Sample confusion and pollution dynamic early warning system for quality control in blood analysis room

Abstract

The invention relates to the technical field of medical examination, and discloses a sample confusion and pollution dynamic early warning system for quality control in a blood analysis room. Comprises: a sample pretreatment module configured to obtain a red blood cell suspension corresponding to a to-be-detected blood sample and to-be-detected serum; the antigen detection module is used for performing a specific binding reaction according to the specific binding reagent and a first antigen preset on the surface of an erythrocyte membrane corresponding to the erythrocyte suspension, and detecting whether a binding signal of the first antigen and the specific binding reagent exists on the surface of the erythrocyte membrane or not in real time; the control module is used for performing comparative analysis on the intensity characteristics and the distribution characteristics of the combined signals and standard antigen information in a standard antigen database to generate sample antigen matching degree data; if an abnormal signal is detected according to the sample antigen matching degree data, sample confusion or pollution early warning information is generated based on a preset early warning rule, and dynamic early warning of confusion or pollution of the to-be-detected blood sample is achieved.

Description

Technical Field

[0001] The present application relates to the field of medical testing technology, and in particular to a dynamic early warning system for sample confusion and contamination used for quality control in blood analysis rooms. Background Art

[0002] In-house quality control (IQC) for blood analysis is a critical step in ensuring the accuracy and reliability of blood analysis results. In medical laboratories, especially when performing blood analysis, blood samples are often difficult to obtain from patients. Internal quality control procedures are crucial to monitor the stability of analytical performance and ensure the reliability of test results.

[0003] Current mainstream clinical laboratory medicine designs all use quality control charts to record and monitor quality control results. Commonly used quality control charts include Levey-Jennings charts, which analyze quality control data using statistical methods such as mean and standard deviation to determine whether there are systematic or accidental errors. In the drawing of relevant charts, blood sample mix-up / contamination is a problem that must be strictly controlled in clinical laboratories, as incorrect sample matching can lead to serious medical consequences, including misdiagnosis and severe adverse blood reactions.

[0004] Common mix-ups and contamination situations include: (1) mislabeling: inaccurate or missing information on the label of the sample tube or specimen container; (2) sample exchange: during sample collection, transportation, or processing, samples are incorrectly exchanged, resulting in cross-contamination; (3) incomplete records: during sample registration or result entry, patient information is incomplete or incorrect, resulting in an inability to correctly match. All of the above mix-ups can be discovered when the laboratory / testing department's results are inconsistent with the patient's medical history or other test results, but they cannot provide early warning when there is no patient history or incomplete records.

[0005] Therefore, there is an urgent need to design a technical solution to solve at least one of the above technical problems. Summary of the Invention

[0006] This application provides a dynamic early warning system for sample mix-up and contamination for quality control in blood analysis rooms, aiming to solve the current mainstream clinical laboratory medicine design, which uses quality control charts to record and monitor quality control results. Common quality control charts include Levey-Jennings Charts, which analyze quality control data by using statistical methods such as mean and standard deviation to determine whether there is systematic error or accidental error. In the drawing of relevant charts, blood sample mix-up / contamination is a problem that must be strictly controlled in clinical laboratories, because incorrect matching of samples may lead to serious medical consequences, including misdiagnosis and serious adverse blood reactions. Common confusion and contamination situations include: (1) labeling errors: inaccurate or missing information on the label of the sample tube or specimen container; (2) sample exchange: during sample collection, transportation or processing, the sample is incorrectly exchanged, resulting in cross-contamination; (3) incomplete records: during the sample registration or result entry stage, the patient information record is incomplete or incorrect, resulting in the inability to correctly match. All of the above confusions can be discovered when the laboratory / testing department's results are inconsistent with the patient's medical history or other test results, but they cannot provide early warning when the patient's medical history is missing or the records are incomplete.

[0007] In a first aspect, the present application provides a dynamic early warning system for sample mix-up and contamination for in-house quality control of blood analysis, comprising:

[0008] A sample pre-processing module is configured to perform red blood cell separation on the blood sample to be tested, and obtain a red blood cell suspension and corresponding serum to be tested;

[0009] An antigen detection module is configured to cause a specific binding reaction between a specific binding reagent and a first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension, so as to incubate the serum to be tested with the red blood cell suspension and detect in real time whether a binding signal between the first antigen and the specific binding reagent is present on the surface of the red blood cell membrane;

[0010] The control module pre-stores a standard antigen database, which stores standard red blood cell membrane surface antigen information corresponding to multiple target detection samples. Based on the intensity characteristics and distribution characteristics of the binding signal and the standard antigen information in the standard antigen database, a comparison analysis is performed to generate sample antigen matching data; based on the sample antigen matching data, it is determined whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current detection process. If the abnormal signal is detected, sample confusion or contamination warning information is generated based on preset warning rules, thereby realizing dynamic warning of confusion or contamination of the blood sample to be tested.

[0011] In some embodiments, the blood sample to be tested is subjected to red blood cell separation processing to obtain a red blood cell suspension and a corresponding serum to be tested, including: performing gradient centrifugation processing on the blood sample to be tested using a centrifugal separation device to separate a blood cell layer and a serum layer; wherein the red blood cell suspension is prepared by washing the blood cell layer with physiological saline and resuspending it, and the serum to be tested is obtained by filtering the serum layer for impurities, and the corresponding association between the original identification information of the sample and the physical sample is maintained during the separation process.

[0012] In some embodiments, the specific binding reagent is a monoclonal antibody labeled with a fluorescent group or an enzyme marker, and the monoclonal antibody is designed for the first antigen epitope preset on the surface of the red blood cell membrane; the specific binding reaction between the specific binding reagent and the first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension comprises: mixing the serum to be tested with the red blood cell suspension under preset incubation conditions, and collecting the fluorescence signal intensity or color reaction intensity on the surface of the red blood cell membrane in real time by a fluorescence imaging device or an enzyme-linked immunosorbent assay device, wherein the binding signal exists in the form of focal fluorescence aggregation or uniform fluorescence label distribution based on the cell surface.

[0013] Exemplarily, the preset incubation conditions include a constant temperature environment of 35-38° C. and an incubation time of 15-30 minutes.

[0014] In some embodiments, the intensity characteristics and distribution characteristics of the binding signal are compared and analyzed with the standard antigen information in the standard antigen database to generate sample antigen matching data, including: converting the intensity value of the binding signal collected in real time into a standardized value, and the standardized value is used to match the first antigen signal intensity threshold range of the corresponding sample type in the standard antigen database; analyzing the distribution ratio of the binding signal in the red blood cell population and the signal distribution uniformity on the surface of a single cell, and calculating the comprehensive matching degree through a preset matching algorithm; the standard antigen database at least includes normal antigen expression patterns corresponding to different patient groups and different test items, and antigen signal characteristic parameters of weakly positive and strongly positive control samples.

[0015] In some embodiments, the method of judging whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current testing process based on the sample antigen matching data includes: when the sample antigen matching is lower than a preset normal matching threshold, or an unexpected antigen binding signal is detected on the red blood cell membrane surface, judging that the abnormal signal exists, and the expected antigen characteristics are comprehensively determined based on the test menu of the current test item, the sample type identifier and the historical test records; the normal matching threshold is dynamically adjusted based on historical quality control data, and at least includes a dual judgment of a signal intensity matching threshold and a distribution uniformity matching threshold; the antigen binding signal is an abnormal antigen epitope binding signal that is not recorded in the standard antigen database.

[0016] In some embodiments, the sample confusion or contamination warning information is generated based on preset warning rules, including: according to the type and severity of the abnormal signal, a warning is simultaneously issued through an audible and visual alarm device, a laboratory information system pop-up window and a sample tube label color change module. The warning information at least includes the time when the abnormal signal occurs, the unique identifier of the sample, the abnormal antigen characteristic description and the recommended processing flow, and at the same time triggers the sample traceability program, associates and traces the sample collection time, operator information, previous processing equipment usage records and adjacent sample test results to form a complete abnormal event log; the severity classification includes suspected contamination, confirmed confusion, and severe cross-contamination.

[0017] In some embodiments, the control module is further used to: at the initial stage of the detection process when the serum contacts the red blood cell membrane, identify unexpected antigen characteristics that may be present in the sample through the specific reaction of the specific binding reagent with the first antigen on the surface of the red blood cell membrane.

