Sample reaction curve processing method and device, storage medium, control terminal and system

Abstract

The invention relates to a sample reaction curve processing method and device, a storage medium, a control terminal and a system, and belongs to the technical field of medical examination.The method comprises the steps that a to-be-recognized reaction curve of a target sample and a target calibration product reaction curve are obtained, the method comprises the steps of obtaining a reaction curve of a calibration product used in target calibration operation of a detection item corresponding to a reaction curve to be recognized, obtaining an anomaly recognition result of the reaction curve to be recognized based on a target calibration product reaction curve, and when the reaction curve to be recognized has an anomaly and the anomaly can be repaired, recognizing the anomaly of the reaction curve to be recognized based on the anomaly recognition result and the target calibration product reaction curve. And obtaining the repaired reaction curve, and displaying the reaction curve to be identified and the repaired reaction curve in the target interface. Therefore, the computer is used for analyzing the curve morphological characteristics to identify the abnormity, a large number of parameters do not need to be manually set and filled, the global mode abnormity can be effectively captured, the abnormal condition can be comprehensively identified, and the abnormal condition and the repairing effect can be visually and clearly reflected for user reference.

Description

Technical Field

[0001] This application relates to the field of medical testing technology, and in particular to a method, apparatus, computer-readable storage medium, control terminal, and sample analysis system for processing sample reaction curves. Background Technology

[0002] For sample testing, reaction curves contain a wealth of information and have a wide range of applications, such as monitoring reaction interference, analyzing reagent quality, and identifying anomalies in instrument status and test results. However, problems such as bubble formation and substrate depletion may occur during the testing process, leading to abnormal reaction curves and consequently inaccurate test results. Therefore, identifying anomalies in reaction curves to ensure their accuracy is crucial.

[0003] In existing technologies, the numerical threshold method is mainly used for anomaly identification. This involves relevant staff manually setting a set of threshold parameters, and then the computer judging whether the corresponding values ​​in the reaction curve are within the parameter range. If they exceed the range, the reaction curve is judged to be abnormal and a prompt is issued.

[0004] However, in existing methods, parameter settings rely on the experience of staff, and a reasonable set of parameters often requires experienced staff to determine through repeated experiments and deliberation. Furthermore, parameters are not interchangeable for different testing items, making parameter determination difficult, and improper parameter settings can easily lead to misjudgments. In addition, there are various reasons for anomalies in response curves in practical applications, and the numerical threshold method can only detect single-point or local values ​​exceeding the range, thus having limited coverage for identifying anomalies. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer-readable storage medium, control terminal, and sample analysis system for processing sample reaction curves to address the technical problems of high false positive rate and limited recognition coverage in the existing reaction curve anomaly identification.

[0006] A method for processing sample response curves, comprising:

[0007] Obtain the target sample's response curve to be identified;

[0008] Obtain the target calibration product response curve; the target calibration product response curve is the response curve of the calibration product used in the target calibration operation for the target detection item, and the target detection item is the detection item corresponding to the response curve to be identified;

[0009] Based on the target calibration product response curve, the anomaly identification result of the response curve to be identified is obtained;

[0010] When the anomaly identification result indicates that the response curve to be identified is an abnormal response curve, it is determined whether the anomaly of the response curve to be identified can be repaired;

[0011] If it can be repaired, the repaired reaction curve is obtained based on the anomaly identification results and the target calibration product reaction curve;

[0012] The target interface displays the reaction curve to be identified and the repaired reaction curve.

[0013] An apparatus for processing sample response curves, comprising:

[0014] The module for acquiring the target curve is used to acquire the target response curve of the target sample.

[0015] The calibration product reaction curve acquisition module is used to acquire the target calibration product reaction curve; the target calibration product reaction curve is the reaction curve of the calibration product used in the target calibration operation for the target detection item, and the target detection item is the detection item corresponding to the reaction curve to be identified.

[0016] The curve anomaly identification module is used to obtain the anomaly identification result of the reaction curve to be identified based on the target calibration product reaction curve;

[0017] The curve repair analysis module is used to determine whether the abnormality of the response curve to be identified can be repaired when the anomaly identification result indicates that the response curve to be identified is an abnormal response curve.

[0018] The repair curve acquisition module is used to obtain the repaired reaction curve based on the anomaly identification result and the target calibration product reaction curve, if repairable.

[0019] The reaction curve display module is used to display the reaction curve to be identified and the repaired reaction curve on the target interface.

[0020] A computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the method described above.

[0021] A control terminal includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the method described above.

[0022] A sample analysis system includes: a target sample analysis instrument and a control terminal, wherein the target sample analysis instrument is communicatively connected to the control terminal;

[0023] The target sample analysis instrument is used to detect the target sample, obtain a detection signal, and send the detection signal to the control terminal;

[0024] The control terminal is used to obtain the target sample's response curve based on the detection signal, and to execute the method described above.

[0025] The above-mentioned sample reaction curve processing method, apparatus, computer-readable storage medium, control terminal, and sample analysis system acquire the target sample's reaction curve to be identified and the target calibration product's reaction curve, i.e., the reaction curve of the calibration product used in the target calibration operation corresponding to the detection item of the reaction curve to be identified. Based on the target calibration product's reaction curve, the abnormal identification result of the reaction curve to be identified is obtained. When the reaction curve to be identified has an abnormality and the abnormality can be repaired, the repaired reaction curve is obtained based on the abnormality identification result and the target calibration product's reaction curve, and the reaction curve to be identified and the repaired reaction curve are displayed on the target interface. Thus, by using the response curve of the target calibration product to determine whether there are any anomalies in the response curve to be identified, this method of identifying anomalies through computer analysis of curve morphology features does not require manual setting and filling in of a large number of parameters, has low dependence on human experience, and can effectively avoid misjudgments caused by improper parameter settings. Moreover, because it focuses on the shape, trend, and fluctuation of the curve, it can break through the limitation of traditional numerical threshold methods that can only detect single-point or local numerical anomalies, effectively capture global pattern anomalies, and thus more comprehensively identify anomalies. At the same time, it displays the response curves before and after repair, which can intuitively and clearly reflect the anomalies and repair effects for user reference. Attached Figure Description

[0026] Figure 1 This is an application environment diagram of the sample response curve processing method in one embodiment;

[0027] Figure 2 This is a flowchart illustrating a method for processing sample response curves in one embodiment;

[0028] Figure 3 This is a schematic diagram of the detection result display interface in one embodiment;

[0029] Figure 4 This is a schematic diagram of the target interface in one embodiment;

[0030] Figure 5 This is a flowchart illustrating the process of obtaining anomaly identification results in one embodiment;

[0031] Figure 6 This is a schematic diagram of the analysis results in one embodiment;

[0032] Figure 7 This is a schematic diagram of the analysis results in one embodiment;

[0033] Figure 8This is a schematic diagram of the analysis results in one embodiment;

[0034] Figure 9 This is a schematic diagram of the analysis results in one embodiment;

[0035] Figure 10 This is a structural block diagram of a device for processing sample reaction curves in one embodiment;

[0036] Figure 11 This is a block diagram of the control terminal in one embodiment. Detailed Implementation

[0037] The terms used in this application, including "comprising," "including," "having," "showing to have," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to the steps or units that are clearly listed, but may also include other steps or units that are not clearly listed or that are inherent to these processes, methods, products, or apparatuses. The terms "multiple," "a plurality of," and any variations thereof refer to at least two, unless otherwise expressly specified. The terms "first," "second," etc., are used to distinguish similar objects by name, but these objects themselves are not limited by these terms and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary or secondary relationship of the indicated technical features.

[0038] To facilitate understanding of the technical implementation of this application, its application will first be explained:

[0039] This application can be applied to the field of sample analysis. Specific application environments include... Figure 1 As shown, it involves a control terminal 110 and a target sample analyzer 120, which are connected in communication.

[0040] Specifically, the target sample analyzer 120 performs corresponding detection on the target sample, obtains the detection signal, and sends it to the control terminal 110. The control terminal 110 obtains the target sample's reaction curve to be identified based on the detection signal, and then acquires the target calibration sample reaction curve, i.e., the reaction curve of the calibration sample used in the target calibration operation corresponding to the detection item of the reaction curve to be identified. Based on the target calibration sample reaction curve, it obtains the anomaly identification result of the reaction curve to be identified. When the anomaly identification result indicates that the reaction curve to be identified is abnormal, it determines whether the anomaly of the reaction curve to be identified can be repaired. If it is determined to be repairable, based on the anomaly identification result and the target calibration sample reaction curve, it obtains the repaired reaction curve and displays the reaction curve to be identified and the repaired reaction curve on the target interface.

