A method and system for clotting assay clot sample screening

Abstract

This invention relates to the field of medical testing technology, specifically to a method and system for screening coagulation clot samples. The screening method includes the following steps: S1: Obtaining routine coagulation test index data of the coagulation sample to be tested, including activated partial thromboplastin time and D-dimer; S2: Inputting the routine coagulation test index data into formula (1), formula (2) or formula (3) to obtain a predictive index that the coagulation sample to be tested belongs to a clot sample, comparing the predictive index with the preset judgment threshold of its corresponding formula, if the predictive index is greater than or equal to the judgment threshold, then the sample to be tested is predicted to be a clot sample, which needs to be further manually verified and confirmed; This method can directly calculate the predictive index reflecting the coagulation risk of the sample, and by comparing the predictive index with the preset threshold, it can quickly and objectively predict whether the sample is a clot sample.

Description

Technical Field

[0001] This invention relates to the field of medical testing technology, specifically to a method and system for screening coagulation clot samples. Background Technology

[0002] Coagulation testing is a core tool for diagnosing bleeding and thrombotic diseases, assessing perioperative bleeding risk, and monitoring anticoagulation therapy. However, sample coagulation in the pre-test stage is a common cause of inaccurate results, accounting for 44.43% of coagulation sample rejections, and can seriously mislead clinical decisions.

[0003] Currently, the identification of agglutinated samples in clinical practice mainly relies on pre-test visual observation and experience-based post-test screening of abnormal results. Visual observation is time-consuming and prone to missing small clots, and is not suitable for fully automated processes such as sample collection, centrifugation, and loading. Post-test screening generally uses activated partial thromboplastin time (APTT) and / or D-dimer (DD) results to screen suspicious samples, then agitates the sample with a cotton swab to further confirm the presence or absence of visible clots. This is generally based on individual experience, and standards vary from person to person. ICSH (International Committee for Standardization in Hematology) recommends checking for clots when APTT is below the lower limit of the reference range by 4 seconds, but this is prone to missed detections. Kim et al. established a screening model using thrombin-antithrombin complex and fibrin monomer complex combined with APTT, achieving high accuracy in identifying in vitro agglutination; however, these molecular markers are not routine tests and cannot be used as routine screening indicators in daily work. Fang et al. used backpropagation neural network analysis to analyze five routine coagulation test indicators: prothrombin time (PT), APTT, thrombin time (TT), fibrinogen (FIB), and DD. However, the resulting neural network model lacked universality. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method and system for screening clot samples in coagulation testing. By analyzing coagulation test data from a large number of clot samples and qualified samples, the inventors discovered that activated partial thromboplastin time (APTT) and the ratio of D-dimer (DD) to APTT are key indicators for identifying in vitro agglutination. Based on this, the inventors derived three concise mathematical formulas using symbolic regression to directly calculate predictive indices reflecting the coagulation risk of a sample. By inputting the APTT and DD of the sample to be tested into one of these three formulas, a predictive index indicating whether the sample is a coagulated sample can be obtained. Comparing this predictive index with a preset threshold in the corresponding formula allows for a rapid and objective prediction of whether a sample is a clot sample.

[0005] Furthermore, the inventors discovered that by inputting the APTT and DD of the sample to be tested into these three formulas, three predictive indices are obtained. These three predictive indices are compared with the preset judgment thresholds of their respective formulas. If at least one predictive index is greater than or equal to the judgment threshold, the coagulation sample to be tested is determined to be a clot sample. Using these three formulas simultaneously to screen whether the sample to be tested belongs to a clot sample can reduce the probability of false negatives, thereby improving the sensitivity of the screening.

[0006] Therefore, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for screening coagulation test clot samples in an optional embodiment, comprising the following steps:

[0008] S1: Obtain routine coagulation test data of the coagulation sample to be tested, including activated partial thromboplastin time and D-dimer;

[0009] S2: Input the conventional coagulation test index data into formula (1), formula (2) or formula (3) to obtain a prediction index that the coagulation sample to be tested belongs to a clot sample. Compare the prediction index with the preset judgment threshold of its corresponding formula. If the prediction index is greater than or equal to the judgment threshold, the coagulation sample to be tested is determined to be a clot sample; or,

[0010] Input the conventional coagulation test index data into formula (1), formula (2) and formula (3) to obtain the prediction index of the three coagulation samples to be tested as clot samples. Compare the prediction index with the preset judgment threshold of the corresponding formula. If at least one prediction index is greater than or equal to the judgment threshold, the coagulation sample to be tested is judged to be a clot sample.

