A method and related apparatus for determining the coagulation time of a blood sample.

Abstract

The application discloses a method for determining the coagulation time of a blood sample and a related device, and relates to the technical field of biomedicine. The method comprises the following steps: for a sample of a first abnormality level, a first coagulation time is obtained by optical detection, and then it is determined whether the first coagulation time is in a corresponding range; if not, a second coagulation time is obtained by automatically starting magnetic bead detection, and the target coagulation time is determined according to the second coagulation time; for the sample of the first abnormality level, the first coagulation time obtained by optical detection and the second coagulation time obtained by magnetic bead detection are fused to determine the target coagulation time, so that the detection result is more reliable, and misjudgment caused by errors possibly generated by a single detection method is avoided; and for a sample of a second abnormality level, the target coagulation time is directly obtained by magnetic bead detection. The automatic switching detection scheme greatly improves the automation degree of the detection process, shortens the time from sample detection to result report, and improves the determination efficiency and accuracy of the coagulation time of the blood sample.

Description

Technical Field

[0001] This application relates to the field of biomedical technology, and in particular to a method and apparatus for determining the coagulation time of a blood sample. Background Technology

[0002] In recent years, hematological disorders such as hemophilia, thrombocytopenia, leukemia, and von Willebrand disease have gradually become important factors affecting people's health. The diagnosis of these diseases relies on the determination of blood clotting time. For example, assessing a patient's coagulation function by measuring clotting time before surgery helps predict the risk of postoperative bleeding; during anticoagulation therapy, regular monitoring of clotting time helps ensure the safety and effectiveness of treatment. Therefore, determining the clotting time of blood samples is particularly important.

[0003] Currently, in traditional blood sample coagulation testing, the coagulation time of blood samples is determined by light source illumination, and manual verification is still required.

[0004] However, when there are abnormalities in the blood sample, the light source alone cannot accurately reflect its true coagulation status, which can easily lead to inaccurate determination of coagulation time. Even after manual verification, the efficiency of determining the coagulation time is also low. Summary of the Invention

[0005] In view of the above problems, this application provides a method and related apparatus for determining the coagulation time of blood samples, in order to improve the accuracy and efficiency of determining the coagulation time of blood samples. The specific solution is as follows:

[0006] The first aspect of this application provides a method for determining the coagulation time of a blood sample, comprising:

[0007] The target components of a blood sample are obtained, and the target light source wavelength and target magnetic field strength corresponding to the target components of the blood sample are determined. The target components of the blood sample are the components in the blood sample that are related to coagulation function.

[0008] Using a machine learning model, the abnormality level of the blood sample is determined. The abnormality level of the blood sample includes a first abnormality level and a second abnormality level, wherein the first abnormality level is lower than the second abnormality level.

[0009] If the abnormality level of the blood sample is the first abnormality level, the blood sample is irradiated with a light source of the target light source wavelength. The time it takes for the light transmittance of the blood sample to change from the first light transmittance to the second light transmittance is taken as the first coagulation time of the blood sample. It is then determined whether the first coagulation time of the blood sample belongs to the coagulation time range corresponding to the first abnormality level. The first light transmittance is determined based on the light transmittance of the blood sample in its initial liquid state, and the second light transmittance is determined based on the light transmittance of the blood sample in its fully coagulated state.

[0010] If the first coagulation time of the blood sample does not fall within the coagulation time range corresponding to the first abnormality level, then the second coagulation time of the blood sample is determined by using magnetic beads in the blood sample that respond to the magnetic field strength of the target magnetic field, and the target coagulation time of the blood sample is determined based on the first coagulation time and the second coagulation time of the blood sample.

[0011] If the abnormality level of the blood sample is the second abnormality level, then the third coagulation time of the blood sample is determined by using magnetic beads in the blood sample that respond to the magnetic field strength of the target magnetic field, and the third coagulation time of the blood sample is taken as the target coagulation time of the blood sample.

[0012] In one possible implementation, determining the target light source wavelength and target magnetic field strength corresponding to the target component of the blood sample includes:

[0013] The target components of the blood sample are scanned across the entire spectrum to obtain the spectral data of the target components;

[0014] The wavelength corresponding to the absorbance peak in the spectral data of the target component is taken as the target light source wavelength;

[0015] The target components of the blood sample are labeled with magnetic beads so that the magnetic beads respond to different magnetic fields in the blood sample, and the response data of the magnetic beads is obtained.

[0016] The magnetic field strength corresponding to the peak responsivity in the response data of the magnetic bead is taken as the target magnetic field strength.

[0017] In one possible implementation, the machine learning model includes a first anomaly level determination model and a second anomaly level determination model;

[0018] The process of using a machine learning model to determine the abnormality level of the blood sample includes:

[0019] The image of the blood sample is obtained and input into a pre-trained first anomaly level determination model to obtain the anomaly level of the blood sample.

[0020] or,

[0021] The blood sample is irradiated with light sources of multiple wavelengths to obtain the absorbance of the blood sample, and the absorbance of the blood sample is input into a pre-trained second abnormality level determination model to obtain the abnormality level of the blood sample.

[0022] In one possible implementation, the determination process of the first anomaly level determination model includes:

[0023] Multiple historical training samples with historical annotation information are obtained, the annotation information including the anomaly level of historical blood samples, and each training sample includes an image of a historical blood sample;

[0024] The initial model is trained based on multiple historical training samples to obtain the first anomaly level determination model. The input of the first anomaly level determination model is the image of the blood sample, and the output of the first anomaly level determination model is the anomaly level of the blood sample.

[0025] In one possible implementation, determining the second coagulation time of the blood sample using magnetic beads in the blood sample that respond to the magnetic field strength of the target magnetic field includes:

[0026] The oscillation amplitude of the magnetic beads is obtained by using magnetic beads in the blood sample that respond to the magnetic field strength of the target magnetic field.

[0027] The time it takes for the swing amplitude of the magnetic bead to change from a first swing amplitude to a second swing amplitude is taken as the second coagulation time of the blood sample. The first swing amplitude is determined based on the swing amplitude of the blood sample in its initial liquid state, and the second swing amplitude is determined based on the swing amplitude of the blood sample in its fully coagulated state.