[0041] The target sample analyzer 120 is an instrument used to detect and analyze sample components, such as a biochemical analyzer and a coagulation analyzer. The control terminal 110 can be, but is not limited to, various personal computers, laptops, smartphones, and tablets.

[0042] The present application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0043] In some embodiments, such as Figure 2 As shown, a method for processing sample response curves is provided. This method is applied to the above... Figure 1 Taking the control terminal 110 as an example, the method may include the following steps S202 to S212.

[0044] S202, Obtain the target sample's response curve to be identified.

[0045] The target sample is the biological sample that has been detected by the target sample analyzer. The target sample can be determined based on user operation; for example, the detection result records can be displayed on the control terminal's display interface, and the user can select a detection result record. The sample corresponding to that detection result record is the target sample. Specifically, as shown below... Figure 3 As shown, the test results display interface displays multiple test result records in a list format. When the user selects the 5th test result record in the list, the corresponding sample number is 7, which is then used as the target sample.

[0046] The reaction curve, with the vertical axis representing the detection signal (e.g., absorbance, fluorescence intensity) and the horizontal axis representing time, reflects the change of the detection signal over time. The reaction curve to be identified, on the other hand, is the reaction curve for anomaly detection; it is generated based on the detection signal output by the target sample analyzer.

[0047] Specifically, the target sample analyzer performs a test on a target sample for a specific test item, obtains a test signal, and outputs it to the control terminal. The control terminal then generates the original reaction curve of the target sample based on the received test signal. When anomaly identification is required on this original reaction curve, it becomes the reaction curve to be identified. For example, a biochemical analyzer performs a test on liver function enzymes on a target sample, performing 32 absorbance measurements on the reaction system within the reaction cycle, outputting 32 absorbance data points to the control terminal. The control terminal can then generate the original reaction curve of the target sample, i.e., a reaction curve including these 32 measurement points. Subsequently, this original reaction curve can be used as the reaction curve to be identified for anomaly identification.

[0048] S204, Obtain the reaction curve of the target calibration sample.

[0049] The target calibration curve is the response curve of the calibration material used in the target calibration operation for the target test item.

[0050] Among them, the target detection item is the detection item corresponding to the target sample's reaction curve to be identified. For example, the reaction curve to be identified is obtained by detecting liver function enzymes in the target sample, and the liver function enzymes are the target detection items.

[0051] Target calibration is a calibration operation that matches the reaction curve to be identified. Specifically, it involves detecting a series of calibrators with known concentrations to establish a quantitative relationship between the instrument's detection signal and the concentration of the analyte. It's important to note that a single sample analyzer can support multiple detection items; for example, a biochemical analyzer typically supports 20 to over 100 detection items. For a single detection item, multiple calibration operations may be required. For instance, the same detection item may have multiple calibration records due to different methodologies or reagent batches. In such cases, the calibration operation that matches the reaction curve to be identified can be selected as the target calibration operation. For example, if the target detection item has undergone multiple calibration operations due to changes in reagent batches, the calibration operation corresponding to the same reagent batch as the reaction curve to be identified can be selected.

[0052] In addition, when performing target calibration, multiple calibrators are usually used, each with its own corresponding reaction curve. Therefore, to obtain the target calibrator reaction curve, specifically, one can obtain the reaction curve corresponding to each calibrator used in the target calibration operation.

[0053] S206, based on the target calibration product reaction curve, obtain the anomaly identification results of the reaction curve to be identified.

[0054] Anomaly identification results are used to characterize the abnormalities of the response curve to be identified. Specifically, the anomaly identification results may include anomaly state information and anomaly type information. The anomaly state information is used to characterize whether the response curve to be identified is abnormal, that is, whether the response curve to be identified is an abnormal response curve or a normal response curve. The anomaly type information is used to characterize the type of anomaly of the response curve to be identified. The anomaly type can be preset based on the actual application situation. For example, it may include curve shape anomalies and curve smoothness anomalies. That is, the anomaly of the response curve to be identified may belong to one or more of the above types.

[0055] Specifically, based on the target calibrator reaction curve, correlation analysis of the curve morphology characteristics of the reaction curve to be identified can be performed to obtain the anomaly identification results of the reaction curve to be identified. It should be noted that the target calibrator reaction curve is obtained by testing a calibrator of known concentration. In principle, the calibrator is a standard, interference-free sample. The reaction curves obtained from subsequent measurements of other samples should also resemble the reaction curves of the corresponding calibrator. Therefore, by referring to the target calibrator reaction curve to perform correlation analysis of the curve morphology characteristics of the reaction curve to be identified, anomalies in the reaction curve to be identified can be effectively identified.

[0056] S208, when the anomaly identification result indicates that the response curve to be identified is an abnormal response curve, determine whether the anomaly of the response curve to be identified can be repaired.

[0057] Specifically, based on pre-set irreparable conditions, it can be determined whether the anomaly of the response curve to be identified is repairable. If the response curve to be identified meets any one of the irreparable conditions, then the anomaly of the response curve to be identified is determined to be irreparable; conversely, if none of the conditions are met, then the anomaly of the response curve to be identified is determined to be repairable. The irreparable conditions can be set according to the actual application.

[0058] S210, if repairable, obtain the repaired reaction curve based on the anomaly identification results and the target calibration product reaction curve.

[0059] The repaired response curve is the response curve obtained after anomaly repair of the response curve to be identified. Specifically, based on the response curve of the target calibration sample and the repair strategy corresponding to the anomaly type of the anomaly identification result, anomaly repair can be performed on the response curve to be identified to obtain the repaired response curve. Different anomaly types can correspond to different repair strategies. Furthermore, by referring to the response curve of the target calibration sample to repair anomalies in the response curve to be identified, the abnormal parts of the response curve to be identified can be effectively restored to the correct shape, improving the accuracy of the repair of the response curve to be identified.

[0060] S212, displays the reaction curve to be identified and the reaction curve after repair in the target interface.

[0061] The target interface is the display interface in the control terminal used to display the reaction curves related to the target sample. Specifically, after obtaining the repaired reaction curve, the target interface can simultaneously display the reaction curve to be identified and the repaired reaction curve of the target sample for comparison by the user. This provides a clear and intuitive representation of the anomaly and the repair effect, assisting the user in reviewing and confirming the final reaction curve.

[0062] In one specific embodiment, a coordinate system can be displayed on the target interface, in which the reaction curve to be identified and the reaction curve after repair are displayed simultaneously. In this way, the difference between the reaction curves before and after repair can be efficiently conveyed through comparison of the same coordinate system.

[0063] The above-described method for processing sample response curves involves obtaining the response curve to be identified for the target sample, as well as the response curve of the target calibration sample (i.e., the response curve of the calibration sample used in the target calibration operation for the corresponding detection item). Based on the target calibration sample response curve, anomaly identification results are obtained for the response curve to be identified. When anomalies exist in the response curve to be identified and are repairable, the repaired response curve is obtained based on the anomaly identification results and the target calibration sample response curve. Both the response curve to be identified and the repaired response curve are then displayed on the target interface. Thus, judging whether anomalies exist in the response curve to be identified based on the target calibration sample response curve is a computer-based method of identifying anomalies through curve morphology analysis. This eliminates the need for manual setting and input of numerous parameters, reducing reliance on human experience and effectively avoiding misjudgments caused by improper parameter settings. Furthermore, because it focuses on the shape, trend, and fluctuation characteristics of the curve, it overcomes the limitations of traditional numerical threshold methods, which can only detect single-point or local numerical anomalies, effectively capturing global pattern anomalies. This allows for a more comprehensive identification of anomalies, and the simultaneous display of the response curves before and after repair provides a clear and intuitive representation of the anomaly and repair effect for user reference.

[0064] In some embodiments, the target interface includes a project detection information display area. Based on this, the sample reaction curve processing method provided in this application may further include the following steps: displaying project detection information corresponding to the reaction curve to be identified and project detection information corresponding to the repaired reaction curve in the project detection information display area; determining the target reaction curve based on the selection operation of the project detection information in the project detection information display area; and determining the valid detection result of the target sample based on the target reaction curve in response to the review trigger operation.