[0011] The formula (1) is:

[0012] ;

[0013] The formula (2) is:

[0014] ;

[0015] The formula (3) is:

[0016] ;

[0017] Where F is the predictive index of whether the coagulation sample to be tested belongs to a clot sample, DD is D-dimer, and APTT is activated partial thromboplastin time.

[0018] The preset judgment threshold of formula (1) is 0.595, the preset judgment threshold of formula (2) is 0.603, and the preset judgment threshold of formula (3) is 0.599.

[0019] Preferably, the method further includes: manually verifying the blood coagulation sample to be tested, which is determined to be a clot sample, to determine whether the sample has agglutinated.

[0020] Secondly, in an optional embodiment, the present invention provides a coagulation test clot sample screening system, the system comprising:

[0021] The data acquisition module is used to acquire routine coagulation test data of the coagulation sample to be tested;

[0022] The prediction module contains formulas (1), (2) and (3), which are used to calculate and output the prediction index of whether the coagulation sample to be tested belongs to the clot sample based on the conventional coagulation test index data.

[0023] The determination module is used to compare the predicted index with a preset determination threshold and output the determination result; and

[0024] The result output module is used to output the judgment result.

[0025] Preferably, formula (1) is:

[0026] ;

[0027] The formula (2) is:

[0028] ;

[0029] The formula (3) is:

[0030] ;

[0031] Where F is the predictive index of whether the coagulation sample to be tested belongs to a clot sample, DD is D-dimer, and APTT is activated partial thromboplastin time.

[0032] The preset judgment threshold of formula (1) is 0.595, the preset judgment threshold of formula (2) is 0.603, and the preset judgment threshold of formula (3) is 0.599.

[0033] Preferably, the system further includes a result prompting module, which prompts that the coagulation sample to be tested needs to be manually reviewed and confirmed when the determination result is that the coagulation sample to be tested belongs to a clot sample.

[0034] Preferably, the manual verification method is as follows: the coagulation sample to be tested is stirred using a stirring device, and the coagulation sample to be tested is judged by visual inspection to determine whether it is a clot sample. If the coagulation sample to be tested contains clots, it is determined that the coagulation sample is a clot sample. The stirring device includes a cotton swab, a blood collection capillary tube, or a wooden stick.

[0035] Thirdly, in an optional embodiment, the present invention provides a computer device including a storage device, a processor, and a computer program stored in the storage device and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0036] Fourthly, in an optional embodiment, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0037] Fifthly, in an optional embodiment, the present invention provides a computer program product including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0038] Compared with the prior art, the present invention has one of the following beneficial effects:

[0039] 1. By analyzing coagulation test data from a large number of clot samples and qualified samples, the inventors discovered that activated partial thromboplastin time (APTT) and the ratio of D-dimer (DD) to APTT are key indicators for identifying in vitro agglutination. Based on this, the inventors derived three simple mathematical formulas using symbolic regression to directly calculate predictive indices reflecting the coagulation risk of a sample. By inputting the APTT and DD of the sample to be tested into one of these three formulas, a predictive index indicating whether the sample is a coagulated sample can be obtained. Comparing this predictive index with the preset threshold of its corresponding formula allows for a rapid and objective prediction of whether a sample is a clot sample.

[0040] 2. The inventors also discovered that by inputting the APTT and DD of the sample to be tested into these three formulas, three predictive indices are obtained. These three predictive indices are compared with the preset judgment thresholds of their respective formulas. If at least one predictive index is greater than or equal to the judgment threshold, the coagulation sample to be tested is determined to be a clot sample. Using these three formulas simultaneously to screen whether the sample to be tested belongs to a clot sample can reduce the probability of false negatives, thereby improving the sensitivity of screening. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram showing the results of feature importance analysis performed using the SHAP method in Embodiment 1 of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0044] The technical solution of the present invention will be described below with reference to embodiments.

[0045] Example 1

[0046] Obtaining and verifying the results of formulas (1), (2), and (3)

[0047] 1. The foundation of mathematical formula derivation

[0048] First, a large amount of coagulation test sample data (including clot samples and qualified samples collected again from the patient) was collected. A high-performance classification model was trained using Lightweight Gradient Boosting Machine (LightGBM) as the base learner. This model achieved a ROC-AUC of 0.984 on the internal test set. Then, feature importance analysis was performed using the SHAP method, which revealed that APTT, DD / APTT ratio, and PT / APTT ratio were the most critical features (see [link to relevant documentation]). Figure 1 Finally, three simplified mathematical equations were extracted from the model using symbolic regression (PySR), which are the formulas (1), (2), and (3) described in this invention.

