Abnormal blood coagulation curve identification method and blood coagulation detection equipment

Abstract

The invention discloses a blood coagulation abnormal curve identification method and blood coagulation detection equipment, and is applied to the blood coagulation detection equipment, and the method comprises the following steps: obtaining to-be-detected first blood coagulation data; according to the first blood coagulation data and a first blood coagulation curve recognition model, a first recognition result corresponding to the first blood coagulation data is obtained, and the first recognition result comprises a first blood coagulation curve type; first feedback information of the user based on the first recognition result is obtained, the first feedback information comprises a second blood coagulation curve type obtained through updating, and the second blood coagulation curve type is different from the first blood coagulation curve type; according to the first feedback information and the first blood coagulation data, a second blood coagulation curve recognition model is determined, and the second blood coagulation curve recognition model is applied to the recognition process of the second blood coagulation data. According to the invention, automatic recognition of the abnormal blood coagulation curve can be realized, and the recognition accuracy can be improved.

Description

Technical Field

[0001] This application relates to the field of detection technology, and specifically relates to a method for identifying abnormal coagulation curves and a coagulation detection device. Background Art

[0002] The coagulation curve refers to the relevant curve reflecting the coagulation process in coagulation function detection. Currently, for the coagulation curves representing abnormal coagulation in the coagulation curve, the method of manually setting the detection threshold is usually used for identification and judgment, that is, when the data represented by the coagulation curve exceeds the set detection threshold, the coagulation curve is marked as an abnormal coagulation curve. The accuracy of identifying abnormal coagulation curves in this way is relatively poor. Summary of the Invention

[0003] Embodiments of this application provide a method for identifying abnormal coagulation curves and a coagulation detection device, in order to achieve automatic identification of abnormal coagulation curves and improve the accuracy of identifying abnormal coagulation curves.

[0004] In a first aspect, embodiments of this application provide a method for identifying abnormal coagulation curves, which is applied to a coagulation detection device. The method includes:

[0005] Obtain first coagulation data to be detected;

[0006] According to the first coagulation data and the first coagulation curve recognition model, obtain a first recognition result corresponding to the first coagulation data, where the first recognition result includes the type of the first coagulation curve;

[0007] Obtain first feedback information of the user based on the first recognition result, where the first feedback information includes an updated second coagulation curve type, and the second coagulation curve type is different from the first coagulation curve type;

[0008] According to the first feedback information and the first coagulation data, determine a second coagulation curve recognition model, so as to apply the second coagulation curve recognition model to the recognition process of second coagulation data.

[0009] Combined with the first aspect, in a possible embodiment, the step of determining a second coagulation curve recognition model according to the first feedback information and the first coagulation data, so as to apply the second coagulation curve recognition model to the recognition process of second coagulation data includes:

[0010] Store the second coagulation curve type and the first coagulation data in the first user dataset;

[0011] Generate the second blood coagulation curve recognition model according to the information in the first user dataset, so that the second blood coagulation curve recognition model and the first blood coagulation curve recognition model are integrated and applied to the recognition process of the second blood coagulation data.

[0012] Combined with the first aspect, in a possible embodiment, during the recognition process of the second blood coagulation data, the method further includes:

[0013] Obtain the second blood coagulation data, and determine a first output result according to the second blood coagulation data and the first blood coagulation curve recognition model;

[0014] Determine a second output result according to the second blood coagulation data and the second blood coagulation curve recognition model;

[0015] Perform an integration operation on the first output result and the second output result to determine a second recognition result corresponding to the second blood coagulation data.

[0016] Combined with the first aspect, in a possible embodiment, the performing an integration operation on the first output result and the second output result to determine a second recognition result corresponding to the second blood coagulation data includes:

[0017] Average the first output result and the second output result to determine the second recognition result; or,

[0018] Perform a weighted average on the first output result and the second output result to determine the second recognition result.

[0019] Combined with the first aspect, in a possible embodiment, the determining the second blood coagulation curve recognition model according to the first feedback information and the first blood coagulation data includes:

[0020] Store the second blood coagulation curve type and the first blood coagulation data in the first user dataset;

[0021] Adjust the weight parameters in at least one hidden layer in the first blood coagulation curve recognition model according to the information in the first user dataset to generate the second blood coagulation curve recognition model.

[0022] Combined with the first aspect, in a possible embodiment, the first blood coagulation curve recognition model includes an input layer, a plurality of hidden layers, and an output layer;

[0023] The adjusting the weight parameters in at least one hidden layer in the first blood coagulation curve recognition model according to the information in the first user dataset to generate the second blood coagulation curve recognition model includes:

[0024] Adjust at least the weight parameters in the hidden layer closest to the output layer in the first blood coagulation curve recognition model according to the information in the first user dataset, so as to generate the second blood coagulation curve recognition model.