[0065] The project testing information display area is the area on the target interface used to display project testing information. Project testing information describes the project testing status of the target sample, and may include sample information, testing item information, testing result information, and correlation curve information, etc. The correlation curve information may include the correlation curve type, that is, the curve type of the reaction curve corresponding to the project testing information. The curve type may include the original reaction curve and the repaired reaction curve.

[0066] In this embodiment, the item detection information display area can simultaneously display item detection information corresponding to the response curve to be identified, as well as item detection information corresponding to the repaired response curve. The detection results in the item detection information are derived from the response curve corresponding to that item. It can be understood that the detection results in the item detection information corresponding to the response curve to be identified correspond to the actual detection performed on the target sample and are considered real detection results, while the detection results in the item detection information corresponding to the repaired response curve correspond to the simulated detection performed based on the repaired response curve and are considered speculative results.

[0067] Furthermore, the specific method of displaying the project testing information in the project testing information display area can adopt any applicable display method, and this application does not limit it. For example, it can be displayed in a list format (i.e., displaying the project testing information corresponding to different reaction curves through different rows in the list) or in a drop-down box format (i.e., displaying the project testing information corresponding to different reaction curves through different drop-down options).

[0068] The selection operation for project testing information in the project testing information display area involves choosing one from the displayed project testing information. Specifically, the project testing information display area simultaneously displays project testing information corresponding to the reaction curve to be identified and project testing information corresponding to the repaired reaction curve. The user can select one from these two sets of project testing information. Furthermore, the selection operation can employ any method applicable in the art. For example, radio buttons can be set in the project testing information display area, and the user can select the project testing information corresponding to that radio button by checking the radio button.

[0069] The target response curve is selected and determined by the user. Specifically, after the control terminal detects the selection operation of the project detection information in the project detection information display area, the response curve corresponding to the selected project detection information is determined as the target response curve.

[0070] The review trigger operation can be used to determine the valid detection result of a target sample. A valid detection result is a result obtained based on the target reaction curve, specifically including the concentration calculated based on the target reaction curve. After the target reaction curve is determined by the user's selection, if the review trigger operation is detected, it indicates that the user believes the determined target reaction curve can accurately reflect the concentration of the target analyte and the reaction process in the target sample. Therefore, a valid detection result for the target sample can be determined based on the target reaction curve. In practice, the review trigger operation can adopt any operation method applicable in the art, such as triggering an review control displayed on the target interface, such as clicking the review control, or a preset gesture operation, etc.

[0071] In some embodiments, after determining the valid detection result of the target sample, the valid detection result can also be displayed in the detection result management interface, so that users can accurately know the final determined detection result.

[0072] In addition, the target interface may also include a reaction curve data display area, which can be linked with the project testing information display area. This area can display the measurement point data of the reaction curve corresponding to the selected project testing information in the project testing information display area. The content of the measurement point data can be set based on actual needs, and may include, for example, the corresponding label, dominant wavelength, secondary wavelength, and dominant-secondary wavelength data. Furthermore, the reaction curve data display area can display the measurement point data of the corresponding reaction curve in a list format, or use any applicable display method; this application does not limit this.

[0073] To facilitate clear understanding, the following is illustrated with a specific example:

[0074] like Figure 4 As shown, the target interface displays the project detection information display area 401, the reaction curve data display area 402, and the audit control 403. The project detection information display area 401 displays the project detection information corresponding to the reaction curve to be identified and the project detection information corresponding to the repaired reaction curve in two rows of lists. The project detection information includes the project abbreviation, sample type, correlation curve type, test type, sample number, test result information (such as reactivity and concentration), experimental circle, dot shape, and cup number, etc. The test result information is calculated based on the corresponding reaction curve.

[0075] If a user selects the item detection information corresponding to the response curve to be identified and then clicks the review control 403, it indicates that the user does not approve of the repair result. In this case, the response curve to be identified will be used as the target response curve, and the detection result information obtained based on the response curve to be identified will be determined as a valid detection result. It should be noted that in this case, since the detection result management interface originally displays this detection result, it should continue to display this detection result in the detection result management interface. In addition, after the user selects the item detection information corresponding to the response curve to be identified, the relevant data of the measurement points included in the response curve to be identified will also be displayed in the response curve data display area 402.

[0076] However, if the user selects the item detection information corresponding to the repaired reaction curve and then clicks the review control 402, it indicates that the user approves the repair result. Therefore, the repaired reaction curve is used as the target reaction curve, and the detection result information obtained based on the repaired reaction curve is determined as a valid detection result. In this case, since the detection result management interface originally displayed the detection results based on the reaction curve to be identified, it needs to be updated to display the valid detection results based on the repaired reaction curve. Furthermore, after the user selects the item detection information corresponding to the repaired reaction curve, the relevant data of the measurement points included in the repaired reaction curve will also be displayed in the reaction curve data display area 402.

[0077] In this embodiment, the target interface displays the item detection information corresponding to the reaction curve to be identified and the item detection information corresponding to the repaired reaction curve. The target reaction curve is determined by detecting the user's selection operation of the item detection information, thereby determining the valid detection result. This supports user review and decision-making and helps to ensure the accuracy of the reaction curve and detection results.

[0078] In some embodiments, after the step of obtaining the anomaly identification result of the reaction curve to be identified based on the target calibration product reaction curve in the sample reaction curve processing method provided in this application, the following step may be included: displaying the anomaly analysis information corresponding to the reaction curve to be identified in the target interface.

[0079] Anomaly analysis information is used to characterize the cause of the anomaly, that is, the reason why the reaction curve to be identified is abnormal, such as a problem with the stirring needle of the target sample analyzer.

[0080] In this embodiment, after determining that there is an anomaly in the response curve to be identified, the anomaly analysis information corresponding to the response curve to be identified is displayed on the target interface. This can help users understand the cause of the anomaly intuitively and in a timely manner, thereby achieving targeted and efficient anomaly handling.

[0081] In some embodiments, the step of obtaining the anomaly identification result of the reaction curve to be identified based on the target calibration product reaction curve in the sample response curve processing method provided in this application may include the following steps: performing similarity calculation based on the reaction curve to be identified and the target calibration product reaction curve to obtain the target similarity; obtaining the curve shape analysis result of the reaction curve to be identified based on the target similarity; and obtaining the anomaly identification result of the reaction curve to be identified based on the curve shape analysis result.

[0082] Target similarity is the similarity between the response curve to be identified and the response curve of the target standard. It characterizes the degree of similarity between the two curves in terms of curve shape, such as shape and trend. There is a one-to-one correspondence between the target similarity and the target standard response curve; that is, for each target standard response curve, its similarity to the response curve to be identified is calculated. Furthermore, the target similarity can be a value within the range of (-∞, 1). The closer the value is to 1, the higher the similarity between the response curve to be identified and the corresponding target standard response curve; conversely, the closer the value is to 1, the lower the similarity.

[0083] The curve morphology analysis results can also correspond one-to-one with the target calibration product response curves. That is, for each target calibration product response curve, the corresponding curve morphology analysis result for the response curve to be identified is obtained based on its similarity to the response curve to be identified. Specifically, the curve morphology analysis result can be normal or abnormal, corresponding to whether the response curve to be identified exhibits an anomaly in its curve morphology when referenced to the corresponding target calibration product response curve. In a specific embodiment, based on the similarity between any target calibration product response curve and the response curve to be identified, the curve morphology analysis result is obtained by determining whether the similarity is lower than a target similarity threshold. If it is lower, it indicates that the response curve to be identified exhibits an anomaly in its curve morphology when referenced to the corresponding target calibration product response curve, and the obtained curve morphology analysis result is abnormal. Conversely, if it is not lower, it indicates that the response curve to be identified does not exhibit an anomaly in its curve morphology when referenced to the corresponding target calibration product response curve, and the obtained curve morphology analysis result is normal. The target similarity threshold can be preset based on actual needs.

[0084] Furthermore, based on the curve morphology analysis results, the anomaly identification result of the reaction curve to be identified can be obtained. In one embodiment, a baseline reaction curve can be determined from the target calibration product reaction curve, and the anomaly identification result of the reaction curve to be identified can be obtained based on the curve morphology analysis result corresponding to the baseline reaction curve. In another specific embodiment, the anomaly identification result of the reaction curve to be identified can be obtained solely based on the corresponding curve morphology analysis result, without performing curve analysis in other dimensions. That is, when the curve morphology analysis result corresponding to the baseline reaction curve is abnormal, the anomaly identification result characterizing the reaction curve to be identified as an abnormal reaction curve is directly obtained; conversely, when the curve morphology analysis result corresponding to the baseline reaction curve is normal, the anomaly identification result characterizing the reaction curve to be identified as a normal reaction curve is obtained. However, in other embodiments, the anomaly identification result of the reaction curve to be identified can also be obtained by combining the curve morphology analysis result with curve analysis results in other dimensions.

