. The tomographic image includes a quality control strip C and at least one test strip. In this embodiment, the tomographic image includes a quality control strip C, a first test strip T1, and a second test strip T2.

The image preprocessing module 4 performs image preprocessing on the tomographic image to obtain a preprocessed tomographic image. The image preprocessing module 4 also amplifies the preprocessed tomographic image to obtain a tomographic image matrix. The tomographic image matrix includes a quality control strip matrix and a test strip matrix. The analysis module 5 determines the test concentration of the test sample according to the ratio of the gray value of the test strip matrix to the quality control strip matrix. The judgment module 6 compares the test concentration with the standard concentration to determine the test result of the test sample. The above is the detection and analysis process of the image analysis system 1 for immunochromatographic detection.

In this embodiment, when calculating the test concentration of the test sample according to the gray value ratio of the test strip matrix and the quality control strip matrix, different test concentration calculation formulas are set according to the experience of the detection concentration calculation. In this embodiment, the first standard formula and the second standard formula are set:

The first standard formula is: first test concentration = T1/C, where T1 represents the image gray value of the first test strip, and C represents the image gray value of the quality control strip;

The second standard formula is; second test density=(T1+T2)/C, where T1 represents the image gray value of the first test strip, T2 represents the image gray value of the second test strip, and C represents The image gray value of the quality control strip.

The image analysis system for immunochromatographic detection can select the first standard formula or the second standard formula to calculate the test concentration.

Or in order to meet the accuracy requirements of different equipment, the fluctuation threshold V is set in this embodiment, and the average value of the first test concentration of multiple reagent cards is calculated when the equipment needs it. When the average value is greater than the fluctuation threshold V, the first test concentration is selected. The second standard formula calculates the test concentration; when the average value is less than the fluctuation threshold V, the first standard formula is selected to calculate the test concentration. That is, one of the first standard formula or the second standard formula is selected and determined by the fluctuation threshold V to calculate the test concentration.

In this embodiment, the image preprocessing module 4 includes an image cropping module 41 and an edge processing module 42.

The image cropping module 41 performs cropping processing on the tomographic image to obtain a cropped tomographic image containing the detection window 13 of the reagent card. The edge processing module 42 performs image edge detection and image edge enhancement processing on the cropped tomographic image based on the Sobel operator.

Please refer to FIG. 1. The image analysis system 1 for immunochromatographic detection can be applied to a high-throughput immunodetection device or a single-channel immunodetection device. In this embodiment, the high-throughput immunodetection device 20 is taken as an example to introduce the immunity. Image analysis method and system for tomographic detection. The high-throughput immunoassay device 20 is provided with a reaction incubation chamber inside, and an interactive operation interface 22 and a collection device 24 for used reagent cards are provided outside. The user inserts the test sample reagent card 10 into the high-throughput immunoassay device, and then automatically enters the reaction incubation and detection process.

Judging from the sequence of the system processing reagent card detection, it includes the reagent card image detection process, the edge positioning process, the detection analysis process, and the abnormal handling process. The detection and analysis process has been introduced.

Please refer to Figure 11 and Figure 12, in order to ensure the accurate incubation of the test sample, the reagent card image detection module 8 includes a reagent card image feature database 81, a reagent card image acquisition module 82, a two-dimensional code recognition module 83, and an orientation judgment module 84 . Among them, the reagent card image feature database 81 includes a handheld image database 111 and a labeling portion database 141. The reagent card image feature database 81 is an image detection model established after image feature extraction of reagent cards inserted in different directions and orientations.

The reagent card image acquisition module 82 acquires the current reagent card image taken by the first camera when the reagent card is inserted, and the current reagent card image includes a handheld image 111 and a labeling image 141.

The two-dimensional code recognition module 83 recognizes the reagent card information and item detection information based on the labeling part image 141.

The orientation determination module 84 uses the KNN nearest neighbor classification algorithm to compare the handheld image 111 and the handheld image database 141 based on the handheld image 111 to determine the correctness of the reagent card insertion direction and orientation.

In a specific operation, when the high-throughput immunodetection device 20 detects that the reagent card 10 is inserted, the high-throughput immunodetection device 20 first takes an image of the reagent card. Perform format conversion on the handheld image 111 or the labeling image 141; preprocess the handheld image 111 or the labeling image 141; perform image detection and analysis on the reagent card image, and each of the reagent card image feature database 81 The models are compared to draw conclusions. The judgment conclusion includes whether the wrong reagent card is inserted. When the wrong reagent card is inserted, the interactive operation interface 22 displays prompt information to the operating user.