[0025] In combination with the first aspect, in a possible embodiment, the determining the second blood coagulation curve recognition model according to the first feedback information and the first blood coagulation data further includes:

[0026] Obtain a second user dataset, where the second user dataset includes third blood coagulation data and its corresponding blood coagulation curve type, and the blood coagulation curve type in the recognition result corresponding to the third blood coagulation data is the same as the blood coagulation curve type in the feedback information corresponding to the third blood coagulation data;

[0027] Adjust the weight parameters in at least one hidden layer in the first blood coagulation curve recognition model according to the information in the first user dataset and the second user dataset, so as to generate the second blood coagulation curve recognition model.

[0028] In combination with the first aspect, in a possible embodiment, during the recognition process of the second blood coagulation data, the method further includes:

[0029] Obtain the second blood coagulation data, and determine a second recognition result corresponding to the second blood coagulation data according to the second blood coagulation data and the second blood coagulation curve recognition model.

[0030] In combination with the first aspect, in a possible embodiment, after determining the second recognition result corresponding to the second blood coagulation data, the method further includes:

[0031] Obtain second feedback information of the user based on the second recognition result, where the second feedback information includes the trigger status of the reset function provided by the blood coagulation detection device, and the trigger status is a triggered status or an untriggered status;

[0032] When the trigger status is the triggered status, reset the second blood coagulation curve recognition model to the first blood coagulation curve recognition model, so as to apply the first blood coagulation curve recognition model to the recognition process of the fourth blood coagulation data.

[0033] In a second aspect, an embodiment of the present application provides a device for recognizing abnormal blood coagulation curves, and the device for recognizing abnormal blood coagulation curves includes:

[0034] A first obtaining unit, configured to obtain first blood coagulation data to be detected;

[0035] A first determination unit, configured to obtain a first recognition result corresponding to the first coagulation data according to the first coagulation data and a first coagulation curve recognition model, where the first recognition result includes a first coagulation curve type;

[0036] A second acquisition unit, configured to acquire first feedback information of a user based on the first recognition result, where the first feedback information includes an updated second coagulation curve type, and the second coagulation curve type is different from the first coagulation curve type;

[0037] A second determination unit, configured to determine a second coagulation curve recognition model according to the first feedback information and the first coagulation data, so as to apply the second coagulation curve recognition model to the recognition process of second coagulation data.

[0038] In a third aspect, an embodiment of the present application provides a coagulation detection device, including a processor, a memory, a communication interface, and one or more programs, where the one or more programs are stored in the memory and are configured to be executed by the processor, and the programs include instructions for executing the steps in the first aspect of the embodiment of the present application.

[0039] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, storing a computer program for electronic data exchange, where the computer program enables a computer to execute some or all of the steps described in the first aspect of this embodiment.

[0040] In a fifth aspect, the present application provides a computer program product, where the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute some or all of the steps described in the first aspect of the present application. The computer program product may be a software installation package.

[0041] It can be seen that in the embodiments of the present application, by obtaining the first coagulation data to be detected; and according to the first coagulation data and the first coagulation curve recognition model, obtaining the first recognition result corresponding to the first coagulation data, where the first recognition result includes the first coagulation curve type; then obtaining the first feedback information of the user based on the first recognition result, where the first feedback information includes the updated second coagulation curve type, and the second coagulation curve type is different from the first coagulation curve type; then determining the second coagulation curve recognition model according to the first feedback information and the first coagulation data, so as to apply the second coagulation curve recognition model to the recognition process of the second coagulation data. The embodiments of the present application can recognize the first coagulation curve through the first coagulation curve recognition model to obtain the first recognition result, and feedback the first recognition result to the user to obtain the first feedback information of the user based on the first recognition result, so as to correct the first coagulation curve type of the first coagulation curve through the second coagulation curve type carried by the first feedback information, and then determine the second coagulation curve recognition model according to the second coagulation curve type and the first coagulation data, and apply the second coagulation curve recognition model to the subsequent recognition process of the second coagulation data. In this way, the user does not need to perform cumbersome parameter settings, and can not only realize the automatic recognition of abnormal coagulation curves, but also make the detection of coagulation data with abnormal coagulation more in line with the user's detection habits, avoiding the influence of differences such as judgment habits corresponding to different regions, detection reagents used, and operator operation habits on the detection results, thereby continuously improving the detection accuracy. Moreover, the embodiments of the present application can recognize the coagulation curve type corresponding to the coagulation data through the coagulation curve recognition model, which is also beneficial to improving the recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0043] Figure 1 It is a schematic diagram of an application scenario of a method for recognizing abnormal coagulation curves provided by an embodiment of the present application;

[0044] Figure 2 It is a schematic diagram of the structure of a coagulation detection device provided by an embodiment of the present application;

[0045] Figure 3 It is a schematic diagram of the structure of another coagulation detection device provided by an embodiment of the present application;

[0046] Figure 4 It is a schematic flowchart of a method for recognizing abnormal coagulation curves provided by an embodiment of the present application;

[0047] Figure 5 It is a schematic diagram of the process for identifying abnormal coagulation curves provided by an embodiment of the present application;

[0048] Figure 6 It is a schematic diagram of the first interaction interface for obtaining the first feedback information provided by an embodiment of the present application;

[0049] Figure 7 It is a schematic diagram of the second interaction interface for prompting parameter abnormalities provided by an embodiment of the present application;

[0050] Figure 8 It is another schematic diagram of the process for identifying abnormal coagulation curves provided by an embodiment of the present application;

[0051] Figure 9 It is a block diagram of the functional units of a device for identifying abnormal coagulation curves provided by an embodiment of the present application;

[0052] Figure 10 It is a block diagram of the functional units of another device for identifying abnormal coagulation curves provided by an embodiment of the present application. Detailed implementation manners

[0053] In order to enable those skilled in the art of the present technology to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0054] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. 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 may optionally further include steps or units not listed, or may optionally further include other steps or units inherent to these processes, methods, products or devices.