[0049] Formula (1) is:

[0050] ;

[0051] The formula (2) is:

[0052] ;

[0053] The formula (3) is:

[0054] ;

[0055] Where F is the predictive index of whether the coagulation sample to be tested belongs to a clot sample, DD is D-dimer, and APTT is activated partial thromboplastin time.

[0056] The results for the threshold, sensitivity, specificity, positive predictive value, and negative predictive value of formulas (1), (2), and (3) are shown in Table 1:

[0057]

[0058] In Table 1, 95% CI refers to the 95% confidence interval. The range values ​​in parentheses in Table 1 indicate that if the same sample is tested 100 times repeatedly, and the sensitivity, specificity, positive predictive value, or negative predictive value is calculated each time, the calculated sensitivity, specificity, positive predictive value, or negative predictive value will fall within the range values ​​in parentheses 95 times.

[0059] 2. To simulate a real clinical scenario (low coagulation rate), the above formula and threshold were applied to an independent external validation set (5812 cases in total, including 56 clot samples, 5756 qualified samples, and a positive rate of 0.96%). The results are shown in Table 2.

[0060]

[0061] Table 2 explains: (1), (2), and (3) represent the verification results obtained by using formulas (1), (2), and (3) individually, respectively. The comprehensive result represents the result of using formulas (1), (2), and (3) simultaneously, and the prediction index of at least one formula is greater than or equal to the judgment threshold, in which case it is judged as a clot sample.

[0062] The above data shows that the three formulas provided by this invention exhibit good generalization ability and stability on different sample sets. Specifically, the specificity is above 98%, meaning an extremely low false positive rate, effectively avoiding unnecessary retesting of qualified samples; simultaneously, the negative predictive value is close to 100%, indicating that there is almost no risk of missed detection for samples deemed qualified. The predicted positive rate is close to the actual positive rate of 0.96% (56 / 5812), suggesting that a large amount of manual verification of clots is unnecessary. Furthermore, using all three formulas simultaneously to screen clot samples can reduce the probability of false negatives and improve the sensitivity of the screening.

[0063] Example 2

[0064] To verify the application of the above formula in a real-world scenario, routine coagulation test data of 5 patients were obtained from a hospital. The APTT and DD data of these 5 patients were then substituted into formulas (1), (2), and (3) respectively to obtain the predictive index of these 5 patients (see Table 3 for the results).

[0065]

[0066] Comparing the predictive indices obtained for patients 1-5 using formulas (1), (2), and (3) with the thresholds of the three formulas, we found that:

[0067] The predictive index of Patient 1 in Formula (1) is 0.943 > 0.595, the predictive index in Formula (2) is 0.914 > 0.603, and the predictive index in Formula (3) is 0.936 > 0.599. Therefore, based on the result of one of Formula (1), Formula (2), or Formula (3), or based on at least one predictive index in Formula (1), Formula (2), and Formula (3) being greater than or equal to the judgment threshold, Patient 1's coagulation sample is determined to be a clot sample, indicating that the coagulation sample needs to be manually reviewed and confirmed.

[0068] The predictive index of Patient 2 in Formula (1) is 0.622 > 0.595, the predictive index in Formula (2) is 0.665 > 0.603, and the predictive index in Formula (3) is 0.641 > 0.599. Therefore, based on the result of one of Formula (1), Formula (2), or Formula (3), or based on at least one predictive index in Formula (1), Formula (2), and Formula (3) being greater than or equal to the judgment threshold, Patient 2's coagulation sample is determined to be a clot sample, indicating that the coagulation sample needs to be manually reviewed and confirmed.

[0069] Patient 3 has a prediction index of 0.555 < 0.595 in formula (1), a prediction index of 0.512 < 0.603 in formula (2), and a prediction index of 0.487 < 0.599 in formula (3). Therefore, based on the result of one of formulas (1), (2), or (3), or based on the fact that none of the prediction indices in formulas (1), (2), and (3) is greater than or equal to the judgment threshold, it is determined that the coagulation sample of Patient 3 is not a clot sample, and the patient's coagulation sample can be reviewed normally and a report can be issued.