[0018] In a second aspect, the present application provides a dynamic early warning method for sample mixup and contamination for quality control in a blood analysis room, characterized in that the method is applied to a control module of a dynamic early warning system for sample mixup and contamination for quality control in a blood analysis room provided in any embodiment of the present application, and comprises:

[0019] The sample pre-processing module performs red blood cell separation on the blood sample to be tested, and obtains the red blood cell suspension and the corresponding serum to be tested;

[0020] Acquiring a binding signal generated by the antigen detection module based on a specific binding reaction between a specific binding reagent and a first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension; the specific binding reaction is used to co-incubate the serum to be tested with the red blood cell suspension and detect in real time whether the binding signal between the first antigen and the specific binding reagent exists on the red blood cell membrane surface;

[0021] Based on the intensity characteristics and distribution characteristics of the binding signal and the standard antigen information in the standard antigen database, a comparison analysis is performed to generate sample antigen matching data; based on the sample antigen matching data, it is determined whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current detection process; if the abnormal signal is detected, sample confusion or contamination warning information is generated based on preset warning rules, thereby realizing dynamic warning of confusion or contamination of the blood sample to be tested; the standard antigen database stores standard red blood cell membrane surface antigen information corresponding to multiple target detection samples.

[0022] In a third aspect, the present application provides a dynamic early warning device for sample mixup and contamination for quality control in a blood analysis room, which is applied to a control module of a dynamic early warning system for sample mixup and contamination for quality control in a blood analysis room provided in any embodiment of the present application, and the device comprises:

[0023] The first acquisition unit is used to obtain the red blood cell suspension and the corresponding serum to be tested by the sample pre-processing module to perform red blood cell separation processing on the blood sample to be tested;

[0024] a second acquisition unit, configured to acquire a binding signal generated by the antigen detection module based on a specific binding reaction between a specific binding reagent and a first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension; the specific binding reaction is used to co-incubate the serum to be tested with the red blood cell suspension and to detect in real time whether the binding signal between the first antigen and the specific binding reagent exists on the surface of the red blood cell membrane;

[0025] The dynamic early warning unit is used to compare and analyze the intensity characteristics and distribution characteristics of the binding signal with the standard antigen information in the standard antigen database to generate sample antigen matching data; based on the sample antigen matching data, it is judged whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current detection process; if the abnormal signal is detected, sample confusion or contamination early warning information is generated based on preset early warning rules, so as to realize dynamic early warning of confusion or contamination of the blood sample to be tested; the standard antigen database stores standard red blood cell membrane surface antigen information corresponding to multiple target detection samples.

[0026] In a fourth aspect, the present application provides a control module, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and implement the method provided in any embodiment of the present application when executing the computer program.

[0027] In a fifth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer-readable instructions are executed by the processor, one or more processors execute the method provided in any embodiment of the present application.

[0028] The dynamic early warning system provided by the present invention focuses on the specific detection of red blood cell surface antigens in blood samples, and constructs a complete technical solution from sample processing to abnormal signal identification. Specifically, it includes the following core contents:

[0029] The sample preprocessing module uses red blood cell (RBC) separation techniques (such as gradient centrifugation, saline washing, and impurity filtration) to obtain a RBC suspension and the serum to be tested. This ensures a one-to-one correspondence between the original identification information and the sample during the physical separation process, providing a pure test sample for subsequent antigen testing. The antigen detection module utilizes a specific binding reagent (such as a monoclonal antibody labeled with a fluorescent group or enzyme) that binds to a pre-determined first antigen on the RBC membrane. This reagent is incubated with the serum to be tested and the RBC suspension, capturing the antigenic signature on the RBC membrane through real-time detection of binding signals (such as fluorescence aggregation and color development). The control module integrates a standard antigen database, storing normal antigen expression patterns and control sample parameters for different sample types. By comparing the intensity and distribution of the real-time detection signal with the standard information in the database, the sample antigen match is calculated. Based on pre-set rules, abnormal signals are identified and an alert is generated. Pre-emptive testing: Antigen-specific reaction testing is performed at the initial stage of the testing process, when the serum contacts the RBC membrane. This allows for the early identification of unexpected antigenic signatures (such as foreign antigen contamination due to sample mix-up), unlike traditional post-testing that relies on comparison of results or medical history. Multi-dimensional signal analysis: Combines signal intensity, distribution ratio within the red blood cell population, and uniformity of individual cell signals to perform comprehensive matching calculations, avoiding misjudgment of a single indicator and improving detection sensitivity and specificity. Dynamic early warning mechanism: Based on the type of abnormal signal (such as labeling errors, sample exchange, abnormal antigen characteristics corresponding to incomplete records) and severity level (suspected contamination, confirmed confusion, severe cross-contamination), real-time early warning is issued through multiple channels such as sound and light alarms, system pop-ups, and label color changes. This triggers the sample traceability program and links operation records to form a complete log.

[0030] In summary, the present invention addresses the limitations of traditional blood analysis quality control, where sample mixup / contamination warning relies on medical history or result comparison, and has the following beneficial effects:

[0031] 1. Early warning to make up for the shortcomings of traditional quality control: Traditional methods (such as the Levy-Jennings quality control chart) judge errors through statistical analysis of test results. Abnormalities can only be detected after the test and rely on the patient's medical history or comparison of multiple groups of results. However, this system directly identifies abnormalities in the sample's own antigenic characteristics through antigen-specific reactions at the initial stage of the testing process. There is no need to rely on external medical history or subsequent results. This solves the problem of "no warning when records are incomplete or medical history is lacking" and achieves advance prevention of confusion / contamination.

[0032] 2. Improve detection reliability and reduce medical risks: Through the precise detection of red blood cell membrane antigens using specific binding reagents, unexpected antigen characteristics caused by sample confusion or contamination (such as foreign antigens introduced by cross-contamination) can be directly captured, avoiding serious consequences such as misdiagnosis and adverse blood reactions caused by incorrect sample matching, thereby improving the accuracy and reliability of test results.

[0033] 3. Optimize processes and reduce sample waste: Complete abnormality screening before serum comes into contact with red blood cells, avoid invalid testing of contaminated samples, reduce the waste of precious clinical samples (especially blood samples that are difficult to obtain), reduce the cost of repeated laboratory testing, and improve overall testing efficiency.

[0034] 4. Multi-dimensional data support enhances traceability: The standard antigen database dynamically adjusts the matching threshold based on historical quality control data to support personalized quality control for different patient groups and testing items. The sample traceability program, triggered synchronously during the early warning, can trace the entire operation process, providing data support for laboratory quality improvement and building a closed-loop quality control system.

[0035] In summary, this invention, through antigen-specific detection and dynamic data comparison, establishes a sample mixup / contamination early warning mechanism independent of medical history. This fills a gap in traditional quality control monitoring during sample pretreatment, ensuring the accuracy of blood analysis from the very beginning, and possesses significant clinical application value and technological innovation.

[0036] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0038] Figure 1 This is a schematic block diagram of the structure of a dynamic early warning system for sample mix-up and contamination for quality control in a blood analysis room provided by an embodiment of the present application;

[0039] Figure 2 This is a schematic flow chart of the steps of a dynamic early warning method for sample mix-up and contamination in a blood analysis room for quality control provided by an embodiment of the present application;

[0040] Figure 3 This is a schematic block diagram of the structure of a control module provided in one embodiment of the present application.

[0041] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0043] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0044] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish between identical or similar items having substantially the same functions and effects. Those skilled in the art will understand that terms such as "first" and "second" do not limit the quantity or order of execution, and that terms such as "first" and "second" do not necessarily define differences.

[0045] It should be understood that the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0046] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0047] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0048] In-house quality control (IQC) for blood analysis is a critical step in ensuring the accuracy and reliability of blood analysis results. In medical laboratories, especially when performing blood analysis, blood samples are often difficult to obtain from patients. Internal quality control procedures are crucial to monitor the stability of analytical performance and ensure the reliability of test results.

[0049] Current mainstream clinical laboratory medicine designs all use quality control charts to record and monitor quality control results. Commonly used quality control charts include Levey-Jennings charts, which analyze quality control data using statistical methods such as mean and standard deviation to determine whether there are systematic or accidental errors. In the drawing of relevant charts, blood sample mix-up / contamination is a problem that must be strictly controlled in clinical laboratories, as incorrect sample matching can lead to serious medical consequences, including misdiagnosis and severe adverse blood reactions.

