Hemolytic recognition method based on thrombus sample image, related device and product

Abstract

The embodiment of the invention relates to the technical field of medical equipment detection, and provides a hemolytic recognition method based on a thrombus sample image, a related device and a product. The method comprises the following steps: acquiring a hemolysis image data set and a non-hemolysis image data set; adjusting an SVM neural network model according to the hemolysis image data set and the non-hemolysis image data set to obtain a sample hemolysis detection model; acquiring to-be-tested thrombus sample image data of a user; and analyzing and processing the image data of the to-be-detected thrombus sample of the user by adopting the sample hemolytic detection model to obtain a hemolytic recognition result. The technical problem that when an existing thrombelastogram is used for thrombus detection, the thrombus detection result is not accurate due to interference of non-physiological factors is solved.

Description

Technical Field

[0001] The present application relates to the field of medical equipment detection technology, and in particular to a hemolytic identification method based on thrombus sample images and related devices and products. Background Art

[0002] In the broad field of clinical medicine, thromboelastograph (TEG), as a precise and efficient coagulation function detection device, is widely used in the diagnosis of various coagulation-related diseases, monitoring of treatment process and formulation of preventive measures.

[0003] The thromboelastography simulates the coagulation and fibrinolysis process in the body, dynamically monitors the entire process of coagulation, thrombus formation and dissolution of blood samples under the action of activators, and provides doctors with a series of key parameters including coagulation reaction time, coagulation rate, fibrin cross-linking strength and maximum thrombus strength. These parameters are of great value in evaluating the patient's coagulation status and predicting the risk of bleeding or thrombosis.

[0004] However, in the actual operation process, it is inevitable to be interfered by non-physiological factors. These non-physiological factors seriously restrict the accuracy and reliability of the detection results of the thromboelastograph, resulting in the thromboelastograph generated by the thromboelastograph being unable to accurately reflect the current coagulation state in the patient.

[0005] In view of this, the inventor proposes a hemolytic identification method based on thrombus sample images, a related device and a product. Summary of the invention

[0006] The embodiment of the present application provides a hemolytic identification method based on a thrombus sample image and a related device to solve the technical problem that when the existing thromboelastography instrument performs thrombus detection, it may be interfered by non-physiological factors, resulting in inaccurate thrombus detection results.

[0007] A first aspect of an embodiment of the present application provides a hemolytic identification method based on a thrombus sample image, the method comprising:

[0008] Acquire a hemolysis image dataset and a non-hemolysis image dataset;

[0009] Adjusting the SVM neural network model according to the hemolytic image data set and the non-hemolytic image data set to obtain a sample hemolytic detection model;

[0010] Obtaining the image data of the thrombus sample to be tested by the user;

[0011] The sample hemolysis detection model is used to analyze and process the image data of the user's thrombus sample to be tested to obtain a hemolysis recognition result.

[0012] Furthermore, the SVM neural network model is adjusted according to the hemolytic image data set and the non-hemolytic image data set to obtain a sample hemolytic detection model, including:

[0013] Merging the hemolysis image dataset and the non-hemolysis image dataset to obtain a total hemolysis image dataset;

[0014] Performing data preprocessing on the total hemolysis image data set to obtain a hemolysis feature data set;

[0015] Dividing the hemolysis characteristic data set into a training set, a validation set and a test set;

[0016] The sample hemolysis detection model is trained using a SVM neural network model according to the training set, validation set and test set until the model converges to obtain a trained sample hemolysis detection model.

[0017] Furthermore, the total hemolysis image data set is preprocessed to obtain a hemolysis feature data set, including:

[0018] Performing data cleaning on the total hemolysis image data set to obtain a cleaned total hemolysis image data set;

[0019] Hemolysis characteristic data is extracted from the cleaned total hemolysis image data set to obtain a hemolysis characteristic data set.