[0028] In one possible implementation, determining the target coagulation time of the blood sample based on the first coagulation time and the second coagulation time of the blood sample includes:

[0029] Based on the transmittance of the blood sample, the transmittance fluctuation coefficient of the blood sample is calculated, and based on the transmittance fluctuation coefficient of the blood sample, the first confidence level of the first coagulation time of the blood sample is determined.

[0030] Based on the oscillation amplitude of the magnetic bead, the signal-to-noise ratio of the oscillation amplitude of the magnetic bead is calculated, and based on the signal-to-noise ratio of the oscillation amplitude of the magnetic bead, the second confidence level of the second coagulation time of the blood sample is determined.

[0031] Based on the first confidence level and the second confidence level, a first weight is assigned to the first coagulation time of the blood sample, and a second weight is assigned to the second coagulation time of the blood sample;

[0032] The target coagulation time of the blood sample is calculated based on the first coagulation time, the first weight, the second coagulation time, and the second weight of the blood sample.

[0033] In one possible implementation, after determining the third coagulation time of the blood sample using magnetic beads in the blood sample that respond to the magnetic field strength of the target magnetic field, the method further includes:

[0034] Determine whether the third coagulation time of the blood sample falls within the coagulation time range corresponding to the second abnormal level;

[0035] If the third coagulation time of the blood sample does not fall within the coagulation time range corresponding to the second abnormal level, the blood sample is irradiated with a light source of the target light source wavelength to obtain the fourth coagulation time of the blood sample.

[0036] The third coagulation time of the blood sample is corrected using the fourth coagulation time of the blood sample so that the corrected third coagulation time of the blood sample falls within the coagulation time range corresponding to the second abnormal level.

[0037] The step of using the third coagulation time of the blood sample as the target coagulation time of the blood sample includes:

[0038] The third coagulation time of the corrected blood sample is taken as the target coagulation time of the blood sample.

[0039] A second aspect of this application provides a computer program product including computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the method for determining the coagulation time of a blood sample as described in the first aspect or any implementation thereof.

[0040] A third aspect of this application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0041] The memory is used to store computer programs;

[0042] The processor is used to execute the computer program so that the electronic device can implement the method for determining the coagulation time of a blood sample as described in the first aspect or any implementation thereof.

[0043] A fourth aspect of this application provides a computer storage medium carrying one or more computer programs that, when executed by an electronic device, enable the electronic device to determine the coagulation time of a blood sample according to the first aspect or any implementation thereof.

[0044] By employing the above technical solution, this application provides a method and related apparatus for determining the coagulation time of blood samples. The method includes: for samples of the first abnormality level, firstly, using optical detection to obtain a first coagulation time, then determining whether it falls within a corresponding range; if not, automatically using magnetic bead detection to obtain a second coagulation time, and determining a target coagulation time accordingly; for samples of the first abnormality level, the target coagulation time is determined by combining the first coagulation time obtained from optical detection and the second coagulation time obtained from magnetic bead detection. This multi-dimensional data combination method can mutually verify and supplement each other, making the detection results more reliable and avoiding misjudgments due to errors that may occur with a single detection method. For samples of the second abnormality level, the target coagulation time is directly obtained using magnetic bead detection. This automated switching of detection schemes greatly improves the automation level of the detection process, shortens the time from sample detection to result report generation, and improves the efficiency and accuracy of determining the coagulation time of blood samples. Attached Figure Description

[0045] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0046] Figure 1 A flowchart illustrating a method for determining the coagulation time of a blood sample, provided in an embodiment of this application;

[0047] Figure 2 A block diagram illustrating an optical detection method provided in an embodiment of this application;

[0048] Figure 3 This is a schematic diagram of the response curve for optical detection provided in an embodiment of this application;

[0049] Figure 4 This is a schematic diagram of a magnetic bead detection structure provided in an embodiment of this application;

[0050] Figure 5 This is a schematic diagram of the oscillation curve of a magnetic bead detection provided in an embodiment of this application;

[0051] Figure 6 This is a schematic diagram of the reaction curve for magnetic bead detection provided in an embodiment of this application;

[0052] Figure 7 This is a schematic diagram of the hardware structure of a device for determining the coagulation time of a blood sample, provided in an embodiment of this application. Detailed Implementation

[0053] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0054] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0055] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0056] To improve the accuracy and efficiency of determining the coagulation time of blood samples, this application provides a method for determining the coagulation time of blood samples. The method for determining the coagulation time of blood samples provided in this application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0057] Please see the appendix Figure 1 , Figure 1 This is a flowchart illustrating a method for determining the coagulation time of a blood sample, provided in an embodiment of this application. The method may include the following steps:

[0058] Step S101: Obtain the target components of the blood sample and determine the target light source wavelength and target magnetic field strength corresponding to the target components of the blood sample.

[0059] It is important to note that the target components of a blood sample are those related to coagulation function. Blood coagulation involves multiple components, primarily including: Coagulation factors: such as coagulation factor I (fibrinogen), coagulation factor II (prothrombin), coagulation factor V, coagulation factor VII, coagulation factor VIII, coagulation factor IX, and coagulation factor X. These factors interact during coagulation to form thrombin, which in turn converts fibrinogen into fibrin, forming a blood clot. Platelets: Platelets play a crucial role in coagulation; they can aggregate at sites of vascular injury, forming platelet emboli and preventing blood loss. Simultaneously, platelets release various coagulation-related substances, promoting the coagulation process. Anticoagulant components: such as antithrombin III and the protein C system, which play an important role in regulating the coagulation process and preventing excessive coagulation. Fibrin degradation products (FDP): Fibrin degradation products are the products generated during the decomposition of fibrin or fibrinogen during fibrinolysis. Their levels can reflect fibrinolytic activity in the body and indirectly reflect the balance between coagulation and fibrinolysis.

