A rapid detection method of biomarker for rheumatoid immunological disease

Abstract

The application relates to the technical field of substance detection, in particular to a rapid biomarker detection method for rheumatism and immunity diseases. The method reflects the substance light intensity value of each position in a detection area by using optical characteristics, and then removes the background position to obtain a substance light intensity distribution diagram of each frame. Then, the substance light intensity distribution diagrams between adjacent frames are matched, the distance between positions and the substance light intensity value difference are considered in the matching process, the block deformation difference is further considered, the comprehensive matching cost and the optimal matching relationship are obtained. The comprehensive matching cost distribution of the optimal matching relationship in the whole process is counted, the substance light intensity values of all positions at the final moment are weighted, and the corrected total amount of antigens of the rheumatism and immunity substance is obtained. By introducing micro deformation analysis and whole process rigidity frequency statistics, the application realizes effective elimination of non-specific aggregation interference in the immunochromatographic detection, and significantly improves the reliability of the biomarker detection result.

Description

Technical Field

[0001] This invention relates to the field of material detection technology, specifically to a rapid detection method for biomarkers in rheumatic and immune diseases. Background Technology

[0002] Endogenous interfering substances such as rheumatoid factor and heterophile antibodies are widely present in blood samples from patients with rheumatic immune diseases. Immunochromatographic test strips utilize capillary action on porous membrane media (such as nitrocellulose membranes) to drive the migration of sample fluid, forming a visible optical signal (such as a colored band) in the detection area through a specific antigen-antibody reaction. Traditional detection methods typically acquire a single static image at a fixed moment after the reaction reaches equilibrium, and calculate the concentration of biomarkers by statistically analyzing the pixel grayscale values ​​or optical density integrals of the detection area.

[0003] In actual testing of blood samples from patients with rheumatic and immune diseases, the sample matrix often contains interfering substances such as rheumatoid factor and heterophilic antibodies. These interfering substances are prone to non-specific physical retention or aggregation within the micropores of the chromatography membrane, forming false-positive signal spots with optical properties similar to specific antigen-antibody complexes. Existing static image analysis methods only focus on the magnitude of the light intensity of substances and cannot distinguish whether the signal source is a stable structure anchored by chemical bonds or a loose aggregate based on physical adsorption. This makes it difficult to identify interfering signals, and the final calculation result of the total antigen amount is easily affected by non-specific background, reducing the accuracy and specificity of the detection. Summary of the Invention

[0004] To address the technical problem of existing technologies failing to distinguish between chemical bonding and physical adsorption as the actual generation of a signal source, leading to interference with the final antigen total calculation and inaccurate detection, the present invention aims to provide a rapid detection method for biomarkers in rheumatic immune diseases. The specific technical solution adopted is as follows: This invention proposes a rapid detection method for biomarkers in rheumatic and immune diseases, the method comprising: Acquire images of the detection area in consecutive frames on the biomarker test strip; For any given frame, the pixel values ​​in the preset neighborhood of each location in the detection area are statistically analyzed to obtain the material light intensity value at each location; background locations are filtered out based on the material light intensity value to obtain the material light intensity distribution map for each frame. For any given frame, the target position in the matter intensity distribution map is matched with the position to be matched in the matter intensity distribution maps of adjacent frames. During the matching process: a basic matching cost is obtained based on the distance between the target position and the position to be matched, as well as the difference in matter intensity values; the neighborhood positions of the target position are statistically analyzed, and the predicted positions of the neighborhood positions on the matter intensity distribution maps of adjacent frames are predicted based on the path from the target position to the position to be matched; the difference in matter intensity values ​​between the neighborhood positions and the predicted positions is obtained; the difference in mass deformation is fused with the basic matching cost to obtain the comprehensive matching cost between the target position and each position to be matched; and the optimal matching relationship between adjacent frames is obtained based on the comprehensive matching cost for each position. At each location, the rigid frequency of the entire process is obtained based on the comprehensive matching cost distribution of the optimal matching relationship; at the final detection moment, the light intensity values ​​of substances at all locations in the detection area are weighted according to the rigid frequency of the entire process to obtain the total amount of corrected antigens of rheumatic immune substances.

[0005] Furthermore, the method for obtaining the light intensity value of the substance includes: For each location, the average gray value within a preset neighborhood is used as the light intensity value of the substance.