[0020] Furthermore, the total hemolysis image data set is cleaned to obtain a cleaned total hemolysis image data set, including:

[0021] Identifying missing values ​​and outliers in the total hemolysis image dataset;

[0022] Interpolation method is used to correct the missing data values ​​and the abnormal values ​​are directly eliminated;

[0023] The corrected data are used to replace the missing data values ​​in the original data set to obtain the cleaned total hemolysis image data set.

[0024] Further, hemolysis feature data is extracted from the cleaned total hemolysis image data set to obtain a hemolysis feature data set, including:

[0025] Extracting hemolysis color feature data from the cleaned total hemolysis image data set using a global color histogram method;

[0026] Extracting hemolysis texture feature data from the cleaned total hemolysis image data set using a gray level co-occurrence matrix method;

[0027] Using the Canny edge detection method to extract the morphological characteristic data of hemolytic cells from the total hemolytic image data set after cleaning;

[0028] The hemolysis color feature data, the hemolysis texture feature data and the hemolysis cell morphology feature data are combined to obtain a hemolysis feature data set.

[0029] A second aspect of an embodiment of the present application provides a thrombus detection device, the device comprising:

[0030] A first acquisition unit is used to acquire a hemolysis image data set and a non-hemolysis image data set;

[0031] The second acquisition unit is used to acquire the thrombus sample data to be tested by the user;

[0032] A first processing unit, configured to adjust the SVM neural network model according to the hemolysis image data set and the non-hemolysis image data set to obtain a sample hemolysis detection model;

[0033] The second processing unit is used to analyze and process the user's thrombus sample data to be tested by using the sample hemolysis detection model to obtain a hemolysis identification result.

[0034] A third aspect of an embodiment of the present application provides an electronic device, comprising a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the hemolytic identification method based on the thrombus sample image as described in any one of the first aspects.

[0035] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the hemolytic identification method based on a thrombus sample image as described in any one of the first aspects.

[0036] A fifth aspect of an embodiment of the present application provides a computer program product, comprising a non-transitory computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the hemolytic identification method based on a thrombus sample image as described in any one of the first aspects.

[0037] Beneficial effects:

[0038] The present invention proposes a hemolytic identification method based on thrombus sample images. First, a hemolytic image data set and a non-hemolytic image data set are obtained, and a sample hemolytic detection model is trained using a SVM neural network model to obtain a trained sample hemolytic detection model, so as to identify the hemolytic degree of the sample to be tested, and then obtain a hemolytic identification result. The hemolytic identification result can be used to eliminate the interference of the sample hemolytic phenomenon on the sample detection by the thromboelastograph. Since the hemolytic phenomenon is the largest non-physiological interference privacy, the technical problem of inaccurate thrombus detection results caused by the interference of non-physiological factors when using a thromboelastograph for thrombus detection in the prior art is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic diagram of the overall process of a hemolytic identification method based on a thrombus sample image is provided for an embodiment of the present application;

[0041] Figure 2 A schematic diagram of the training process of a sample hemolytic detection model of a hemolytic identification method based on a thrombus sample image is provided for an embodiment of the present application;

[0042] Figure 3 A schematic diagram of the data preprocessing process of a training sample hemolytic detection model for a hemolytic identification method based on thrombus sample images is provided for an embodiment of the present application;

[0043] Figure 4 A schematic diagram of a process for acquiring a hemolysis characteristic data set in the process of training a hemolysis detection model of a hemolysis identification method based on a thrombus sample image is provided for an embodiment of the present application;

[0044] Figure 5 A structural schematic diagram of a hemolytic identification device based on a thrombus sample image is provided for an embodiment of the present application;

[0045] Figure 6 A schematic diagram of the structure of an electronic device is provided for an embodiment of the present application;

[0046] Reference numerals:

[0047] First acquisition unit-001, second acquisition unit-002, first processing unit-003, second processing unit-004. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0049] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0050] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0051] In order to better understand the hemolytic identification method based on thrombus sample images provided in an embodiment of the present application, the scenario of applying the business product determination method is first briefly introduced below.