[0060] There are several ways to obtain the target components from a blood sample. The following describes some of the methods:

[0061] (1) Whole blood separation: After directly collecting a blood sample, it may be necessary to separate the blood sample according to the testing requirements. For example, the blood sample may be separated into blood sample or serum by centrifugation. Blood sample is the liquid part of blood after blood cells are removed. It contains a variety of clotting factors and proteins, and is an important sample source for detecting components related to coagulation function. Serum is the liquid that separates after blood coagulates, and also contains some coagulation-related components.

[0062] (2) Immunoprecipitation: This method utilizes the specific binding reaction between antigens and antibodies to obtain the target component. Specific antibodies against the target component are added to the blood sample, and the antibodies bind to the target component to form immune complexes. Then, magnetic or precipitable substances (such as protein A / G coupled magnetic beads) are added to bind with the antibodies, and the immune complexes are precipitated using methods such as magnetic fields or centrifugation, thereby achieving the separation and enrichment of the target component.

[0063] (3) Affinity chromatography: A separation technique based on the specific binding between a target component and a specific ligand. Ligands that specifically bind to the target component are immobilized on the packing material of a chromatography column. When a blood sample passes through the column, the target component binds to the ligand and remains in the column, while other components elute with the mobile phase. Subsequently, by changing the elution conditions, the target component is eluted from the ligand, thereby achieving the purification and enrichment of the target component.

[0064] In this application, after identifying the target component of the blood sample, firstly, the target component can be scanned across the entire wavelength range to obtain its spectral data. Then, the wavelength corresponding to the absorbance peak in the target component's spectral data can be used as the target light source wavelength. Next, the target component of the blood sample can be labeled with magnetic beads, causing the beads to respond to different magnetic fields in the blood sample, thus obtaining the bead response data. Finally, the magnetic field strength corresponding to the responsivity peak in the magnetic bead response data is used as the target magnetic field strength.

[0065] Specifically, first, the blood plasma is placed in an optically transparent cuvette (usually made of quartz or optical plastic) to ensure sufficient light transmittance for detection. The container material must be transparent to the target light source wavelength and free from fluorescence interference to facilitate subsequent spectral analysis. A spectrometer can be used to scan the target components of the blood sample across the entire wavelength range (e.g., 200-1000 nm), covering the ultraviolet, visible, and near-infrared regions. This light passes through the blood sample; some of the light is absorbed by various components in the blood sample, while the transmitted light is received by the spectrometer's detector. Based on the intensity of the transmitted light received by the spectrometer, the absorbance value at each wavelength can be calculated, thus obtaining the full-band absorption spectrum curve of the blood sample. The absorbance value reflects the degree of absorption of different wavelengths of light by the blood, and its calculation formula is: A = lg(I / I0), where I0 is the incident light intensity and I is the transmitted light intensity.

[0066] Each substance has its unique spectral characteristics, meaning it has a strong absorption of light at specific wavelengths. In blood samples, the characteristic absorption peaks corresponding to different components are as follows: Hemoglobin: has a characteristic absorption peak at 405 nm. If hemolysis occurs in the plasma, hemoglobin is released into the plasma, resulting in a significant absorption peak at this wavelength; Bilirubin: has a characteristic absorption peak in the 450-500 nm range. When the bilirubin content in the plasma increases, such as in the case of jaundice, the absorbance in this wavelength range will increase significantly; Lipids: have a characteristic absorption peak in the near-infrared band, such as at 660 nm. In cases of lipemia, the lipid content in the plasma increases, and its absorbance at this wavelength will also increase accordingly.

[0067] Then, on the obtained full-band absorption spectrum curve, find the point with the highest absorbance, i.e., the absorbance peak. The wavelength corresponding to this peak is the characteristic wavelength of the main absorbing component in the plasma sample, and can be used as the target light source wavelength. For example, if hemolysis is present in the plasma, the absorbance peak may appear at 405 nm; if jaundice is present, the peak may be at some wavelength in the range of 450-500 nm; if it is lipemia, the peak may be at 660 nm, etc.

[0068] Furthermore, second-derivative spectroscopy involves performing two derivative operations on the original absorption spectrum curve. This mathematical processing amplifies subtle changes in the spectral curve, separating and clarifying previously overlapping or nearly overlapping absorption peaks, while also eliminating background interference such as slow-changing interference caused by light scattering and instrument noise. In practice, specialized spectral analysis software is used to calculate the second derivative of the spectral data obtained from a full-band scan. The software processes the absorbance value at each wavelength point to obtain the second-derivative spectral curve. On this curve, absorption peaks that were not very prominent in the original spectrum become more prominent and sharper, making it easier to accurately determine the wavelength position corresponding to the absorbance peak, thus more precisely locking onto the target light source wavelength. For example, for characteristic absorption peaks of components that are difficult to distinguish in the original spectrum due to background interference or spectral overlap, second-derivative spectroscopy can help to more accurately identify and locate them.

[0069] Magnetic beads can then be used to label target components in blood samples. Magnetic beads are tiny magnetic particles with chemically modified surfaces that can specifically bind to target components. The magnetic bead labeling process typically involves the following steps: selecting magnetic beads with specific antibodies or ligands on their surfaces based on the properties of the target component. These magnetic beads can specifically bind to target components (such as clotting factors, platelets, etc.); mixing the magnetic beads with the blood sample and reacting them under appropriate conditions (such as temperature, pH, reaction time, etc.) to ensure sufficient binding between the magnetic beads and the target component; using a magnetic field to separate the magnetic beads bound to the target component from the sample, followed by washing with an appropriate buffer solution to remove unbound impurities and non-specifically bound substances.

[0070] Finally, through magnetic bead labeling, the magnetic beads can respond to different magnetic field strengths and frequencies in blood samples. To obtain the response data of the magnetic beads, magnetic analysis instruments (such as magnetic bead analyzers or magnetic resonance imaging equipment) can be used for detection. The specific steps are as follows: A series of magnetic fields of different intensities are applied to the container containing the labeled sample. The magnetic field strength can be gradually increased from low to high to observe the response of the magnetic beads under different magnetic fields; at each magnetic field strength, the magnetic responsivity of the magnetic beads is measured. The responsivity can be measured in various ways, such as the movement speed of the magnetic beads, the degree of aggregation, and the magnetization intensity. These data reflect the behavioral characteristics of the magnetic beads in the magnetic field. The responsivity of the magnetic beads at each magnetic field strength is recorded to form a curve relating magnetic field strength and responsivity. The recorded response data is plotted on a coordinate system, with the horizontal axis representing magnetic field strength and the vertical axis representing the responsivity of the magnetic beads. Usually, a curve with one or more peaks is obtained. On the response curve, the point where the responsivity reaches its maximum value is found, i.e., the responsivity peak. The magnetic field strength corresponding to this peak is the target magnetic field strength. The target magnetic field strength reflects the strongest magnetic response of the magnetic beads after they bind with the target component. This indicates that the interaction between the magnetic beads and the target component is most significant under this magnetic field strength, enabling effective detection and analysis of the target component.