[0006] Furthermore, the method for determining the background position includes: If the light intensity value of the substance is less than the preset background threshold, then the corresponding position is determined to be the background position.

[0007] Furthermore, the method for obtaining the basic matching cost includes: For a pair of locations to be matched consisting of a target location and a location to be matched, the coordinate distance and the difference in material light intensity between the pair are obtained. After normalizing the coordinate distance and the difference in material light intensity, the basic matching cost is obtained by weighted summation according to a preset fusion weight.

[0008] Furthermore, the method for obtaining the degree of difference in agglomerate deformation includes: For any neighborhood location, the material light intensity value of the neighborhood location is used as the difference weight. The difference weight is used to weight the difference in material light intensity values ​​between the neighborhood location and the predicted location to obtain the initial deformation difference degree of each neighborhood location. The deformation difference of the clumping is obtained by statistically analyzing the initial deformation differences of all neighboring locations.

[0009] Furthermore, the method for obtaining the comprehensive matching cost includes: The degree of deformation difference of the clumps is mapped using an exponential function to obtain a non-rigid penalty factor. The product of the basic matching cost and the non-rigid penalty factor is used as the comprehensive matching cost.

[0010] Furthermore, the method for obtaining the optimal matching relationship includes: The comprehensive matching cost between the target location and all locations to be matched constitutes a comprehensive matching cost sequence. The comprehensive matching cost sequence of all locations in the material light intensity distribution map of each frame is obtained, forming an initial matching matrix. The initial matching matrix is ​​augmented and expanded with a preset maximum element value as the fill value to obtain a matching matrix. The Hungarian algorithm is used to optimize the matching matrix to obtain the optimal matching relationship.

[0011] Furthermore, the method for obtaining the rigid frequency of the entire process includes: For any given position, count the number of frames whose overall matching cost is less than a preset cost threshold across all frames. Use the counted number of frames as the numerator and the number of frames at the non-background position as the denominator to obtain the rigid frequency of the entire process. If the number of frames at the non-background position is 0, then set the rigid frequency of the entire process to 0.

[0012] Furthermore, the step of weighting the light intensity values ​​of the substance at all locations in the detection area based on the rigid frequency of the entire process to obtain the corrected total antigen includes: If the rigidity frequency of the entire process is 0, then the confidence weight is set to 0; If the rigid frequency of the entire process is not 0, the frequency difference between the rigid frequency of the entire process and the preset high confidence frequency threshold is obtained, and the frequency difference is normalized to obtain the confidence weight. At the final moment of detection, the light intensity values ​​of the substance at all locations in the detection area are weighted and summed using the confidence weight to obtain the total amount of the corrected antigen.

[0013] Furthermore, the method for obtaining the predicted location includes: The vector between the target position and the position to be matched is used as the fluid migration vector; along the direction of the fluid migration vector, starting from the neighborhood position, a prediction vector with the same magnitude as the fluid migration vector is constructed, and the endpoint of the prediction vector is the prediction position.

[0014] The present invention has the following beneficial effects: This invention first obtains the material light intensity value by statistically analyzing the pixel values ​​in the preset neighborhood of each location in the detection area. Background locations are then filtered out based on these material light intensity values ​​to obtain a material light intensity distribution map for each frame. Invalid background areas without material distribution are excluded, focusing on the observation point containing the material, thereby reducing computational load and improving the efficiency and accuracy of subsequent matching. Furthermore, when matching the target location with adjacent frames, not only is the basic matching cost based on distance and material light intensity value differences calculated, but the difference in material light intensity values ​​between the neighborhood and the predicted location on the predicted path is also calculated using the predicted location of the target location's neighborhood, obtaining the clumping deformation difference degree. This clumping deformation difference degree can quantify the ability of material clumps to maintain their microstructure during migration, effectively identifying non-specific aggregates with high optical intensity but unstable structures, providing crucial micro-dynamic basis for distinguishing between specific aggregates and non-specific interfering substances. By comprehensively considering position migration, light intensity changes, and structural deformation, the most physically consistent material correspondences can be selected from a global perspective. Finally, the rigidity frequency of each position throughout the entire process is obtained based on the comprehensive matching cost distribution of the optimal matching relationship. This frequency is then used to weight the material light intensity value at the final detection moment to obtain the corrected total antigen amount. This method utilizes the dynamic stability history throughout the entire cycle as a confidence gating mechanism to suppress non-specific background signals in the final static image, thereby effectively reducing false positive interference and improving the specificity and accuracy of antigen quantitative detection. This invention, by introducing microscopic deformation analysis and full-process rigidity frequency statistics, effectively eliminates non-specific aggregation interference in immunochromatographic detection, significantly improving the reliability of biomarker detection results. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of a rapid detection method for biomarkers in rheumatic immune diseases, provided as an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a rapid detection method for biomarkers in rheumatic immune diseases proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following describes in detail, with reference to the accompanying drawings, a specific scheme for a rapid detection method of biomarkers for rheumatic immune diseases provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a flowchart of a rapid detection method for biomarkers in rheumatic immune diseases according to an embodiment of the present invention. The method includes: Step S1: Obtain images of the detection area in consecutive frames on the biomarker test strip.