[0052] The specific scenario involved in this embodiment is that when a thromboelastograph is used to detect thrombus on a sample to be tested, it may be interfered by non-physiological factors during the actual operation. These non-physiological factors seriously restrict the accuracy and reliability of the test results of the thromboelastograph, resulting in the thromboelastograph generated by the thromboelastograph being unable to accurately reflect the current coagulation state in the patient. Among them, since improper operation may occur in the early processing of the sample, the sample may be hemolyzed, so hemolysis is the most likely non-physiological factor. In addition, in the process of processing the sample, if the medical staff violently shakes or stirs the sample, the temperature is controlled too high or too low during storage, and the centrifugation rate is too fast or too long, these common phenomena may cause red blood cell damage and rupture, thereby inducing hemolysis. Finally, when the sample to be tested is placed in the thromboelastograph for testing, the accuracy of the test result of the thromboelastograph is affected due to the hemolysis of the current sample.

[0053] In summary, this embodiment solves the technical problem that when the existing thromboelastograph performs thrombus detection, the thrombus detection result is inaccurate due to interference from non-physiological factors. This embodiment preferably proposes a hemolytic identification method based on thrombus sample images, such as Figure 1 As shown, including:

[0054] S1. Obtain a hemolytic image dataset and a non-hemolytic image dataset. In this embodiment, preferably, the hemolytic image dataset and the non-hemolytic image dataset are obtained from the TCIA medical image database for subsequent training of a neural network.

[0055] S2. According to the hemolytic image data set and the non-hemolytic image data set, the SVM neural network model is adjusted to obtain a sample hemolytic detection model. In this embodiment, preferably, Figure 2 As shown, including:

[0056] S201, merging the hemolysis image dataset and the non-hemolysis image dataset to obtain a total hemolysis image dataset, and merging the two datasets into one dataset to facilitate simultaneous processing of the dataset.

[0057] S202, performing data preprocessing on the total hemolysis image data set to obtain a hemolysis feature data set, which preferably includes:

[0058] S2021, performing data cleaning on the total hemolysis image dataset to obtain a cleaned total hemolysis image dataset, specifically comprising:

[0059] S20211. Identifying missing values ​​and abnormal values ​​in the total hemolysis image dataset;

[0060] S20212. Correct the missing data values ​​using an interpolation method, and directly remove the abnormal values;

[0061] S20213, replacing the missing data values ​​in the original data set with the corrected data to obtain the cleaned total hemolysis image data set. Through the above data cleaning steps, a complete total hemolysis image data set can be obtained without missing or abnormal values, thereby increasing the accuracy of subsequent model training.

[0062] S2022, extracting hemolysis characteristic data from the cleaned total hemolysis image data set to obtain a hemolysis characteristic data set, as a preferred solution of this embodiment, including:

[0063] S20221, extracting hemolysis color feature data from the cleaned total hemolysis image data set using a global color histogram method. Specifically, as a preferred solution of this embodiment, the method includes:

[0064] Read image data one by one from the cleaned total hemolysis image data set;

[0065] Convert the original color space of image data to a more suitable color analysis space;

[0066] Calculate the global color histogram of each image in the selected color space;

[0067] Statistical features are extracted from the color histogram, wherein the statistical features include mean, median, standard deviation, skewness, and kurtosis, and the statistical features are used as hemolysis color feature data.

[0068] S20222, extracting hemolysis texture feature data from the cleaned total hemolysis image data set using a gray level co-occurrence matrix method. Specifically, as a preferred solution of this embodiment, the method includes:

[0069] First, convert the color image into a grayscale image;

[0070] Set the distance and direction, and calculate the gray-level co-occurrence matrix for each gray-level image;

[0071] Texture features are extracted from the gray-level co-occurrence matrix, where the texture features include contrast, energy, entropy, and homogeneity.