[0055] Referring to "embodiment" in this article means that the specific features, structures or characteristics described in connection with the embodiment may be included in at least one embodiment of the present application. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0056] The embodiments of the present application will be described below with reference to the accompanying drawings.

[0057] Please refer to Figure 1 , Figure 1 , which is a schematic diagram of an application scenario of a method for identifying coagulation abnormality curves provided by an embodiment of the present application. In this application scenario, there are a coagulation detection device 100 and a user 200. The user 200 can perform human-computer interaction with the coagulation detection device 100 to implement the method for identifying coagulation abnormality curves of the present application. Among them, the coagulation examination device is an instrument for detecting indexes related to human blood coagulation, mainly used for evaluating coagulation function, diagnosing hemorrhagic or thrombotic diseases, etc. Exemplarily, in a clinical scenario, an operator can add a detection reagent to a blood sample to be detected to prepare a detection sample, and analyze the transmitted light or scattered light obtained by irradiating the detection sample with light through methods such as the clotting method, the synthetic substrate method, the immunoturbidimetry method, or the agglutination method, so as to realize the relevant analysis of blood coagulation ability.

[0058] In specific implementation, the coagulation detection device 100 can be configured with a main control mechanism, such as a processor. This main control mechanism can be used to execute the steps in the method for identifying coagulation abnormality curves provided by the present application. In some embodiments, the coagulation detection device 100 can also be configured with or externally connected to a display mechanism and an operation mechanism. The display mechanism is used to display contents such as human-computer interaction information and identification results for the user 200 to view. The user 200 can perform human-computer interaction operations through the operation mechanism. Specifically, the display mechanism and the operation mechanism can be configured independently. For example, the coagulation detection device 100 is externally connected to a display, a mouse, and / or a keyboard, etc. Alternatively, the display mechanism and the operation mechanism can also be integrated into a single structure. For example, the coagulation detection device 100 can be externally connected to a touch display, etc.

[0059] Specifically, please refer to Figure 2 , Figure 2 , which is a schematic structural diagram of a coagulation detection device provided by an embodiment of the present application. As shown in Figure 2 , in this example, the coagulation detection device 100 can include a processor 110, a memory 120, a communication interface 130, and one or more programs 121. Among them, the one or more programs 121 are stored in the above-mentioned memory 120 and are configured to be executed by the above-mentioned processor 110. The one or more programs 121 include instructions for executing any step in the following method embodiments. In specific implementation, the processor 110 is used to execute any step in the following method embodiments, and when performing data transmission such as sending and receiving, it can optionally call the communication interface 130 to complete the corresponding operation.

[0060] Alternatively, in specific implementation, the coagulation detection device 100 can also be configured with at least a main control structure and a display structure. Please refer toFigure 3 , Figure 3 is a schematic structural diagram of another blood coagulation detection device provided by an embodiment of the present application. As shown in Figure 3 , in this example, the blood coagulation detection device 100 includes a processor 110, a memory 120, a communication interface 130, a display 140, and one or more programs 121. Among them, the one or more programs 121 are stored in the above-mentioned memory 120 and are configured to be executed by the above-mentioned processor 110. The one or more programs 121 include instructions for executing any step in the following method embodiments. In a specific implementation, the processor 120 is used to execute any step in the following method embodiments, and when performing data transmission processes such as receiving and sending, the communication interface 130 can be selectively called to complete the corresponding operations. When performing operations such as displaying, it can be completed by the display 140. No further limitations are imposed on the specific structure of the blood coagulation detection device 100 here.

[0061] In a specific implementation, when the user has a need to detect a blood sample, a blood coagulation detection request can be initiated to instruct the blood coagulation detection device to execute the steps in the blood coagulation abnormality curve recognition method described below. In some embodiments, the user can input blood coagulation data to the blood coagulation detection device. The blood coagulation detection device can generate a blood coagulation detection request carrying the blood coagulation data according to the blood coagulation data, and execute the steps in the blood coagulation abnormality curve recognition method described below according to the blood coagulation detection device. Alternatively, in some embodiments, the application scenario architecture of the blood coagulation abnormality curve recognition method may further include a user device or a user device and a server. When the user has a need to detect a blood sample, the user can input blood coagulation data to the user device to generate a blood coagulation detection request carrying the blood coagulation data, and directly send the blood coagulation detection request to the blood coagulation detection device, or directly upload the blood coagulation detection request to the server and send it to the blood coagulation detection device through the server. Thus, the blood coagulation detection device is instructed to execute the steps in the blood coagulation abnormality curve recognition method described below. Among them, the user device can be other devices related to blood sample detection or terminal devices used by the user, etc., and no further limitations are imposed here.