[0021] In order to eliminate errors caused by physical agglomeration on the test strip, this embodiment of the invention considers that the chromatographic test strip should remain stationary relative to the optical sensor during the detection process, while the sample fluid migrates unidirectionally inside the porous membrane. Therefore, by detecting the detection area images of consecutive frames in the fixed detection area, the fluid motion analysis can be transformed into a numerical transfer problem between fixed spatial units. In subsequent steps, the reliability of the information at each position in the detection area on the current test strip can be determined through the analysis of consecutive frames.

[0022] In the specific implementation of this invention, the detection area refers to the fluid flow area including the detection line and the quality control line. Since the image field of view is fixed, the detection area image can be directly extracted from the initial image obtained by shooting. The specific location of the extracted area needs to be set according to the specific implementation scenario, which will not be elaborated here.

[0023] It should be noted that each pixel in the detection area image represents the same physical location in consecutive frames. The subsequent processes in this embodiment of the invention are described in terms of location, that is, a location corresponds to a pixel in consecutive frames. This embodiment of the invention aims to analyze the final information reliability of each location in the entire detection process.

[0024] In one specific implementation of this invention, considering the optical reflection and uneven texture of the test strip film material itself, and the need to filter out background locations in the detection area in subsequent steps, it is necessary to determine the light intensity benchmark when there is no material distribution before detecting biomarkers for rheumatic and immune diseases. In this implementation, in the initial stage before the sample fluid flows through the detection area, the average gray value of the detection area is read as a background threshold for background location filtering in subsequent steps.

[0025] Step S2: For any frame, statistically analyze the pixel value information in the preset neighborhood of each position in the detection area to obtain the material light intensity value at each position; filter out background positions based on the material light intensity value to obtain the material light intensity distribution map of each frame.

[0026] The distribution and concentration of biomarkers on the chromatographic membrane are directly reflected in the grayscale values ​​acquired by the optical sensor. Therefore, the light intensity value of the substance at each location in each frame can be determined based on the pixel values ​​representing the detection area image. The light intensity value can characterize the substance distribution at the corresponding location. Since the background locations are where the fluid does not flow or has not yet flowed, the light intensity values ​​they represent are significantly different from those of the substances present. Therefore, background locations can be filtered out based on the light intensity values ​​to obtain the light intensity distribution map for each frame. That is, the light intensity distribution map of each frame can be considered as containing only irregular images from non-background locations. Similarly, the shape of the light intensity distribution map may also differ between different frames.

[0027] Preferably, in some implementations of the present invention, in order to avoid random errors caused by individual pixel information at each location, the average gray value within a preset neighborhood is used as the material light intensity value for each location.

[0028] Furthermore, in a further implementation of this invention, after statistically analyzing the grayscale values ​​of a preset neighborhood at each location, an image preprocessing algorithm is used to eliminate noise in the grayscale values ​​within the neighborhood, and then the average value is calculated to obtain a more accurate material light intensity value. Specifically, Gaussian smoothing or mean filtering can be used for this implementation, and the size of the neighborhood can be set to 8 neighborhoods. These are techniques well-known to those skilled in the art and will not be elaborated upon or limited here.

[0029] Preferably, in some implementations of the present invention, if the light intensity value of the substance is less than a preset background threshold, it indicates that the location has not adsorbed enough markers in the current frame, and the corresponding location is determined to be a background location and deleted from the detection area image of the current frame. Conversely, it is determined that there is a significant substance distribution at the location in the current frame.