[0072] S20223, extracting hemolytic cell morphological feature data from the cleaned total hemolytic image data set using the Canny edge detection method, which is preferred in this embodiment and specifically includes:

[0073] Use the Canny edge detection algorithm to perform edge detection on the grayscale image to obtain an edge image;

[0074] Perform morphological operations on edge images to improve edge continuity and remove noise;

[0075] The morphological features are extracted from the improved edge image as the hemolytic cell morphological feature data, wherein the morphological features include edge length, edge number, complexity of edge distribution and edge curvature, and the hemolytic cell morphological feature data is used to reflect the morphological changes of the cells.

[0076] S20224, merging the hemolysis color feature data, the hemolysis texture feature data and the hemolysis cell morphology feature data to obtain a hemolysis feature data set. Through the above steps, the hemolysis feature data is extracted from the total hemolysis image data set after cleaning, and a hemolysis feature data set for subsequent analysis is constructed, and the hemolysis feature data set is used as an input for training the SVM neural network model, so that the trained SVM neural network model can accurately detect the hemolysis of the sample.

[0077] S203, dividing the hemolysis characteristic data set into a training set, a validation set and a test set;

[0078] S204, training a sample hemolysis detection model using a SVM neural network model according to the training set, the validation set and the test set until the model converges to obtain a trained sample hemolysis detection model. Specifically, this embodiment preferably includes:

[0079] The RBF kernel is used as the kernel parameter of the SVM neural network model. In this step, the RBF kernel is used to enable the SVM model to capture the nonlinear characteristics in the image data, thereby more accurately classifying the hemolytic image and the non-hemolytic image;

[0080] Initialize the regularization parameter and gamma parameter in the SVM neural network model. It should be noted that the regularization parameter C is used to control the balance between the complexity of the model and the training error, and the gamma parameter affects the range of action of the RBF kernel.

[0081] The SVM model is trained using a training set that contains a large amount of hemolytic image data and non-hemolytic image data. Through training, the SVM model can learn the feature representation and classification rules in these data, laying the foundation for subsequent classification tasks.

[0082] K-fold cross validation was used to evaluate the model performance under different parameter combinations, and the optimal parameter combination was selected. The above steps were iterated to obtain a preliminary sample hemolytic detection model to reduce the risk of overfitting and help find the optimal parameter combination, thereby improving the generalization ability of the model;

[0083] The preliminary sample hemolysis detection model was validated using the validation set to evaluate the generalization ability of the model;

[0084] According to the validation results, the preliminary sample hemolysis detection model was fine-tuned to obtain a further optimized sample hemolysis detection model;

[0085] Finally, the validation set is used to validate the further optimized sample hemolytic detection model, and the trained hemolytic detection model is obtained for validation. Through the above steps, a well-trained hemolytic detection model with excellent performance is obtained to perform hemolytic identification on the thrombus sample to be detected.

[0086] S3, obtaining the image data of the thrombus sample to be tested by the user;

[0087] S4. Using the sample hemolysis detection model to analyze and process the image data of the user's thrombus sample to be tested, to obtain a hemolysis recognition result.

[0088] That is, this embodiment proposes a hemolytic identification method based on thrombus sample images. First, a hemolytic image data set and a non-hemolytic image data set are obtained, and a sample hemolytic detection model is trained using a SVM neural network model to obtain a trained sample hemolytic detection model, so as to identify the hemolytic degree of the sample to be tested, and then obtain a hemolytic identification result. The hemolytic identification result can be used to eliminate the interference of the sample hemolytic phenomenon on the sample detection by the thromboelastograph. Since the hemolytic phenomenon is the largest non-physiological interference privacy, the technical problem of inaccurate thrombus detection results caused by the interference of non-physiological factors when using a thromboelastograph for thrombus detection in the prior art is solved.

[0089] In addition, as a preferred solution of this embodiment, it also includes:

[0090] S5. Determine the thrombus detection mode based on the hemolytic identification results.

