HBsAg Detection Method and Related Equipment Based on Multiple Wavelengths

Abstract

The present invention provides a multi-wavelength-based HBsAg detection method, including: detecting the light absorption value of a sample to be tested for HBsAg by a sandwich ELISA method using a multi-wavelength microplate reader to obtain the light absorption data of the sample to be tested for HBsAg at multiple wavelengths; determining the OD value of the sample, the OD value of the negative control, and the OD value of the positive control at dual wavelengths based on the light absorption data at wavelengths of 450 nm and 630 nm; determining whether the first detection result is valid within the preset concentration range of HBsAg based on the OD value of the negative control and the OD value of the positive control; if the first detection result is invalid within the preset concentration range of HBsAg, determining the second detection result of the sample to be tested for HBsAg based on the mapping relationship between the light absorption data at multiple wavelengths and the detection result of chemiluminescence method. The present invention improves the sensitivity, accuracy, and reliability of hepatitis B surface antigen detection at low concentrations.

Description

Technical Field

[0001] The present invention relates to the technical field of immune detection of hepatitis B surface antigen, and in particular to a multi-wavelength-based HBsAg detection method and related equipment. Background Art

[0002] Hepatitis B surface antigen (HBsAg) is an important marker of hepatitis B virus infection, and its immunodetection is of great significance for the diagnosis, treatment and prevention of hepatitis B. At present, the immunodetection methods of HBsAg mainly include enzyme-linked immunosorbent assay (ELISA), chemiluminescence assay, radioimmunoassay, etc. Among them, the double antibody sandwich method is a commonly used ELISA technology, which is widely used because of its simple operation and low cost.

[0003] However, in practical applications, the double antibody sandwich method has some problems, such as low sensitivity and susceptibility to interference factors, which sometimes lead to inaccurate test results. Especially in the detection of low-concentration HBsAg samples, the presence of interfering substances or slight changes in operating conditions may cause deviations in test results, resulting in false negative or false positive results, which affect the accuracy of the test. Therefore, there is an urgent need for a solution to improve the detection accuracy of hepatitis B surface antigen at low concentrations. Summary of the invention

[0004] The embodiment of the present invention provides a HBsAg detection method based on multiple wavelengths, which aims to improve the detection accuracy of hepatitis B surface antigen at low concentrations. The multi-wavelength microplate reader can be used to detect the light absorption value of the HBsAg sample to be tested by the double antibody sandwich method, and the first detection result is determined based on the light absorption data at dual wavelengths (450nm and 630nm). Then, the negative control OD value and the positive control OD value are used to verify whether the detection result is valid within the preset concentration range of HBsAg. When the first detection result is invalid within the preset concentration range of HBsAg, the second detection result of the sample to be tested is determined based on the mapping relationship between the light absorption data at multiple wavelengths and the chemiluminescence detection result. The multi-wavelength detection, the double verification mechanism and the correction means of the chemiluminescence result mapping are adopted to effectively solve the problems existing in the traditional double antibody sandwich method in the immunological detection of hepatitis B surface antigen, and improve the sensitivity, accuracy and reliability of the detection of hepatitis B surface antigen at low concentrations.

[0005] In a first aspect, an embodiment of the present invention provides a multi-wavelength-based HBsAg detection method, the multi-wavelength-based HBsAg detection method comprising: The HBsAg sample to be tested by the double antibody sandwich method is tested for light absorption value by a multi-wavelength microplate reader to obtain light absorption data of the HBsAg sample to be tested at multiple wavelengths, wherein the multiple wavelengths include multiple wavelengths between 450 nm and 630 nm; Based on the light absorption data at the wavelength of 450 nm and the wavelength of 630 nm, the sample OD value, the negative control OD value and the positive control OD value at the wavelength of 450 nm and the wavelength of 630 nm are determined; Determine a first test result of the HBsAg sample based on the sample OD value, the negative control OD value, and the positive control OD value at the 450nm wavelength and the 630nm wavelength, and determine whether the first test result is valid within a preset concentration range of HBsAg based on the negative control OD value and the positive control OD value; If the first test result is invalid within the preset concentration range of HBsAg, the second test result of the HBsAg sample to be tested is determined based on the mapping relationship between the light absorption data at the multiple wavelengths and the chemiluminescence test result.

[0006] Optionally, the number of the negative control OD values ​​is multiple, the number of the positive control OD value is 1, and the step of determining whether the first test result is valid within a preset concentration range of HBsAg based on the negative control OD value and the positive control OD value includes: Obtaining a first critical value of the 450nm wavelength within a preset concentration range of HBsAg, and obtaining a second critical value of the 630nm wavelength within a preset concentration range of HBsAg; Based on the sample OD value, the positive control OD value, the first critical value and the second critical value, determine whether the first detection result is valid within a preset concentration range of HBsAg.

[0007] Optionally, the step of determining whether the first test result is valid within a preset concentration range of HBsAg based on the sample OD value, the positive control OD value, the first critical value, and the second critical value includes: Determining whether the first critical value is less than a preset first critical value threshold, and determining whether the second critical value is less than a preset second critical value threshold, the first critical value threshold being greater than the second critical value threshold; If the first critical value is less than a preset first critical value threshold, and the second critical value is less than a preset second critical value threshold, it is determined that the negative control OD value is valid within the preset concentration range of HBsAg, otherwise, it is determined that the negative control OD value is invalid within the preset concentration range of HBsAg; Determine whether the positive control OD value is greater than a preset positive value threshold; If the positive control OD value is greater than the preset positive value threshold, the positive control OD value is determined to be valid within the preset concentration range of HBsAg; otherwise, the positive control OD value is determined to be invalid within the preset concentration range of HBsAg; If the negative control OD value is valid within the preset concentration range of HBsAg, and the positive control OD value is valid within the preset concentration range of HBsAg, then the first test result is determined to be valid within the preset concentration range of HBsAg; otherwise, the first test result is determined to be invalid within the preset concentration range of HBsAg.

[0008] Optionally, the step of determining a second test result of the sample to be tested based on the light absorption data at multiple wavelengths includes: Determining a deviation value of the first detection result within a preset concentration range of HBsAg; Acquiring light absorption data at the multiple wavelengths, and determining a light absorption value sequence sorted from small to large by wavelength based on the light absorption data at the multiple wavelengths; Based on the deviation value and the light absorption value sequence, a second detection result of the HBsAg test sample is determined.

[0009] Optionally, the step of determining a deviation value of the first detection result within a preset concentration range of HBsAg includes: Determining a first negative deviation value of the first detection result within a preset concentration range of HBsAg based on a difference between the first critical value and the first critical value threshold; Determining a second negative deviation value of the first detection result within a preset concentration range of HBsAg based on a difference between the second critical value and the second critical value threshold; Determining a positive deviation value of the first test result within a preset concentration range of HBsAg based on a difference between the positive control OD value and the positive value threshold; Based on the first negative deviation value, the second negative deviation value and the positive deviation value, a deviation value of the first detection result within a preset concentration range of HBsAg is determined.