[0030] Step S3: For any frame, match the target position in the material intensity distribution map with the position to be matched in the material intensity distribution map of the adjacent frame. During the matching process: obtain the basic matching cost based on the distance between the target position and the position to be matched and the difference in material intensity value; count the neighborhood positions of the target position, predict the predicted position of the neighborhood positions on the material intensity distribution map of the adjacent frame based on the path between the target position and the position to be matched, and obtain the clumping deformation difference degree based on the difference in material intensity value between the neighborhood positions and the predicted positions; fuse the clumping deformation difference degree and the basic matching cost to obtain the comprehensive matching cost between the target position and each position to be matched; obtain the optimal matching relationship between the positions in the adjacent frames based on the comprehensive matching cost of each position.

[0031] To analyze the dynamic mechanical behavior of substances in a microfluidic environment and address the difficulty in distinguishing between specific chemical bindings and non-specific physical aggregations based on optical characteristics, this invention requires dynamic analysis of each frame. Considering that specific immune complexes are anchored by chemical bonds and exhibit stable rigid structures, while rheumatoid factor aggregates, non-specific aggregates, are soft matter, essentially formed by material adhesion, and deform under the combined effects of fluid drag and fiber pore shearing, this invention performs a matching analysis between each frame and adjacent frames to calculate the instantaneous dynamic state at each location.

[0032] Taking any frame as an example, in the material intensity distribution map of the target frame, each position is matched with its position in the material intensity distribution maps of adjacent frames. Any position in the target frame is chosen as the target position, and the positions in adjacent frames become the positions to be matched against the target position. Each position to be matched forms a matching group with the target position, and a matching cost is calculated. During the matching process, for any matching group, the path from the target position to the matching position can be considered a physical assumption: the material mass at the target position moves to the matching position by translation under fluid propulsion. For this physical assumption of movement, if the assumption is a rigid characteristic characterized by specific immune complexes, the material intensity values ​​between the target position and the matching position are relatively similar; and the time interval between adjacent frames is short, so the translation will not produce a large distance, therefore the distance between the target position and the matching position will not be too far. Based on this, the basic matching cost can be obtained by statistically analyzing the distance between the target position and the matching position and the difference in material intensity values. In other words, for the basic matching cost, the greater the distance and the greater the difference in light intensity of the substances, the less the physical assumption of the group to be matched does not conform to the rigid characteristics characterized by the specific immune complex. Therefore, the basic matching cost is greater, and the group to be matched is less likely to be a moving matching result of a clear specific immune complex.

[0033] This invention further considers the need to quantify the structural retention capability of material clumps under the assumed migration path to distinguish between rigid binders and soft matter aggregates. If the material is rigid, its grayscale distribution pattern in its neighborhood should remain unchanged after translation; if the material deforms or disintegrates, its neighborhood distribution will not match the prediction after translation. Therefore, for the neighborhood position of the target location, the predicted position of the neighborhood position on the material intensity distribution map of the adjacent frame can be predicted based on the path between the target location and the location to be matched. The greater the difference in material intensity value between the neighborhood position and its predicted position, the less it conforms to the overall translation characteristics of the clump containing the target location, and the greater the corresponding matching cost should be. Therefore, the clump deformation difference degree can be obtained by statistically analyzing the difference in material intensity value between the neighborhood position and the predicted position. This clump deformation difference degree directly reflects the degree of loss of microstructural integrity of the material during migration. That is, the greater the clump deformation difference degree, the less the clump conforms to the overall movement characteristics of the rigid binder of the specific immune complex during movement, and the less it conforms to the non-uniform movement characteristics of the soft matter aggregate. Therefore, the difference in mass deformation and the basic matching cost can be fused to obtain the comprehensive matching cost between the target location and each location to be matched. The smaller the comprehensive matching cost, the more the group to be matched conforms to the rigidity characteristics of the specific immune complex. Finally, the optimal matching relationship between adjacent frames can be obtained based on the comprehensive matching cost for each location.

[0034] In the specific implementation of this invention, to avoid invalid searches across the entire map, fluid dynamics constraints are introduced. For the target location, a circular search area is constructed with the target location as the center and the maximum flow velocity constraint radius as the radius. The next frame of the target frame is used as the adjacent frame for matching, and all positions within the corresponding search area in the next frame are selected as the positions to be matched. This eliminates physically impossible long-distance jumps. It should be noted that the maximum flow velocity constraint radius needs to be set according to the specific camera parameters, shooting distance, etc., in the specific implementation scenario, and will not be elaborated or limited here.