[0085] In other embodiments, the curve morphology analysis results corresponding to the reaction curves of each target calibration product can also be analyzed. When all the curve morphology analysis results are abnormal, an abnormal identification result is obtained that characterizes the reaction curve to be identified as an abnormal reaction curve. Conversely, if one or more curve morphology analysis results are normal, the reaction curve to be identified is determined to be a normal reaction curve, that is, an abnormal identification result is obtained that characterizes the reaction curve to be identified as a normal reaction curve.

[0086] It should be noted that the target calibration curve is obtained by testing a calibration sample of known concentration. The calibration sample is, in principle, a standard, interference-free sample, and the reaction curves obtained from subsequent measurements of other samples should also resemble the corresponding calibration curve. Therefore, by calculating the similarity between the reaction curve to be identified and the target calibration curve, and analyzing the similarity in curve morphology such as shape and trend, the curve morphology analysis of the reaction curve to be identified can be performed with reference to the target calibration curve, reliably identifying anomalies in the reaction curve to be identified.

[0087] In some embodiments, the step of obtaining anomaly identification results of the reaction curve to be identified based on the target calibration product reaction curve in the sample response curve processing method provided in this application may include the following steps: determining the curve model corresponding to the target calibration product reaction curve based on the curve type of the target calibration product reaction curve; fitting the curve model corresponding to the reaction curve to be identified and the target calibration product reaction curve based on a preset fitting model to determine the target model parameters of the curve model corresponding to the target calibration product reaction curve; obtaining a smooth reaction curve based on the curve model corresponding to the target calibration product reaction curve and its target model parameters; calculating the difference between the smooth reaction curve and the reaction curve to be identified to obtain the difference result; obtaining the curve smoothness analysis result of the reaction curve to be identified based on the difference result; and obtaining the anomaly identification result of the reaction curve to be identified based on the curve smoothness analysis result.

[0088] The curve type can be the shape type corresponding to the target standard response curve, such as linear, exponential, and polynomial shapes. The curve model is a mathematical expression describing the relationship between measurement points, corresponding to the curve type, such as a linear model, exponential model, and polynomial model. In a specific embodiment, the correspondence between the target standard response curve, curve type, and curve model can be one-to-one; in other words, for a target standard response curve, a curve type and a corresponding curve model can be determined.

[0089] A pre-set fitting model is used to determine the model parameters of the curve model. These parameters can be preset according to actual needs, such as using least squares (least squares model) and RANSACRegressor (Random Sample Consensus Regressor). Specifically, the pre-set fitting model fits the curve model corresponding to the response curve to be identified and the response curve of the target calibration product. During the fitting process, the curve model is used to define the mathematical expression of the curve, and the response curve to be identified serves as the benchmark data for fitting. The pre-set fitting model adjusts the parameters of the curve model based on the response curve to be identified. Specifically, the parameters of the curve model are optimized by comparing the error between the model's predicted values ​​and the actual values ​​corresponding to the response curve to be identified, thus obtaining the optimized parameters, i.e., the target model parameters.

[0090] After obtaining the target model parameters, they are substituted into the curve model to obtain the smooth response curve. The measurement points in the smooth response curve correspond one-to-one with the measurement points in the response curve to be identified. Following the previous example, if the response curve to be identified includes 32 measurement points, then the smooth response curve also includes 32 measurement points.

[0091] Then, the difference between the smooth response curve and the response curve to be identified is calculated to obtain the difference result. Specifically, the difference between each measurement point in the smooth response curve and the corresponding measurement point in the response curve to be identified can be calculated separately to obtain multiple difference values.

[0092] Then, based on the difference results, the curve smoothness analysis results of the reaction curve to be identified are obtained. In a specific embodiment, the curve smoothness analysis results can correspond one-to-one with the target calibration product reaction curves. That is, for each target calibration product reaction curve, the above steps from determining the curve type to calculating the difference results can be performed to obtain its corresponding curve smoothness analysis results. For example, in the target calibration operation of the target detection project, three calibration products are used, namely O1, O2 and O3. Each calibration product corresponds to a target calibration product response curve, namely S01, S02 and S03. For the target calibration product response curve S01, a preset fitting model is used to fit the curve model corresponding to the response curve to be identified and the target calibration product response curve S01 to determine the target model parameters of the curve model corresponding to the target calibration product response curve S01. Then, based on the curve model corresponding to the target calibration product response curve S01 and its target model parameters, a smooth response curve is obtained. Then, the difference between the smooth response curve and the response curve to be identified is calculated to obtain the difference result. Based on the difference result, the curve jump point analysis result C12 of the response curve to be identified is obtained. The above operation can also be performed on the target calibration product response curves S02 and S03 respectively to obtain the curve jump point analysis results C22 and C32 respectively.

[0093] Specifically, the curve smoothness analysis result can be normal or abnormal, corresponding to whether the response curve to be identified has an abnormality or no abnormality in terms of curve smoothness when the response curve of the corresponding target calibration product is used as a reference. In a specific embodiment, the curve smoothness analysis result may include the curve jump point analysis result, corresponding to whether the response curve to be identified has or does not have measurement points with significant discontinuities or unreasonable jumps when the response curve of the corresponding target calibration product is used as a reference.

[0094] In one specific embodiment, the curve smoothness analysis result includes the curve jump point analysis result. Based on abnormal differences in the difference results, the curve jump point analysis result of the response curve to be identified can be obtained. An abnormal difference is a difference whose absolute value exceeds a preset absolute value threshold. Specifically, the number of abnormal differences can be compared with a preset number threshold. Based on the comparison result, the curve jump point analysis result is obtained. If the number of abnormal differences exceeds the preset number threshold, the curve jump point analysis result is "abnormal"; otherwise, if it does not exceed the threshold, the curve jump point analysis result is "normal". For example, both the smooth response curve and the response curve to be identified include 32 measurement points, and the difference results correspondingly include 32 differences. If the preset number threshold is 0 and the preset absolute value threshold is 3, then the absolute values ​​of these 32 differences exceeding 3 are abnormal differences. If one or more abnormal differences exist among the 32 differences, the curve jump point analysis result is "abnormal"; if no abnormal differences exist, the curve jump point analysis result is "normal".

[0095] Furthermore, based on the curve smoothness analysis results, the anomaly identification results of the reaction curve to be identified are obtained. In a specific embodiment, a benchmark reaction curve can be determined from the target calibration product reaction curve. Then, based on the curve smoothness analysis results corresponding to the benchmark reaction curve, the anomaly identification results of the reaction curve to be identified are obtained. That is, when the curve smoothness analysis result corresponding to the benchmark reaction curve is abnormal, an anomaly identification result is obtained that characterizes the reaction curve to be identified as an abnormal reaction curve; conversely, when the curve smoothness analysis result corresponding to the benchmark reaction curve is normal, an anomaly identification result is obtained that characterizes the reaction curve to be identified as a normal reaction curve.

[0096] In other embodiments, the smoothness analysis results of the curves corresponding to the reaction curves of each target calibration product can also be analyzed. When all the smoothness analysis results of each curve are abnormal, an abnormal identification result is obtained that characterizes the reaction curve to be identified as an abnormal reaction curve. Conversely, if one or more curve smoothness analysis results are normal, an abnormal identification result is obtained that characterizes the reaction curve to be identified as an abnormal reaction curve.

[0097] In this embodiment, based on the curve model corresponding to the target calibration product's reaction curve, the smoothness of the reaction curve to be identified is analyzed, which can identify the smoothness of the reaction curve to be identified, such as jump points, thereby reliably realizing the anomaly detection of the reaction curve to be identified.

[0098] It should be noted that, in some embodiments, the step of obtaining the anomaly identification result of the reaction curve to be identified based on the target calibration product reaction curve in the sample reaction curve processing method provided in this application may simultaneously include the related steps of "curve morphology analysis and curve smoothness analysis" described above, for example... Figure 5 As shown.