Among them, in the above format conversion, the image captured by the CCD of the first camera is a 24-bit BMP image, which needs to be converted into an 8-bit BMP image first. In the above image preprocessing, the preprocessing includes image edge processing and normalization processing based on the Sobel operator, which enlarges or smoothes the features of the hand-held image 111 or the labeling image 141 Then, the reagent card image feature database 81 is used as the judgment threshold, and the currently photographed handheld image 111 or the labeling image 141 is normalized. After the real-time handheld image 111 or the labeling image 141 is preprocessed, the KNN nearest neighbor classification algorithm is used to compare the images, calculate the distance between the model and the current data, and draw the image detection judgment conclusion.

Please also refer to FIG. 6, the edge positioning module 9 includes an edge data storage module 91 and a position determination module 92.

The edge data storage module 91 stores an edge matrix data model. The edge position determining module 92 scans the tomographic image matrix using the edge matrix data model to obtain the edge position coordinates of the image of the detection window 13 of the reagent card in the tomographic image matrix.

In a specific operation, after the reagent card dropped into the test sample is incubated inside the high-throughput immunoassay device 20, the high-throughput immunoassay device 20 uses the CCD lens of the second camera to capture the reagent card detection window 13 Tomographic image.

Since the reagent card enters the card slot of the high-throughput immune detection device 20 or the rotating motor is positioned, there is an uncontrollable slight deviation. Therefore, when the image is captured by the CCD lens, the non-image calculation part will be included in the tomographic image. Increase the image noise of image analysis. Therefore, in the actual detection and analysis process, it is necessary to position the actual position of the detection window 13 of the reagent card to ensure the accuracy of the final calculation result.

As shown in FIG. 3, there is an obvious border around the detection window 13 of the reagent card, and the tomographic image of the detection window 13 is edge-processed based on the Sobel operator. Then, zero-value processing is performed on the non-edge part of the tomographic image to obtain edge data greater than zero.

The edge matrix data model is an edge matrix data model established according to the image size of the detection window 13. The edge position determination module 92 scans the tomographic image using the edge matrix data model, compares the image data extracted in real time with the rectangular model, and obtains the edge position coordinates of the image of the detection window 13 in the tomographic image.

The process of edge positioning is as follows: in order to speed up the image processing, the tomographic image is cropped first to obtain the cropped tomographic image containing the detection window 13 of the reagent card, that is, the portion of the image within the range of the detection window 13 is obtained.

The preprocessed tomographic image is enlarged to obtain a tomographic image matrix, and the tomographic image matrix includes a quality control strip matrix and a test strip matrix. The zero value method is used to remove tomographic image data that does not have a continuous relationship.

After obtaining the tomographic image matrix, use the established edge matrix data model to perform calculation processing such as scanning and translation to obtain the actual accurate edge position coordinate system.

Please refer to FIG. 8, after the reagent card is added to the test sample in the sample loading window 12, there may be uneven distribution or lumps or foreign objects falling onto the detection window 13 of the reagent card during the chromatography process. These abnormal conditions will affect the accuracy of the image analysis results. In this embodiment, the abnormal processing flow of the image background is established by setting the abnormal processing module 7, thereby improving the accuracy of analysis.

The abnormality processing module 7 includes an abnormality model storage module 71, a scanning judgment module 72 and an abnormality processing module 73.

The abnormal model storage module 71 stores a data model of the shape or color change difference of the foreign object. The scanning judgment module 72 scans the tomographic image matrix according to the edge position coordinates and the foreign body shape or color change difference data model, and judges abnormal values in the tomographic image matrix row by row. The abnormality processing module 73 marks or restores the abnormal value.

In the image recognition of the detection window 13, the shape of the frame and various strips of the detection window 13 are all regular shapes. The abnormality processing module 73 is used for judging the image characteristics of irregular shapes, combining the difference of the color change to establish a data model of the shape of the foreign body or the difference of the color change and the corresponding adjustment threshold.

The abnormal handling process is as follows: firstly, based on a large amount of test data in the early stage, establish a data model of the difference in the shape or color of the foreign body. Calculate and count each column of test data based on the standard edge coordinate system, and determine the corresponding mean threshold, maximum threshold, minimum threshold, and standard square deviation threshold. Scan the current tomographic image matrix, and make judgments and comparisons based on the pre-stored data model of foreign body shape or color change difference. This scan needs to calculate the value of each pixel, and when determining the occurrence of abnormal image features, calculate the neighboring value of the corresponding pixel at the same time. At the edge of the color difference, the adjacent value is used for calculation to determine whether it is an abnormal value. When an abnormal value occurs, the abnormality processing module 7 marks or restores the abnormal value.