[0050] Common mix-ups and contamination situations include: (1) mislabeling: inaccurate or missing information on the label of the sample tube or specimen container; (2) sample exchange: during sample collection, transportation, or processing, samples are incorrectly exchanged, resulting in cross-contamination; (3) incomplete records: during sample registration or result entry, patient information is incomplete or incorrect, resulting in an inability to correctly match. All of the above mix-ups can be discovered when the laboratory / testing department's results are inconsistent with the patient's medical history or other test results, but they cannot provide early warning when there is no patient history or incomplete records.

[0051] Therefore, there is an urgent need to design a technical solution to solve at least one of the above technical problems.

[0052] To solve the above problems, please refer to Figure 1The present application provides a dynamic early warning system for sample confusion and contamination for in-house quality control of blood analysis, comprising: a sample preprocessing module configured to perform red blood cell separation on a blood sample to be tested, thereby obtaining a red blood cell suspension and a corresponding serum to be tested; an antigen detection module configured to induce a specific binding reaction between a specific binding reagent and a first antigen preset on the red blood cell membrane surface corresponding to the red blood cell suspension, thereby incubating the serum to be tested with the red blood cell suspension and detecting in real time whether a binding signal between the first antigen and the specific binding reagent exists on the red blood cell membrane surface; a control module pre-stored with a standard antigen database containing information on standard red blood cell membrane surface antigens corresponding to a plurality of target test samples, and generating sample antigen matching data by comparing and analyzing the intensity and distribution characteristics of the binding signal with the standard antigen information in the standard antigen database; and determining, based on the sample antigen matching data, whether an abnormal signal in the blood sample to be tested does not match the expected antigen characteristics of the current test process. If the abnormal signal is detected, generating sample confusion or contamination warning information based on preset warning rules, thereby achieving a dynamic early warning of confusion or contamination of the blood sample to be tested.

[0053] Specifically, by physically pre-treating the blood sample to be tested, using centrifugal separation technology (such as low-speed centrifugation, usually at a speed of 2000-3000 rpm, for 5-10 minutes), and utilizing the difference in blood component density (red blood cell density > serum), the blood sample is separated into an upper serum layer and a lower red blood cell layer.

[0054] The preparation of red blood cell suspension is done by discarding part of the upper serum after separation, retaining the red blood cell layer, adding physiological saline or buffer (such as PBS) for washing, repeating centrifugation and washing 2-3 times to remove interfering components (such as free antibodies and impurities) in the residual serum, and finally preparing a red blood cell suspension of a certain concentration (such as the concentration is adjusted to 1×10 6 -5×10 6 cells / μL).

[0055] The serum to be tested is retained after separation and stored separately as a sample for subsequent incubation with the red blood cell suspension to simulate the antigen-antibody reaction environment in the body. Physical separation technology is used to obtain pure red blood cell suspension and serum to be tested, eliminating interference from non-specific components and providing a standardized sample for subsequent antigen testing.

[0056] Selection of specific binding reagents: Based on the detection requirements (such as ABO blood group antigens, Rh blood group antigens, etc.), select specific reagents targeting the preset first antigen on the red blood cell membrane surface (such as fluorescently labeled monoclonal antibodies, antibody Fab segments targeting antigen epitopes, and Fc segments coupled to fluorescent dyes such as FITC and PE).

[0057] The co-incubation reaction involves mixing the prepared red blood cell suspension with the serum to be tested in a predetermined ratio (e.g., a 1:10 volume ratio) and incubating at an appropriate temperature (e.g., 37°C) and time (e.g., 15-30 minutes) to simulate the binding process between serum antibodies and red blood cell antigens in vivo. Contamination or confusion with the sample may introduce abnormal antigens or antibodies, leading to unintended binding reactions.

[0058] Optical detection techniques (such as flow cytometry and fluorescence microscopy) are used to monitor binding signals on the red blood cell membrane surface in real time. The intensity of the fluorescent signal is used to determine the number of bound antibodies (the higher the signal intensity, the more specific reagents are bound); the uniformity of the signal distribution on the cell surface (such as whether it is diffusely distributed or clustered) is used to determine binding specificity (non-specific binding typically results in uneven signal distribution or abnormal intensity). Using the antigen-antibody specific binding reaction, the antigenic characteristics of the red blood cells in the sample are converted into quantifiable optical signals, and abnormal antigen expression can be identified through the signal characteristics.

[0059] The standard antigen database is constructed by pre-collecting historically compliant samples (known to be free of contamination and confusion). Using the same testing process, data such as the signal intensity and distribution patterns of red blood cell surface antigens are obtained. This data is then used to construct standard antigen information (including signal intensity threshold ranges and typical distribution patterns for different antigen types, such as the characteristic signal models for ABO blood group antigens A, B, and O). The database supports dynamic updates, incorporating new standard sample data to optimize the model.

[0060] The real-time detected binding signal intensity characteristics (such as mean fluorescence intensity MFI) and distribution characteristics (such as signal uniformity coefficient) are compared with the corresponding antigen information in the standard antigen database, and the sample antigen matching degree is calculated using a similarity algorithm (such as Euclidean distance, cosine similarity) (such as matching degree = 1-distance value, the higher the value, the closer it is to the standard antigen).

[0061] Abnormal signal judgment and warning generation are achieved by setting warning thresholds (e.g., a match of <90% triggers a first-level warning, and <80% triggers a second-level warning). If a match is detected below the threshold or an atypical pattern appears in the signal distribution (e.g., a standard type A blood sample detects a type B blood characteristic signal), it is determined to be an "antigen characteristic mismatch abnormal signal." Based on preset warning rules (e.g., a single abnormal signal triggers a prompt, and two consecutive abnormalities trigger an alarm), the software system generates visual warning information (e.g., pop-up windows, sound and light alarms), and records the abnormal sample number, detection time, and abnormality type (confusion or contamination) for review by laboratory personnel. Through statistical comparison and pattern recognition, the antigen detection signal is converted into a quantifiable quality control indicator, enabling real-time verification of sample authenticity.

[0062] Traditional methods rely on medical history or record verification, and cannot provide early warning when information is incomplete. This system directly detects red blood cell antigen characteristics (such as blood type antigens are inherent attributes of the sample), without the need to associate with external medical history. Even if the sample is incorrectly labeled or the record is missing, it can directly identify abnormalities through antigen matching (such as a type A blood sample detecting a type B antigen signal, which directly determines confusion). Labeling error: By comparing the actual antigen test results with the expected labeling information of the sample (such as the blood type marked on the application form), labeling errors are discovered (such as misplaced labels resulting in antigen mismatches); unexpected antigen signals are detected (such as double antigen signals due to mixing in other people's red blood cells), and cross-contamination is identified; independent of registration information, the sample authenticity is directly verified through antigen characteristics to avoid misjudgment of results due to entry errors.

[0063] Combined with the real-time signal acquisition of antigen detection and the instant data analysis of the control module, dynamic monitoring of the entire process from sample processing to result output is achieved. Compared with traditional post-verification (such as tracing back after discovering abnormal results through quality control charts), the early warning time is brought forward to the detection process, greatly shortening the problem discovery cycle.

[0064] By replacing subjective judgment with matching data (quantitative indicators) and combining it with the standardized model of the standard antigen database, human errors can be reduced and the consistency of results between laboratories can be improved; it supports long-term storage of quality control data, facilitates the drawing of trend charts (such as the extended application of Levey-Jennings charts), and monitors the stability of the analysis system.

[0065] Directly block misdiagnosis caused by sample confusion (such as transfusion reactions caused by incorrect ABO blood typing) and abnormal results caused by contamination (such as cross-contamination leading to misjudgment of infection markers), ensuring the reliability of test results from the source and avoiding serious medical consequences. Sample pretreatment, antigen detection and control modules can be integrated into existing blood analysis pipelines and are compatible with mainstream testing equipment (such as flow cytometers and fully automatic blood typing analyzers) without the need for large-scale modification of laboratory hardware. The standard antigen database can be flexibly expanded (such as adding HLA antigens, rare blood type antigens, etc.) to suit different testing scenarios (routine blood typing, special antigen screening), improving the versatility of the system.

[0066] In summary, through the technical path of "physical separation - antigen-specific reaction - intelligent data comparison", this system establishes a real-time verification mechanism independent of sample labeling and medical history, filling the loophole of traditional quality control's reliance on external information and achieving "detection is verification" for sample mix-up and contamination. Its core value lies in moving the quality control node forward into the testing process. Through quantitative antigen matching indicators and dynamic early warning rules, it provides dual guarantees for the accuracy of blood analysis (traditional quality control charts monitor analytical performance + this system monitors sample authenticity), representing a significant technological innovation in clinical laboratory risk management.