[0099] Furthermore, in one specific embodiment, a baseline reaction curve can first be determined from the target calibration product reaction curve based on the similarity between the reaction curve to be identified and the target calibration product reaction curve. Then, the curve shape analysis results and curve smoothness analysis results corresponding to the baseline reaction curve are obtained. Subsequently, based on the curve shape analysis results and curve smoothness analysis results corresponding to the baseline reaction curve, the anomaly identification result of the reaction curve to be identified is obtained. The baseline reaction curve can be determined based on the similarity between the reaction curve to be identified and the target calibration product reaction curve; for example, the target calibration product reaction curve with the highest similarity to the reaction curve to be identified can be determined as the baseline reaction curve.

[0100] In another specific embodiment, the above-mentioned curve morphology analysis and curve smoothness analysis can also be performed on each target calibration product reaction curve to obtain the analysis results of the reaction curve to be identified in the above two dimensions. Then, based on all the obtained analysis results, a benchmark reaction curve is determined. Then, based on the curve morphology analysis results and curve smoothness analysis results corresponding to the benchmark reaction curve, the anomaly identification results of the reaction curve to be identified are obtained.

[0101] Following the previous example, three calibration products, 01, 02, and 03, are used in the target calibration operation for the target detection project. Each calibration product corresponds to a target calibration product response curve, namely S01, S02, and S03. For the target calibration product response curve S01, the aforementioned curve morphology analysis and curve smoothness analysis are performed to obtain the curve morphology analysis result C11 and the curve smoothness analysis result C12. Similarly, for the target calibration product response curves S02 and S03, the corresponding curve morphology analysis results C21 and C31 and the curve smoothness analysis results C22 and C32 are obtained. All the analysis results are as follows. Figure 6 As shown. Furthermore, based on all the analysis results C11 to C31 and C12 to C32, the anomaly identification results of the reaction curve to be identified are obtained.

[0102] In one specific embodiment, the method of determining the baseline response curve based on all obtained analysis results can be implemented by first analyzing the curve morphology analysis results corresponding to the response curves of each target calibration product. If the curve morphology analysis results corresponding to all target calibration product response curves are abnormal, then the calibration product response curves whose similarity to the response curve to be identified is not lower than the similarity lower limit are identified, and the one with the highest similarity is taken as the baseline response curve. If there is a target calibration product response curve with a normal curve morphology analysis result, and only one corresponding curve morphology analysis result is normal, then this target calibration product response curve is the baseline response curve. If multiple corresponding curve morphology analysis results are normal, then the curve smoothness analysis result is determined from these multiple target calibration product response curves. If only one target calibration product response curve with a normal curve smoothness analysis result is identified, then this is the baseline response curve. If there are zero or more, then the similarity between each target calibration product response curve and the response curve to be identified is further compared, and the one with the highest similarity is determined as the baseline response curve.

[0103] Furthermore, as described above, in one specific embodiment, the anomaly identification result may include anomaly state information and anomaly type information. The anomaly state information is used to characterize whether the response curve to be identified is abnormal, that is, whether the response curve to be identified is an abnormal curve or a normal curve. The anomaly type information is used to characterize the anomaly type of the response curve to be identified. For example, after obtaining the curve morphology analysis results and curve smoothness analysis results of the reaction curve to be identified based on each target calibration product reaction curve, a baseline reaction curve can be determined from the target calibration product reaction curves based on all the obtained analysis results. Then, based on the curve morphology analysis results and curve smoothness analysis results corresponding to the baseline reaction curve, the anomaly identification results of the reaction curve to be identified can be obtained. The anomaly identification results include anomaly state information and anomaly type information. When at least one of the curve morphology analysis results and curve smoothness analysis results corresponding to the baseline reaction curve is abnormal, the obtained anomaly state information indicates that the reaction curve to be identified is an abnormal curve. The anomaly type represented by the obtained anomaly type information corresponds to the anomaly analysis results in these two dimensions. For example, if the curve morphology analysis result corresponding to the baseline reaction curve is normal and the curve smoothness analysis result is abnormal, then the anomaly type represented by the obtained anomaly type information includes curve smoothness anomaly.

[0104] Following the example above, let's assume all the analysis results are as follows: Figure 7 As shown, the curve shape analysis results C11 to C31 corresponding to the three target calibration product reaction curves S01 to S03 are analyzed first. Only C31 is normal. Therefore, the target calibration product reaction curve S03 is directly determined as the benchmark reaction curve.

[0105] Let's assume all the analysis results are as follows Figure 8 As shown, the curve shape analysis results C11 to C31 corresponding to the three target calibration product reaction curves S01 to S03 are analyzed first. C11 and C31 are normal. Therefore, the curve smoothness analysis results corresponding to the target calibration product reaction curves S01 and S03 are determined to be normal. Only the curve smoothness analysis result corresponding to S03 is normal. Therefore, the target calibration product reaction curve S03 is determined as the benchmark reaction curve.

[0106] Let's assume all the analysis results are as follows Figure 9 As shown, the curve shape analysis results C11 to C31 corresponding to the three target calibration product reaction curves S01 to S03 are analyzed first. C11 and C21 are normal. Then, the curve smoothness analysis results corresponding to the target calibration product reaction curves S01 and S02 are determined to be normal. However, both are abnormal. Then, the similarity between the two and the reaction curve to be identified is further compared. Assuming that S01 has a higher similarity to the reaction curve to be identified, S01 is determined as the baseline reaction curve.

[0107] In some embodiments, the sample response curve processing method provided in this application may include the following steps: segmenting the target calibration sample response curve based on the experimental configuration information corresponding to the target detection item to obtain multiple sub-calibration sample response curves; and determining the curve type corresponding to each sub-calibration sample response curve through a curve classification model. Based on this, the step of determining the curve model corresponding to the target calibration sample response curve based on its curve type may include the following steps: determining the curve model corresponding to each sub-calibration sample response curve based on its corresponding curve type.

[0108] Experimental configuration information can be used to determine the segmentation method of the target calibration sample reaction curve. Specifically, the experimental configuration information may include at least one of the following: the number of reagents corresponding to the target detection item and the information on the light-collecting points. The number of reagents refers to the number of types of reagents used in the target detection item, and the information on the light-collecting points may include the light-collecting time points and light-collecting coordinates corresponding to the target detection item.

[0109] In one specific embodiment, the number of segments and segment positions are determined based on the number of reagents and the information of the light-collecting points. Then, based on the number of segments and segment positions, the target calibration product reaction curve is segmented to obtain multiple sub-calibration product reaction curves. The number of segments refers to the total number of segments, i.e., the total number of sub-calibration product reaction curves obtained after segmentation. The number of segments can be the same as the number of reagents; that is, the number of sub-calibration product reaction curves corresponds to the number of reagents used in the target detection item. The segment position is the boundary point of the segmentation; that is, the target calibration product reaction curve is divided at the segment position to obtain multiple sub-calibration product reaction curves. For example, in the detection process of a biochemical detection item, two reagents, R1 and R2, are used. Reagent R1 is added first to construct the reaction environment, and then reagent R2 is added to trigger signal generation. In this case, the target calibration product reaction curve can be divided into two segments based on the segment position corresponding to the information of the light-collecting points, resulting in two sub-calibration product reaction curves.

[0110] A curve classification model is used to determine the curve type corresponding to the curves input into the model. Specifically, it can be implemented using a random forest model, but any other suitable model can also be used. In this embodiment, multiple sub-calibrator response curves are input into the curve classification model, which classifies each sub-calibrator response curve to obtain the curve type corresponding to each sub-calibrator response curve. Specifically, multiple candidate curve types can be pre-set. For each sub-calibrator response curve input into the curve classification model, one of the multiple candidate curve types is determined as the curve type corresponding to that sub-calibrator response curve.

[0111] Then, based on the curve type corresponding to each sub-calibrator reaction curve, the curve model corresponding to each sub-calibrator reaction curve is determined. In a specific embodiment, unlike the previous embodiment where a curve type and a curve model are determined for a single target calibrator reaction curve, instead, a curve type and a curve model are first determined for each sub-calibrator reaction curve. Since a target calibrator reaction curve is divided into multiple sub-calibrator reaction curves, multiple curve types and multiple curve models can be determined for a single target calibrator reaction curve.

[0112] Furthermore, based on the curve models corresponding to the reaction curves of each sub-calibrator, the curve smoothness analysis results of the reaction curve to be identified are obtained. In a specific embodiment, in addition to segmenting the target calibrator reaction curve, the reaction curve to be identified can also be segmented accordingly to obtain multiple sub-reaction curves to be identified. The curve models corresponding to each sub-calibrator reaction curve also correspond one-to-one with each sub-reaction curve to be identified. Furthermore, based on the curve models corresponding to each sub-calibrator reaction curve and their respective corresponding sub-reaction curves to be identified, the curve smoothness analysis results of the reaction curve to be identified are obtained.