[0035] Preferably, in this embodiment of the invention, the method for obtaining the basic matching cost includes: For a pair consisting of a target location and a location to be matched, the coordinate distance and the difference in material light intensity between the pair are obtained. After normalizing the coordinate distance and the difference in material light intensity, they are weighted and summed according to a preset fusion weight to obtain the basic matching cost. Normalization effectively fuses two data points with different dimensions, and the introduction of a weighted fusion mechanism allows implementers to select the weights according to specific needs, focusing on either changes in light intensity or the distance of the flow.

[0036] As a specific example, in the specific implementation of this embodiment, since the position to be matched is a point in the search area, the distance between the target position and the position to be matched will not be greater than the maximum flow velocity constraint radius. Therefore, the maximum flow velocity constraint radius can be used as the denominator, and the distance as the numerator, to achieve maximum maximization and normalization. Similarly, this specific implementation sets a maximum difference in material light intensity value to 255, using 255 as the denominator and the difference in material light intensity value as the numerator, to also achieve maximum maximization and normalization.

[0037] As a specific example, in the specific implementation of this embodiment of the invention, considering that the material light intensity value is a representative of the material quantity and has greater importance in the matching process, the weight of the difference in the normalized material light intensity value is set to 0.6 and the weight of the normalized coordinate distance is set to 0.4 in the process of obtaining the basic matching cost by weighted summation.

[0038] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the predicted location includes: The vector between the target position and the position to be matched is used as the fluid migration vector; along the direction of the fluid migration vector, starting from the neighborhood position, a prediction vector with the same magnitude as the fluid migration vector is constructed, and the endpoint of the prediction vector is the prediction position.

[0039] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the degree of difference in agglomerate deformation includes: For any neighborhood location, the light intensity value of the material at that location is used as a difference weight. This difference weight is then used to weight the difference in light intensity values ​​between the neighborhood location and the predicted location to obtain the initial deformation difference degree for each neighborhood location. By weighting the difference weights, the deformation of high-concentration regions can contribute more to the final agglomerate deformation difference degree, thus better reflecting the degree of loss of microstructural integrity characterized by the agglomerate deformation difference degree.

[0040] Finally, the initial deformation difference of all neighboring locations is statistically analyzed to obtain the deformation difference of the clumping.

[0041] As a specific example, in one implementation of this invention, the degree of difference in agglomerate deformation is expressed by the formula: ;in Let K be the clumping deformation difference between the target position n and the position u to be matched, and K be the number of neighboring positions. To use t-1 as the light intensity value of the material at the k-th neighborhood position in the target frame, Let l be the predicted light intensity value of the material at position l corresponding to the k-th neighborhood position in the next frame of the target frame. This is a preset hyperparameter. In this specific implementation, the number of neighborhood positions is 8, meaning that the analysis selects 8 neighborhoods of the target position; the preset hyperparameter is set to 1 to prevent the denominator from being 0.

[0042] In the above formula, weighting is achieved by multiplication and divided by the cumulative value of the difference weights. Essentially, it is based on the difference weights to calculate the average of the differences in light intensity values ​​of substances in all neighboring locations. This weighted fusion method can effectively retain the contribution of differences reflected in high-concentration locations.

[0043] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the comprehensive matching cost includes: The degree of clumping deformation difference is mapped using an exponential function to obtain a non-rigid penalty factor. This exponential function mapping allows for a larger non-rigid penalty factor corresponding to a greater degree of clumping deformation difference, thus achieving a sensitive and significant penalty. The product of the basic matching cost and the non-rigid penalty factor is then used as the comprehensive matching cost.

[0044] As a specific example, in one implementation of this invention, the comprehensive matching cost can be expressed by the formula: ;in The total matching cost between the target position n and the position u to be matched. The weights of the normalized coordinate distances, The weighting of the differences in the normalized light intensity values ​​of the substances. Let be the fluid migration vector between the target location and the location to be matched. Let be the magnitude of the fluid migration vector, i.e. This represents the coordinate distance between two locations. The radius constrained by the maximum flow velocity. Let be the light intensity value of the material at target position n in the target frame at time t-1. Let be the light intensity value of the material at the location to be matched at time t. The exp function is an exponential function with the natural constant as the base, representing the preset maximum difference in material light intensity. The preset gain factor, The difference in clumping deformation between the target position n and the position u to be matched is denoted as ...