[0062] Please refer to Figure 4 , Figure 4 is a schematic flowchart of a blood coagulation abnormality curve recognition method provided by an embodiment of the present application. As shown in Figure 4 , the blood coagulation abnormality curve recognition method includes the following steps:

[0063] S410, obtain first blood coagulation data to be detected.

[0064] Among them, the coagulation data is data associated with the coagulation curve, which includes reaction time, detection indicators, and other contents. The first coagulation data obtained by the coagulation detection device can be presented in the form of a chart image or a data set. Among them, the chart image is used to characterize the relationship between the reaction time (such as the content shown on the abscissa in the chart image shown in Figures 5 to 7 ), and the detection indicator (such as the content shown on the ordinate in the chart image shown in Figures 5 to 7 ). The data set includes multiple sets of data corresponding to the reaction time and the detection indicator. Exemplarily, the data in the data set can be presented in the form of structured text, where the structured text is a data representation form with a specific structure and format. The first coagulation data is used to characterize the change of the target value over time during the blood coagulation process. Among them, the target value can be determined according to the detection method, detection indicators, and other contents. For example, in the prothrombin time (PT) test or the activated partial thromboplastin time (APTT) test, the target value can be the light signal intensity or absorbance of blood coagulation, etc. No further restrictions are imposed on the content corresponding to the target value here.

[0065] S420, according to the first coagulation data and the first coagulation curve recognition model, obtain the first recognition result corresponding to the first coagulation data.

[0066] Among them, the first recognition result includes the first coagulation curve type. The coagulation curve type includes a normal curve and an abnormal curve. Among them, the abnormal curve can specifically be no coagulation, mild coagulation, sudden increase in the curve, sudden drop in the curve, noise, plateau curve, slow reaction, drift, prozone, postzone, gentle curve, or stepped curve, etc. Optionally, the coagulation curve type is used to characterize that the first coagulation data is a normal curve, or the specific type corresponding to the first coagulation data being an abnormal curve.

[0067] Among them, the first coagulation curve recognition model is a pre-trained model, or a model updated according to the previous coagulation data. Among them, the previous one refers to the first or the Nth one arranged before the first coagulation data, and N is a positive integer greater than or equal to 2.

[0068] In a specific implementation, the first coagulation data can be used as input data and input into the first coagulation curve recognition model for processing, and the first recognition result can be used as the output data of the first coagulation curve recognition model. Exemplarily, referring to Figure 5 , the first coagulation data can be input into the first coagulation curve recognition model in the form of data or an icon image. After the first coagulation curve recognition model processes the first coagulation data, it can identify the first coagulation curve type corresponding to the first coagulation data from multiple coagulation curve types, that is, Figure 5 the stepped curve identified in

[0069] Specifically, when using the first blood coagulation data as the input data for the first blood coagulation curve recognition model, the following steps need to be performed: Determine the actual reaction duration corresponding to the first blood coagulation data; Compare the actual reaction duration with the preset reaction duration; When the actual reaction duration is greater than or equal to the preset reaction duration, intercept the data within the preset reaction duration from the first blood coagulation data, and determine this data as the input data for the first blood coagulation curve recognition model; When the actual reaction duration is less than the preset reaction duration, fill in the preset data at each time node within the time difference after the actual reaction of the first blood coagulation data ends, and determine the first blood coagulation data and the filled preset data as the input data for the first blood coagulation curve recognition model. The time difference is the difference between the preset reaction duration and the actual reaction duration.

[0070] Among them, the actual reaction duration refers to the time length of the reaction time recorded by the first blood coagulation data. The preset reaction duration refers to the reaction duration preset and adapted to the first blood coagulation curve recognition model. Each time node refers to the node corresponding to each unit time within the duration corresponding to the difference. The preset data can be set according to requirements. For example, the preset data can be 0. Among them, when the first blood coagulation data is an icon image, the first blood coagulation curve recognition model can directly perform image recognition to obtain the actual reaction duration. If the first blood coagulation data is data in a dataset, the first blood coagulation curve recognition model can sort the data corresponding to the duration in the dataset from large to small to determine the actual reaction duration.

[0071] When the actual reaction duration is equal to the preset reaction duration, the first blood coagulation curve recognition model can directly process the first blood coagulation data. When the actual reaction duration is greater than the preset reaction duration, data with the same time length as the preset reaction duration can be intercepted from the first blood coagulation data. For example, data within the first preset reaction duration starting from the start reaction moment can be intercepted from the first blood coagulation data. And the intercepted data is processed by the first blood coagulation curve recognition model. When the actual reaction duration is less than the preset reaction duration, first determine the duration corresponding to the difference between the actual reaction duration and the preset reaction duration, and configure and fill in the preset data at each time node within the duration corresponding to the difference after the actual reaction time ends, so that the reaction duration of the filled total data (i.e., the first blood coagulation data and all filled data) is equal to the preset reaction duration, and then the total data is processed by the first blood coagulation curve recognition model.