[0067] In some embodiments, the blood sample to be tested is subjected to red blood cell separation processing to obtain a red blood cell suspension and a corresponding serum to be tested, including: performing gradient centrifugation processing on the blood sample to be tested using a centrifugal separation device to separate a blood cell layer and a serum layer; wherein the red blood cell suspension is prepared by washing the blood cell layer with physiological saline and resuspending it, and the serum to be tested is obtained by filtering the serum layer for impurities, and the corresponding association between the original identification information of the sample and the physical sample is maintained during the separation process.

[0068] Use a centrifuge with temperature control function (such as a medical-grade low-speed centrifuge) to perform gradient centrifugation on the blood sample to be tested. The centrifugation parameters are set as follows: speed 2000-3000 rpm, centrifugation time 8-10 minutes, and temperature control at 4-25°C (to avoid cell rupture). After centrifugation, the blood is stratified into an upper serum layer, a middle white blood cell / platelet layer, and a lower red blood cell layer. Use a pipette to accurately separate the serum layer (pipette 2 / 3 of the volume of the upper layer) and the red blood cell layer (compacted red blood cells in the lower layer), and avoid aspirating the middle white blood cell layer to reduce contamination. Add 3-5 times the volume of normal saline (0.9% NaCl solution) to the red blood cell layer, gently blow and resuspend, centrifuge at 1500-2000 rpm for 5 minutes, discard the supernatant (wash to remove residual serum proteins), repeat washing 2-3 times, and finally adjust the red blood cell concentration to 1×10 6 -5×10 6 pcs / μL.

[0069] The separated serum layer is filtered through a 0.22μm filter (to remove impurities such as cell debris and blood clots) to ensure that the serum is free of particulate matter that could interfere with subsequent antigen-antibody reactions. During the separation process, a barcode scanner (or RFID chip reader) records the unique sample identifier (e.g., a barcode number generated by the LIS system) in real time. This unique identifier is then mapped to the centrifuge tube, serum tube, and red blood cell suspension tube used for the physical sample, ensuring traceability of each component throughout the subsequent process.

[0070] Gradient centrifugation combined with saline washing effectively removes white blood cells, platelets, and serum impurities, preventing nonspecific reactions (such as interference of white blood cell antigens with red blood cell antigen detection). Filtering serum through a membrane eliminates optical interference of particulate impurities on signal detection. Automated identification and correlation technology ensures that the separated red blood cell suspension and serum correspond to the original sample, preventing sample confusion caused by manual manipulation (such as misplacing a centrifuge tube) from the source and providing basic data for subsequent tracing of abnormal signals.

[0071] In some embodiments, the specific binding reagent is a monoclonal antibody labeled with a fluorescent group or an enzyme marker, and the monoclonal antibody is designed for the first antigen epitope preset on the surface of the red blood cell membrane; the specific binding reaction between the specific binding reagent and the first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension comprises: mixing the serum to be tested with the red blood cell suspension under preset incubation conditions, and collecting the fluorescence signal intensity or color reaction intensity on the surface of the red blood cell membrane in real time by a fluorescence imaging device or an enzyme-linked immunosorbent assay device, wherein the binding signal exists in the form of focal fluorescence aggregation or uniform fluorescence label distribution based on the cell surface.

[0072] Monoclonal antibodies (prepared through hybridoma technology) are designed against target antigens on the surface of red blood cell membranes (such as the A and B antigens of the ABO blood group system and the D antigen of the Rh blood group system). The Fab segment of the antibody targets the antigen epitope, and the Fc segment is conjugated to a fluorescent group (such as FITC, PE, APC) or an enzyme marker (such as HRP, AP). For example, an anti-A blood group monoclonal antibody is labeled with FITC (excitation light 488nm, emission light 525nm).

[0073] The co-incubation reaction was performed by mixing the serum to be tested with the red blood cell suspension at a volume ratio of 10:1 (100 μL of serum + 10 μL of red blood cell suspension), adding the mixture to the reaction tube containing the specific antibody, vortexing and then placing the mixture in a constant temperature incubator (37°C ± 1°C) for 15-30 minutes (preset conditions in Example 3).

[0074] Fluorescence imaging involves using a flow cytometer (such as the BD FACSCanto II) to detect the fluorescence signal of a single red blood cell, obtaining forward scattered light (FSC, reflecting cell size), side scattered light (SSC, reflecting cell granularity), and fluorescence channel signal intensity; or using a fluorescence microscope (equipped with a CCD camera) to image cell smears and calculate the fluorescence intensity and distribution of a single cell using image analysis software (such as ImageJ).

[0075] Enzyme-linked immunosorbent assay includes using enzyme-labeled antibodies, adding a substrate (such as TMB) after incubation, and detecting the absorbance value using a microplate reader (450nm wavelength) to reflect the intensity of the color reaction.

[0076] Signal feature recognition is manifested as uniform fluorescent label distribution (antibodies uniformly bound to the red blood cell membrane surface) through normal specific binding. If there is contamination or confusion, focal fluorescence aggregation (such as foreign red blood cells carrying different antigens, resulting in localized intensive antibody binding) or abnormal fluorescence signals (such as the presence of A antigen fluorescence signals in O-type blood samples that are expected to be free of A antigen) may occur.

[0077] Monoclonal antibodies target a single antigen epitope, reducing cross-reactivity (e.g., anti-A antibodies only recognize antigen A and do not bind to antigen B), ensuring signal authenticity. Fluorescence / enzyme labeling enables signal quantification, avoiding subjective interpretation errors. Signal distribution characteristics (uniformity and aggregation) can be used to directly distinguish specific binding from nonspecific adsorption (e.g., random fluorescent spots caused by impurity adsorption). For example, focal aggregation suggests possible contamination with foreign red blood cells (sample exchange), while uniform but abnormal intensity suggests labeling errors or abnormal antigen expression.

[0078] Exemplarily, the preset incubation conditions include a constant temperature environment of 35-38° C. and an incubation time of 15-30 minutes.

[0079] Use incubation equipment with temperature control function (such as water bath, constant temperature metal bath) to accurately control the temperature at 35-38°C (preferably 37°C, to simulate human body temperature), with a temperature fluctuation range of ≤±0.5°C, to ensure that the antigen-antibody reaction proceeds at the optimal temperature.

[0080] The reaction time is set to 15-30 minutes (preferably 20 minutes), and the timing is automatically set by the device timer or laboratory information system (LIS). When the time is reached, the signal collection process is automatically triggered to avoid manual timing errors.

[0081] 37°C is the optimal temperature for antigen-antibody reactions, accelerating specific binding while preventing protein denaturation caused by high temperatures or reaction stagnation caused by low temperatures, ensuring stable and measurable signal strength. Fixed incubation conditions serve as a key quality control point, eliminating the impact of temperature / time differences on test results (e.g., too short a time leads to insufficient binding, too long an incubation time leads to increased nonspecific binding), thereby improving the consistency of testing across different batches.

[0082] In some embodiments, the intensity characteristics and distribution characteristics of the binding signal are compared and analyzed with the standard antigen information in the standard antigen database to generate sample antigen matching data, including: converting the intensity value of the binding signal collected in real time into a standardized value, and the standardized value is used to match the first antigen signal intensity threshold range of the corresponding sample type in the standard antigen database; analyzing the distribution ratio of the binding signal in the red blood cell population and the signal distribution uniformity on the surface of a single cell, and calculating the comprehensive matching degree through a preset matching algorithm; the standard antigen database at least includes normal antigen expression patterns corresponding to different patient groups and different test items, and antigen signal characteristic parameters of weakly positive and strongly positive control samples.

[0083] Convert fluorescence intensity values ​​(e.g., MFI, mean fluorescence intensity) detected by flow cytometry or absorbance values ​​on a microplate reader to a normalized Z-score: Z = (X - μ) / σ; where μ is the mean signal intensity of the corresponding antigen in a standard antigen database, and σ is the standard deviation. This normalized value is then compared to a pre-defined threshold range in the database (e.g., μ ± 2σ is the normal range).

[0084] Distribution feature analysis includes: RBC population distribution ratio: Calculate the proportion of cells with positive signals (e.g., testing 10,000 RBCs and counting the proportion of cells with fluorescence intensity >2 times that of the negative control), determine whether there is a mixed antigen sample (e.g., AB type blood should have both A and B antigen-positive cells; the presence of positive cells in type O blood is abnormal). Single cell signal uniformity: Calculate the standard deviation / mean ratio (uniformity coefficient) of the fluorescence signal on the surface of a single RBC through image analysis. Smaller values ​​indicate a more uniform distribution, while exceeding the threshold (e.g., >0.3) indicates focal binding (nonspecific adsorption).