[0010] Optionally, determining the second test result of the HBsAg sample to be tested based on the deviation value and the light absorption value sequence includes: Inputting the deviation value and the light absorption value sequence into a pre-trained prediction model, wherein the prediction model includes a first feature extraction network, a second feature extraction network, a feature fusion network, and an output network, wherein the output ends of the first feature extraction network and the second feature extraction network are both connected to the input end of the feature fusion network, and the output end of the feature fusion network is connected to the input end of the output network; Performing a deconvolution operation on the deviation value through the first feature extraction network to obtain a deviation feature corresponding to the deviation value; Performing a convolution operation on the light absorption value sequence through the second feature extraction network to obtain a light absorption feature of the light absorption value sequence, wherein the deviation feature has the same resolution size as the light absorption feature; Obtaining a fusion feature by using the fusion network to perform fusion on the deviation feature and the light absorption feature; The fusion feature is predicted and processed by the output network to obtain a prediction result as the second detection result of the HBsAg sample to be tested.

[0011] Optionally, before the step of inputting the deviation value and the light absorption value sequence into a pre-trained prediction model, the method further includes: Obtaining a first sample test result and a second sample test result of the same HBsAg sample, wherein the first sample test result is a test result of a double antibody sandwich method, and the second sample test result is a test result of a chemiluminescence method, wherein the sample double antibody sandwich method test result is light absorption data at multiple wavelengths corresponding to an invalid state of the first test result, and the HBsAg sample is an HBsAg sample within a preset concentration range of HBsAg; Based on the first sample detection result, determine a sample deviation value corresponding to the first sample detection result, and determine the first sample detection result and the sample deviation value as a two-tuple of sample data; Determine the second sample detection result as label data of the sample data; Constructing a training data set based on the sample data and the label data; A pre-trained model is obtained, and the pre-trained model is trained using the training data set, and a trained prediction model is obtained after the training is completed.

[0012] In a second aspect, an embodiment of the present invention further provides a multi-wavelength HBsAg detection device, the multi-wavelength HBsAg detection device comprising: The detection module is used to detect the light absorption value of the HBsAg sample to be tested by the double antibody sandwich method through a multi-wavelength microplate reader to obtain the light absorption data of the HBsAg sample to be tested at multiple wavelengths, wherein the multiple wavelengths include multiple wavelengths between 450nm and 630nm; A first processing module is used to determine the sample OD value, the negative control OD value and the positive control OD value at the wavelength of 450nm and the wavelength of 630nm based on the light absorption data at the wavelength of 450nm and the wavelength of 630nm; A second processing module is used to determine a first detection result of the HBsAg sample to be tested based on the sample OD value, the negative control OD value and the positive control OD value at the wavelength of 450nm and the wavelength of 630nm, and determine whether the first detection result is valid within a preset concentration range of HBsAg based on the negative control OD value and the positive control OD value; The third processing module is used to determine the second detection result of the HBsAg sample to be tested based on the mapping relationship between the light absorption data at multiple wavelengths and the chemiluminescence detection result if the first detection result is invalid within the preset concentration range of HBsAg.

[0013] The present invention also provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the multi-wavelength based HBsAg detection method as described in any one of the embodiments of the present invention are implemented.

[0014] The present invention also provides a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the multi-wavelength HBsAg detection method as described in any one of the embodiments of the present invention are implemented.

[0015] By adopting the multi-wavelength-based HBsAg detection method of the present invention, the light absorption value of the HBsAg sample to be tested by the double antibody sandwich method can be detected by a multi-wavelength microplate reader, and a first detection result is determined based on the light absorption data under the dual wavelengths (450nm and 630nm), and then the negative control OD value and the positive control OD value are used to verify whether the detection result is valid within the preset concentration range of HBsAg. When the first detection result is invalid within the preset concentration range of HBsAg, a second detection result of the sample to be tested is determined based on the mapping relationship between the light absorption data under the multi-wavelengths and the chemiluminescence detection result, and the multi-wavelength detection, double verification mechanism and chemiluminescence result mapping correction means are adopted to effectively solve the problems existing in the traditional double antibody sandwich method in the hepatitis B surface antigen immunoassay, and improve the sensitivity, accuracy and reliability of the hepatitis B surface antigen detection at low concentrations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1It is a schematic diagram of a process of HBsAg detection based on multi-wavelength provided by an embodiment of the present invention; Figure 2 It is a structural schematic diagram of a multi-wavelength HBsAg detection device provided by an embodiment of the present invention; Figure 3 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] like Figure 1 As shown, Figure 1 The present invention provides a flow chart of a multi-wavelength HBsAg detection method. It is used for the determination of HBsAg (hepatitis B surface antigen) results. The multi-wavelength HBsAg detection method includes: 101. Use a multi-wavelength microplate reader to detect the light absorption value of the HBsAg test sample of the double antibody sandwich method to obtain the light absorption data of the HBsAg test sample at multiple wavelengths.

[0020] In an embodiment of the present invention, the light absorption value of the HBsAg sample to be tested by the double antibody sandwich method is detected by a multi-wavelength microplate reader. The multi-wavelength microplate reader can simultaneously detect the light absorption at multiple wavelengths. The wavelength range selected in this embodiment is 450nm to 630nm. The selection of this wavelength range is based on the change in optical properties produced after HBsAg binds to the antibody. The light absorption values ​​at different wavelengths can reflect the concentration difference of HBsAg in the sample.

[0021] During the detection process, the sample to be tested reacts with the microplate coated with specific antibodies to form an antigen-antibody complex, and then the enzyme-labeled antibody is added to form a double antibody sandwich structure. After the addition of the substrate, the enzyme catalyzes the substrate to produce a color reaction, and its light absorption value is proportional to the concentration of HBsAg in the sample. Specifically, an anti-HBs coated reaction plate is used, and after incubation, the sample to be tested is added, and anti-HBS-HRP is added. When HBsAg is present in the sample, the HBsAg binds to the coated anti-HBs and the anti-HBs-HRP to form an anti-HBs-HBsAg-anti-HBsHRP complex. The addition of TMB substrate produces a color reaction, otherwise there is no color reaction.

[0022] By using a multi-wavelength microplate reader to detect the light absorption value of this color development reaction at different wavelengths, the light absorption data of the sample to be tested at multiple wavelengths can be obtained. By adjusting the filter of the multi-wavelength microplate reader, the multi-wavelength microplate reader can be adjusted to collect light absorption data at different wavelengths. Different filters correspond to different wavelengths.

[0023] Specifically, the process of obtaining the light absorption data of the HBsAg sample to be tested at multiple wavelengths by the double antibody sandwich method is as follows: 1. Prepare working concentration washing solution (use purified water to make 25 times dilution. For example: take 480ml purified water, add 20ml washing solution and mix thoroughly to make 500ml working concentration washing solution, set aside).

[0024] 2. According to the experimental requirements, select a certain amount of reaction strips.

[0025] 3. Add 75 µL of the sample to be tested and negative and positive controls into the reaction wells (reserve 3 wells for negative control, 1 well for positive control, and it is recommended to reserve 1 well for blank control).

[0026] 4. Cover the reaction plate with sealing paper and incubate the reaction plate at 37°C for 60 minutes.

[0027] 5. Take out the reaction plate, tear off the seal, and add 50uL of enzyme conjugate to the wells where the samples to be tested and the negative and positive controls have been added.