[0072] It can be seen that in this example, the input of the first blood coagulation curve recognition model is curve data with a fixed time length. By means of data interception or filling in preset data, the first blood coagulation data to be processed can meet the time length requirements of the first blood coagulation curve recognition model, and the data used for processing has coherence and integrity, which is conducive to improving the reliability of the output result of the first blood coagulation curve recognition model.

[0073] S430, obtain first feedback information of the user based on the first recognition result.

[0074] Among them, the first feedback information is used to characterize the user's judgment on the accuracy of the first recognition result.

[0075] In some embodiments, the first feedback information includes an updated second blood coagulation curve type, which is different from the first blood coagulation curve type. The second blood coagulation curve type is the content after the user corrects the first blood coagulation curve type in the first recognition result. For example, when the first blood coagulation curve type is a normal curve, the second blood coagulation curve type can be the specific type corresponding to an abnormal curve. Or, when the first blood coagulation curve type is the specific type corresponding to an abnormal curve, the second blood coagulation curve type can be a type other than the first blood coagulation curve type corresponding to a normal curve or an abnormal curve.

[0076] In a specific implementation, after the blood coagulation detection device processes the first blood coagulation data to obtain the first recognition result, it can feedback the first recognition result to the user so that the user can learn about the situation of the first blood coagulation data detection, and the user can confirm whether the first recognition result is accurate. For example, the first recognition result can be presented to the user in the form of including the first blood coagulation data and the content of the first blood coagulation curve type. Then the user can upload the confirmation result to the blood coagulation detection device so that the blood coagulation detection device can obtain the first feedback information when the detected first blood coagulation curve type is incorrect. And when the first blood coagulation curve type detected by the blood coagulation detection device is correct, it can obtain the third feedback information indicating that the first blood coagulation curve type is correct.

[0077] Specifically, when the first blood coagulation curve type characterizes that the first blood coagulation data is a certain abnormal curve or a normal curve, the blood coagulation detection device can display the first recognition result through a display for the user to check. If the user confirms that the first blood coagulation curve type is incorrect during the check, the first blood coagulation curve type can be corrected, and the first feedback information including the corrected second blood coagulation curve type can be uploaded. Specifically, the user can upload the second blood coagulation curve type by text input or by selecting and confirming a preset option. Exemplarily, when the blood coagulation detection device identifies that the first blood coagulation curve type is an abnormal curve such as "sudden drop of the curve", it can be displayed through the display as Figure 6The first interaction interface described above. The first interaction interface displays the currently recognized coagulation data (i.e., the first coagulation data), the current recognition result (i.e., the first coagulation curve type), and multiple selectable preset options, etc. When the user confirms that the coagulation data and the current recognition result do not match according to the content displayed on the first interaction interface, the user can select a preset option that matches the coagulation data from the multiple preset options, and by triggering the "OK" control, generate and upload the first feedback information carrying the second coagulation curve type according to the selected preset option, so that the coagulation agent detection device can obtain the first feedback information.

[0078] In some embodiments, when the first coagulation curve type represents that the first coagulation data is a certain abnormal curve or normal curve, if the user confirms that the first coagulation curve type is correct during the inspection, the third feedback information carrying the third coagulation curve type, which is the same as the first coagulation curve type, can also be generated and uploaded in the above-mentioned manner to indicate that the first coagulation curve type is correct. Or, in some embodiments, if the coagulation detection device does not obtain the first feedback information or other information within the preset time, the third feedback information carrying the third coagulation curve type can be default obtained.

[0079] In specific implementation, while the coagulation detection device performs curve abnormality recognition on the first coagulation data, it also verifies the coagulation index parameters corresponding to the first coagulation data. For example, by comparing the detected result value of the coagulation index parameter with the reference value range corresponding to the coagulation index parameter, it is determined whether the coagulation index parameter is within the normal range. Specifically, when the first coagulation curve type represents that the first coagulation data is a normal curve, if it is detected that the coagulation index parameter does not match its corresponding reference value range, that is, the coagulation index parameter is not within the normal range, the coagulation detection device will display an abnormal prompt message through the display to remind the user to check the first recognition result. Exemplarily, see Figure 7 , when it is detected that the coagulation index parameter does not match its corresponding reference value range, even if the first coagulation curve type represents that the first coagulation data is a normal curve, the display can also display an abnormal prompt message in the form of a pop-up window or the like. By verifying the coagulation index parameters while performing coagulation curve abnormality recognition, the coagulation detection device is beneficial to improving the accuracy of recognizing abnormal situations in coagulation detection.

[0080] S440, determine a second coagulation curve recognition model according to the first feedback information and the first coagulation data, so as to apply the second coagulation curve recognition model to the recognition process of the second coagulation data.

[0081] Among them, the second blood coagulation curve recognition model is a model for recognizing second blood coagulation data. The second blood coagulation data refers to the blood coagulation data to be detected arranged after the first blood coagulation data. Specifically, the second blood coagulation data can be the first or the Mth blood coagulation data to be detected arranged after the first blood coagulation data, where M is a positive integer greater than or equal to 2.