[0070] Patient 4's predictive index in formula (1) is 0.591 < 0.595, in formula (2) it is 0.595 < 0.603, and in formula (3) it is 0.539 < 0.599. Therefore, based on the result of one of formulas (1), (2), or (3), or based on the fact that none of the predictive indices in formulas (1), (2), and (3) is greater than or equal to the judgment threshold, it is determined that Patient 4's coagulation sample is not a clot sample, and the patient's coagulation sample can be reviewed normally and a report can be issued.

[0071] Patient 5 has a prediction index of 0.515 < 0.595 in formula (1), a prediction index of 0.470 < 0.603 in formula (2), and a prediction index of 0.444 < 0.599 in formula (3). Therefore, based on the result of one of formulas (1), (2), or (3), or based on the fact that none of the prediction indices in formulas (1), (2), and (3) is greater than or equal to the judgment threshold, it is determined that the coagulation sample of Patient 5 is not a clot sample, and the patient's coagulation sample can be reviewed normally and a report can be issued.

[0072] To further evaluate the formula's performance in the real world, the formula and threshold were applied to routine working conditions for validation (a total of 2470 cases, including 15 clot samples and 2455 qualified samples, with a positive rate of 0.61%). Samples predicted as clots and negative by the formula were manually verified. All results are shown in Table 4.

[0073]

[0074] Table 4 explains: (1), (2), and (3) represent the verification results obtained by using formulas (1), (2), and (3) individually, respectively. The comprehensive result represents the result of using formulas (1), (2), and (3) simultaneously, and the prediction index of at least one formula is greater than or equal to the judgment threshold, in which case it is judged as a clot sample.

[0075] The above data shows that the three formulas provided by this invention, used individually or in combination, exhibit good generalization ability and stability in the real world.

[0076] Example 3

[0077] The coagulation test clot sample screening system of the present invention can be integrated into the hospital's laboratory information system (LIS). The specific workflow is as follows:

[0078] After the coagulation testing instrument completes the sample test, it transmits the APTT and DD results of the coagulation sample to the LIS.

[0079] The system's data acquisition module automatically receives these two values ​​from the coagulation sample to be tested.

[0080] The prediction module calculates the DD / APTT ratio of the coagulation sample to be tested, and substitutes the DD / APTT ratio and APTT into the above mathematical formulas (1), (2) or (3) to calculate a prediction index for the coagulation sample to be tested. Alternatively, the DD / APTT ratio and APTT are substituted into the above mathematical formulas (1), (2) and (3) to calculate three prediction indices for the coagulation sample to be tested.

[0081] The judgment module compares a predicted index of the coagulation sample to be tested, calculated by the prediction module, with the threshold of formula (1), formula (2), or formula (3): If the predicted index of the coagulation sample to be tested is calculated by formula (1), then the predicted index is compared with the threshold of formula (1). If F1≥0.595, then the coagulation sample to be tested is marked as a "suspected clot sample". If F1<0.595, then the coagulation sample to be tested is marked as "qualified". If the predicted index of the coagulation sample to be tested is calculated by formula (2), then the predicted index is compared with the threshold of formula (2). If F2≥0.603, then the coagulation sample to be tested is marked as a "suspected clot sample". If F2<0.603, then the coagulation sample to be tested is marked as "qualified". If the predicted index of the coagulation sample to be tested is calculated by formula (3), then the predicted index is compared with the threshold of formula (3). If F3 ≥ 0.599, the coagulation sample to be tested is marked as a "suspected clot sample". If F3 < 0.599, the coagulation sample to be tested is marked as "qualified". Alternatively, the judgment module compares the three predicted indices of the coagulation sample to be tested calculated by the prediction module with the thresholds of the corresponding formulas (1), (2) and (3). If the predicted index of at least one formula is greater than or equal to the judgment threshold, the coagulation sample to be tested is marked as a "suspected clot sample". If the predicted index of any formula is greater than or equal to the judgment threshold, the coagulation sample to be tested is marked as "qualified".

[0082] The results output module displays the judgment results on the review interface, and automatically pops up a prompt for the coagulation sample marked as "suspected clot sample": "The sample may contain microclots, manual confirmation is recommended".

[0083] Inspectors only need to review samples marked as "suspected clot samples," while other samples can be reviewed and reports issued normally. The entire judgment process is fully automated, requiring no manual intervention, significantly improving the efficiency of pre-inspection quality control.