[0085] The matching algorithm implementation includes the comprehensive signal intensity Z-score (weight 60%), the positive cell ratio (weight 20%), and the uniformity coefficient (weight 20%). The comprehensive matching degree is calculated by weighted summation: matching degree = 0.6×S+0.2×P+0.2×U; among them, S is the intensity matching score (0-1 points after standardization), P is the distribution ratio matching score, and U is the uniformity matching score (the smaller the value, the higher the score).

[0086] The standard antigen database contains normal antigen expression patterns of different patient groups (such as adults, newborns, and people with rare blood types) (such as weak expression of ABO antigens in newborns and a wide range of signal intensity thresholds); weak positive controls (such as subtype A 10 signal intensity threshold of type A blood), signal characteristic parameters (MFI range, uniformity coefficient standard) of strong positive control (standard type A blood).

[0087] Combining three dimensions: intensity, population distribution, and single cell uniformity, we avoid missed detections based on a single indicator (e.g., measuring intensity alone may miss low-proportion contamination, or measuring distribution alone may overlook weakly positive samples). The database incorporates characteristics of diverse patient populations to address misjudgment of special samples, such as those from newborns and subtyped blood. Weakly positive / strongly positive control parameters provide a basis for critical value determination, reducing controversy surrounding gray zone results.

[0088] In some embodiments, the method of judging whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current testing process based on the sample antigen matching data includes: when the sample antigen matching is lower than a preset normal matching threshold, or an unexpected antigen binding signal is detected on the red blood cell membrane surface, judging that the abnormal signal exists, and the expected antigen characteristics are comprehensively determined based on the test menu of the current test item, the sample type identifier and the historical test records; the normal matching threshold is dynamically adjusted based on historical quality control data, and at least includes a dual judgment of a signal intensity matching threshold and a distribution uniformity matching threshold; the antigen binding signal is an abnormal antigen epitope binding signal that is not recorded in the standard antigen database.

[0089] The current test items (such as the "ABO+Rh blood typing" menu corresponds to the detection of A, B, and D antigens), sample type identifiers (such as "venous blood" and "umbilical cord blood", the latter may have weaker Rh antigen expression) and historical test records (such as the patient's last test was type A blood), generate expected antigen characteristics (such as type A blood should be A antigen positive, B antigen negative, and D antigen is determined based on historical records). The dual judgment threshold setting includes a signal intensity matching threshold: the preset value is ≥90% (that is, the Z-score is within μ±1.645σ, corresponding to a 95% confidence interval); the distribution uniformity threshold includes a uniformity coefficient ≤0.2 (if exceeded, the distribution is judged to be abnormal). Both thresholds must be met at the same time to be considered normal, and if either one is not met, abnormal signal detection is triggered.

[0090] Abnormal signal identification includes the detection of an antigen binding signal that is not recorded in the standard antigen database (such as the simultaneous appearance of A, B, and C antigen signals, and the C antigen is an abnormal epitope not included in the database), or a matching degree of less than 80% (severely deviating from the normal range), which is determined to be an abnormal signal.

[0091] By triple-limiting the expected antigens based on the test menu, sample type, and historical records, we can avoid cross-project misjudgments (such as directly judging an HLA antigen signal as abnormal in a blood typing test); dual thresholds can reduce false alarms caused by fluctuations in a single indicator (such as occasional noise affecting the intensity value, but no warning is triggered if the distribution uniformity is normal). We can specifically identify antigen epitopes not recorded in the database (such as unknown antigens carried by new pollutants), making up for the limitation of traditional methods that can only detect known antigens, and improving the system's sensitivity to rare pollution events.

[0092] In some embodiments, the sample confusion or contamination warning information is generated based on preset warning rules, including: according to the type and severity of the abnormal signal, a warning is simultaneously issued through an audible and visual alarm device, a laboratory information system pop-up window and a sample tube label color change module. The warning information at least includes the time when the abnormal signal occurs, the unique identifier of the sample, the abnormal antigen characteristic description and the recommended processing flow, and at the same time triggers the sample traceability program, associates and traces the sample collection time, operator information, previous processing equipment usage records and adjacent sample test results to form a complete abnormal event log; the severity classification includes suspected contamination, confirmed confusion, and severe cross-contamination.

[0093] The warning levels and trigger mechanisms include: Level 1 (suspected contamination): A match of 80%-90%, or a weak positive abnormal signal is detected (such as a low proportion of A antigen-positive cells in expected O-type blood), triggering an audible and visual alarm (a short buzzer beep + a yellow pop-up window on the computer screen). Level 2 (confirmed confusion): A match of 70%-80%, or the signal intensity is completely opposite to the expected antigen (such as a strong positive signal for A antigen appearing in an expected B-type blood sample), triggering a red pop-up window in the laboratory information system (LIS) and printing an abnormal label (containing the words "sample confusion pending review"). Level 3 (serious cross-contamination): A match of less than 70%, or the detection of mixed signals of multiple antigens (such as A, B, and D antigens are simultaneously positive and abnormal in intensity), triggering an audible and visual alarm + LIS locking the sample process + sample tube label LED color change (flashing red), prohibiting the sample from entering the subsequent testing process.

[0094] Multi-dimensional early warning methods include: sound and light alarm devices: alarm lights (different colors correspond to different levels) and buzzers (different frequencies correspond to different severity) are installed on-site in the laboratory; LIS pop-up windows: real-time early warning information pops up at the inspector's workstation, including sample ID, abnormal time, and antigen feature description (such as "A antigen intensity exceeds the threshold by 3 times"); label color change module: through RFID electronic tags or thermal printing technology, a red warning mark is displayed on the sample tube label, indicating abnormalities at the physical level.

[0095] The traceability program and event log include: When an alert is triggered, the system automatically retrieves the sample collection time (linked to the hospital's HIS system), the operator's employee number (tracked through the login account), the use of previous processing equipment (such as centrifuge number and usage timestamp), and the test results of adjacent samples (to determine whether there is a possibility of continuous contamination). The abnormal event log includes: sample ID, alert time, abnormality type (confusion / contamination), matching data, traceability information, and supports Excel export and long-term archiving (at least 2 years).

[0096] Different treatment measures are taken according to the degree of risk to avoid the "crying wolf" effect caused by excessive alarms (level one warning prompts re-examination, level three warning directly blocks the process) and improve laboratory efficiency. By correlating the data of the entire collection, processing, and testing process, confusion / contamination links can be quickly located (such as determining whether the wrong sample was drawn during collection or the sample tube was misplaced during centrifugation), facilitating subsequent process improvements (such as optimizing the placement of centrifuge tubes). Label color change is combined with system pop-up windows to prevent inspectors from ignoring warning information (especially in busy scenarios) and ensure that abnormal samples are not mistakenly identified as normal.

[0097] In some embodiments, the control module is further used to: at the initial stage of the detection process when the serum contacts the red blood cell membrane, identify unexpected antigen characteristics that may be present in the sample through the specific reaction of the specific binding reagent with the first antigen on the surface of the red blood cell membrane.

[0098] Before the serum and red blood cell suspension are mixed and reacted (i.e., at the beginning of the incubation step), a baseline signal test is performed first: a fluorescently labeled negative control antibody (IgG with no specific antigen binding ability) is added to the red blood cell suspension alone, and the background fluorescence signal is detected to confirm that there is no autofluorescence or nonspecific adsorption.

[0099] Specific binding reagents are then added, and signal changes are monitored in real time within the first 5 minutes after the reaction begins to capture early binding signals (normal reaction signals gradually increase over time. If a high-intensity signal appears in the initial stage, it indicates that there may be pre-existing abnormal antigen-antibody complexes, that is, the sample has been contaminated before collection).

[0100] Compare the baseline signal with the specific reaction signal. If the signal intensity increase within the initial 5 minutes exceeds the normal range (such as >2 times the standard deviation of the mean), or an antigen signal not recorded in the database appears (such as an early strong signal of D antigen in an Rh-negative sample that is expected to have no D antigen), it is determined that an unexpected antigen feature exists, and the detection process is immediately terminated and an early warning is triggered.