[0045] In the above formula, the non-rigid penalty factor is fused with the basic matching cost using a product. Furthermore, a preset gain factor is introduced into the non-rigid penalty factor to further amplify the penalty term. Through the mapping of an exponential function, the penalty term rapidly increases as the difference in clumping deformation increases, causing the resulting comprehensive matching cost to tend towards infinity, thereby optimizing the matching efficiency. This indicates that the path from the target location to the location to be matched does not conform to the physical characteristics of specific binding. The preset gain factor can be set to 5, which will not be elaborated or limited here.

[0046] Preferably, in some specific implementations of the embodiments of the present invention, the method for obtaining the optimal matching relationship includes: The combined matching cost between the target location and all locations to be matched constitutes a combined matching cost sequence. The combined matching cost sequence for all locations in each frame's material light intensity distribution map is obtained, forming the initial matching matrix.

[0047] It should be noted that in one specific implementation of this invention, since the positions to be matched correspond to different positions, the lengths of the resulting comprehensive matching cost sequences are different, and the coordinates of the positions to be matched corresponding to the same position in the sequence are also different. Therefore, when constructing the initial matching matrix, it is necessary to count the coordinates of all positions to be matched. For example, the initial matching matrix can be constructed with all positions to be matched as rows and all target positions as columns. If the corresponding element in the initial matching matrix has a comprehensive matching cost, then the element value of that element is the corresponding comprehensive matching cost; if it does not exist, for example, if the position to be matched corresponding to the element is not a position to be matched within the search range of the target position, then the element value of the corresponding element is set to the preset maximum element value, which means that the corresponding information is meaningless information.

[0048] Using a preset maximum element value as the fill value, the initial matching matrix is ​​augmented and expanded to obtain a matching square matrix, where the length and width of the matching square matrix are the same. The purpose is to solve the problem using the minimum weight matching of a bipartite graph. Therefore, the Hungarian algorithm can be used to optimize the matching square matrix to obtain the optimal matching relationship. The preset maximum element value can be set to infinity, and finally, the optimal one-to-one matching relationship between each position on the current frame's material light intensity distribution map and its adjacent frames can be obtained.

[0049] Step S4: At each location, obtain the rigid frequency of the entire process for each location based on the comprehensive matching cost distribution of the optimal matching relationship; at the final detection moment, weight the material light intensity values ​​of all locations in the detection area based on the rigid frequency of the entire process to obtain the total amount of corrected antigen of rheumatic immune substances.

[0050] In a given frame, a smaller overall matching cost for the optimal matching relationship at a given location indicates that the substance at that location has maintained positional stability and structural rigidity under fluid impact, consistent with the characteristics of a specific immune complex. Conversely, a larger overall matching cost indicates that the substance at that location has undergone significant drift or internal deformation, consistent with the characteristics of a non-specific aggregate. Therefore, for each location, the distribution of the overall matching cost of the optimal matching relationship across all frames throughout the entire detection process can be statistically analyzed to obtain the overall rigidity frequency for each location. A higher overall rigidity frequency indicates that the location exhibits non-specific rigidity information for most of the detection process, resulting in higher confidence in the substance content at that location. Conversely, a lower overall rigidity frequency indicates that the location exhibits non-rigid characteristics for most of the process, representing non-specific data caused by physical viscosity. The substance characteristics reflected by optical features at this location are unreliable and do not represent true substance data.

[0051] Therefore, at the final moment of detection, the light intensity values ​​of the material at all locations in the detection area can be weighted according to the frequency of the steel core throughout the process, reducing the influence of non-specific feature locations on the material quality information, and thus obtaining the corrected total amount of antigen that truly only contains the specific binding results.

[0052] It should be noted that after weighting the light intensity values ​​of substances, the final result is the total corrected light intensity value of substances. By using a dose-response standard curve calibrated with known concentration standards, and by finding the coordinates corresponding to the total corrected light intensity value of substances, the actual total amount of corrected antigens of rheumatic and immune disease biomarkers can be obtained, that is, the final result is an actual concentration information.