[0084] Example 4

[0085] This embodiment discloses an intelligent screening method for coagulation clot samples based on middleware for testing instruments. This method relies on independently deployed middleware to build a clot sample screening function system. The middleware is an independent mid-level software platform deployed between the front-end coagulation testing instrument and the hospital's LIS / HIS business system. It is neither attached to the local workstation of the coagulation testing instrument nor part of the hospital's inherent business system. It plays a core hub role in data interface between heterogeneous devices, data protocol translation, business rule scheduling, and execution of quality control logic. This effectively breaks down data interaction barriers between the underlying coagulation testing equipment and the upper-level hospital information system, enabling fully automated screening of coagulation clot abnormalities and improving the quality control level of the entire testing process.

[0086] The coagulation test clot sample screening system in this embodiment is integrated into the middleware of the testing instrument and operates independently of the front-end testing equipment and the hospital LIS / HIS system. The specific implementation workflow is as follows:

[0087] Step 1: Uplink transmission of raw test data. After the hospital's front-end coagulation testing instrument completes the automated testing of a batch of coagulation samples, it uploads the core test results of APTT (Activated Partial Thromboplastin Time) and DD (D-dimer) for each sample to the middleware of the testing instrument via the device's proprietary communication protocol, completing the one-way output of raw data from the underlying device.

[0088] Step Two: Middleware Automatic Data Acquisition and Parsing. The data acquisition module built into the middleware of the testing instrument monitors the data transmission link of the front-end coagulation testing instrument in real time, automatically receiving the raw APTT and DD test values ​​of each coagulation sample to be tested. Through the middleware's protocol translation capabilities, it uniformly converts the device's proprietary protocol data into the hospital information system's common standard data format, completing data cleaning, parsing, and standardized storage, thus avoiding recognition anomalies caused by inconsistent data formats from heterogeneous devices.

[0089] Step 3: Middleware calculates and generates clot prediction indices. The middleware's built-in prediction module retrieves the standardized APTT and DD values, first automatically calculating the DD / APTT ratio of the coagulation sample to be tested, and then substituting the DD / APTT ratio and the corresponding APTT test value into preset mathematical formulas (1), (2), and (3) to complete the prediction index calculation. This embodiment supports two calculation modes: first, selectively substituting into a single formula to obtain a single prediction index F1, F2, or F3 corresponding to the sample to be tested; second, simultaneously substituting into three sets of formulas for parallel calculation to generate three prediction indices F1, F2, and F3 corresponding to the sample to be tested at once, and the calculation logic can be flexibly switched according to the hospital's laboratory quality control needs.

[0090] Step 4: Middleware intelligently determines the abnormal state of sample clumps. The middleware's built-in determination module retrieves the predicted index generated by the prediction module and combines it with the specific thresholds corresponding to each formula to complete an automated comparison and determination. The specific determination rules are as follows:

[0091] If a single formula calculation mode is used: when the sample prediction index is F1 calculated by formula (1), F1 is compared with the threshold 0.595. If F1≥0.595, the sample to be tested is marked as a "suspected clot sample", and if F1<0.595, it is marked as a "qualified sample". When the sample prediction index is F2 calculated by formula (2), F2 is compared with the threshold 0.603. If F2≥0.603, it is marked as a "suspected clot sample", and if F2<0.603, it is marked as a "qualified sample". When the sample prediction index is F3 calculated by formula (3), F3 is compared with the threshold 0.599. If F3≥0.599, it is marked as a "suspected clot sample", and if F3<0.599, it is marked as a "qualified sample".

[0092] If a three-formula parallel operation mode is adopted: the generated F1, F2, and F3 are matched with the thresholds of each formula and compared. As long as any one of the three sets of prediction indices reaches the corresponding judgment threshold, the sample to be tested is immediately marked as a "suspected clot sample"; if none of the three prediction indices reach the corresponding threshold, it is marked as a "qualified sample".

[0093] Step 5: Result Push and Visual Early Warning. The middleware's built-in result output module synchronously pushes the final judgment result to the hospital's LIS review interface, completing the standardized transmission of data to the upper-level business system. For all samples marked as "suspected clot samples," the system automatically triggers a pop-up early warning, displaying the message: "The sample may contain microclots; manual confirmation is recommended," prominently reminding laboratory personnel to conduct a thorough review.

[0094] Step Six: Differentiated Sample Review and Flow. Based on the intelligent screening and judgment results of the middleware, the inspectors only need to conduct manual review and secondary verification of abnormal samples marked as "suspected clot samples". All specimens marked as "qualified samples" can be directly reviewed and issued test reports through the LIS system.