[0113] Specifically, for each sub-calibrator response curve, a preset fitting model is used to fit the curve model corresponding to the sub-calibrator response curve and its corresponding sub-response curve to be identified, thereby determining the target model parameters of the curve model corresponding to the sub-calibrator response curve. Then, based on the curve model corresponding to the sub-calibrator response curve and its target model parameters, a sub-smoothed response curve is obtained. Subsequently, the sub-smoothed response curves can be merged to obtain a smoothed response curve. Then, the difference between the smoothed response curve and the response curve to be identified is calculated to obtain the difference result. Based on the difference result, the curve smoothness analysis result of the response curve to be identified is obtained.

[0114] In this embodiment, the response curve of the target calibration product is segmented, and the curve type and curve model are determined for each sub-calibration product response curve. Compared with the method of directly determining the curve type and curve model for the complete response curve of the target calibration product, the amount of model data generated is greatly reduced, which is beneficial to improving the efficiency of anomaly identification.

[0115] In some embodiments, the step of determining whether the anomaly of the reaction curve to be identified is repairable in the sample response curve processing method provided in this application may include the following steps: determining whether the reaction curve to be identified meets any one of the unrepairable conditions; if it meets the conditions, determining that the anomaly of the reaction curve to be identified is unrepairable; if it does not meet the conditions, determining that the anomaly of the reaction curve to be identified is repairable; wherein, the unrepairable conditions include at least one of the following: the reaction curve to be identified is associated with the target fault information reported by the target sample analyzer, the number of effective photometers in the reaction curve to be identified is less than the photometer threshold, and the similarity between the reaction curve to be identified and the target calibration product reaction curve is lower than the similarity lower limit.

[0116] Regarding the association between the target reaction curve to be identified and the target fault information reported by the target sample analyzer, the target fault information describes the target fault that occurs during the detection process. Specifically, if a target fault occurs during the detection of the target sample by the target sample analyzer, the target sample analyzer sends a fault notification carrying the target fault information to the control terminal. The control terminal then associates the target fault information with the target reaction curve to be identified. Therefore, when determining whether the target reaction curve meets the unrepairable condition, if the target reaction curve is associated with target fault information, it is determined that the target reaction curve meets the unrepairable condition, and its abnormality is irreparable. The target fault can be preset based on actual application conditions, and may include, but is not limited to, abnormal termination of the experiment, reagent empty aspiration, and sample empty aspiration.

[0117] Regarding the number of effective photometers in the response curve to be identified being less than the photometer threshold, effective photometers are measurement positions or time points used to obtain reliable data during the detection process. They usually need to meet conditions such as signal stability, low background interference, and compliance with experimental design requirements. If it is determined that the number of effective photometers in the response curve to be identified is less than the photometer threshold, it indicates that there are insufficient effective photometers in the response curve to be identified and there is no reliable calculation range. Therefore, the response curve to be identified is judged to meet the unrepairable condition, and its abnormality cannot be repaired.

[0118] Regarding the similarity between the response curve to be identified and the target calibration curve being below the lower similarity limit, specifically, the similarity between the response curve to be identified and the target calibration curve can be calculated, i.e., the target similarity described above. If the similarity is below the lower similarity limit, the response curve to be identified is determined to meet the unrepairable condition, and its anomaly cannot be repaired. The lower similarity limit can be preset based on actual application conditions. It should be noted that if the lower similarity limit is below the target similarity threshold described above, that is, when the similarity between the response curve to be identified and the target calibration curve is below the target similarity threshold, it indicates that the response curve to be identified has an anomaly, but its anomaly may be repairable. However, if it further falls below the lower similarity limit, the anomaly is unrepairable.

[0119] In this way, by setting the aforementioned unrepairable conditions, various unrepairable anomalies can be accurately identified, thereby avoiding interference from invalid data and improving the reliability of the analysis results.

[0120] In some embodiments, the sample response curve processing method provided in this application may further include the following step before obtaining the repaired response curve based on the anomaly identification result and the target calibration product response curve: determining a baseline response curve from the target calibration product response curve. Based on this, the step of obtaining the repaired response curve based on the anomaly identification result and the target calibration product response curve may include the following step: obtaining the repaired response curve based on the baseline response curve and the anomaly type corresponding to the anomaly identification result.

[0121] A baseline response curve is a target calibration curve established as a reference standard. It can be used to determine the type of anomaly in a response curve to be identified and to repair the curve. The method for determining the baseline response curve from the target calibration curve is as described above and will not be repeated here.

[0122] In this embodiment, the repaired response curve can be obtained based on the baseline response curve and the anomaly type corresponding to the anomaly identification result. Specifically, the baseline response curve is used as the basis for repairing the response curve to be identified, and the anomaly type corresponding to the anomaly identification result is used to determine the repair strategy for the response curve to be identified. Each anomaly type has its corresponding repair strategy.

[0123] In this way, based on the baseline response curve and the anomaly type of the response curve to be identified, targeted and differentiated anomaly strategies can be implemented, resulting in high repair accuracy.

[0124] In some embodiments, the step of obtaining the repaired response curve based on the baseline response curve and the anomaly type corresponding to the anomaly identification result in the sample response curve processing method provided in this application may include the following steps: determining the anomaly type with the highest repair priority from the anomaly types corresponding to the anomaly identification result, wherein the repair priority of curve shape anomaly is higher than the repair priority of curve smoothness anomaly; when the anomaly type with the highest repair priority is curve shape anomaly, scaling the baseline response curve and fusing the response curve to be identified and the scaled baseline response curve to obtain the repaired response curve; when the anomaly type with the highest repair priority is curve smoothness anomaly, using the smooth response curve as the repaired response curve.

[0125] Repair priority can characterize the importance of repair, and can be preset based on actual needs. In a specific embodiment, the repair priority for abnormal curve shape is higher than the repair priority for abnormal curve smoothness.

[0126] From the anomaly identification results, the anomaly type with the highest repair priority is determined, and then the response curve to be identified is repaired based on the repair strategy corresponding to that anomaly type. For example, when the anomaly identification results include both curve shape anomalies and curve smoothness anomalies, the anomaly type with the highest repair priority is determined to be curve shape anomalies. As another example, when the anomaly identification results only include curve smoothness anomalies, the anomaly type with the highest repair priority is determined to be curve smoothness anomalies.

[0127] Specifically, when the highest priority anomaly type to be repaired is curve morphology anomaly, the baseline response curve is scaled down, and the response curve to be identified and the scaled-down baseline response curve are fused together to obtain the repaired response curve.

[0128] The scaling process involves adjusting the shape of the baseline response curve through mathematical transformations to match the abnormal region features of the response curve to be identified. This provides a more suitable correction benchmark for subsequent fusion. Mathematical transformations include linear / nonlinear scaling, translation, or weighting. The fusion process involves integrating the scaled baseline and the response curve to be identified in the abnormal region, using methods such as weighted averaging or interpolation to obtain the repaired response curve. In this way, for cases of abnormal curve shape, the repaired response curve is obtained by combining the characteristics of the baseline and the response curve to be identified, allowing for targeted anomaly repair with high accuracy.

[0129] Furthermore, when the highest priority anomaly type for repair is curve smoothness anomaly, the smoothed reaction curve is directly used as the repaired reaction curve. As described above, the smoothed reaction curve is obtained based on the curve model corresponding to the baseline reaction curve and its target model parameters. The target model parameters are obtained by fitting a preset fitting model based on the curve models corresponding to the reaction curve to be identified and the baseline reaction curve; specific details are not elaborated here.

[0130] Thus, precise and reasonable repair strategies are adopted to address important anomalies such as abnormal curve smoothness and abnormal curve shape, ensuring the reliability of the detection.

[0131] In some embodiments, such as Figure 10 As shown, a sample reaction curve processing device 1000 is also provided, which may include the following modules 1002 to 1012.

[0132] The module 1002 for acquiring the target curve is used to acquire the target response curve of the target sample.

[0133] The calibration sample reaction curve acquisition module 1004 is used to acquire the target calibration sample reaction curve; the target calibration sample reaction curve is the reaction curve of the calibration sample used in the target calibration operation for the target detection item, and the target detection item is the detection item corresponding to the reaction curve to be identified.

[0134] The curve anomaly identification module 1006 is used to obtain the anomaly identification result of the reaction curve to be identified based on the reaction curve of the target calibration product.