[0101] Identifying anomalies at the initial stage of the testing process (early incubation) reduces the warning time by 10-15 minutes compared to traditional methods that require testing after the reaction is complete, thus avoiding reagent waste and subsequent ineffective testing steps. This initial screening allows for timely detection of sample mixups during collection or transportation (e.g., sample tubes that were mixed up during collection), preventing contaminated samples from entering subsequent analysis and preventing cross-contamination of other samples (e.g., autosampler carryover).

[0102] In some embodiments, by combining convolutional neural networks (CNN) with transfer learning technology, an intelligent recognition model for red blood cell antigen signals is constructed, breaking through the limitations of traditional rule engines and automatically mining abnormal features in complex signal patterns.

[0103] Training data was constructed by collecting fluorescence microscopy images of over 10,000 clinical samples (including normal, weakly positive, contaminated, and confused images), annotating signal distribution characteristics (such as uniformity, aggregation morphology, and fluorescence intensity gradient). Data augmentation (rotation, scaling, and Gaussian noise addition) was performed on the images to construct a training set containing over 100,000 images.

[0104] The neural network architecture design uses ResNet-18 as the basic network and adds an attention module after the convolutional layer of the traditional CNN to focus on the signal distribution details at the edge of the red blood cell membrane. The output layer is designed for multi-label classification (simultaneously identifying multiple antigen states and abnormal types such as ABO / Rh), and the loss function uses focal loss (FocalLoss) to balance the ratio of positive and negative samples.

[0105] During the test, a fluorescence microscope automatically collects red blood cell images (at least 500 cells per sample), which are input into the trained model after preprocessing; the model output includes the probability value of the antigen type (such as 98% for antigen A and 1% for antigen B) and the confidence level of the abnormal type (such as 95% confidence level for "sample confusion"). When the confidence level of any abnormality is greater than 80%, an early warning is triggered.

[0106] While traditional rules rely on preset thresholds (e.g., a uniformity coefficient greater than 0.3 is considered abnormal), deep learning can automatically identify subtle signal differences that are difficult to discern with the naked eye (e.g., focal weak fluorescence accumulation in subtype blood), increasing the abnormality detection rate from 92% to 98%. Through transfer learning, the model can quickly adapt to newly discovered antigen epitopes (e.g., rare blood type antigens), eliminating the need for manual rule base updates and addressing detection blind spots in unknown contamination scenarios. Single-sample image analysis takes less than 10 seconds, and supports multi-channel parallel processing (simultaneously analyzing five fluorescently labeled antigens), resulting in an efficiency improvement of over 10 times compared to traditional manual interpretation.

[0107] In some embodiments, by designing a fully integrated microfluidic chip, the four major steps of blood gradient centrifugation, red blood cell washing, antigen-antibody reaction, and signal detection are integrated on a single chip, realizing fully automated detection of "sample in - result out".

[0108] The chip consists of three layers: the upper layer is the sample loading area (including the barcode scanning window), the middle layer is a microchannel network (integrated with a gradient centrifugation chamber, a washing and mixing pool, and a fluorescence detection pool), and the lower layer is the pneumatic control channel (driving liquid flow). The gradient centrifugation chamber uses a spiral microchannel design and uses a centrifugal field gradient (achieved by chip rotation) to automatically separate the red blood cell layer from the serum layer, eliminating the need for an external centrifuge.

[0109] By injecting 200μL of whole blood into the chip, the code scanning module automatically reads the sample ID and associates it with the LIS system; the chip is placed on a rotating drive platform (rotational speed 500-1500 rpm), and blood cell stratification is completed within 1 minute. The serum automatically flows into the detection pool, and the red blood cells enter the washing chamber (normal saline is injected through a micro pump and washed twice); specific antibody reagents are pre-packaged in the chip reagent compartment and automatically injected into the detection pool to mix with the red blood cell suspension. The chip's built-in heating module controls incubation at 37°C for 20 minutes; an integrated micro fluorescence sensor (wavelength range 400-700nm) scans the detection pool, and data is uploaded to the analysis system in real time.

[0110] The sample is enclosed in the chip from loading to testing, avoiding cross-contamination caused by manual pipetting (the transfer contamination rate of traditional centrifuge tubes is about 0.5%, while the chip system is reduced to 0.01%). The chip size is only 5cm×5cm and can be integrated into bedside testing (POCT) equipment. It is suitable for emergency and field medical scenarios, solving the space limitations of traditional large equipment. The cost of the chip consumed for a single test is less than 10 yuan, which is 60% lower than the cost of the traditional centrifuge tube + reagent plate combination, and no cleaning and reuse are required, avoiding the risk of residual contamination.

[0111] In some embodiments, by storing the entire sample testing process data on the chain, the tamper-proof nature of the blockchain is used to build a trusted traceability system, and smart contracts are used to achieve automatic cross-system response to abnormal warnings.

[0112] Using a consortium blockchain architecture, it connects to the hospital's HIS system, laboratory LIS system, and equipment management system, with each node deploying a consensus mechanism (such as PBFT, Practical Byzantine Fault Tolerance). Data uploaded to the blockchain includes: sample ID (hash value), collection time (accurate to the second), operator ID (digital signature), centrifugation parameters (speed / time), test equipment number, antigen matching data, and early warning event records.

[0113] The preset three-level warning corresponding to the smart contract includes the following: Level 1 warning (suspected contamination): automatically freeze the sample detection process and trigger the "manual review" work order of the LIS system; Level 2 warning (confirmed confusion): the smart contract calls the sample library interface to trace adjacent samples collected in the same batch (such as samples within ±10 minutes of the collection time), and marks them as "to be investigated"; Level 3 warning (serious contamination): the contract automatically sends an alarm to the hospital infection control department and locks the involved detection equipment (such as centrifuge number X-03), prohibiting its continued use until the disinfection verification is passed.

[0114] Patient privacy data (such as name and medical record number) is hashed and uploaded to the chain. The original data is encrypted and stored in the hospital's private cloud, and only summary information is stored on the chain. A blockchain browser is provided for audit by regulatory authorities, and it supports second-level tracing of abnormal events (such as from the location of early warning information to the timestamp of the nurse's operation video).

[0115] Blockchain evidence storage addresses the risk of tampering with traditional traceability systems (e.g., manual modification of equipment usage records). This technology ensures traceability of testing processes meets judicial-grade evidence standards. Smart contracts replace manual intervention, reducing the response time for Level 3 alerts from an average of 10 minutes to real-time triggering (e.g., immediate equipment deactivation in the event of severe contamination to prevent contamination from spreading to subsequent samples). Through a blockchain browser, hospital quality control departments can remotely monitor all testing nodes in real time, reducing the time required to audit abnormal events from hours to minutes, meeting the stringent traceability requirements of certifications such as ISO 15189.

[0116] In some embodiments, by using hyperspectral imaging technology (full spectral range of 200-1000nm) to collect spectral signals on the surface of red blood cell membranes, a "spectral fingerprint" database of antigen epitopes is constructed to achieve the identification and positioning of unknown antigen epitopes.

[0117] Hyperspectral data were collected by scanning the red blood cell suspension smear using a push-broom hyperspectral camera (spectral resolution 2 nm, spatial resolution 1 μm), and the reflection / fluorescence signals of 500+ spectral channels were acquired for each cell.

[0118] The signal is preprocessed: background noise is removed (polynomial fitting baseline correction) and the spectrum is normalized (correcting for changes in signal intensity caused by differences in cell size). The spectral fingerprint library is constructed by collecting spectra of labeled antibodies for known antigens (such as A, B, and D antigens), extracting characteristic wavelengths (such as FITC-labeled antibodies have a strong emission peak at 525nm, and HRP enzyme substrates have an absorption peak at 450nm), and constructing a spectral fingerprint library containing 100+ known antigens. The spectral angle matching (SAM) algorithm is used to calculate the spectral angle between the unknown signal and the fingerprint library. An angle of less than 10° is determined to be a known antigen, and an angle of greater than 30° is determined to be an unknown antigen epitope. When a signal with a spectral angle greater than 30° is detected, the specific red blood cell is automatically located (with submicron accuracy), and the spatial distribution of abnormal signals is displayed through image overlay (for example, the unknown spectral feature appears only in a certain area of ​​the red blood cell membrane, indicating contamination by foreign red blood cell debris).