[0082] Specifically, the recognition process of the second blood coagulation data can refer to the recognition process of the above-mentioned first blood coagulation data, and no further limitation is made here.

[0083] In some embodiments, the blood coagulation detection device may be configured with an update switch, and the update switch is used to control the start and stop of the update operation of the first blood coagulation curve recognition model. If the update switch is in the on state, after obtaining the first feedback information, the second blood coagulation curve recognition model can be determined according to the first feedback information and the first blood coagulation data, and the second blood coagulation curve recognition model is applied to the recognition process of the second blood coagulation data. If the update switch is in the off state, after obtaining the first feedback information, the blood coagulation detection device will not automatically determine the second blood coagulation curve recognition model, and the first blood coagulation curve recognition model will still be applied to the recognition process of the second blood coagulation data in the future. This is beneficial to improving the controllability of the recognition of abnormal blood coagulation curves and is beneficial to differentially meeting the needs of different users.

[0084] It can be seen that in the embodiments of the present application, by obtaining the first coagulation data to be detected; and according to the first coagulation data and the first coagulation curve recognition model, obtaining the first recognition result corresponding to the first coagulation data, wherein the first recognition result includes the type of the first coagulation curve; then obtaining the first feedback information of the user based on the first recognition result, wherein the first feedback information includes the updated second coagulation curve type, and the second coagulation curve type is different from the first coagulation curve type; then according to the first feedback information and the first coagulation data, determining the second coagulation curve recognition model to apply the second coagulation curve recognition model to the recognition process of the second coagulation data. The embodiments of the present application can recognize the first coagulation data through the first coagulation curve recognition model to obtain the first recognition result, and feedback the first recognition result to the user to obtain the first feedback information of the user based on the first recognition result, so as to correct the first coagulation curve type of the first coagulation data through the second coagulation curve type carried in the first feedback information, and then determine the second coagulation curve recognition model according to the second coagulation curve type and the first coagulation data, and apply the second coagulation curve recognition model to the subsequent recognition process of the second coagulation data. In this way, the user does not need to perform cumbersome parameter settings, and can automatically recognize abnormal coagulation curves while making the detection of coagulation data with abnormal coagulation more in line with the user's detection habits, avoiding the influence of differences such as judgment habits corresponding to different regions, detection reagents used, and operator operation habits on the detection results, thereby continuously improving the accuracy of detection. Moreover, the embodiments of the present application can recognize the type of the coagulation curve corresponding to the coagulation data through the coagulation curve recognition model, which is also beneficial to improving the accuracy of recognition.

[0085] In some embodiments, the second coagulation curve recognition model can be integrated with the first coagulation curve recognition model for application. Determining the second coagulation curve recognition model according to the first feedback information and the first coagulation data to apply the second coagulation curve recognition model to the recognition process of the second coagulation data includes: storing the second coagulation curve type and the first coagulation data in the first user data set; generating the second coagulation curve recognition model according to the information in the first user data set, so that the second coagulation curve recognition model and the first coagulation curve recognition model are integrated for application in the recognition process of the second coagulation data.

[0086] Among them, the first user data set is used to store relevant data when the user corrects the recognition result of the first coagulation curve recognition model. Specifically, the first user data set includes at least one set of coagulation data and the updated coagulation curve type corresponding to the coagulation data.

[0087] In some embodiments, after the coagulation detection device obtains the first feedback information, it may store the currently identified first coagulation data and the second coagulation curve type in the first feedback information in the first user dataset, and establish a correspondence between the first coagulation data and the second coagulation curve type. Then, a second coagulation curve recognition model is generated according to the information in the first user dataset. Since the generated second coagulation curve recognition model is more suitable for the judgment habits corresponding to the user's region, the detection reagents used, and the operation habits of the operators, etc., in the subsequent recognition process of the second coagulation data, the second coagulation curve recognition model and the first coagulation curve recognition model can be integrated for use to improve the accuracy and reliability of recognition.

[0088] In some embodiments, the coagulation detection device can establish a connection with the cloud. After storing the second coagulation curve type and the first coagulation data in the first user dataset, the first user dataset is uploaded to the cloud to train and generate a lightweight model, that is, the second coagulation curve recognition model, in the cloud. Then, the second coagulation curve recognition model sent by the cloud is obtained, so that the first coagulation curve recognition model and the second coagulation curve recognition model are integrated and applied to the recognition process of the second coagulation data. In this way, the data processing volume of the coagulation detection device can be reduced, the configuration of the coagulation detection device can be simplified, and costs can be saved.

[0089] Specifically, when integrating and using the first coagulation curve recognition model and the second coagulation curve recognition model, fusion can be performed based on the network structures of the first coagulation curve recognition model and the second coagulation curve recognition model to obtain a comprehensive coagulation curve recognition model. After the second coagulation data is processed by this comprehensive coagulation curve recognition model, the corresponding second recognition result can be obtained.