[0028] 6. Shake on a microwell shaker for 10 seconds, or shake gently by hand for 10 seconds.

[0029] 7. Cover the reaction plate with sealing paper and incubate the reaction plate at 37°C for 30 minutes.

[0030] 8. Take out the reaction plate, tear off the sealing paper, and wash the reaction plate 5 times. Manual plate washing: discard the liquid in the wells, fill each well with the working concentration washing solution prepared in step 1, let it stand for 30-60 seconds, spin dry and repeat 5 times, and pat dry on clean absorbent paper. Plate washing machine: select the 5-time washing program, fill each well with the working concentration washing solution prepared in step 1, and the working concentration washing solution stays in the microwells of the reaction plate for 30-60 seconds each time, and ensure that there is no residue each time. Pat dry on clean absorbent paper after washing.

[0031] 9. Immediately after washing, add 50uL of color developer A and 50uL of color developer B into all wells and mix well.

[0032] 10. Shake on a microwell shaker for 10 seconds, or shake gently by hand for 10 seconds.

[0033] 11. Cover the reaction plate with sealing paper and incubate the reaction plate at 37°C for 30 minutes.

[0034] 12. Add 50uL of stop solution to all wells and shake the reaction plate for 5 seconds to mix thoroughly.

[0035] 13. Read the samples with a multi-wavelength microplate reader at wavelengths from 450nm to 630nm. If the colorimetric blank needs to be subtracted, first calibrate the control well with the colorimetric blank, and then read the OD value (optical density) of each well.

[0036] 102. Based on the light absorption data at a wavelength of 450 nm and a wavelength of 630 nm, determine the sample OD value, the negative control OD value, and the positive control OD value at a wavelength of 450 nm and the wavelength of 630 nm.

[0037] In an embodiment of the present invention, based on the light absorption data at a wavelength of 450 nm and a wavelength of 630 nm, the sample OD value S, the negative control OD value NCn, and the positive control OD value PC at a wavelength of 450 nm and the wavelength of 630 nm are determined. The OD value is a way of expressing the light absorption value, and its size is proportional to the concentration of the substance to be tested in the sample. The negative control and the positive control are two control methods designed to verify the accuracy and reliability of the enzyme-linked immunosorbent assay (ELISA). The negative control is usually a sample that does not contain HBsAg, and the negative control OD value is used to determine the background noise and the detection limit; the positive control is a sample known to contain the substance to be tested, and the positive control OD value is used to verify the sensitivity and specificity of the experiment.

[0038] The following index data can be obtained through the readings of the multi-wavelength microplate reader: NCn: Negative control OD value, n can be equal to 3.

[0039] PC: positive control OD value.

[0040] In this embodiment, by comparing the relationship between the sample OD value and the negative control OD value and the positive control OD value, it can be preliminarily determined whether the concentration of HBsAg in the sample to be tested is within a preset concentration range.

[0041] It can be understood that, among the above-mentioned 450nm and 630nm wavelengths, the 450nm wavelength is a wavelength that absorbs color development, and the 630nm wavelength is a wavelength that is insensitive to specific color development. Furthermore, the light absorption data obtained at a wavelength of 630 nm is non-specific, and comes from the absorption caused by fingerprints, dust, dirt, etc. on the plate holes.

[0042] 103. Based on the OD values of the sample, negative control, and positive control at 450 nm wavelength and the 630 nm wavelength, determine the first test result of the HBsAg sample to be tested, and based on the OD values of the negative control and positive control, determine whether the first test result is valid within the preset concentration range of HBsAg.

[0043] In the embodiments of the present invention, the above first test result is determined by the light absorption data at 450 nm wavelength and 630 nm wavelength. The specific steps to determine the first test result of the HBsAg sample to be tested through the light absorption data at 450 nm wavelength and 630 nm wavelength are as follows: NCx: Average OD value of the negative control.

[0044] NCn: OD value of the negative control, where n can be equal to 3.

[0045] PC: OD value of the positive control.

[0046] Calculate NCx: NCx = (NC1 + NC2 + …… + NCn) ÷ n. If NCx < 0, then calculate as 0.

[0047] Calculate COV (Cut - Off Value reference value): COV = NCx + a, where a can be equal to 0.100.

[0048] S: OD value of the sample to be tested.

[0049] S / COV: Ratio of the sample to be tested and COV. When S / COV > b, it indicates that the HBsAg result of the sample to be tested is positive, that is, the first test result is positive, indicating the presence of hepatitis B surface antigen. b can be equal to 1.0.

[0050] When S / COV < b, it indicates that the HBsAg result of the sample to be tested is negative, that is, the first test result is negative, indicating the absence of hepatitis B surface antigen.

[0051] The detection validity of the first test result within the preset concentration range is used to illustrate whether the corresponding HBsAg of the sample to be tested may be a low - concentration HBsAg sample (since the HBsAg of the sample to be tested cannot be quantitatively detected in advance, the first test result obtained by the double - antibody sandwich method may have false negative or false positive results).

[0052] If NCx ≤ e and PC > f, the test result is valid. Conversely, if NCx > e and PC ≤ f, the test result is invalid. Among them, e and f can be determined according to the empirical values of the preset concentration range. For example, e can be equal to 0.100 and f can be equal to 1.000.

[0053] Color reagent blank: In the dual wavelength reading of the multi-wavelength microplate reader, if the colorimetric blank is ≤ c, c can be equal to 0.040, then the test result is valid.

[0054] In a possible embodiment, a machine learning model may be used to predict the sample OD value, the negative control OD value, and the positive control OD value to predict whether the first test result is valid within a preset concentration range of HBsAg.

[0055] The above-mentioned preset concentration range is the concentration range in which the multi-wavelength microplate reader can determine effective detection results, for example, it can be the concentration range corresponding to the hepatitis B virus amount within 10 to the power of 2.

[0056] 104. If the first test result is invalid within the preset concentration range of HBsAg, a second test result of the sample to be tested is determined based on a mapping relationship between the light absorption data at multiple wavelengths and the chemiluminescence test result.

[0057] In the embodiment of the present invention, when the first test result is invalid within the preset concentration range of HBsAg, that is, the OD value of the sample to be tested exceeds the range of the negative control and the positive control, it means that the concentration of HBsAg in the sample may be too low to be directly accurately measured by the double antibody sandwich method. At this time, based on the mapping relationship between the light absorption data at multiple wavelengths and the chemiluminescence test results, the second test result of the sample to be tested is determined.

[0058] Chemiluminescence is a highly sensitive detection method. Its principle is to use the light energy generated by chemical reactions to detect the concentration of the substance to be tested. It has high sensitivity and can perform quantitative detection of HBsAg, but the detection process is relatively complicated and takes a longer time. In this embodiment, through the pre-established mapping relationship between the multi-wavelength light absorption data and the chemiluminescence detection results, the multi-wavelength light absorption data of the sample to be tested can be converted into the detection results of the chemiluminescence method, thereby achieving accurate determination of low-concentration samples. The purpose of this step is to make up for the shortcomings of the double antibody sandwich method in the low-concentration detection range through the high sensitivity of the chemiluminescence method, and to improve the accuracy and reliability of the detection.