[0053] Preferably, in this embodiment of the invention, the method for obtaining the rigidity frequency throughout the entire process includes: For any given location, count the number of frames across all frames where the overall matching cost is less than a preset cost threshold. A cost less than the preset threshold indicates that the location at that frame has a high confidence level in terms of material quality, and the information is reliable. Use the counted number of frames as the numerator and the number of frames at that location that are not background positions as the denominator to obtain the rigid frequency of the entire process. If the number of frames at non-background positions is 0, it means that that location is a background position throughout the entire process, and the rigid frequency of the entire process is set to 0.

[0054] It should be noted that the cost threshold is an empirical constant, which can be obtained by using the upper limit of reliable specific features in statistical historical data, and will not be elaborated or limited here.

[0055] Preferably, in this embodiment of the invention, the total amount of corrected antigen is obtained by weighting the light intensity values ​​of the substance at all locations in the detection area according to the rigid frequency of the entire process, including: If the rigidity frequency of the entire process is 0, then the confidence weight is set to 0; this indicates that the position is always a background position, and the light intensity value of the material reflected there is no reference value. The confidence weight is directly set to 0 to extract the information of the background position.

[0056] If the rigid frequency of the entire process is not 0, the frequency difference between the rigid frequency of the entire process and the preset high confidence frequency threshold is obtained, and the frequency difference is normalized to obtain the confidence weight. By introducing the high confidence frequency threshold, the final confidence weight result of higher rigid frequencies of the entire process can be amplified. That is, only rigid frequencies of the entire process that are greater than the preset high confidence frequency threshold are judged as high confidence signals, and the confidence weight approaches 1. For lower rigid frequencies of the entire process, the corresponding confidence weight approaches 0.

[0057] At the final moment of detection, a corresponding confidence weight can be obtained for each location in the detection area. The material light intensity values ​​at all locations in the detection area are weighted and summed using the confidence weights to obtain the total amount of the modified antigen.

[0058] As a specific example, in this embodiment of the invention, the confidence weight is expressed by the formula: ;in Let be the confidence weight for the target position n, and exp be an exponential function with base 1000. For mapping coefficients, Let n be the rigid frequency throughout the entire process at target position n. This is the high-confidence frequency threshold.

[0059] In the above formula, the mapping coefficient is used to adjust the steepness of the weight transformation. To represent frequency differences, a negative correlation mapping is performed using an exponential function with the natural constant as the base, followed by a mapping using the reciprocal form. Since the negative correlation mapping result of the exponential function is data between 0 and 1, adding 1 to the denominator results in a final confidence weight result that is also data between 0 and 1. Furthermore, the greater the frequency difference... The closer a value is to 0, the closer the confidence weight is to 1. The high-confidence frequency threshold is set to 0.8, meaning that only substances that remain stable for more than 80% of the time are considered high-confidence signals. The mapping coefficient can be set to 10 to significantly amplify higher frequency differences, thereby obtaining the expected confidence weight result.

[0060] In summary, this invention utilizes optical features to reflect the light intensity value of substances at each location in the detection area, thereby filtering out background locations to obtain a light intensity distribution map for each frame. The light intensity distribution maps of adjacent frames are then matched, taking into account the distance between locations and the difference in light intensity values, further considering the difference in clumping deformation, to obtain a comprehensive matching cost and an optimal matching relationship. By statistically analyzing the comprehensive matching cost distribution of the optimal matching relationship throughout the entire process, the light intensity values ​​at all locations at the final moment are weighted to obtain the total corrected antigen amount of rheumatic immune substances. This invention, by introducing microscopic deformation analysis and full-process rigidity frequency statistics, effectively eliminates non-specific aggregation interference in immunochromatographic detection, significantly improving the reliability of biomarker detection results.

[0061] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0062] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A rapid detection method for biomarkers in rheumatic immune diseases, characterized in that, The method includes: Acquire images of the detection area in consecutive frames on the biomarker test strip; For any given frame, the pixel values ​​in the preset neighborhood of each location in the detection area are statistically analyzed to obtain the material light intensity value at each location; background locations are filtered out based on the material light intensity value to obtain the material light intensity distribution map for each frame. For any given frame, the target position in the matter intensity distribution map is matched with the position to be matched in the matter intensity distribution maps of adjacent frames. During the matching process: a basic matching cost is obtained based on the distance between the target position and the position to be matched, as well as the difference in matter intensity values; the neighborhood positions of the target position are statistically analyzed, and the predicted positions of the neighborhood positions on the matter intensity distribution maps of adjacent frames are predicted based on the path from the target position to the position to be matched; the difference in matter intensity values ​​between the neighborhood positions and the predicted positions is obtained; the difference in mass deformation is fused with the basic matching cost to obtain the comprehensive matching cost between the target position and each position to be matched; and the optimal matching relationship between adjacent frames is obtained based on the comprehensive matching cost for each position. At each location, the rigid frequency of the entire process is obtained based on the comprehensive matching cost distribution of the optimal matching relationship; at the final detection moment, the light intensity values ​​of substances at all locations in the detection area are weighted according to the rigid frequency of the entire process to obtain the total amount of corrected antigens of rheumatic immune substances.