Example 2

Please also refer to FIG. 4 and FIG. 5 for the specific software processing flow of the image analysis method for shooting immunochromatographic detection in this embodiment.

The image analysis method of immunochromatographic detection mainly includes the following steps:

Step 101: Establish a standard concentration corresponding to the test item;

Step 103: Obtain a tomographic image of the test sample on the reagent card, the tomographic image including a quality control strip and at least one test strip;

Step 104: Perform image preprocessing on the tomographic image to obtain a preprocessed tomographic image;

Step 105: Enlarge the preprocessed tomographic image to obtain a tomographic image matrix, the tomographic image matrix including a quality control strip matrix and a test strip matrix;

Step 108: Determine the test concentration of the test sample according to the gray value ratio of the test strip matrix and the quality control strip matrix;

Step 109: Compare the test concentration with the standard concentration to determine the test result of the test sample.

In the complete image analysis method, it also includes:

Step 102: a reagent card image detection step, which is started when the reagent card is inserted into the high-throughput immunodetection device 20.

Step 106: an edge positioning step, which is started when the tomographic image matrix is reviewed after the preprocessing is completed.

Step 107: Background anomaly processing step, which is started when the tomographic image matrix is reviewed after the preprocessing is completed.

Performing cropping processing on the tomographic image to obtain a cropped tomographic image containing the detection window 13 of the reagent card;

Please refer to FIG. 7, the image detection step of the reagent card is set before the step of acquiring the tomographic image of the test sample on the reagent card. The image detection steps of the reagent card include:

Step 1021: Establish a reagent card image feature database, including a handheld image database and a labeling department database;

Step 1022: Acquire a current reagent card image, the current reagent card image includes an image of the hand-held part and an image of the labeling part;

Step 1023: Recognize the reagent card information and item detection information based on the image of the labeling part;

Step 1024: Based on the hand-held image, the KNN nearest neighbor classification algorithm is used to compare the hand-held image and the hand-held image database to determine the correctness of the insertion direction and orientation of the reagent card.

Please refer to Figure 7, the edge positioning steps include:

Step 1061: Establish an edge matrix data model;

Step 1062: Scan the tomographic image matrix using the edge matrix data model to obtain the edge position coordinates of the detection window image of the reagent card in the tomographic image matrix.

Please refer to Figure 8. The background exception handling steps include:

Step 1071: Establish a data model for the difference in the shape or color of the foreign body;

Step 1072: Scan the tomographic image matrix according to the edge position coordinates and the foreign body shape or color change difference data model, and judge the abnormal value in the tomographic image matrix row by row;

Step 1073: Mark or restore the abnormal value.

The image analysis method and system for immunochromatographic detection of this embodiment establishes the standard concentration corresponding to the test item in advance, and calculates the tomographic image quality control band and the detection band based on the tomographic image of the test sample on the reagent card. Determine the test concentration of the test sample, and then compare the test concentration with the standard concentration to determine the test result of the test sample. Only the tomographic image needs to be obtained during the test to quickly and accurately analyze the detection of the pattern sample As a result, the detection efficiency and detection accuracy are improved.

Example 3

FIG. 10 is a device of an image analysis system for immunochromatographic detection provided by an embodiment of the present application, such as a schematic diagram of the hardware structure of a device 600. As shown in FIG. 10, the device 600 of the system includes:

One or more processors 610, a memory 620, and an image acquisition component 650. One processor 610 is taken as an example in FIG. 10. The memory 620 stores instructions that can be executed by the at least one processor 610, that is, the computer program 640. When the instructions are executed by the at least one processor, the image data of the detection window 13 is acquired through the image acquisition component 650, so that the At least one processor can execute the image analysis method of immunochromatographic detection.

The processor 610, the memory 620, and the communication component 650 may be connected through a bus or in other ways. In FIG. 10, the connection through a bus is taken as an example.

As a non-volatile computer-readable storage medium, the memory 620 can be used to store non-volatile software programs, non-volatile computer-executable programs and modules, such as the image analysis of immunochromatographic detection in the embodiment of the present application. The program instruction/module corresponding to the method. The processor 610 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions, and modules stored in the memory 620, that is, realizing the image analysis method of immunochromatographic detection in the above method embodiment .