[0071] Step S102: Use a machine learning model to determine the abnormality level of the blood sample. The abnormality level of the blood sample includes a first abnormality level and a second abnormality level, where the first abnormality level is lower than the second abnormality level.

[0072] It should be noted that the machine learning model includes a first anomaly level determination model and a second anomaly level determination model.

[0073] One possible implementation is to acquire an image of a blood sample and input the image into a pre-trained first anomaly level determination model to obtain the anomaly level of the blood sample.

[0074] In this application, images of blood samples can be acquired using specialized medical imaging equipment, such as microscope cameras, flow cytometers, or specific blood analysis imaging systems. These images clearly present information such as the morphology, distribution, and concentration of components in the blood sample, including cells and particles, such as the morphology and aggregation state of red blood cells, white blood cells, and platelets. These visual features may be closely related to the degree of blood abnormality. The acquired blood sample images are input into a pre-trained first abnormality level determination model. This model has learned to identify and analyze various features in images and associate them with corresponding abnormality levels through a large number of training samples with historical annotation information. Internally, the model performs feature extraction and processing on the input images, such as identifying abnormal morphology, aggregation degree, or other key visual indicators of bleeding cells, and then determines the abnormality level of the blood sample based on these features. Finally, the model outputs a prediction result, namely the first abnormality level of the blood sample, thus providing doctors or laboratory personnel with a reference for the degree of abnormality based on the sample's appearance features.

[0075] The process of determining the first anomaly level determination model may include: obtaining multiple historical training samples with historical annotation information, including the anomaly level of historical blood samples, and each training sample including an image of a historical blood sample. The initial model is trained based on these multiple historical training samples to obtain the first anomaly level determination model. The input of the first anomaly level determination model is the image of the blood sample, and the output of the first anomaly level determination model is the anomaly level of the blood sample.

[0076] Specifically, historical training samples are collected: samples with accurate abnormality level labels are selected from a large number of historical blood test records as training data. Each training sample consists of two parts: first, an image of the historical blood sample, which is acquired using professional equipment and preprocessed to ensure that its quality and clarity meet the requirements of feature analysis; second, the corresponding labeling information, i.e., the true abnormality level of the historical blood sample, which is usually determined by professional medical laboratory personnel or doctors based on clinical diagnosis results.

[0077] Preprocessing and Feature Extraction: Before training begins, the collected image data needs to be preprocessed, such as by standardizing image size, converting to grayscale, or normalizing pixel values, to eliminate differences between data and improve model training efficiency. Simultaneously, feature extraction is a crucial step. Traditional image processing techniques (such as edge detection and texture analysis) can be used to extract manually designed features, or convolutional neural networks (CNNs) in deep learning can automatically learn high-level feature representations from images. These features can more accurately capture visual information related to the level of abnormality in blood sample images, such as changes in cell morphology and abnormal distribution of components.

[0078] Training the initial model: Preprocessed image data and their corresponding anomaly level labels are input into the initial model for training. The initial model can be built based on artificial neural networks, support vector machines (SVM), or other machine learning algorithms suitable for image classification tasks. During training, the model continuously adjusts its parameters to minimize the error between the predicted anomaly level and the actual labeled anomaly level. This process typically requires multiple epochs. By continuously forward propagating the input data, calculating the loss function, and backpropagating to update the parameters, the model gradually learns the complex nonlinear relationship between image features and anomaly levels. Simultaneously, regularization techniques and validation sets are employed during training to prevent overfitting, ensuring the model has good generalization ability and can accurately predict anomaly levels on unseen new samples.

[0079] Validation and Optimization: The trained model is evaluated using an independent validation set. The validation set consists of a subset of historical samples not used in training. Predictive performance is objectively measured by calculating metrics such as accuracy, recall, and F1 score on the validation set. If the model performs poorly on the validation set, adjustments to its structure may be necessary, such as increasing or decreasing the number of network layers, changing the number of neurons, adjusting hyperparameters (e.g., learning rate, regularization strength), or employing different feature extraction methods. After repeated adjustments and optimizations, a first-order anomaly level determination model is finally obtained that can accurately and stably predict anomaly levels based on blood sample images.

[0080] As another possible implementation, a blood sample can be irradiated with a light source of multiple wavelengths to obtain the absorbance of the blood sample, and the absorbance of the blood sample can be input into a pre-trained second abnormality level determination model to obtain the abnormality level of the blood sample.

[0081] In this application, optical instruments such as a spectrophotometer can be used to sequentially illuminate a blood sample with multiple light sources of different wavelengths, measuring the absorbance of the sample at each wavelength. The selection of these specific wavelengths is typically based on the spectral characteristics of different components in the blood; for example, components such as hemoglobin and bilirubin have characteristic absorption peaks at specific wavelengths. By measuring the absorbance at multiple wavelengths, the relative content and optical properties of various components in the blood sample can be obtained. This data can reflect certain biochemical states and abnormalities in the blood sample. The obtained blood sample absorbance data is input into a pre-trained second anomaly level determination model. This model analyzes the absorbance data of new samples by learning the mapping relationship between absorbance characteristics and anomaly levels in historical data. It identifies absorbance patterns related to the anomaly level, such as abnormal increases or decreases in absorbance at certain wavelengths, and correlations between absorbance at different wavelengths. Based on these characteristics, the model calculates and outputs the second anomaly level of the blood sample, providing another dimension based on biochemical optical characteristics for anomaly assessment of blood samples.