[0059] It is worth noting that the present embodiment also involves multiple settings of negative control OD values ​​and positive control OD values ​​and the judgment of critical values. The number of negative control OD values ​​is multiple, which is to more accurately determine the range of background noise and detection limit; the number of positive control OD values ​​is 1, which is to verify the sensitivity and specificity of the experiment. When judging whether the first test result is valid within the preset concentration range of HBsAg, it is necessary to obtain the critical values ​​of 450nm wavelength and 630nm wavelength within the preset concentration range of HBsAg and compare them with the sample OD value. At the same time, it is also necessary to judge whether the negative control OD value and the positive control OD value are within the preset threshold range to ensure the accuracy and reliability of the experimental results.

[0060] It should be noted that the present embodiment also introduces a prediction model to correct and predict the deviation value. When the first detection result is invalid, the introduction of the prediction model can further improve the accuracy and reliability of the detection, especially for those samples that cannot be accurately determined in the first detection result.

[0061] In the process of training the prediction model, it is necessary to obtain the first sample test result and the second sample test result of the same HBsAg sample. The first sample test result is the test result of the double antibody sandwich method, and the second sample test result is the test result of the chemiluminescence method mapped by the light absorption data under multiple wavelengths. By constructing a training data set and training the pre-trained model, a trained prediction model can be obtained. The accuracy and generalization ability of the detection are improved through machine learning methods, so that the prediction model can better adapt to the changes of different samples and experimental conditions.

[0062] The multi-wavelength HBsAg detection method combines the advantages of the double antibody sandwich method and the chemiluminescence method. The multi-wavelength microplate reader detects the light absorption value of the sample under test at different wavelengths, and the deviation value is corrected and predicted in combination with the prediction model, thereby achieving accurate determination of HBsAg. This method has the advantages of high sensitivity, strong specificity, and simple operation.

[0063] Embodiment In the embodiment of the present invention, the light absorption value of the HBsAg sample to be tested by the double antibody sandwich method can be detected by a multi-wavelength microplate reader, and a first detection result is determined based on the light absorption data at dual wavelengths (450nm and 630nm), and then the negative control OD value and the positive control OD value are used to verify whether the detection result is valid within the preset concentration range of HBsAg. When the first detection result is invalid within the preset concentration range of HBsAg, the second detection result of the sample to be tested is determined based on the mapping relationship between the light absorption data at multiple wavelengths and the chemiluminescence detection result. The multi-wavelength detection, double verification mechanism and chemiluminescence result mapping correction method are adopted to effectively solve the problems existing in the traditional double antibody sandwich method in the hepatitis B surface antigen immunoassay, and improve the sensitivity, accuracy and reliability of hepatitis B surface antigen detection at low concentrations.

[0064] Optionally, the number of negative control OD values ​​is multiple, the number of positive control OD values ​​is 1, and the step of determining whether the first test result is valid within the preset concentration range of HBsAg based on the negative control OD value and the positive control OD value includes: obtaining a first critical value of a wavelength of 450nm within the preset concentration range of HBsAg, and obtaining a second critical value of a wavelength of 630nm within the preset concentration range of HBsAg; based on the sample OD value, the positive control OD value, the first critical value and the second critical value, determining whether the first test result is valid within the preset concentration range of HBsAg.

[0065] In the embodiments of the present invention, it should be noted that in immunological testing, negative controls and positive controls are the basis for evaluating the validity of experimental results. Negative controls are usually samples that do not contain the target antigen and are used to determine the background signal level; while positive controls are samples known to contain the target antigen and are used to verify the sensitivity and accuracy of the detection system. In this embodiment, the setting of the negative control OD value and the positive control OD value is intended to establish a reference framework for accurate interpretation of the sample OD value of the HBsAg sample to be tested.

[0066] There are multiple negative control OD values, while there is only one positive control OD value. This design is based on statistical principles. Multiple negative controls can reduce the impact of accidental errors and improve the stability and reliability of the results. Multiple negative control OD values ​​can form a distribution to better estimate the range and average value of background noise. In contrast, as a known standard, the positive control has a higher consistency, so only one is needed to meet the verification requirements.

[0067] Specifically, in this embodiment, in order to determine whether the OD value of the sample to be tested falls within the preset concentration range of HBsAg, the first critical value and the second critical value of the wavelength of 450nm and the wavelength of 630nm within the preset concentration range of HBsAg can be determined first. The selection of these two wavelengths is based on the principle of spectral analysis. 450nm is usually located in the blue light region of visible light, while 630nm is located in the red light region. They have different absorption characteristics for different types of chemical bonds and molecular structures, and therefore can provide more information for distinguishing the components in the sample. The above critical value can be the ratio PC / NCx between the OD value of the positive control and the average OD value of the negative control. The above first critical value and the second critical value are only used to represent the judgment conditions in this double antibody sandwich method, and are only related to the readings of the multi-wavelength microplate reader in this double antibody sandwich method.

[0068] The effectiveness of negative control and positive control can be evaluated by comparing the actual measured OD value with the preset critical value and threshold value. Specifically, it can be judged whether the first critical value is less than the preset first critical value threshold value, and whether the second critical value is less than the preset second critical value threshold value. Here, the first critical value threshold value and the second critical value threshold value are pre-set judgment criteria, and the first critical value threshold value and the second critical value threshold value represent the maximum allowable value that the negative control OD value can be considered to be effective at this wavelength. Due to the different light absorption characteristics of different wavelengths, different threshold values ​​are set.

[0069] If both the first critical value and the second critical value are less than the corresponding threshold value, it means that the negative control OD value is valid within the preset concentration range of HBsAg, that is, the background noise is at an acceptable level. Otherwise, it means that the negative control is invalid, which may be caused by unstable experimental conditions, instrument errors or sample contamination.

[0070] For the positive control, it is necessary to determine whether its OD value is greater than the preset positive value threshold. This threshold is the standard for determining whether the positive control can be correctly identified. If the OD value of the positive control is greater than the positive value threshold, it means that the detection system can accurately identify samples containing HBsAg and the positive control is effective. Otherwise, it indicates that the detection system may have insufficient sensitivity.

[0071] The effectiveness of the negative control and the positive control are comprehensively considered to determine whether the first test result is valid within the preset concentration range of HBsAg. Only when both the negative control and the positive control are valid can the first test result be considered reliable. Otherwise, the first test result cannot be considered reliable.

[0072] Optionally, the step of determining whether the first test result is valid within the preset concentration range of HBsAg based on the sample OD value, the positive control OD value, the first critical value and the second critical value includes: determining whether the first critical value is less than a preset first critical value threshold, and determining whether the second critical value is less than a preset second critical value threshold, and the first critical value threshold is greater than the second critical value threshold; if the first critical value is less than the preset first critical value threshold, and the second critical value is less than the preset second critical value threshold, then determining that the negative control OD value is valid within the preset concentration range of HBsAg, otherwise, determining the negative control OD value Invalid within the preset concentration range of HBsAg; determine whether the positive control OD value is greater than the preset positive value threshold; if the positive control OD value is greater than the preset positive value threshold, determine that the positive control OD value is valid within the preset concentration range of HBsAg, otherwise, determine that the positive control OD value is invalid within the preset concentration range of HBsAg; if the negative control OD value is valid within the preset concentration range of HBsAg, and the positive control OD value is valid within the preset concentration range of HBsAg, then determine that the first test result is valid within the preset concentration range of HBsAg, otherwise, determine that the first test result is invalid within the preset concentration range of HBsAg.