[0119] Traditional methods can only detect antigens corresponding to preset antibodies, while hyperspectral technology can identify unknown antigens outside the database (such as abnormal protein expression caused by contamination from new pathogens), filling the detection gap of "unknown unknown" risks. Combining spectral and spatial information, it can distinguish between "overall sample confusion" (all cells have abnormal spectra) and "local contamination" (only 5% of cells have abnormal membrane areas), providing accurate evidence for source of contamination analysis (such as determining whether it is foreign cells mixed in during sample collection, or impurity contamination during the reagent production process). It supports the simultaneous detection of more than 5 fluorescent-labeled antibodies (each antibody corresponds to a unique spectral fingerprint), which is 3 times more efficient than traditional single-color detection, and avoids signal crosstalk caused by overlapping fluorescence spectra (mixed signals are separated by spectral unmixing algorithms).

[0120] In some embodiments, by developing smart labeling reagents based on upconversion nanoparticles (UCNPs), the conformational changes of red blood cell membrane antigens during the detection process are monitored in real time, and antigen damage or contamination caused by sample pretreatment is identified through conformational abnormalities.

[0121] The smart labeling reagent preparation involves synthesizing NaYF4:Yb,Er upconversion nanoparticles (excitation at 980 nm, dual-wavelength emission at 525 nm / 660 nm), coupling the surface with an anti-A monoclonal antibody, and modifying the nanoparticles with pH-responsive polyethylene glycol (PEG) chains. When the antibody binds to the antigen, conformational changes in the PEG chains alter the spacing between the nanoparticles, triggering a shift in the 525 nm / 660 nm emission intensity ratio (I525 / I660) (normal binding is 1.2 ± 0.1, while abnormal antigen conformation results in a ratio greater than 1.5 or less than 1.0).

[0122] The dynamic conformation monitoring process involves mixing the red blood cell suspension with the smart reagent, placing it under a 980nm laser scanning confocal microscope, and collecting dual-wavelength signals every 30 seconds; plotting the I525 / I660 change curve over time. The normal sample curve should fluctuate within the range of 1.2±0.1. If there is a continuous upward / downward trend (such as excessive centrifugal force causing the spatial structure of the antigen to be destroyed, and the ratio continues to drop to 0.8), or a sudden peak (contamination causes nonspecific binding, and the ratio suddenly rises to 1.8), it is judged as a conformational abnormality.

[0123] The traditional method only detects whether the antigen is present or not, while the present embodiment can monitor the integrity of the spatial conformation of the antigen (such as antigen denaturation caused by sample freeze-thaw), and avoid false negative / false positive results due to antigen damage (according to statistics, the proportion of abnormal antigen conformation caused by improper sample handling in clinical practice is about 3%). Through the conformational change curve, the abnormality can be traced back to the collection link (such as insufficient anticoagulant in the blood collection tube causing red blood cell rupture), the transportation link (such as low-temperature storage temperature fluctuations of >5°C) or laboratory processing (such as excessive centrifugation speed), to achieve quality control of the entire chain before, during and after detection. Upconversion nanoparticles use near-infrared excitation light (980nm) to avoid interference from biological autofluorescence (mainly in the visible light range), and the detection sensitivity is 10 times higher than that of traditional fluorescent labels, which is suitable for the accurate detection of low-abundance antigens (such as weakly expressed antigens in newborns).

[0124] See also Figure 2 , Figure 2 This is a schematic flow chart of a method for dynamic early warning of sample mixup and contamination for quality control in a blood analysis room, provided in one embodiment of the present application. The method is performed by a device that is a control module of a dynamic early warning system for sample mixup and contamination for quality control in a blood analysis room, provided in any embodiment of the present application.

[0125] like Figure 2 As shown, the provided method includes steps S101 to S103. The control module can be a handheld terminal, a laptop computer, a wearable device, or a robot, etc., for implementing steps S101 to S103 and their corresponding embodiments.

[0126] Step S101. The sample pre-processing module performs red blood cell separation on the blood sample to be tested, and obtains a red blood cell suspension and the corresponding serum to be tested;

[0127] Step S102. Obtaining a binding signal generated by the antigen detection module based on a specific binding reaction between a specific binding reagent and a first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension; the specific binding reaction is used to incubate the serum to be tested with the red blood cell suspension and detect in real time whether the binding signal between the first antigen and the specific binding reagent exists on the surface of the red blood cell membrane;

[0128] Step S103. Generate sample antigen matching data based on the comparison and analysis of the intensity characteristics and distribution characteristics of the binding signal with the standard antigen information in the standard antigen database; determine whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current detection process based on the sample antigen matching data; if the abnormal signal is detected, generate sample confusion or contamination warning information based on preset warning rules, and realize dynamic warning of confusion or contamination of the blood sample to be tested; the standard antigen database stores standard red blood cell membrane surface antigen information corresponding to multiple target detection samples.

[0129] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity of description, the above-described dynamic early warning method for sample confusion and contamination for indoor quality control of blood analysis rooms and the specific working process of each step can refer to the corresponding processes in the embodiments of the dynamic early warning system for sample confusion and contamination for indoor quality control of blood analysis rooms described in the above-mentioned embodiments, and will not be repeated here.

[0130] The embodiments of the present application also provide a dynamic early warning device for sample confusion and contamination for indoor quality control of blood analysis rooms. This dynamic early warning device for sample confusion and contamination for indoor quality control of blood analysis rooms is used to execute the steps of the dynamic early warning method for sample confusion and contamination for indoor quality control of blood analysis rooms shown in the above embodiments. This dynamic early warning device for sample confusion and contamination for indoor quality control of blood analysis rooms can be a single server or a server cluster, or this dynamic early warning device for sample confusion and contamination for indoor quality control of blood analysis rooms can be a terminal, which can be a handheld terminal, a laptop computer, a wearable device, a robot, etc.

[0131] Dynamic early warning devices for sample mix-up and contamination used for quality control in blood analysis rooms include:

[0132] The first acquisition unit is used to obtain the red blood cell suspension and the corresponding serum to be tested by the sample pre-processing module to perform red blood cell separation processing on the blood sample to be tested;

[0133] a second acquisition unit, configured to acquire a binding signal generated by the antigen detection module based on a specific binding reaction between a specific binding reagent and a first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension; the specific binding reaction is used to co-incubate the serum to be tested with the red blood cell suspension and to detect in real time whether the binding signal between the first antigen and the specific binding reagent exists on the surface of the red blood cell membrane;

[0134] The dynamic early warning unit is used to compare and analyze the intensity characteristics and distribution characteristics of the binding signal with the standard antigen information in the standard antigen database to generate sample antigen matching data; based on the sample antigen matching data, it is judged whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current detection process; if the abnormal signal is detected, sample confusion or contamination early warning information is generated based on preset early warning rules, so as to realize dynamic early warning of confusion or contamination of the blood sample to be tested; the standard antigen database stores standard red blood cell membrane surface antigen information corresponding to multiple target detection samples.

[0135] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity of description, the specific working processes of the dynamic early warning device for sample confusion and contamination for quality control in a blood analysis room and each unit described above can refer to the corresponding processes in the embodiments of the dynamic early warning method for sample confusion and contamination for quality control in a blood analysis room described in the above embodiments, and will not be repeated here.

[0136] The above-mentioned dynamic early warning method for sample confusion and contamination for in-room quality control of blood analysis is implemented in the form of a computer program, which can be run on the above-mentioned device.

[0137] See also Figure 3 , Figure 3 1 is a schematic block diagram of the structure of a control module provided in an embodiment of the present application. The control module includes a processor, a memory and a network interface connected via a device bus, wherein the memory may include a storage medium and an internal memory.

[0138] The storage medium can store an operating device and a computer program. The computer program includes program instructions, which, when executed, can cause a processor to execute any embodiment of a dynamic early warning method for sample mix-up and contamination for quality control in a blood analysis room.

[0139] The processor is used to provide computing and control capabilities and support the operation of the entire control module.

[0140] The internal memory provides an environment for the operation of a computer program in a non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any one of the dynamic early warning system methods for sample confusion and contamination for quality control in a blood analysis room.