[0082] The process involves collecting historical training samples: obtaining multiple historical training samples with historical annotation information, including the anomaly level of historical blood samples and the absorbance of each training sample. The initial model is then trained based on these historical training samples to obtain a second anomaly level determination model. The input to the second anomaly level determination model is the absorbance of the blood samples, and the output of the second anomaly level determination model is the anomaly level of the blood samples.

[0083] Specifically, similarly, historical blood sample data with anomaly level labels are collected. However, this sample data mainly includes absorbance measurements of blood samples at multiple light source wavelengths and their corresponding anomaly levels. This data can be obtained from previous blood spectral detection records, ensuring the accuracy of absorbance data for each sample and the reliability of anomaly level labeling.

[0084] Data preprocessing and feature engineering: Preprocessing absorbance data includes noise removal, normalization, and standardization to ensure data quality and consistency. Feature engineering also plays a crucial role in this process, potentially involving feature extraction and transformation of the raw absorbance data. For example, calculating differences or ratios between absorbance values ​​at different wavelengths, or the derivative of the absorbance curve after smoothing, helps to better capture patterns and trends related to anomaly levels in the absorbance data.

[0085] Training the initial model: Preprocessed absorbance feature data and their corresponding anomaly level labels are input into the initial model for training. Selectable model algorithms include linear regression, random forest, neural networks, and other machine learning methods suitable for handling numerical features and classification tasks. During training, the model adjusts its internal parameters based on the input data, learning the relationship between absorbance features and anomaly levels. Through multiple iterations of training, the model gradually improves its accuracy in predicting anomaly levels. A validation set is also used to monitor for overfitting, ensuring good model performance on new data.

[0086] Validation and optimization: The model's performance is evaluated using a validation set, and relevant evaluation metrics are calculated. Based on the evaluation results, necessary optimizations and adjustments are made to the model, such as adjusting the model's complexity, selecting different feature subsets, or adopting ensemble learning methods, to improve the model's predictive ability and generalization performance, ultimately obtaining a reliable second anomaly level determination model.

[0087] If the blood sample has an abnormality level of Level 1, then step S103 can be executed.

[0088] Step S103: Irradiate the blood sample with a light source of the target wavelength, and take the time when the light transmittance of the blood sample changes from the first light transmittance to the second light transmittance as the first coagulation time of the blood sample, and determine whether the first coagulation time of the blood sample belongs to the coagulation time range corresponding to the first abnormal level.

[0089] It should be noted that when a blood sample is classified as having the highest level of abnormality, it means that the abnormality is relatively mild. In this case, the abnormality will not significantly interfere with coagulation function testing, and the optical detection channel can be used for testing. The first transmittance is determined based on the transmittance of the initially liquid blood sample, and the second transmittance is determined based on the transmittance of the fully coagulated blood sample.

[0090] In this application, a target light source wavelength for the blood sample has been determined, which typically matches the spectral characteristics of components in the blood sample related to coagulation function. A light source of the selected wavelength (such as an LED or laser) is irradiated onto the blood sample. In the optical detection channel, the light emitted by the light source passes through the sample, generating signals such as transmitted light and scattered light based on the sample's optical properties. Before the blood sample is treated with coagulation reagent and the coagulation reaction begins, its transmittance in its initial liquid state is recorded, i.e., the first transmittance. This transmittance reflects the optical transparency of the sample before coagulation. The optical detection system continuously monitors changes in the intensity of transmitted light and converts them into a transmittance change curve. When the blood sample is completely coagulated, its transmittance is recorded, i.e., the second transmittance. This transmittance reflects the degree of light scattering and absorption by the fibrin clot formed after blood coagulation.

[0091] As the coagulation process proceeds, fibrinogen in the blood is gradually converted into insoluble fibrin, leading to increased blood turbidity and decreased light transmittance. For a clearer understanding, please refer to [link to relevant documentation]. Figure 2 , Figure 2 This is a block diagram illustrating an optical detection method provided in an embodiment of this application. Transmitted light is converted into an electrical signal by a photoelectric signal conversion circuit. After processing by an I / V conversion circuit and a voltage amplification circuit, the signal is input to an AD conversion chip, converted into a digital signal, and then acquired by an AD acquisition circuit board to form a response curve (AD sequence). For easier understanding, please refer to the following details. Figure 3 , Figure 3 This is a schematic diagram of a reaction curve for optical detection provided in an embodiment of this application. The reaction curve shows the change of transmitted light intensity over time. The horizontal axis represents the solidification time, and the vertical axis represents the intensity of transmitted light. The starting point of the curve corresponds to the initial transmittance (i.e., the first transmittance), and the plateau period of the curve corresponds to the final transmittance (i.e., the second transmittance). From the reaction curve of optical detection, the time point when the transmitted light intensity reaches a predetermined percentage of the solidification point is found and taken as the solidification time.

[0092] In clinical practice, a coagulation time range corresponding to the first level of abnormality can be determined based on coagulation time data from a large number of normal and mildly abnormal samples. The measured first coagulation time of a blood sample can be compared with this range. If the first coagulation time falls within this range, it can be preliminarily determined that the sample has a mild abnormality in coagulation function, meeting the characteristics of the first level of abnormality, and the first coagulation time of the blood sample can be directly used as the target coagulation time.

[0093] If the first coagulation time of the blood sample does not fall within the coagulation time range corresponding to the first abnormal level, then step S104 can be executed.

[0094] Step S104: Using magnetic beads in the blood sample that respond to the magnetic field strength of the target magnetic field, determine the second coagulation time of the blood sample, and determine the target coagulation time of the blood sample based on the first coagulation time and the second coagulation time of the blood sample.