[0135] The curve repair analysis module 1008 is used to determine whether the abnormality of the response curve to be identified can be repaired when the anomaly identification result indicates that the response curve to be identified is an abnormal response curve.

[0136] The repair curve acquisition module 1010 is used to obtain the repaired reaction curve based on the anomaly identification result and the target calibration product reaction curve if repair is possible.

[0137] The reaction curve display module 1012 is used to display the reaction curve to be identified and the repaired reaction curve on the target interface.

[0138] The above-mentioned sample reaction curve processing method, apparatus, computer-readable storage medium, control terminal, and sample analysis system acquire the target sample's reaction curve to be identified and the target calibration product's reaction curve, i.e., the reaction curve of the calibration product used in the target calibration operation corresponding to the detection item of the reaction curve to be identified. Based on the target calibration product's reaction curve, the abnormal identification result of the reaction curve to be identified is obtained. When the reaction curve to be identified has an abnormality and the abnormality can be repaired, the repaired reaction curve is obtained based on the abnormality identification result and the target calibration product's reaction curve, and the reaction curve to be identified and the repaired reaction curve are displayed on the target interface. Thus, by using the response curve of the target calibration product to determine whether there are any anomalies in the response curve to be identified, this method of identifying anomalies through computer analysis of curve morphology features does not require manual setting and filling in of a large number of parameters, has low dependence on human experience, and can effectively avoid misjudgments caused by improper parameter settings. Moreover, because it focuses on the shape, trend, and fluctuation of the curve, it can break through the limitation of traditional numerical threshold methods that can only detect single-point or local numerical anomalies, effectively capture global pattern anomalies, and thus more comprehensively identify anomalies. At the same time, it displays the response curves before and after repair, which can intuitively and clearly reflect the anomalies and repair effects for user reference.

[0139] In some embodiments, the target interface includes an item detection information display area. Based on this, the sample reaction curve processing device 1000 provided in this application may further include the following modules: an item detection information display module, used to display item detection information corresponding to the reaction curve to be identified and item detection information corresponding to the repaired reaction curve in the item detection information display area; a target reaction curve determination module, used to determine the target reaction curve based on the selection operation of the item detection information in the item detection information display area; and an effective detection result determination module, used to determine the effective detection result of the target sample based on the target reaction curve in response to the review trigger operation.

[0140] In some embodiments, the sample response curve processing device 1000 provided in this application may further include the following module: an anomaly analysis information display module, used to display anomaly analysis information corresponding to the response curve to be identified in the target interface; the anomaly analysis information is used to characterize the cause of the anomaly.

[0141] In some embodiments, the sample response curve processing device 1000 provided in this application may further include the following modules: a target similarity calculation module, used to perform similarity calculation based on the response curve to be identified and the target calibration product response curve to obtain the target similarity; a morphology analysis result acquisition module, used to obtain the curve morphology analysis result of the response curve to be identified based on the target similarity; and a first abnormal result acquisition module, used to obtain the abnormal identification result of the response curve to be identified based on the curve morphology analysis result.

[0142] In some embodiments, the sample response curve processing apparatus 1000 provided in this application may further include the following modules: a curve model determination module, used to determine the curve model corresponding to the target calibration product response curve based on the curve type of the target calibration product response curve; a target model parameter determination module, used to determine the target model parameters of the curve model corresponding to the target calibration product response curve by fitting the curve model corresponding to the response curve to be identified and the curve model corresponding to the target calibration product response curve using a preset fitting model; a smooth response curve generation module, used to obtain a smooth response curve based on the curve model corresponding to the target calibration product response curve and its target model parameters; a difference result calculation module, used to calculate the difference between the smooth response curve and the response curve to be identified, and obtain a difference result; a smoothness analysis result generation module, used to obtain the curve smoothness analysis result of the response curve to be identified based on the difference result; and a second abnormal result acquisition module, used to obtain the abnormal identification result of the response curve to be identified based on the curve smoothness analysis result.

[0143] In some embodiments, the sample response curve processing apparatus 1000 provided in this application may further include the following modules:

[0144] The segmentation processing module is used to segment the target calibration product response curve based on the experimental configuration information corresponding to the target detection project, and obtain multiple sub-calibration product response curves.

[0145] The curve type determination module is used to determine the curve type corresponding to each sub-calibrator reaction curve through a curve classification model.

[0146] Based on this, the curve model determination module is specifically used to determine the curve model corresponding to each sub-calibrator reaction curve based on the curve type corresponding to each sub-calibrator reaction curve.

[0147] In some embodiments, in the sample response curve processing apparatus 1000 provided in this application, the curve repair analysis module 1008 can be specifically used to determine whether the response curve to be identified meets any one of the unrepairable conditions; if it meets the conditions, it is determined that the abnormality of the response curve to be identified is unrepairable; if it does not meet the conditions, it is determined that the abnormality of the response curve to be identified is repairable; wherein, the unrepairable conditions include at least one of the following: the response curve to be identified is associated with the target fault information reported by the target sample analyzer, the number of effective photometers in the response curve to be identified is less than the photometer threshold, and the similarity between the response curve to be identified and the target calibration product response curve is lower than the similarity lower limit.

[0148] In some embodiments, the sample reaction curve processing apparatus 1000 provided in this application may further include a baseline curve determination module, used to determine a baseline reaction curve from the target calibration product reaction curve. Based on this, the repair curve acquisition module 1010 is specifically used to obtain the repaired reaction curve based on the baseline reaction curve and the anomaly type corresponding to the anomaly identification result.

[0149] In some embodiments, the sample response curve processing apparatus 1000 provided in this application may include a repair curve acquisition module 1010, which may include the following units: a repair priority analysis unit, used to determine the abnormal type with the highest repair priority from the abnormal types corresponding to the abnormal identification results; the repair priority of curve morphology abnormality is higher than the repair priority of curve smoothness abnormality; a morphology repair unit, used to scale the baseline response curve when the abnormal type with the highest repair priority is curve morphology abnormality, and to perform fusion processing based on the response curve to be identified and the scaled baseline response curve to obtain the repaired response curve; and a smoothness repair unit, used to use the smooth response curve as the repaired response curve when the abnormal type with the highest repair priority is curve smoothness abnormality; wherein, the smooth response curve is obtained based on the curve model corresponding to the baseline response curve and its target model parameters, and the target model parameters are obtained by fitting a preset fitting model based on the curve model corresponding to the response curve to be identified and the baseline response curve.

[0150] It should be noted that the specific limitations of the sample response curve processing device 1000 can be found in the limitations of the corresponding sample response curve processing methods described above, and will not be repeated here. Each module in the sample response curve processing device 1000 can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in the control terminal, or stored in software in the memory of the control terminal, so that the processor can call and execute the corresponding operations of each module.

[0151] In some embodiments, a computer-readable storage medium is provided storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the sample response curve processing method provided in any embodiment of this application.

[0152] In some embodiments, a control terminal is provided, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the sample response curve processing method provided in any embodiment of this application.

[0153] Specifically, the aforementioned control terminal can be Figure 1 The control terminal 110 shown in the figure has an internal structure as follows: Figure 10 As shown, the system includes a processor, memory, network interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs, while the internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. When executed by the processor, the computer programs implement the steps of the sample response curve processing method provided in any embodiment of this application. The network interface is used for communication with external terminals via a network connection. The display screen can be a liquid crystal display (LCD) or an electronic ink display. The input device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad mounted on the control terminal casing, or an external keyboard, touchpad, or mouse, etc.

[0154] In some embodiments, the sample response curve processing apparatus 1000 provided in this application can be implemented as a computer program, which can be configured as follows: Figure 11 The control terminal shown operates on this device. The control terminal's memory can store the various program modules that make up the device, for example, Figure 10 The examples shown include the curve acquisition module 1002 to be identified and the calibration sample reaction curve acquisition module 1004. The computer program, comprised of these modules, causes the processor to execute the steps of the sample reaction curve processing method provided in any embodiment of this application, for example... Figure 11 The control terminal shown can be accessed via, for example... Figure 10 The sample reaction curve processing device 1000 shown in the figure executes step S202 through the target curve acquisition module 1002 and step S1004 through the calibration product reaction curve acquisition module 1004.

[0155] Furthermore, it is understandable that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the solution of this application, and does not constitute a limitation on the control terminal to which the solution of this application is applied. Specific control terminals may include, but are not limited to, those shown in the diagram. Figure 11 The diagram shows more or fewer components, or combinations of certain components, or different component arrangements.