2. The rapid detection method for biomarkers in rheumatic immune diseases according to claim 1, characterized in that, The method for obtaining the light intensity value of the substance includes: For each location, the average gray value within a preset neighborhood is used as the light intensity value of the substance.

3. The rapid detection method for biomarkers in rheumatic immune diseases according to claim 1, characterized in that, The methods for determining the background location include: If the light intensity value of the substance is less than the preset background threshold, then the corresponding position is determined to be the background position.

4. The rapid detection method for biomarkers in rheumatic immune diseases according to claim 1, characterized in that, The method for obtaining the basic matching cost includes: For a pair of locations to be matched consisting of a target location and a location to be matched, the coordinate distance and the difference in material light intensity between the pair are obtained. After normalizing the coordinate distance and the difference in material light intensity, the basic matching cost is obtained by weighted summation according to a preset fusion weight.

5. The rapid detection method for biomarkers in rheumatic immune diseases according to claim 1, characterized in that, The method for obtaining the degree of difference in agglomerate deformation includes: For any neighborhood location, the material light intensity value of the neighborhood location is used as the difference weight. The difference weight is used to weight the difference in material light intensity values ​​between the neighborhood location and the predicted location to obtain the initial deformation difference degree of each neighborhood location. The deformation difference of the clumping is obtained by statistically analyzing the initial deformation differences of all neighboring locations.

6. The rapid detection method for biomarkers in rheumatic immune diseases according to claim 1, characterized in that, The method for obtaining the comprehensive matching cost includes: The degree of deformation difference of the clumps is mapped using an exponential function to obtain a non-rigid penalty factor. The product of the basic matching cost and the non-rigid penalty factor is used as the comprehensive matching cost.

7. The rapid detection method for biomarkers in rheumatic immune diseases according to claim 1, characterized in that, The method for obtaining the optimal matching relationship includes: The comprehensive matching cost between the target location and all locations to be matched constitutes a comprehensive matching cost sequence. The comprehensive matching cost sequence of all locations in the material light intensity distribution map of each frame is obtained, forming an initial matching matrix. The initial matching matrix is ​​augmented and expanded with a preset maximum element value as the fill value to obtain a matching matrix. The Hungarian algorithm is used to optimize the matching matrix to obtain the optimal matching relationship.

8. The rapid detection method for biomarkers in rheumatic immune diseases according to claim 1, characterized in that, The method for obtaining the rigid frequency of the entire process includes: For any given position, count the number of frames whose overall matching cost is less than a preset cost threshold across all frames. Use the counted number of frames as the numerator and the number of frames at the non-background position as the denominator to obtain the rigid frequency of the entire process. If the number of frames at the non-background position is 0, then set the rigid frequency of the entire process to 0.

9. A rapid detection method for biomarkers in rheumatic immune diseases according to claim 8, characterized in that, The step of weighting the light intensity values ​​of substances at all locations in the detection area based on the rigid frequency of the entire process to obtain the total corrected antigen includes: If the rigidity frequency of the entire process is 0, then the confidence weight is set to 0; If the rigid frequency of the entire process is not 0, the frequency difference between the rigid frequency of the entire process and the preset high confidence frequency threshold is obtained, and the frequency difference is normalized to obtain the confidence weight. At the final moment of detection, the light intensity values ​​of the substance at all locations in the detection area are weighted and summed using the confidence weight to obtain the total amount of the corrected antigen.

10. A rapid detection method for biomarkers in rheumatic immune diseases according to claim 1, characterized in that, The method for obtaining the predicted location includes: The vector between the target position and the position to be matched is used as the fluid migration vector; along the direction of the fluid migration vector, starting from the neighborhood position, a prediction vector with the same magnitude as the fluid migration vector is constructed, and the endpoint of the prediction vector is the prediction position.

Images