[0090] In some embodiments, with reference to Figure 8 , for the recognition of the second coagulation data, it specifically includes: obtaining the second coagulation data, determining a first output result according to the second coagulation data and the first coagulation curve recognition model; determining a second output result according to the second coagulation data and the second coagulation curve recognition model; performing an integration operation on the first output result and the second output result to determine the second recognition result corresponding to the second coagulation data.

[0091] Among them, both the first output result and the second output result are used to characterize the adaptability of the second coagulation data to each coagulation curve type. For example, this adaptability can be represented by parameters such as probability or score.

[0092] In a specific implementation, when identifying the second coagulation data, the second coagulation data can be input into the first coagulation curve recognition model and the second coagulation curve recognition model respectively to obtain a first output result and a second output result correspondingly. Then, the first output result and the second data result are integrated and calculated, so as to realize the integrated application of the first coagulation curve recognition model and the second coagulation curve recognition model, and obtain a second recognition result corresponding to the second coagulation data. This is beneficial to reducing the calculation cost and the risk of overfitting, and improving the output efficiency of the second recognition result.

[0093] In some embodiments, when integrating and calculating the first output result and the second output result, a new model can be trained based on the second coagulation data, the first output result and the second output result for quadratic regression calculation, and the prediction result of the new model is determined as the second recognition result corresponding to the second coagulation data.

[0094] In some embodiments, integrating and calculating the first output result and the second output result to determine a second recognition result corresponding to the second coagulation data includes: averaging the first output result and the second output result to determine the second recognition result; or performing a weighted average on the first output result and the second output result to determine the second recognition result.

[0095] Specifically, the fitness degrees corresponding to the same type of coagulation curve in the first output result and the second output result can be averaged to obtain the comprehensive fitness degree after integration of each type of coagulation curve. Then, the comprehensive fitness degrees of each type of coagulation curve can be compared, and the type of coagulation curve with the highest comprehensive fitness degree is determined as the second recognition result.

[0096] Exemplarily, taking the coagulation curve types that can be recognized by the first coagulation curve recognition model and the second coagulation curve recognition model including a flat curve, a step curve, a sudden increase in the curve, and a sudden drop in the curve as an example. The first output result can be: the fitness degree of the flat curve is 0.15, the fitness degree of the step curve is 0.4, the fitness degree of the sudden increase in the curve is 0.15, and the fitness degree of the sudden drop in the curve is 0.3. The second output result can be: the fitness degree of the flat curve is 0.1, the fitness degree of the step curve is 0.5, the fitness degree of the sudden increase in the curve is 0.1, and the fitness degree of the sudden drop in the curve is 0.3. The comprehensive fitness degrees after averaging the first output result and the second output result are respectively: flat curve 0.125, step curve 0.45, sudden increase in the curve 0.125, and sudden drop in the curve 0.3. Among them, the comprehensive fitness degree of the step curve is the largest, so the step curve is determined as the second recognition result. This is beneficial to reducing the difficulty of the integrated application of the first coagulation curve recognition model and the second coagulation curve recognition model, and improving the efficiency of the application.

[0097] Alternatively, in a specific implementation, the integrated operation performed on the first output result and the second output result may be: performing a weighted average on the first output result and the second output result. Specifically, a first preset weight may be configured for the first blood coagulation curve recognition model, and a second preset weight may be configured for the second blood coagulation curve recognition model, so as to perform a weighted average calculation on the fitness degrees corresponding to the same type of blood coagulation curve in the first output result and the second output result according to the first preset weight and the second preset weight, thereby obtaining the comprehensive fitness degree after integration of each type of blood coagulation curve. Then, the comprehensive fitness degrees of each type of blood coagulation curve may be compared, and the type of blood coagulation curve with the highest comprehensive fitness degree may be determined as the second recognition result.

[0098] Exemplarily, taking the example that the types of blood coagulation curves recognizable by the first blood coagulation curve recognition model and the second blood coagulation curve recognition model include a flat curve, a stepped curve, a sudden increase in the curve, and a sudden drop in the curve, and the first preset weight is 0.7 and the second preset weight is 0.3. When the first output result and the second output result are the data in the above example, the comprehensive fitness degrees after performing a weighted average calculation on the first output result and the second output result are respectively: flat curve 0.0675, stepped curve 0.215, sudden increase in the curve 0.0675, and sudden drop in the curve 0.15. Among them, the comprehensive fitness degree of the stepped curve is the largest, so the stepped curve is determined as the second recognition result. In this way, the weight ratio of the first output result and the second output result can be adjusted differentially, improving the matching degree between the second recognition result and the user.

[0099] It can be seen that in this example, by using the information in the first user dataset to generate the second blood coagulation curve recognition model, the generated second blood coagulation curve recognition model can be more adapted to the judgment habits corresponding to the user's region, the detection reagents used, and the operation habits of the operators, etc. By integrating and applying the second blood coagulation curve recognition model and the first blood coagulation curve recognition model to the recognition process of the second blood coagulation data, it is beneficial to optimize the second recognition result in the direction of adapting to the judgment habits corresponding to the user's region, the detection reagents used, and the operation habits of the operators, etc., while reducing the influence of individual data differences on the accuracy, and ensuring the accuracy and reliability of the recognition of abnormal blood coagulation curves.