[0156] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The aforementioned program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0157] In some embodiments, a sample analysis system is provided, including a target sample analysis instrument and a control terminal, wherein the target sample analysis instrument and the control terminal are communicatively connected. Based on this, the target sample analysis instrument is used to detect a target sample, obtain a detection signal, and send the detection signal to the control terminal; the control terminal is used to obtain a target sample reaction curve based on the detection signal, and to execute the steps of the sample reaction curve processing method provided in any embodiment of this application.

[0158] The aforementioned computer-readable storage medium, control terminal, and sample analysis system acquire the target sample's reaction curve to be identified, as well as the target calibration product's reaction curve, i.e., the reaction curve of the calibration product used in the target calibration operation corresponding to the detection item of the reaction curve to be identified. Based on the target calibration product's reaction curve, the system obtains the anomaly identification result of the reaction curve to be identified. When the reaction curve to be identified has an anomaly and the anomaly is repairable, the system obtains the repaired reaction curve based on the anomaly identification result and the target calibration product's reaction curve, and displays the reaction curve to be identified and the repaired reaction curve on the target interface. Thus, by using the response curve of the target calibration product to determine whether there are any anomalies in the response curve to be identified, this method of identifying anomalies through computer analysis of curve morphology features does not require manual setting and filling in of a large number of parameters, has low dependence on human experience, and can effectively avoid misjudgments caused by improper parameter settings. Moreover, because it focuses on the shape, trend, and fluctuation of the curve, it can break through the limitation of traditional numerical threshold methods that can only detect single-point or local numerical anomalies, effectively capture global pattern anomalies, and thus more comprehensively identify anomalies. At the same time, it displays the response curves before and after repair, which can intuitively and clearly reflect the anomalies and repair effects for user reference.

[0159] It is stated that the technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0160] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for processing sample response curves, characterized in that, include: Obtain the target sample's response curve to be identified; Obtain the target calibration product response curve; the target calibration product response curve is the response curve of the calibration product used in the target calibration operation for the target detection item, and the target detection item is the detection item corresponding to the response curve to be identified; Based on the target calibration product response curve, the anomaly identification result of the response curve to be identified is obtained; When the anomaly identification result indicates that the response curve to be identified is an abnormal response curve, it is determined whether the anomaly of the response curve to be identified can be repaired; If it can be repaired, the repaired reaction curve is obtained based on the anomaly identification results and the target calibration product reaction curve; The target interface displays the reaction curve to be identified and the repaired reaction curve.

2. The method as described in claim 1, characterized in that, The target interface includes a project detection information display area, and the method further includes: In the project detection information display area, project detection information corresponding to the reaction curve to be identified and project detection information corresponding to the repaired reaction curve are displayed; Based on the selection operation of the project detection information in the project detection information display area, the target response curve is determined; In response to the audit trigger operation, the valid detection result of the target sample is determined based on the target response curve.

3. The method as described in claim 1, characterized in that, After obtaining the anomaly identification result of the reaction curve to be identified based on the target calibration product reaction curve, the method further includes: The target interface displays anomaly analysis information corresponding to the reaction curve to be identified; the anomaly analysis information is used to characterize the cause of the anomaly.

4. The method as described in claim 1, characterized in that, The step of obtaining the anomaly identification result of the response curve to be identified based on the target calibration product response curve includes: The similarity is calculated based on the reaction curve to be identified and the reaction curve of the target calibrator to obtain the target similarity. Based on the target similarity, the curve shape analysis results of the response curve to be identified are obtained; Based on the curve morphology analysis results, the anomaly identification results of the reaction curve to be identified are obtained.

5. The method as described in claim 1, characterized in that, The step of obtaining the anomaly identification result of the response curve to be identified based on the target calibration product response curve includes: Based on the curve type of the target calibration product reaction curve, determine the curve model corresponding to the target calibration product reaction curve; By using a preset fitting model, the target model parameters of the curve model corresponding to the reaction curve to be identified and the reaction curve of the target calibrator are determined by fitting the curve model corresponding to the reaction curve of the target calibrator. Based on the curve model corresponding to the reaction curve of the target calibrator and its target model parameters, a smooth reaction curve is obtained; The difference is calculated based on the smooth reaction curve and the reaction curve to be identified to obtain the difference result. Based on the difference results, the curve smoothness analysis results of the reaction curve to be identified are obtained; Based on the curve smoothness analysis results, the anomaly identification results of the reaction curve to be identified are obtained.

6. The method as described in claim 5, characterized in that, The method further includes: Based on the experimental configuration information corresponding to the target detection item, the target calibration product reaction curve is segmented to obtain multiple sub-calibration product reaction curves. The curve classification model is used to determine the curve type corresponding to the reaction curves of each of the sub-calibrators. The step of determining the curve model corresponding to the target standard reaction curve based on the curve type of the target standard reaction curve includes: Based on the curve type corresponding to the reaction curves of each of the sub-calibrators, the curve model corresponding to the reaction curves of each of the sub-calibrators is determined.

7. The method according to any one of claims 1 to 6, characterized in that, The step of determining whether the anomaly of the response curve to be identified can be repaired includes: Determine whether the reaction curve to be identified satisfies any one of the irreparable conditions; If the condition is met, the anomaly of the reaction curve to be identified is determined to be irreparable. If the conditions are not met, it is determined that the anomaly in the reaction curve to be identified can be repaired; The unrepairable conditions include at least one of the following: the reaction curve to be identified is associated with the target fault information reported by the target sample analyzer; the number of effective photometers in the reaction curve to be identified is less than the photometer threshold; and the similarity between the reaction curve to be identified and the target calibration product reaction curve is lower than the similarity lower limit.

8. The method according to any one of claims 1 to 6, characterized in that, Before obtaining the repaired response curve based on the anomaly identification result and the target calibration product response curve, the method further includes: Determine the baseline reaction curve from the target calibrator reaction curve; The process of obtaining the repaired response curve based on the anomaly identification result and the target calibration product response curve includes: Based on the baseline response curve and the anomaly type corresponding to the anomaly identification result, the repaired response curve is obtained.

9. The method as described in claim 8, characterized in that, The process of obtaining the repaired response curve based on the baseline response curve and the anomaly type corresponding to the anomaly identification result includes: From the anomaly identification results, determine the anomaly type with the highest repair priority; among them, the repair priority of curve shape anomalies is higher than that of curve smoothness anomalies. When the anomaly type with the highest repair priority is curve morphology anomaly, the baseline reaction curve is scaled down, and a fusion process is performed based on the reaction curve to be identified and the scaled-down baseline reaction curve to obtain the repaired reaction curve. When the anomaly type with the highest repair priority is curve smoothness anomaly, the smooth reaction curve is used as the repaired reaction curve; wherein, the smooth reaction curve is obtained based on the curve model corresponding to the benchmark reaction curve and its target model parameters, and the target model parameters are obtained by fitting a preset fitting model based on the curve model corresponding to the reaction curve to be identified and the benchmark reaction curve.

10. A device for processing sample response curves, characterized in that, include: The module for acquiring the target curve is used to acquire the target response curve of the target sample. The calibration product reaction curve acquisition module is used to acquire the target calibration product reaction curve; the target calibration product reaction curve is the reaction curve of the calibration product used in the target calibration operation for the target detection item, and the target detection item is the detection item corresponding to the reaction curve to be identified. The curve anomaly identification module is used to obtain the anomaly identification result of the reaction curve to be identified based on the target calibration product reaction curve; The curve repair analysis module is used to determine whether the abnormality of the response curve to be identified can be repaired when the anomaly identification result indicates that the response curve to be identified is an abnormal response curve. The repair curve acquisition module is used to obtain the repaired reaction curve based on the anomaly identification result and the target calibration product reaction curve, if repairable. The reaction curve display module is used to display the reaction curve to be identified and the repaired reaction curve on the target interface.

11. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for processing the sample response curve according to any one of claims 1 to 9.

12. A control terminal, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for processing the sample response curve according to any one of claims 1 to 9.

13. A sample analysis system, characterized in that, It includes a target sample analyzer and a control terminal, wherein the target sample analyzer and the control terminal are communicatively connected; The target sample analysis instrument is used to detect the target sample, obtain a detection signal, and send the detection signal to the control terminal; The control terminal is used to obtain the target sample's reaction curve based on the detection signal, and to perform the steps of the sample reaction curve processing method according to any one of claims 1 to 9.

Images