[0073] In the embodiment of the present invention, light absorption data at two wavelengths of 450nm and 630nm are obtained, which is obtained by detecting the HBsAg sample to be tested by the double antibody sandwich method using a multi-wavelength microplate reader. Based on the light absorption data at two wavelengths of 450nm and 630nm, the multi-wavelength microplate reader can calculate the sample OD value, the negative control OD value and the positive control OD value as readings. Among them, the number of negative control OD values ​​is multiple, which is used to reflect the level of nonspecific reaction during the detection process, and the number of positive control OD values ​​is 1, which is used to verify the effectiveness of the detection system.

[0074] Furthermore, the above critical value can be the ratio PC / NCx between the OD value of the positive control and the average OD value of the negative control. The above first critical value and second critical value are only used to represent the judgment conditions in this double antibody sandwich method, and are only related to the readings of the multi-wavelength microplate reader in this double antibody sandwich method. Determining whether the first test result is valid within the preset concentration range of HBsAg mainly depends on the setting of two critical value thresholds: the first critical value threshold corresponds to a wavelength of 450nm, and the second critical value threshold corresponds to a wavelength of 630nm. These two critical value thresholds are obtained through statistical analysis of a large amount of experimental data, and represent the standards for the limits of light absorption values ​​when the HBsAg concentration reaches a preset range at different wavelengths.

[0075] After determining the critical value, a comparison operation can be performed. Check whether the first critical value is less than the preset first critical value threshold, and whether the second critical value is less than the preset second critical value threshold. It should be noted that the first critical value threshold is set to be greater than the second critical value threshold, and the variation law and sensitivity of the light absorption value may be different at different wavelengths. By setting different thresholds, the accuracy of the detection result can be controlled more accurately.

[0076] If both the first critical value and the second critical value are less than their respective corresponding threshold values, then it can be determined that the negative control OD value is valid within the preset concentration range of HBsAg. This further indicates that the nonspecific reaction during the detection process is controlled within an acceptable range and will not significantly affect the accuracy of the test results. On the contrary, if any critical value is greater than or equal to the corresponding threshold value, then the negative control OD value can be determined to be invalid, which may be due to excessive nonspecific reactions or other interfering factors during the detection process.

[0077] Further, the positive control OD value can be checked for validity. Specifically, it can be determined whether the positive control OD value is greater than a preset positive value threshold. The positive value threshold is also obtained through statistical analysis of experimental data, representing the minimum light absorption value of the positive control under effective conditions. If the positive control OD value is greater than the positive value threshold, it means that the test can correctly identify the positive sample, that is, the first test result is valid. On the contrary, if the positive control OD value is less than or equal to the positive value threshold, it may mean that there is a problem or failure in the first test result.

[0078] Furthermore, the validity of the first test result can be determined by combining the validity of the negative control OD value and the positive control OD value. Only when the negative control OD value and the positive control OD value are both valid, the first test result can be determined to be valid within the preset concentration range of HBsAg. The double verification mechanism can ensure the accuracy and reliability of the test results and reduce the risk of false positives or false negatives.

[0079] If the first test result is determined to be invalid, further measures can be taken to determine the second test result of the sample to be tested. Based on the mapping relationship between the light absorption data at multiple wavelengths and the chemiluminescence test result, a relatively accurate second test result can still be obtained using the light absorption data at multiple wavelengths when the first test result is invalid.

[0080] In this embodiment, by reasonably setting the critical value and threshold value, and making a comprehensive judgment based on the sample OD value, the negative control OD value, and the positive control OD value, the accuracy and reliability of the detection result can be effectively improved.

[0081] Optionally, the step of determining a second test result of the sample to be tested based on light absorption data at multiple wavelengths includes: determining a deviation value of the first test result within a preset concentration range of HBsAg; acquiring the light absorption data at multiple wavelengths, and determining a sequence of light absorption values ​​sorted from small to large wavelengths based on the light absorption data at multiple wavelengths; and determining the second test result of the sample to be tested based on the deviation value and the sequence of light absorption values.

[0082] In an embodiment of the present invention, a deviation value of the first test result within a preset HBsAg concentration range can be determined. The above deviation value reflects the degree of difference between the first test result and the preset concentration range, and can be used to subsequently determine the second test result. In order to calculate this deviation value, the difference between the first critical value, the second critical value, and the positive control OD value and each threshold value can be comprehensively considered to obtain a value that fully reflects the degree of deviation of the test result.

[0083] The light absorption data under multiple wavelengths can be obtained. The light absorption data under multiple wavelengths is obtained by testing the HBsAg test sample with a multi-wavelength microplate reader, covering multiple wavelengths from 450nm to 630nm. The light absorption data under multiple wavelengths can reflect the light absorption characteristics of the HBsAg test sample at different wavelengths.

[0084] After obtaining the light absorption data at multiple wavelengths, the light absorption data at multiple wavelengths can be processed to determine a light absorption value sequence sorted from small to large wavelengths. The construction of the above light absorption value sequence is to facilitate subsequent data analysis and feature extraction, and ensure the consistency and comparability of the data.

[0085] After obtaining the deviation value and the light absorption value sequence, the second test result of the sample to be tested can be further determined. Specifically, the deviation value and the light absorption value sequence can be input into a pre-trained prediction model for prediction, thereby obtaining the second test result of the HBsAg sample to be tested. The above prediction model can be constructed by a convolutional neural network CNN, a recurrent neural network RNN, etc.

[0086] In this embodiment, the advantages of light absorption data at multiple wavelengths and deep learning technology are fully utilized to improve the accuracy and reliability of detection.

[0087] Optionally, the step of determining the deviation value of the first test result within the preset concentration range of HBsAg includes: determining a first negative deviation value of the first test result within the preset concentration range of HBsAg based on the difference between the first critical value and the first critical value threshold; determining a second negative deviation value of the first test result within the preset concentration range of HBsAg based on the difference between the second critical value and the second critical value threshold; determining a positive deviation value of the first test result within the preset concentration range of HBsAg based on the difference between the positive control OD value and the positive value threshold; determining the deviation value of the first test result within the preset concentration range of HBsAg based on the first negative deviation value, the second negative deviation value and the positive deviation value.

[0088] In an embodiment of the present invention, the difference between the first critical value and the first critical value threshold value can be calculated first, and this difference is the first negative deviation value. The first critical value is a limit value of the preset concentration range of HBsAg at a wavelength of 450nm, and the first critical value threshold value is a preset standard of this limit value. By comparing the difference between the two, the degree of deviation of the first detection result from the expected concentration range at a wavelength of 450nm can be quantified.

[0089] At the same time, the difference between the second critical value and the second critical value threshold is calculated to obtain a second negative deviation value. Similar to the first negative deviation value, the second negative deviation value reflects the deviation of the first detection result from the preset concentration range at a wavelength of 630nm. The second critical value and the second critical value threshold are respectively the limit value of the preset concentration range of HBsAg at a wavelength of 630nm and its preset standard.