[0141] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 3The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the terminal to which the solution of the present application is applied. The specific control module may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0142] It should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0143] In one embodiment, the processor is configured to execute a computer program stored in the memory to implement the following steps:

[0144] The sample pre-processing module performs red blood cell separation on the blood sample to be tested, and obtains the red blood cell suspension and the corresponding serum to be tested;

[0145] Acquiring a binding signal generated by the antigen detection module based on a specific binding reaction between a specific binding reagent and a first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension; the specific binding reaction is used to co-incubate the serum to be tested with the red blood cell suspension and detect in real time whether the binding signal between the first antigen and the specific binding reagent exists on the red blood cell membrane surface;

[0146] Based on the intensity characteristics and distribution characteristics of the binding signal and the standard antigen information in the standard antigen database, a comparison analysis is performed to generate sample antigen matching data; based on the sample antigen matching data, it is determined whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current detection process; if the abnormal signal is detected, sample confusion or contamination warning information is generated based on preset warning rules, thereby realizing dynamic warning of confusion or contamination of the blood sample to be tested; the standard antigen database stores standard red blood cell membrane surface antigen information corresponding to multiple target detection samples.

[0147] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the processor described above can refer to the corresponding process in the method embodiments described in the above embodiments, and will not be repeated here.

[0148] A computer-readable storage medium is also provided in an embodiment of the present application. The computer-readable storage medium stores a computer program, and the computer program includes program instructions. The processor executes the program instructions to implement the steps of the dynamic early warning method for sample confusion and contamination for indoor quality control of blood analysis provided in the above embodiments of the present application.

[0149] The computer-readable storage medium may be an internal storage unit of the control module described in the aforementioned embodiment, such as a hard disk or memory of the control module. The computer-readable storage medium may also be an external storage device of the control module, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the control module.

[0150] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A dynamic early warning system for sample mix-up and contamination for quality control in blood analysis rooms, characterized in that: include: A sample pre-processing module is configured to perform red blood cell separation on the blood sample to be tested, and obtain a red blood cell suspension and corresponding serum to be tested; An antigen detection module is configured to cause a specific binding reaction between a specific binding reagent and a first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension, so as to incubate the serum to be tested with the red blood cell suspension and detect in real time whether a binding signal between the first antigen and the specific binding reagent is present on the surface of the red blood cell membrane; The control module pre-stores a standard antigen database, which stores standard red blood cell membrane surface antigen information corresponding to multiple target detection samples. Based on the intensity characteristics and distribution characteristics of the binding signal and the standard antigen information in the standard antigen database, a comparison analysis is performed to generate sample antigen matching data; based on the sample antigen matching data, it is determined whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current detection process. If the abnormal signal is detected, sample confusion or contamination warning information is generated based on preset warning rules, thereby realizing dynamic warning of confusion or contamination of the blood sample to be tested.

2. The system according to claim 1, wherein: The blood sample to be tested is subjected to red blood cell separation processing to obtain a red blood cell suspension and corresponding serum to be tested, including: Performing gradient centrifugation on the blood sample to be tested by a centrifugal separation device to separate the blood cell layer and the serum layer; The red blood cell suspension is prepared by washing the blood cell layer with physiological saline and resuspending it, and the serum to be tested is obtained by filtering the serum layer for impurities, and the corresponding association between the original identification information of the sample and the physical sample is maintained during the separation process.

3. The system according to claim 1, wherein: The specific binding reagent is a monoclonal antibody labeled with a fluorescent group or an enzyme marker, and the monoclonal antibody is designed to target the first antigen epitope preset on the red blood cell membrane surface; the specific binding reagent reacts specifically with the first antigen preset on the red blood cell membrane surface corresponding to the red blood cell suspension, including: The serum to be tested is mixed with the red blood cell suspension for reaction under preset incubation conditions, and the fluorescence signal intensity or color reaction intensity on the red blood cell membrane surface is collected in real time by a fluorescence imaging device or an enzyme-linked immunosorbent assay device. The binding signal exists in the form of focal fluorescence aggregation or uniform fluorescence label distribution based on the cell surface.

4. The system according to claim 3, characterized in that The preset incubation conditions include a constant temperature environment of 35-38° C. and an incubation time of 15-30 minutes.

5. The system according to claim 1, wherein: The comparing and analyzing the intensity characteristics and distribution characteristics of the binding signal with the standard antigen information in the standard antigen database to generate sample antigen matching data includes: Converting the intensity value of the binding signal collected in real time into a standardized value, wherein the standardized value is used to match the first antigen signal intensity threshold range of the corresponding sample type in the standard antigen database; The distribution ratio of the binding signal in the red blood cell population and the uniformity of the signal distribution on the surface of a single cell are analyzed, and the comprehensive matching degree is calculated using a preset matching algorithm; the standard antigen database includes at least normal antigen expression patterns corresponding to different patient populations and different test items, and antigen signal characteristic parameters of weakly positive and strongly positive control samples.

6. The system according to claim 1, wherein: The step of determining whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current testing process based on the sample antigen matching data includes: When the sample antigen matching degree is lower than the preset normal matching threshold, or an unexpected antigen binding signal is detected on the red blood cell membrane surface, it is determined that the abnormal signal exists, and the expected antigen feature is comprehensively determined based on the test menu of the current test item, the sample type identifier and the historical test records; The normal matching threshold is dynamically adjusted according to historical quality control data, and at least includes a dual determination of a signal strength matching threshold and a distribution uniformity matching threshold; The antigen binding signal is an abnormal antigen epitope binding signal not recorded in the standard antigen database.

7. The system according to claim 1, wherein: The generation of sample confusion or contamination warning information based on preset warning rules includes: Based on the type and severity of the abnormal signal, an early warning is issued simultaneously through the sound and light alarm device, the laboratory information system pop-up window, and the sample tube label color change module. The early warning information at least includes the time when the abnormal signal occurs, the sample unique identifier, the abnormal antigen feature description, and the recommended processing flow. At the same time, the sample traceability program is triggered, and the sample collection time, operator information, previous processing equipment usage records, and adjacent sample test results are linked and traced to form a complete abnormal event log; The severity levels include suspected contamination, confirmed mix-up, and severe cross-contamination.

8. The system according to claim 1, wherein: The control module is also used to: in the initial stage of the detection process when the serum contacts the red blood cell membrane, identify unexpected antigen characteristics that may exist in the sample through the specific reaction of the specific binding reagent with the first antigen on the surface of the red blood cell membrane.

9. A dynamic early warning method for sample mix-up and contamination for quality control in blood analysis rooms, characterized in that: A control module for a dynamic early warning system for sample mix-up and contamination for in-room quality control of blood analysis according to any one of claims 1 to 8, the method comprising: The sample pre-processing module performs red blood cell separation on the blood sample to be tested, and obtains the red blood cell suspension and the corresponding serum to be tested; Acquiring a binding signal generated by the antigen detection module based on a specific binding reaction between a specific binding reagent and a first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension; the specific binding reaction is used to co-incubate the serum to be tested with the red blood cell suspension and detect in real time whether the binding signal between the first antigen and the specific binding reagent exists on the red blood cell membrane surface; Based on the intensity characteristics and distribution characteristics of the binding signal and the standard antigen information in the standard antigen database, a comparison analysis is performed to generate sample antigen matching data; based on the sample antigen matching data, it is determined whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current detection process; if the abnormal signal is detected, sample confusion or contamination warning information is generated based on preset warning rules, thereby realizing dynamic warning of confusion or contamination of the blood sample to be tested; the standard antigen database stores standard red blood cell membrane surface antigen information corresponding to multiple target detection samples.

10. A dynamic early warning device for sample mix-up and contamination for quality control in a blood analysis room, characterized in that: A control module for a dynamic early warning system for sample mix-up and contamination for in-room quality control of blood analysis according to any one of claims 1 to 8, the device comprising: The first acquisition unit is used to obtain the red blood cell suspension and the corresponding serum to be tested by the sample pre-processing module to perform red blood cell separation processing on the blood sample to be tested; a second acquisition unit, configured to acquire a binding signal generated by the antigen detection module based on a specific binding reaction between a specific binding reagent and a first antigen preset on the surface of the red blood cell membrane corresponding to the red blood cell suspension; the specific binding reaction is used to co-incubate the serum to be tested with the red blood cell suspension and to detect in real time whether the binding signal between the first antigen and the specific binding reagent exists on the surface of the red blood cell membrane; The dynamic early warning unit is used to compare and analyze the intensity characteristics and distribution characteristics of the binding signal with the standard antigen information in the standard antigen database to generate sample antigen matching data; based on the sample antigen matching data, it is judged whether there is an abnormal signal in the blood sample to be tested that does not match the expected antigen characteristics of the current detection process; if the abnormal signal is detected, sample confusion or contamination early warning information is generated based on preset early warning rules, so as to realize dynamic early warning of confusion or contamination of the blood sample to be tested; the standard antigen database stores standard red blood cell membrane surface antigen information corresponding to multiple target detection samples.

Images