[0095] In this application, firstly, the oscillation amplitude of a magnetic bead responding to a target magnetic field strength in a blood sample can be obtained. Then, the time it takes for the oscillation amplitude of the magnetic bead to change from a first oscillation amplitude to a second oscillation amplitude can be used as the second coagulation time of the blood sample. The first oscillation amplitude is determined based on the initial liquid state of the blood sample, and the second oscillation amplitude is determined based on the fully coagulated blood sample. Next, a transmittance fluctuation coefficient of the blood sample can be calculated based on its transmittance values, and a first confidence level for the first coagulation time of the blood sample can be determined based on this coefficient. Then, a signal-to-noise ratio (SNR) of the oscillation amplitude of the magnetic bead can be calculated based on its oscillation amplitude, and a second confidence level for the second coagulation time of the blood sample can be determined based on this SNR. Finally, a first weight is assigned to the first coagulation time of the blood sample, and a second weight is assigned to the second coagulation time of the blood sample, based on both the first and second confidence levels. Finally, the target coagulation time of a blood sample can be calculated based on the first coagulation time, the first weight, the second coagulation time, and the second weight of the blood sample.

[0096] Specifically, first, magnetic beads can be placed in a blood sample. The beads will respond to changes in the magnetic field. When the blood is liquid, the beads swing more significantly under the influence of the magnetic field; as the blood coagulates, the movement of the beads is restricted, and the swing amplitude decreases. For a clearer understanding, please refer to [link to relevant documentation]. Figure 4 , Figure 4 This is a schematic diagram of a magnetic bead detection structure provided in an embodiment of this application. Figure 4 The diagram shows an LED (1), a light-receiving circuit (2), a driving coil (3), a magnetic bead (4), and a reaction vessel (5). The LED emits light that passes through a blood sample in the reaction vessel. If a steel bead is present in the sample, the light is partially blocked. The light-receiving circuit then receives the light after it has passed through the blood sample and the steel bead, converting the optical signal into an electrical signal. This electrical signal reflects the extent to which the steel bead blocks the transmitted light path. Simultaneously, the driving coil generates an alternating electromagnetic field, which drives the steel bead to oscillate within the reaction vessel. As the steel bead oscillates, the time it blocks the transmitted light path changes; this time is called the blocking time t. There is a relationship between the oscillation amplitude A of the steel bead and the blocking time t; by monitoring the oscillation amplitude A, the blocking time t can be determined. For a clearer understanding, please refer to [reference needed]. Figure 5 , Figure 5 This is a schematic diagram of the oscillation curve of a magnetic bead detection provided in an embodiment of this application.

[0097] As blood coagulates, fibrinogen gradually transforms into cross-linked fibrin, causing changes in the viscosity and fluidity of the reaction system. This significantly hinders the oscillation of the steel bead until it can no longer oscillate. The oscillation amplitude of the magnetic bead can be obtained by analyzing the signal output from the light-receiving circuit. This change in the oscillation amplitude reflects the alteration in the blood coagulation state. As blood coagulates, the oscillation amplitude of the magnetic bead gradually decreases from an initial large value (first oscillation amplitude) until it reaches a stable small value (second oscillation amplitude). This process can be monitored in real time by a detection device. For a clearer understanding, please refer to [link to relevant documentation]. Figure 6 , Figure 6 This is a schematic diagram of the reaction curve for magnetic bead detection provided in an embodiment of this application. The horizontal axis represents the reaction time, and the vertical axis represents the light-blocking time of the magnetic bead. In the initial stage of the reaction curve for magnetic bead detection, the light-blocking time is short and fluctuates little. As the blood coagulates, the movement of the magnetic bead is restricted, and the light-blocking time gradually increases. When the blood has completely coagulated, the magnetic bead may stop at the position of the light-transmitting aperture or within a range on both sides of the light-transmitting aperture. At this time, the light-blocking time of the magnetic bead changes abruptly, indicating that the magnetic bead tends to stop. The time point at which the light-blocking time of the magnetic bead changes significantly can be found, that is, the time required for the swing amplitude of the magnetic bead to change from the first swing amplitude to the second swing amplitude can be determined from the curve. This time is the second coagulation time of the blood sample.

[0098] Based on the changes in transmittance of blood samples during the coagulation process, a transmittance fluctuation coefficient can be calculated. This coefficient reflects the degree of transmittance fluctuation over time and can be used to assess the reliability of the first coagulation time measured based on transmittance. The magnitude of the transmittance fluctuation coefficient determines the first reliability of the first coagulation time. A smaller transmittance fluctuation coefficient indicates more stable transmittance changes, and a higher reliability of the first coagulation time; conversely, a larger coefficient indicates lower reliability. Based on the swing amplitude data of a steel ball, the swing amplitude signal-to-noise ratio (SNR) can be calculated. The SNR reflects the relative intensity of the useful signal and the noise signal and can be used to assess the reliability of the second coagulation time measured based on the swing amplitude of the steel ball. The magnitude of the SNR determines the second reliability of the second coagulation time. A higher SNR indicates a stronger useful signal and less noise interference, and a higher reliability of the second coagulation time; conversely, a lower SNR indicates lower reliability.

[0099] Finally, weights can be assigned to the first and second coagulation times based on their respective confidence levels. Coagulation times with higher confidence levels will receive larger weights, reflecting their importance in the final coagulation time calculation. Combining the first and second coagulation times with their respective weights, and using a weighted average or other suitable mathematical method, the final target coagulation time of the blood sample is calculated. This target coagulation time comprehensively considers the results of both transmittance and steel ball oscillation amplitude detection methods, resulting in higher reliability and accuracy.

[0100] If the blood sample has an abnormality level of the second abnormality level, then step S105 can be executed.

[0101] Step S105: Using magnetic beads in the blood sample that respond to the magnetic field strength of the target magnetic field, determine the third coagulation time of the blood sample, and use the third coagulation time of the blood sample as the target coagulation time of the blood sample.

[0102] In this application, when the blood sample's abnormality level is the second level, indicating a relatively severe abnormality, the interaction between magnetic beads and a magnetic field can be used to more accurately monitor the blood coagulation process. The magnetic beads are ferromagnetic steel balls that can be magnetized in a magnetic field and respond to changes in the field. The movement of the magnetic beads within the blood sample can be controlled by an external magnetic field. The oscillation of the magnetic beads within the blood sample can be used to monitor the blood coagulation process. The oscillation amplitude and duration of the magnetic beads vary depending on the coagulation state of the blood sample. An alternating electromagnetic field is applied through drive coils on both sides of the reaction cup. The strength and frequency of the magnetic field generated by the drive coils can be adjusted as needed. The alternating electromagnetic field causes the magnetic beads to oscillate within the reaction cup. When the blood sample is not coagulated, the magnetic beads can oscillate freely; as the blood sample coagulates, the oscillation of the magnetic beads is restricted.