[0095] This embodiment relies on independent testing instrument middleware to complete the entire process of data processing, calculation, judgment, and result scheduling. All data parsing, index calculation, anomaly detection, and early warning push are completed automatically by the middleware without manual intervention, and without consuming the computing power of the front-end testing equipment or the hospital's LIS system operating resources. It not only solves the data interoperability problem between heterogeneous coagulation testing equipment and hospital business systems, but also enables accurate pre-screening of coagulation sample microclot anomalies, significantly reducing the workload of manual review and significantly improving the automation level and overall work efficiency of pre-coagulation testing quality control.

[0096] Although the principles of the present invention have been described in detail above with reference to preferred embodiments, those skilled in the art should understand that the above embodiments are merely illustrative explanations of the implementation of the present invention and are not intended to limit the scope of the present invention. The details in the embodiments do not constitute a limitation on the scope of the present invention. Any obvious changes, such as equivalent transformations or simple substitutions, based on the technical solutions of the present invention without departing from the spirit and scope of the present invention fall within the protection scope of the present invention.

Claims

1. A method for screening coagulation clot samples in coagulation testing, characterized in that, Includes the following steps: S1: Obtain routine coagulation test data of the coagulation sample to be tested, including activated partial thromboplastin time and D-dimer; S2: Input the conventional coagulation test index data into formula (1), formula (2) or formula (3) to obtain a prediction index of whether the coagulation sample to be tested belongs to a clot sample. Compare the prediction index with the preset judgment threshold of its corresponding formula. If the prediction index is greater than or equal to the judgment threshold, the coagulation sample to be tested is determined to be a clot sample. or, Input the conventional coagulation test index data into formula (1), formula (2) and formula (3) to obtain the prediction index of the three coagulation samples to be tested as clot samples. Compare the prediction index with the preset judgment threshold of the corresponding formula. If at least one prediction index is greater than or equal to the judgment threshold, the coagulation sample to be tested is judged to be a clot sample. The formula (1) is: ； The formula (2) is: ； The formula (3) is: ； Where F is the predictive index of whether the coagulation sample to be tested belongs to a clot sample, DD is D-dimer, and APTT is activated partial thromboplastin time. The preset judgment threshold of formula (1) is 0.595, the preset judgment threshold of formula (2) is 0.603, and the preset judgment threshold of formula (3) is 0.

599.

2. The method for screening coagulation clot samples according to claim 1, characterized in that, The method further includes: The coagulation samples identified as clot samples are manually reviewed and confirmed to determine whether the samples have agglutinated.

3. A coagulation test clot sample screening system, characterized in that, The system includes: The data acquisition module is used to acquire routine coagulation test data of the coagulation sample to be tested; The prediction module contains formulas (1), (2) and (3), which are used to calculate and output the prediction index of whether the coagulation sample to be tested belongs to the clot sample based on the conventional coagulation test index data. The determination module is used to compare the predicted index with a preset determination threshold and output the determination result; and The result output module is used to output the judgment result.

4. The coagulation test clot sample screening system according to claim 3, characterized in that, The formula (1) is: ； The formula (2) is: ； The formula (3) is: ； Where F is the predictive index of whether the coagulation sample to be tested belongs to a clot sample, DD is D-dimer, and APTT is activated partial thromboplastin time. The preset judgment threshold of formula (1) is 0.595, the preset judgment threshold of formula (2) is 0.603, and the preset judgment threshold of formula (3) is 0.

599.

5. The coagulation test clot sample screening system according to claim 3, characterized in that, The system also includes a result prompting module, which prompts that the coagulation sample to be tested needs to be manually reviewed and confirmed when the determination result is that the coagulation sample to be tested belongs to a clot sample.

6. The coagulation test clot sample screening system according to claim 5, characterized in that, The manual verification method is as follows: the coagulation sample to be tested is stirred with a stirring device, and the coagulation sample to be tested is judged by visual inspection to see if it is a clot sample. If the coagulation sample to be tested contains clots, it is determined that the coagulation sample is a clot sample. The agitating device includes a cotton swab, a blood collection capillary tube, or a wooden stick.

7. A computer device, comprising a storage unit, a processor, and a computer program stored in the storage unit and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method of claim 1 or 2.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method of claim 1 or 2.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in claim 1 or 2.

Images