The memory 620 may include a storage program area and a storage data area. The storage program area can store an operating system and at least one application program required by a function; the storage data area can store data created based on the use of an image analysis system for immunochromatographic detection. Data etc. In addition, the memory 620 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the memory 620 may optionally include memories remotely provided with respect to the processor 610, and these remote memories may be connected to the robot through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The one or more modules are stored in the memory 620, and when executed by the one or more processors 610, the above-mentioned immunochromatographic detection image analysis method is executed, for example, the above-described image analysis method in FIG. 4 is executed. Method step 101 to step 109, or execute method step 1021 to step 1024 in FIG. 7, or execute method step 1071 to step 1073 in FIG. 8; implement Figure 6 edge positioning module 9, image preprocessing module 4, analysis module 5. The functions of the judgment module 6, the abnormality processing module 7, and the reagent card image detection module 8.

The above-mentioned products can execute the methods provided in the embodiments of the present application, and have functional modules and beneficial effects corresponding to the execution methods. For technical details not described in detail in this embodiment, please refer to the method provided in the embodiment of this application.

The embodiment of the present application provides a non-volatile computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by one or more processors, for example, execute the above Steps 101 to 109 of the method in Fig. 4 described, or steps 1021 to 1024 of the method in Fig. 7, or steps 1071 to 1073 of the method in Fig. 8 are performed; the edge positioning module 9 and image preview in Fig. 6 are implemented The functions of the processing module 4, the analysis module 5, the judgment module 6, the abnormality processing module 7 and the reagent card image detection module 8.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, not to limit them; under the idea of this application, the above embodiments or the technical features in different embodiments can also be combined. The steps can be implemented in any order, and there are many other variations in different aspects of the application as described above. For the sake of brevity, they are not provided in the details; although the application has been described in detail with reference to the foregoing embodiments, the ordinary in the field The technical personnel should understand that: they can still modify the technical solutions described in the foregoing embodiments, or equivalently replace some of the technical features; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the implementations of this application. Examples of the scope of technical solutions.

Claims

An image analysis method for immunochromatographic detection is characterized in that it comprises the following steps:

Establish the standard concentration corresponding to the test item;

Acquiring a tomographic image of the test sample on the reagent card, the tomographic image including a quality control strip and at least one test strip;

Performing image preprocessing on the tomographic image to obtain a preprocessed tomographic image;

Enlarging the preprocessed tomographic image to obtain a tomographic image matrix, the tomographic image matrix including a quality control strip matrix and a test strip matrix;

Determine the test concentration of the test sample according to the ratio of the gray value of the test strip matrix to the quality control strip matrix;

The test concentration is compared with the standard concentration to determine the test result of the test sample.
The image analysis method for immunochromatographic detection according to claim 1, wherein while the step of establishing the standard concentration corresponding to the test item is carried out, it further comprises an edge positioning step, and the edge positioning step comprises:

Establish an edge matrix data model;

The edge matrix data model is used to scan the tomographic image matrix to obtain the edge position coordinates of the detection window image of the reagent card in the tomographic image matrix.
The image analysis method for immunochromatographic detection according to claim 2, wherein the step of performing image preprocessing on the tomographic image comprises:

Performing cropping processing on the tomographic image to obtain a cropped tomographic image containing the detection window of the reagent card;

Image edge detection and image edge enhancement processing are performed on the cropped tomographic image based on the Sobel operator.
The image analysis method for immunochromatographic detection according to claim 3, further comprising a background abnormality processing step on the cropped tomographic image, and the background abnormality processing step comprises:

Establish a data model for the difference of foreign body shape or color change;

Scanning the tomographic image matrix according to the edge position coordinates and the foreign body shape or color change difference data model, and judging the abnormal values in the tomographic image matrix row by row;

Mark or restore the abnormal value.
The image analysis method for immunochromatographic detection according to claim 4, wherein the foreign body shape or color change difference data model includes an average threshold, a maximum threshold, a minimum threshold, and a standard square deviation threshold.
The image analysis method for immunochromatographic detection according to claim 1, wherein the reagent card comprises a main body, a hand-held part, a sample application window, a detection window, and a labeling part provided on the main body, and The identification code is affixed on the labeling section.
The image analysis method for immunochromatographic detection according to claim 6, characterized in that, before the step of obtaining the tomographic image of the test sample on the reagent card, it further comprises a reagent card image detection step, and the reagent card image The detection steps include:

Establish a database of image characteristics of reagent cards, including handheld image database and labeling department database;

Acquiring a current reagent card image, the current reagent card image including an image of the hand-held part and an image of the labeling part;