[0091] S6. extracting a hemolysis percentage value from the hemolysis identification result;

[0092] S7, comparing the hemolytic percentage value with a preset first threshold value to determine a thrombus detection mode, specifically, this embodiment preferably includes:

[0093] The hemolysis percentage value is compared with the preset first threshold value. If the hemolysis percentage value is higher than the preset first threshold value, it is confirmed that the thrombus detection mode is not in compliance with the thrombus detection link mode. At this time, it is considered that the current thrombus sample to be detected has a high degree of hemolysis, which will have a greater impact on the thrombus detection results. This assists medical staff in making the decision to discard the current sample and avoid sending the sample to the thromboelastography instrument for thrombus detection, wasting medical testing resources. That is, when the hemolysis percentage value is higher than the preset first threshold value, medical staff can quickly determine that the sample is not suitable for thrombus detection, thereby avoiding possible misdiagnosis or missed diagnosis. This instant information feedback helps medical staff make more reasonable and scientific clinical decisions.

[0094] The hemolysis percentage value is compared with the preset first threshold value. If the hemolysis percentage value is lower than the preset first threshold value, the thrombus detection mode is confirmed to enter the thrombus detection link mode directly. In this step, since the hemolysis percentage value is lower than the preset first threshold value, it indicates that the hemolysis degree of the sample is low and will not have a significant impact on the subsequent thrombus detection results. Therefore, directly sending such a sample to the thrombus detection link can save unnecessary evaluation or preprocessing steps, thereby speeding up the detection process and improving detection efficiency.

[0095] For the above embodiments, please refer to Figure 6 , Figure 6A structural schematic diagram of an electronic device provided in an embodiment of the present application, as shown in the figure, includes a processor, an input device, an output device and a memory, the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, the processor is configured to call the program instructions, and the program includes instructions for executing the following steps;

[0096] Acquire a hemolysis image dataset and a non-hemolysis image dataset;

[0097] Adjusting the SVM neural network model according to the hemolytic image data set and the non-hemolytic image data set to obtain a sample hemolytic detection model;

[0098] Obtaining the image data of the thrombus sample to be tested by the user;

[0099] The sample hemolysis detection model is used to analyze and process the image data of the user's thrombus sample to be tested to obtain a hemolysis recognition result.

[0100] In this example, by acquiring a hemolytic image dataset and a non-hemolytic image dataset, a sample hemolytic detection model is trained using a SVM neural network model to obtain a trained sample hemolytic detection model, so as to identify the hemolytic degree of the sample to be tested, and then obtain a hemolytic identification result. The hemolytic identification result can be used to eliminate the interference of the sample hemolysis phenomenon on the sample detection by the thromboelastograph. Since the hemolytic phenomenon is the largest non-physiological interference privacy, the technical problem of inaccurate thrombus detection results caused by interference from non-physiological factors when using a thromboelastograph for thrombus detection in the prior art is solved.

[0101] The above mainly introduces the scheme of the embodiment of the present application from the perspective of the execution process on the method side. It is understandable that in order to realize the above functions, the terminal includes a hardware structure and / or software module corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0102] The embodiment of the present application can divide the electronic device into functional units according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0103] In line with the above, see Figure 5 , Figure 5 The present application provides a schematic diagram of a structure for hemolytic identification based on a thrombus sample image. Figure 5 As shown, the device comprises:

[0104] A first acquisition unit 001 is used to acquire a hemolysis image data set and a non-hemolysis image data set;

[0105] The second acquisition unit 002 is used to acquire the thrombus sample data to be tested by the user;

[0106] The first processing unit 003 is used to adjust the SVM neural network model according to the hemolysis image data set and the non-hemolysis image data set to obtain a sample hemolysis detection model;

[0107] The second processing unit 004 is used to analyze and process the user's thrombus sample data to be tested by using the sample hemolysis detection model to obtain a hemolysis identification result.