[0090] At the same time, the difference between the positive control OD value and the positive value threshold is also calculated, and this difference is the positive deviation value. The positive control OD value is the test result of the HBsAg sample of known concentration set during the experiment, and the positive value threshold is the standard for judging whether the test result is valid. By calculating the positive deviation value, the degree of conformity between the positive control part in the first test result and the expected value can be evaluated.

[0091] After obtaining the first negative deviation value, the second negative deviation value and the positive deviation value, the three deviation values ​​can be combined to determine the total deviation value of the first test result within the preset HBsAg concentration range. The above total deviation value is a comprehensive indicator, which integrates the information of the negative control and the positive control under dual wavelengths, and can fully reflect the overall deviation between the first test result and the expected concentration range.

[0092] The total deviation value not only quantitatively evaluates the validity of the first test result, but also provides a key input for determining the second test result based on the light absorption data at multiple wavelengths. In some possible embodiments, by accurately calculating the deviation value, it is possible to more accurately identify those samples to be tested whose results deviate from the expected range in the first test due to various reasons (such as too low sample concentration, experimental operation error, etc.).

[0093] Let the first negative deviation be , the second negative deviation value is , the positive deviation value is , is the average of the first negative deviation value, the second negative deviation value and the positive deviation value. The influencing factor of the preset concentration range is ( is a constant determined according to experimental conditions, sample characteristics and other factors, used to adjust the calculation of the deviation value). Then the deviation value of the first test result within the preset HBsAg concentration range is It can be calculated as:

[0094] in, It means taking the logarithm after adding 1 to each deviation value, so as to handle the case where the deviation value is 0, and at the same time make the increase of the deviation value produce a nonlinear effect in the formula, thereby improving the nonlinear response capability of the deviation value within the preset HBsAg concentration range. It means taking the ratio of three deviation values ​​to the average value as the input of the exponential function, which can dynamically amplify or reduce the impact of the deviation value, depending on the average level of the deviation value. As an influencing factor for the preset concentration range.

[0095] In a possible embodiment, in addition to the above-mentioned method based on the difference between the critical value and the threshold value, other statistical methods or machine learning algorithms can also be considered to evaluate the difference between the first test result and the expected value. For example, statistics such as standard deviation and coefficient of variation can be used to measure the degree of discreteness of the test results; or a special machine learning model can be trained to predict and evaluate the deviation value. Of course, the selection of these methods should be determined according to the specific application scenario, data characteristics and actual needs.

[0096] In this embodiment, the method based on the difference between the critical value and the threshold value is used to determine the deviation value, which has its unique advantages. It is simple and easy to operate, has high computational efficiency, and can intuitively reflect the degree of deviation between the detection result and the expected concentration range.

[0097] Optionally, based on the deviation value and the light absorption value sequence, determining the second detection result of the sample to be tested includes: inputting the deviation value and the light absorption value sequence into a pre-trained prediction model, the prediction model including a first feature extraction network, a second feature extraction network, a feature fusion network, and an output network, the output ends of the first feature extraction network and the second feature extraction network are both connected to the input end of the feature fusion network, and the output end of the feature fusion network is connected to the input end of the output network; performing a deconvolution operation on the deviation value through the first feature extraction network to obtain a deviation feature corresponding to the deviation value; performing a convolution operation on the light absorption value sequence through the second feature extraction network to obtain a light absorption feature of the light absorption value sequence, the deviation feature and the light absorption feature having the same resolution size; performing a fusion operation on the deviation feature and the light absorption feature through the fusion network to obtain a fusion feature; performing prediction processing on the fusion feature through the output network to obtain a prediction result as the second detection result of the sample to be tested.

[0098] In an embodiment of the present invention, the system needs to input the previously calculated deviation value and the light absorption value sequence sorted from small to large by wavelength as input data into a pre-trained prediction model. The above prediction model can be a deep learning model, and the above prediction model can include a first feature extraction network, a second feature extraction network, a feature fusion network and an output network. The above prediction model is intended to fully extract and fuse the feature information in the deviation value and the light absorption value sequence to improve the accuracy of the second detection result.

[0099] The first feature extraction network performs a deconvolution operation on the deviation value. Deconvolution operation is a commonly used upsampling technique that can upsample the input data through the learned filter to obtain a higher resolution feature map. In this embodiment, the deviation feature corresponding to the deviation value can be obtained through the deconvolution operation, and the deviation feature reflects the distribution and change of the deviation value in space.

[0100] At the same time, the second feature extraction network performs a convolution operation on the light absorption value sequence. Convolution operation is one of the most commonly used feature extraction methods in deep learning, and can perform local feature extraction on the input data by means of a sliding window. In this embodiment, the light absorption features in the light absorption value sequence can be extracted through the convolution operation, and the light absorption features reflect the changing law and internal connection of the light absorption values ​​at different wavelengths. It is worth noting that in order to ensure that the deviation feature and the light absorption feature can be effectively fused in subsequent steps, it is necessary to ensure that the deviation feature and the light absorption feature have the same resolution size. When the resolution sizes of the deviation feature and the light absorption feature are different, the deviation feature and / or the light absorption feature can be linearly transformed according to the linear matrix to obtain deviation features and light absorption features with the same resolution size.

[0101] The deviation feature and the light absorption feature are fused through the feature fusion network. Feature fusion can fuse features from different sources or different levels to obtain a more comprehensive and rich feature representation. Feature fusion can be splicing fusion or superposition fusion. In this embodiment, through the feature fusion operation, the useful information in the deviation feature and the light absorption feature is integrated to obtain a fused feature. The fused feature contains both the spatial distribution and change of the deviation value, and the change law and internal connection of the light absorption value at different wavelengths, so it has stronger expression and generalization capabilities.

[0102] The fused features are predicted and processed through the output network to obtain the predicted result as the second test result of the sample to be tested. This second test result is a continuous value or a classification label, and the second test result reflects the concentration or existence state of HBsAg in the sample to be tested. In order to obtain an accurate second test result, the output network usually uses a fully connected layer or a classifier to achieve the mapping and conversion of the fused features. At the same time, in order to improve the robustness and generalization ability of the model, regularization, batch normalization and other techniques can also be used to optimize the training process of the model.

[0103] In this embodiment, a prediction method based on a deep learning model is used to determine the second detection result of the sample to be tested, and the accuracy and reliability of the second detection result are improved by fully extracting and fusing the characteristic information in the deviation value and light absorption value sequence.

[0104] Optionally, before the step of inputting the deviation value and the light absorption value sequence into the pre-trained prediction model, it also includes: obtaining a first sample test result and a second sample test result of the same HBsAg sample, the first sample test result is the test result of the double antibody sandwich method, and the second sample test result is the test result of the chemiluminescence method, the sample double antibody sandwich method test result is the light absorption data at multiple wavelengths corresponding to the invalid state of the first test result, and the HBsAg sample is an HBsAg sample within a preset concentration range of HBsAg; based on the first sample test result, determining the sample deviation value corresponding to the first sample test result, and determining the first sample test result and the sample deviation value as a binary group of sample data; determining the second sample test result as label data of the sample data; based on the sample data and the label data, constructing a training data set; obtaining a pre-trained model, and training the pre-trained model through the training data set, and obtaining a trained prediction model after the training is completed.