Recognizing the reagent card information and item detection information based on the image of the labeling part;

Based on the handheld image, the KNN nearest neighbor classification algorithm is used to compare the handheld image and the handheld image database to determine the correctness of the insertion direction and orientation of the reagent card.
The image analysis method for immunochromatographic detection according to claim 1, wherein the ratio of the gray value of the test strip matrix and the quality control strip matrix is used to determine the test concentration of the test sample The steps include:

Set the first standard formula and the second standard formula;

A fluctuation threshold is set, and the first standard formula or the second standard formula is selected through the fluctuation threshold to calculate the test concentration.
An image analysis system for immunochromatographic detection, which is characterized by comprising a standard storage module, an image acquisition module, an image preprocessing module, an analysis module, and a judgment module,

The standard storage module is used to store the standard concentration corresponding to the test item;

The image acquisition module is used to acquire a tomographic image of the test sample on the reagent card, and the tomographic image includes a quality control strip and at least one test strip;

The image preprocessing module is used to perform image preprocessing on the tomographic image to obtain a preprocessed tomographic image; the image preprocessing module is also used to amplify the preprocessed tomographic image to obtain a tomographic image matrix, so The tomographic image matrix includes a quality control strip matrix and a test strip matrix;

The analysis module is used to determine the test concentration of the test sample according to the gray value ratio of the test strip matrix and the quality control strip matrix; and

The judgment module is used to compare the test concentration with the standard concentration to determine the test result of the test sample.
The image analysis system for immunochromatographic detection according to claim 9, further comprising an edge positioning module, the edge positioning module comprising an edge data storage module and a position determination module:

The edge data storage module is used to store an edge matrix data model;

The edge position determination module is used to scan the tomographic image matrix using the edge matrix data model to obtain the edge position coordinates of the detection window image of the reagent card in the tomographic image matrix.
The image analysis system for immunochromatographic detection according to claim 9, wherein the preprocessing module includes an image cropping module and an edge processing module,

The image cropping module is configured to perform cropping processing on the tomographic image to obtain a cropped tomographic image containing the detection window of the reagent card;

The edge processing module is used to perform image edge detection and image edge enhancement processing on the cropped tomographic image based on the Sobel operator.
The image analysis system for immunochromatographic detection according to claim 11, further comprising an abnormality processing module for performing background abnormality processing on the cropped tomographic image; the abnormality processing module includes an abnormal model storage module, Scanning judgment module and exception handling module:

The abnormal model storage module is used to store the foreign body shape or color change difference data model;

The scanning judgment module is configured to scan the tomographic image matrix according to the edge position coordinates and the foreign body shape or color change difference data model, and judge the abnormal value in the tomographic image matrix row by row;

The abnormality processing module is used to mark or restore the abnormal value.
The image analysis system for immunochromatographic detection according to claim 12, wherein the foreign body shape or color change difference data model includes an average threshold, a maximum threshold, a minimum threshold, and a standard square deviation threshold.
The image analysis system for immunochromatographic detection according to claim 9, wherein the reagent card comprises a main body, a hand-held part, a sample addition window, a detection window, and a labeling part provided on the main body, and The identification code is affixed on the labeling section.
The image analysis system for immunochromatographic detection according to claim 14, further comprising a reagent card image detection module, and the reagent card image detection module includes a reagent card image feature database, a reagent card image acquisition module, and a two-dimensional A code recognition module and an orientation judgment module, wherein the reagent card image feature database includes a handheld image database and a labeling portion database;

The reagent card image acquisition module is used to acquire the current reagent card image when the reagent card is inserted, and the current reagent card image includes the image of the handheld part and the image of the labeling part;

The two-dimensional code recognition module is used for recognizing the reagent card information and item detection information based on the image of the labeling part;

The orientation determination module is used for comparing the handheld image with the handheld image database based on the handheld image and using the KNN nearest neighbor classification algorithm to determine the correctness of the inserting direction and orientation of the reagent card.
An image analysis system for immunochromatographic detection, characterized in that it comprises at least one processor, a memory communicatively connected with the at least one processor, and an image acquisition device;

The memory stores instructions that can be executed by the at least one processor, and when the instructions are executed by the at least one processor, the image acquisition device acquires image data and the at least one processor executes the claims The method described in any one of 1-8.
A computer program product, wherein the computer program product includes a computer program stored on a non-volatile computer-readable storage medium, the computer program includes program instructions, and when the program instructions are executed by a computer, The computer is caused to execute the method according to any one of claims 1-8.