[0103] When the blood sample is in a liquid state, the magnetic bead oscillates along a curved track at the bottom of the reaction cup under the drive of an alternating electromagnetic field. At this stage, the oscillation amplitude of the magnetic bead is large, and the oscillation time is short. The oscillation of the magnetic bead is monitored by an optical detection system (including an LED and a transmitted light receiving circuit). During oscillation, the magnetic bead periodically blocks the transmitted light path, causing changes in the intensity of the transmitted light. The transmitted light receiving circuit converts these changes into electrical signals and records the time the magnetic bead blocks the light-transmitting aperture. As the blood sample undergoes a clotting reaction, fibrinogen is gradually converted into cross-linked fibrin, and the viscosity and fluidity of the blood sample change. The oscillation of the magnetic bead is gradually hindered, the oscillation amplitude decreases, and the oscillation time prolongs. When the blood sample is completely coagulated, the oscillation of the magnetic bead tends to stop. At this point, the magnetic bead may stop at the light-transmitting aperture or within a range on either side of the aperture. A significant change in the time the magnetic bead blocks the light-transmitting aperture indicates that the magnetic bead is approaching a stationary state. By recording the change in the time the magnetic bead blocks the light-transmitting aperture during the oscillation process and combining it with algorithmic analysis, the time from oscillation to approaching a stationary state can be determined, which is the third coagulation time.

[0104] Furthermore, after determining the third coagulation time of the blood sample using magnetic beads that respond to the magnetic field strength of the target magnetic field in the blood sample, the method may further include: first, determining whether the third coagulation time of the blood sample falls within the coagulation time range corresponding to the second abnormality level; then, if the third coagulation time of the blood sample does not fall within the coagulation time range corresponding to the second abnormality level, irradiating the blood sample with a light source of the target wavelength to obtain the fourth coagulation time of the blood sample; finally, correcting the third coagulation time of the blood sample using the fourth coagulation time of the blood sample so that the corrected third coagulation time of the blood sample falls within the coagulation time range corresponding to the second abnormality level.

[0105] At this point, using the third coagulation time of the blood sample as the target coagulation time of the blood sample can be used as the corrected third coagulation time of the blood sample.

[0106] Specifically, based on extensive clinical data and experimental statistics, it was first determined that for blood samples with the second level of abnormality (more severe abnormality), the clotting time will fall within a specific range. This range reflects the performance of blood clotting function under more severe abnormal conditions. The measured third clotting time is then compared with this range. If the third clotting time is within the range, it is directly used as the target clotting time, indicating that the blood sample's clotting function is in a more severe abnormal state, and no further correction is required.

[0107] If the readings are outside the acceptable range, it indicates potential data bias or interference, requiring further processing. The target light source wavelength, determined in the preceding steps, typically matches the spectral characteristics of coagulation-related components in the blood sample, effectively reflecting the coagulation state. Irradiating the blood sample with the selected wavelength allows it to interact with the coagulation components, generating a detectable signal. As the blood coagulates, fibrinogen is converted to fibrin, increasing blood turbidity and decreasing transmittance. Changes in transmitted light intensity are monitored using an optical detection system, recording the transmittance change curve. The time when the transmittance level reaches the predetermined percentage of the coagulation point is identified from the transmittance change curve and designated as the fourth coagulation time, reflecting the blood sample's coagulation function.

[0108] Finally, when the third coagulation time exceeds the range corresponding to the second abnormality level, it indicates the possible presence of interfering factors or measurement errors, causing the third coagulation time to fail to accurately reflect the true coagulation state of the blood sample. In this case, using the fourth coagulation time for correction can improve the accuracy of the coagulation time. Based on the relationship between the third and fourth coagulation times, such as by calculating the average, weighted average, or other mathematical methods, the third coagulation time is adjusted to fall within the coagulation time range corresponding to the second abnormality level, resulting in the corrected third coagulation time.

[0109] In summary, this application provides a method for determining the coagulation time of blood samples. The method includes: for samples of the first abnormality level, firstly, using optical detection to obtain a first coagulation time, then determining whether it falls within a corresponding range; if not, automatically using magnetic bead detection to obtain a second coagulation time, and determining the target coagulation time accordingly. For samples of the first abnormality level, the target coagulation time is determined by combining the first coagulation time obtained from optical detection and the second coagulation time obtained from magnetic bead detection. This multi-dimensional data combination method allows for mutual verification and supplementation, making the detection results more reliable and avoiding misjudgments due to errors that may occur with a single detection method. For samples of the second abnormality level, the target coagulation time is directly obtained using magnetic bead detection. This automated switching of detection schemes significantly improves the automation level of the detection process, shortens the time from sample detection to result report generation, and improves the efficiency and accuracy of determining the coagulation time of blood samples.

[0110] This application also provides an electronic device in its embodiments. (See reference...) Figure 7 The diagram illustrates a structural schematic suitable for implementing the electronic device in the embodiments of this application. The electronic device in the embodiments of this application may include, but is not limited to, fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 7 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0111] like Figure 7 As shown, the electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage device 708 into a random access memory (RAM) 703. When the electronic device is powered on, the RAM 703 also stores various programs and data required for the operation of the electronic device. The processing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0112] Typically, the following devices can be connected to I / O interface 705: input devices 706 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 707 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 708 including, for example, memory cards, hard drives, etc.; and communication devices 709. Communication device 709 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 7 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have alternatively.

[0113] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement any of the methods for determining the coagulation time of a blood sample provided in this application.

[0114] This application also provides a computer-readable storage medium carrying one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any of the methods for determining the coagulation time of a blood sample provided in this application.

[0115] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0116] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0117] In the above embodiments, the implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, in the form of a computer program product.