[0105] In an embodiment of the present invention, two different test results of the same HBsAg sample may be obtained first, namely, a first sample test result and a second sample test result. In this embodiment, for HBsAg samples collected from the same human body at the same time, the double antibody sandwich method and the chemiluminescence method are used for detection respectively to obtain the corresponding first sample test result and second sample test result, the first sample test result is obtained by the double antibody sandwich method, and the second sample test result is obtained by the chemiluminescence method.

[0106] It should be noted that these HBsAg samples are within the preset concentration range of HBsAg (the low concentration range of HBsAg), and the first test results of these HBsAg samples (i.e., the results of the double antibody sandwich method) are invalid. This means that these HBsAg samples may not obtain accurate results in the conventional double antibody sandwich method due to some reasons (such as too low concentration, interfering substances, etc.). Therefore, it is necessary to use these special samples, combined with their corresponding multi-wavelength light absorption data, to train the prediction model to improve the accuracy and robustness of the model when dealing with similar complex situations.

[0107] Based on the first sample test results (i.e., invalid double antibody sandwich method results), the sample deviation values ​​corresponding to these results are determined. The calculation method of the above deviation values ​​is the same as the calculation method of the above deviation values, which is used to reflect the difference between the actual sample results and the expected results.

[0108] The first sample detection result and its corresponding sample deviation value are combined into a two-tuple sample data. The organization of the two-tuple sample data helps the model to pay attention to both the original detection data of the sample and its deviation during the learning process.

[0109] The second sample test result (i.e., the result obtained by chemiluminescence) is determined as the label data of the corresponding binary sample data. Since chemiluminescence usually has higher sensitivity and specificity, its results can be regarded as the true value or standard value of these samples. Using the results of chemiluminescence as label data can guide the prediction model to gradually approach these true values ​​during the learning process, thereby improving the prediction accuracy of the model.

[0110] After completing data preparation and label determination, a training data set can be constructed based on the sample data and the corresponding label data. The training data set will contain multiple binary sample data and their corresponding label data.

[0111] A pre-trained model will be obtained and further trained with the training data set constructed above. The pre-trained model may be a deep learning model that has been preliminarily trained on large-scale data and has certain feature extraction and representation learning capabilities. By further training on the training data set of the current specific task (i.e., multi-wavelength-based HBsAg detection), the model can learn more features and rules related to the current task, thereby improving its performance in processing similar tasks.

[0112] During the training process, a variety of optimization algorithms and learning strategies can be used to adjust the parameters and structure of the model to minimize the difference between the model prediction results and the true labels. For example, the gradient descent algorithm can be used to update the weights and bias terms of the model; regularization techniques can be used to prevent the model from overfitting; and the learning rate decay strategy can be used to dynamically adjust the learning step size. This ensures that the model can converge stably during the training process and achieve a high prediction accuracy.

[0113] After the training is completed, a trained prediction model can be obtained. The trained prediction model has fully learned the characteristics and rules in the multi-wavelength HBsAg detection task and has the ability to accurately predict new samples. In subsequent practical applications, this trained model can be directly used to determine the second test result of the sample to be tested, thereby improving the accuracy and reliability of the entire detection method.

[0114] It should be noted that the above-mentioned second test result is closer to the result of the chemiluminescence method, and can be used to verify whether the first test result is accurate. For example, if the first test result is negative and the second test result is also negative, it can be determined that the first test result is accurate. If the first test result is negative and the second test result is positive, it can be determined that the first test result is a false negative, and then it can be determined that the first test result is inaccurate.

[0115] It should also be noted that since the second test result is only obtained when the first test result is invalid within the preset concentration range of HBsAg, the detection speed is not affected for the test results outside the preset concentration range. At the same time, since the mapping relationship between the light absorption data under multiple wavelengths and the chemiluminescence test results is used, in actual use, there is no chemiluminescence detection process, no additional collection of HBsAg test samples, and no need to prepare a chemiluminescence detection environment. The equipment only needs to support multiple wavelengths of the microplate reader. Therefore, a more sensitive HBsAg test result can be obtained at a lower cost.

[0116] like Figure 2 As shown, Figure 21 is a schematic diagram of the structure of a multi-wavelength HBsAg detection device provided by an embodiment of the present invention. The multi-wavelength HBsAg detection device comprises: The detection module 201 is used to detect the light absorption value of the HBsAg sample to be tested by the double antibody sandwich method through a multi-wavelength microplate reader to obtain the light absorption data of the HBsAg sample to be tested at multiple wavelengths, wherein the multiple wavelengths include multiple wavelengths between 450nm and 630nm; A first processing module 202 is used to determine the sample OD value, the negative control OD value and the positive control OD value at the wavelength of 450nm and the wavelength of 630nm based on the light absorption data at the wavelength of 450nm and the wavelength of 630nm; The second processing module 203 is used to determine the first detection result of the HBsAg sample to be tested based on the sample OD value, the negative control OD value and the positive control OD value at the wavelength of 450nm and the wavelength of 630nm, and determine whether the first detection result is valid within a preset concentration range of HBsAg based on the negative control OD value and the positive control OD value; The third processing module 204 is used to determine the second test result of the sample to be tested based on the mapping relationship between the light absorption data at multiple wavelengths and the chemiluminescence test result if the first test result is invalid within the preset concentration range of HBsAg.

[0117] The multi-wavelength HBsAg detection device provided in the embodiment of the present invention can implement each process implemented by the multi-wavelength HBsAg detection method in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0118] See also Figure 3 , Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, it includes: a memory 302, a processor 301, and a computer program based on a multi-wavelength HBsAg detection method stored in the memory 302 and executable on the processor 301, wherein: The processor 301 is used to call the computer program stored in the memory 302 and execute the following steps: The HBsAg sample to be tested by the double antibody sandwich method is tested for light absorption value by a multi-wavelength microplate reader to obtain light absorption data of the HBsAg sample to be tested at multiple wavelengths, wherein the multiple wavelengths include multiple wavelengths between 450 nm and 630 nm; Based on the light absorption data at the wavelength of 450 nm and the wavelength of 630 nm, the sample OD value, the negative control OD value and the positive control OD value at the wavelength of 450 nm and the wavelength of 630 nm are determined; Determine a first test result of the HBsAg sample based on the sample OD value, the negative control OD value, and the positive control OD value at the 450nm wavelength and the 630nm wavelength, and determine whether the first test result is valid within a preset concentration range of HBsAg based on the negative control OD value and the positive control OD value; If the first test result is invalid within the preset concentration range of HBsAg, the second test result of the HBsAg sample to be tested is determined based on the mapping relationship between the light absorption data at the multiple wavelengths and the chemiluminescence test result.

[0119] The electronic device provided in the embodiment of the present invention can implement each process implemented by the multi-wavelength HBsAg detection method in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.