[0108] An embodiment of the present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program enables a computer to execute part or all of the steps of any one of the hemolytic identification methods based on thrombus sample images recorded in the above method embodiments.

[0109] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0110] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0111] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be electrical or other forms.

[0112] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0113] In addition, the functional units in the various embodiments of the application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software program modules.

[0114] If the integrated unit is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, disk or optical disk and other media that can store program codes.

[0115] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, which can include: a flash drive, a read-only memory, a random access memory, a magnetic disk or an optical disk, etc.

[0116] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A hemolytic identification method based on thrombus sample images, characterized in that: include: Acquire a hemolysis image dataset and a non-hemolysis image dataset; Adjusting the SVM neural network model according to the hemolytic image data set and the non-hemolytic image data set to obtain a sample hemolytic detection model; Obtaining the image data of the thrombus sample to be tested by the user; The sample hemolysis detection model is used to analyze and process the image data of the user's thrombus sample to be tested to obtain a hemolysis recognition result.

2. The hemolytic identification method based on thrombus sample image according to claim 1, characterized in that: The SVM neural network model is adjusted according to the hemolytic image data set and the non-hemolytic image data set to obtain a sample hemolytic detection model, including: Merging the hemolysis image dataset and the non-hemolysis image dataset to obtain a total hemolysis image dataset; Performing data preprocessing on the total hemolysis image data set to obtain a hemolysis feature data set; Dividing the hemolysis characteristic data set into a training set, a validation set and a test set; The sample hemolysis detection model is trained using a SVM neural network model according to the training set, validation set and test set until the model converges to obtain a trained sample hemolysis detection model.

3. The hemolytic identification method based on thrombus sample image according to claim 2, characterized in that: The total hemolysis image data set is preprocessed to obtain a hemolysis feature data set, including: Performing data cleaning on the total hemolysis image data set to obtain a cleaned total hemolysis image data set; Hemolysis characteristic data is extracted from the cleaned total hemolysis image data set to obtain a hemolysis characteristic data set.

4. The hemolytic identification method based on thrombus sample image according to claim 3, characterized in that: The total hemolysis image data set is cleaned to obtain a cleaned total hemolysis image data set, including: Identifying missing values ​​and outliers in the total hemolysis image dataset; Interpolation method is used to correct the missing data values ​​and the abnormal values ​​are directly eliminated; The corrected data are used to replace the missing data values ​​in the original data set to obtain the cleaned total hemolysis image data set.

5. The hemolytic identification method based on thrombus sample image according to claim 3, characterized in that: Extracting hemolysis characteristic data from the cleaned total hemolysis image data set to obtain a hemolysis characteristic data set, including: Extracting hemolysis color feature data from the cleaned total hemolysis image data set using a global color histogram method; Extracting hemolysis texture feature data from the cleaned total hemolysis image data set using a gray level co-occurrence matrix method; Using the Canny edge detection method to extract the morphological characteristic data of hemolytic cells from the total hemolytic image data set after cleaning; The hemolysis color feature data, the hemolysis texture feature data and the hemolysis cell morphology feature data are combined to obtain a hemolysis feature data set.

6. A thrombus detection device, characterized in that: The device comprises: A first acquisition unit is used to acquire a hemolysis image data set and a non-hemolysis image data set; The second acquisition unit is used to acquire the thrombus sample data to be tested by the user; A first processing unit, configured to adjust the SVM neural network model according to the hemolysis image data set and the non-hemolysis image data set to obtain a sample hemolysis detection model; The second processing unit is used to analyze and process the user's thrombus sample data to be tested by using the sample hemolysis detection model to obtain a hemolysis identification result.

7. An electronic device, characterized in that: The invention comprises a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the hemolytic identification method based on the thrombus sample image according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the hemolytic identification method based on a thrombus sample image according to any one of claims 1 to 5.

9. A computer program product, characterized in that The computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program enables a computer to execute part or all of the steps of the hemolytic identification method based on thrombus sample images according to any one of claims 1 to 5.

Images