[0118] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

Claims

1. A method of determining the clotting time of a blood sample, characterized in that, The method comprises the following steps: acquiring a target component of a blood sample and determining a target light source wavelength and a target magnetic field strength corresponding to the target component of the blood sample, the target component of the blood sample being a component related to blood coagulation function in the blood sample; determining an abnormality level of the blood sample by using a machine learning model, the abnormality level of the blood sample comprising a first abnormality level and a second abnormality level, the first abnormality level being lower than the second abnormality level, the machine learning model comprising a first abnormality level determination model and a second abnormality level determination model; if the abnormality level of the blood sample is the first abnormality level, irradiating the blood sample by using a light source of the target light source wavelength, changing a light transmittance of the blood sample from a first light transmittance to a second light transmittance to obtain a first coagulation time of the blood sample, and determining whether the first coagulation time of the blood sample belongs to a coagulation time range corresponding to the first abnormality level, the first light transmittance being a light transmittance determined according to the blood sample in an initial liquid state, and the second light transmittance being a light transmittance determined according to the blood sample in a complete coagulation state; if the first coagulation time of the blood sample does not belong to the coagulation time range corresponding to the first abnormality level, determining a second coagulation time of the blood sample by using magnetic beads in the blood sample responding to a magnetic field of the target magnetic field strength, and determining a target coagulation time of the blood sample based on the first coagulation time of the blood sample and the second coagulation time of the blood sample; if the abnormality level of the blood sample is the second abnormality level, determining a third coagulation time of the blood sample by using the magnetic beads in the blood sample responding to the magnetic field of the target magnetic field strength, and taking the third coagulation time of the blood sample as the target coagulation time of the blood sample; the method of determining the abnormality level of the blood sample by using the machine learning model comprises: acquiring a picture of the blood sample, inputting the picture of the blood sample into a pre-trained first abnormality level determination model, and obtaining the abnormality level of the blood sample; or, irradiating the blood sample by using light sources of multiple light source wavelengths, obtaining an absorbance of the blood sample, and inputting the absorbance of the blood sample into a pre-trained second abnormality level determination model to obtain the abnormality level of the blood sample; the method of determining the target coagulation time of the blood sample based on the first coagulation time of the blood sample and the second coagulation time of the blood sample comprises: calculating a light transmittance fluctuation coefficient of the blood sample based on each light transmittance of the blood sample, and determining a first credibility of the first coagulation time of the blood sample based on the light transmittance fluctuation coefficient of the blood sample; calculating a swing amplitude signal-to-noise ratio of the magnetic beads based on each swing amplitude of the magnetic beads, and determining a second credibility of the second coagulation time of the blood sample based on the swing amplitude signal-to-noise ratio of the magnetic beads; assign a first weight to a first clotting time of the blood sample and a second weight to a second clotting time of the blood sample based on the first credibility and the second credibility; calculate a target clotting time of the blood sample based on the first clotting time of the blood sample, the first weight, the second clotting time of the blood sample, and the second weight.

2. The method of determining the clotting time of a blood sample according to claim 1, wherein, The determination of the target component of the blood sample corresponds to the target light source wavelength and the target magnetic field strength, comprising: full-band scanning the target component of the blood sample to obtain spectral data of the target component; the wavelength corresponding to the absorbance peak in the spectral data of the target component is taken as the target light source wavelength; magnetic bead labeling the target component of the blood sample to make the magnetic beads respond to different magnetic fields in the blood sample to obtain response data of the magnetic beads; the magnetic field strength corresponding to the response degree peak in the response data of the magnetic beads is taken as the target magnetic field strength.

3. The method of determining the clotting time of a blood sample according to claim 1, wherein, The determination process of the first abnormality level determination model comprises: obtaining a plurality of historical training samples with historical annotation information, the annotation information including the abnormality level of the historical blood sample, each of the training samples including a picture of the historical blood sample; training an initial model based on a plurality of the historical training samples to obtain the first abnormality level determination model, the input of the first abnormality level determination model being the picture of the blood sample, and the output of the first abnormality level determination model being the abnormality level of the blood sample.

4. The method of determining the clotting time of a blood sample according to claim 1, wherein, The determination of the second clotting time of the blood sample by using the magnetic beads responding to the target magnetic field strength in the magnetic field of the blood sample comprises: using the magnetic beads responding to the target magnetic field strength in the magnetic field of the blood sample to obtain the oscillation amplitude of the magnetic beads; the time for the oscillation amplitude of the magnetic beads to change from a first oscillation amplitude to a second oscillation amplitude is taken as the second clotting time of the blood sample, the first oscillation amplitude being the oscillation amplitude determined according to the initial liquid state of the blood sample, and the second oscillation amplitude being the oscillation amplitude determined according to the completely coagulated blood sample.

5. The method of determining the clotting time of a blood sample according to claim 1, wherein, After the determination of the third clotting time of the blood sample by using the magnetic beads responding to the target magnetic field strength in the magnetic field of the blood sample, the method further comprises: determining whether the third clotting time of the blood sample belongs to the clotting time range corresponding to the second abnormality level; if the third clotting time of the blood sample does not belong to the clotting time range corresponding to the second abnormality level, irradiating the blood sample by using the light source of the target light source wavelength to obtain a fourth clotting time of the blood sample; correcting the third clotting time of the blood sample by using the fourth clotting time of the blood sample so that the corrected third clotting time of the blood sample belongs to the clotting time range corresponding to the second abnormality level; The third clotting time of the blood sample is taken as the target clotting time of the blood sample, comprising: the third clotting time of the corrected blood sample is taken as the target clotting time of the blood sample.

6. A computer program product, characterised in that, Computer readable instructions are included, which, when run on an electronic device, enable the electronic device to implement the method for determining the clotting time of a blood sample according to any one of claims 1 to 5.

7. An electronic device, comprising: The electronic device comprises at least one processor and a memory connected to the processor, wherein: The memory is configured to store a computer program; The processor is configured to execute the computer program to enable the electronic device to implement the method for determining the clotting time of a blood sample according to any one of claims 1 to 5.

8. A computer storage medium, characterized in that The storage medium carries one or more computer programs, which, when executed by an electronic device, enable the electronic device to implement the method for determining the clotting time of a blood sample according to any one of claims 1 to 5.

Images