[0120] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the multi-wavelength HBsAg detection method provided by the embodiment of the present invention are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0121] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the computer-readable storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0122] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A multi-wavelength HBsAg detection method, characterized in that: The multi-wavelength-based HBsAg detection method comprises: The HBsAg sample to be tested by the double antibody sandwich method is tested for light absorption value by a multi-wavelength microplate reader to obtain light absorption data of the HBsAg sample to be tested at multiple wavelengths, wherein the multiple wavelengths include multiple wavelengths between 450 nm and 630 nm; Based on the light absorption data at the wavelength of 450 nm and the wavelength of 630 nm, the sample OD value, the negative control OD value and the positive control OD value at the wavelength of 450 nm and the wavelength of 630 nm are determined; Determine a first test result of the HBsAg sample based on the sample OD value, the negative control OD value, and the positive control OD value at the 450nm wavelength and the 630nm wavelength, and determine whether the first test result is valid within a preset concentration range of HBsAg based on the negative control OD value and the positive control OD value; If the first test result is invalid within the preset concentration range of HBsAg, the second test result of the HBsAg sample to be tested is determined based on the mapping relationship between the light absorption data at the multiple wavelengths and the chemiluminescence test result.

2. The multi-wavelength HBsAg detection method according to claim 1, characterized in that: The number of the negative control OD values ​​is multiple, the number of the positive control OD values ​​is 1, and the step of determining whether the first test result is valid within a preset concentration range of HBsAg based on the negative control OD value and the positive control OD value comprises: Obtaining a first critical value of the 450nm wavelength within a preset concentration range of HBsAg, and obtaining a second critical value of the 630nm wavelength within a preset concentration range of HBsAg; Based on the sample OD value, the positive control OD value, the first critical value and the second critical value, determine whether the first detection result is valid within a preset concentration range of HBsAg.

3. The multi-wavelength HBsAg detection method according to claim 1, characterized in that: The step of determining whether the first test result is valid within a preset concentration range of HBsAg based on the sample OD value, the positive control OD value, the first critical value, and the second critical value comprises: Determining whether the first critical value is less than a preset first critical value threshold, and determining whether the second critical value is less than a preset second critical value threshold, the first critical value threshold being greater than the second critical value threshold; If the first critical value is less than a preset first critical value threshold, and the second critical value is less than a preset second critical value threshold, it is determined that the negative control OD value is valid within the preset concentration range of HBsAg, otherwise, it is determined that the negative control OD value is invalid within the preset concentration range of HBsAg; Determine whether the positive control OD value is greater than a preset positive value threshold; If the positive control OD value is greater than the preset positive value threshold, the positive control OD value is determined to be valid within the preset concentration range of HBsAg; otherwise, the positive control OD value is determined to be invalid within the preset concentration range of HBsAg; If the negative control OD value is valid within the preset concentration range of HBsAg, and the positive control OD value is valid within the preset concentration range of HBsAg, then the first test result is determined to be valid within the preset concentration range of HBsAg; otherwise, the first test result is determined to be invalid within the preset concentration range of HBsAg.

4. The multi-wavelength HBsAg detection method according to claim 3, characterized in that: The step of determining the second detection result of the HBsAg sample to be tested based on the light absorption data at multiple wavelengths comprises: Determining a deviation value of the first detection result within a preset concentration range of HBsAg; Acquiring light absorption data at the multiple wavelengths, and determining a light absorption value sequence sorted from small to large by wavelength based on the light absorption data at the multiple wavelengths; Based on the deviation value and the light absorption value sequence, a second detection result of the HBsAg test sample is determined.

5. The multi-wavelength HBsAg detection method according to claim 4, characterized in that: The step of determining the deviation value of the first detection result within the preset concentration range of HBsAg comprises: Determining a first negative deviation value of the first detection result within a preset concentration range of HBsAg based on a difference between the first critical value and the first critical value threshold; Determining a second negative deviation value of the first detection result within a preset concentration range of HBsAg based on a difference between the second critical value and the second critical value threshold; Determining a positive deviation value of the first test result within a preset concentration range of HBsAg based on a difference between the positive control OD value and the positive value threshold; Based on the first negative deviation value, the second negative deviation value and the positive deviation value, a deviation value of the first detection result within a preset concentration range of HBsAg is determined.

6. The multi-wavelength HBsAg detection method according to claim 4 or 5, characterized in that: Determining the second detection result of the HBsAg sample to be tested based on the deviation value and the light absorption value sequence includes: Inputting the deviation value and the light absorption value sequence into a pre-trained prediction model, wherein the prediction model includes a first feature extraction network, a second feature extraction network, a feature fusion network, and an output network, wherein the output ends of the first feature extraction network and the second feature extraction network are both connected to the input end of the feature fusion network, and the output end of the feature fusion network is connected to the input end of the output network; Performing a deconvolution operation on the deviation value through the first feature extraction network to obtain a deviation feature corresponding to the deviation value; Performing a convolution operation on the light absorption value sequence through the second feature extraction network to obtain a light absorption feature of the light absorption value sequence, wherein the deviation feature has the same resolution size as the light absorption feature; Obtaining a fusion feature by using the fusion network to perform fusion on the deviation feature and the light absorption feature; The fusion feature is predicted and processed by the output network to obtain a prediction result as the second detection result of the HBsAg sample to be tested.

7. The multi-wavelength HBsAg detection method according to claim 6, characterized in that: Before the step of inputting the deviation value and the light absorption value sequence into a pre-trained prediction model, the method further comprises: Obtaining a first sample test result and a second sample test result of the same HBsAg sample, wherein the first sample test result is a test result of a double antibody sandwich method, and the second sample test result is a test result of a chemiluminescence method, wherein the sample double antibody sandwich method test result is light absorption data at multiple wavelengths corresponding to an invalid state of the first test result, and the HBsAg sample is an HBsAg sample within a preset concentration range of HBsAg; Based on the first sample detection result, determine a sample deviation value corresponding to the first sample detection result, and determine the first sample detection result and the sample deviation value as a two-tuple of sample data; Determine the second sample detection result as label data of the sample data; Constructing a training data set based on the sample data and the label data; A pre-trained model is obtained, and the pre-trained model is trained using the training data set, and a trained prediction model is obtained after the training is completed.

8. A multi-wavelength HBsAg detection device, characterized in that: The multi-wavelength HBsAg detection device comprises: The detection module is used to detect the light absorption value of the HBsAg sample to be tested by the double antibody sandwich method through a multi-wavelength microplate reader to obtain the light absorption data of the HBsAg sample to be tested at multiple wavelengths, wherein the multiple wavelengths include multiple wavelengths between 450nm and 630nm; A first processing module is used to determine the sample OD value, the negative control OD value and the positive control OD value at the wavelength of 450nm and the wavelength of 630nm based on the light absorption data at the wavelength of 450nm and the wavelength of 630nm; A second processing module is used to determine a first detection result of the HBsAg sample to be tested based on the sample OD value, the negative control OD value and the positive control OD value at the wavelength of 450nm and the wavelength of 630nm, and determine whether the first detection result is valid within a preset concentration range of HBsAg based on the negative control OD value and the positive control OD value; The third processing module is used to determine the second detection result of the HBsAg sample to be tested based on the mapping relationship between the light absorption data at multiple wavelengths and the chemiluminescence detection result if the first detection result is invalid within the preset concentration range of HBsAg.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the multi-wavelength based HBsAg detection method as described in any one of claims 1 to 7 are implemented.

10. A computer storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the multi-wavelength based HBsAg detection method according to any one of claims 1 to 7 are implemented.

Images