[0100] In a specific application, the second blood coagulation curve recognition model may be directly used as the updated first blood coagulation curve recognition model. In a possible example, the determining the second blood coagulation curve recognition model according to the first feedback information and the first blood coagulation data includes: storing the second type of blood coagulation curve and the first blood coagulation data in the first user dataset; adjusting the weight parameters in at least one hidden layer in the first blood coagulation curve recognition model according to the information in the first user dataset to generate the second blood coagulation curve recognition model.

[0101] In a specific implementation, the first blood coagulation curve recognition model can be a fully connected network model or a convolutional network model. The network structure for the first blood coagulation curve recognition includes an input layer, multiple hidden layers, and an output layer arranged in sequence. Among them, the input layer is used to process the input first blood coagulation data, and the output layer is used to output the fitness of the first blood coagulation data corresponding to each blood coagulation curve type, and output the first recognition result according to the fitness. Among them, the fitness is the model prediction value and can be represented by a probability.

[0102] Specifically, when the first blood coagulation curve recognition model is a fully connected network model, the multiple hidden layers are fully connected layers. The blood coagulation detection device can optimize the first blood coagulation curve recognition model by adjusting at least one of the multiple fully connected layers to generate a second blood coagulation curve recognition model.

[0103] Specifically, the fully connected network model can include at least 7 layers, including 1 input layer, 1 output layer, and at least 5 hidden layers. The activation function uses the rectified linear unit (ReLU for short). In this way, deep learning of the first blood coagulation curve recognition model can be realized to meet the requirements of complex data processing. Exemplarily, when the fully connected network model includes 7 layers, the number of nodes in the input layer can be 50, which is used to represent a total of 50 s of data collected, corresponding to one data per second, that is, the preset reaction duration is 50 s. When the fully connected network model includes 7 layers, the nodes corresponding to the 5 hidden layers can be 40, 40, 30, 30, 20 respectively, and the output layer has only one node, and this node can be configured as a Softmax classifier to identify and output the probabilities corresponding to the first blood coagulation data and each blood coagulation curve type.

[0104] Specifically, when the first blood coagulation curve recognition model is a convolutional network model, the multiple hidden layers include multiple convolutional layers and a pooling layer, where the pooling layer is adjacent to the output layer. The blood coagulation detection device can optimize the first blood coagulation curve recognition model by adjusting at least one of the multiple convolutional layers and the pooling layer to generate a second blood coagulation curve recognition model.

[0105] Specifically, the convolutional network model may include at least 7 layers, including 1 input layer and 1 output layer, and at least 5 hidden layers. Among them, at least 5 hidden layers include at least 4 convolutional layers and 1 pooling layer. The convolutional dimension of the convolutional network model in the time direction is 4, that is, the window size between two adjacent convolutional layers is 4, and the activation function uses ReLU. In this way, deep learning of the first blood coagulation curve recognition model can be realized to meet the needs of complex data processing. Exemplarily, the size of the input layer can be 1*50, which is used to represent a total of 50 seconds of collected data, that is, the preset reaction duration is 50s. The sizes of the hidden layers are 20*46, 30*42, 30*38, 20*34, 20*1 respectively, and the output layer has only one node corresponding to it. This node can be configured as a Softmax classifier to identify and output the probabilities corresponding to the first blood coagulation data and each blood coagulation curve type.

[0106] It can be seen that in this example, by adjusting the weight parameters of at least one hidden layer in the first blood coagulation curve recognition model according to the information in the first user dataset, the first blood coagulation curve recognition model can be optimized in the direction of adapting to the judgment habits corresponding to the user's region, the detection reagents used, and the operation habits of the operator, etc. according to the information in the first user dataset, so as to improve the matching degree between the optimized second blood coagulation curve recognition model and the user.

[0107] In a possible example, the first blood coagulation curve recognition model includes an input layer, multiple hidden layers, and an output layer; adjusting the weight parameters of at least one hidden layer in the first blood coagulation curve recognition model according to the information in the first user dataset to generate the second blood coagulation curve recognition model includes: adjusting at least the weight parameters of the hidden layer closest to the output layer in the first blood coagulation curve recognition model according to the information in the first user dataset to generate the second blood coagulation curve recognition model.

[0108] In specific implementation, the adjustment of the weight parameters of the hidden layer closest to the output layer in the first blood coagulation curve recognition model can be realized locally on the blood coagulation detection device. At this time, parameter freezing is performed on other hidden layers that are not adjusted. In this way, the number of parameters that need to be trained can be reduced, the hardware calculation requirements for the blood coagulation detection device can be reduced, and costs can be saved.

[0109] Specifically, when the first blood coagulation curve recognition model is a fully connected network model, the model can be updated by adjusting the weight parameters of the fully connected layer adjacent to the output layer to generate the second blood coagulation curve recognition model. Or, when the first blood coagulation curve recognition model is a convolutional network model, the model can be updated by adjusting the weight parameters of the pooling layer to generate the second blood coagulation curve recognition model.

[0110] It can be seen that in this example, by adjusting the weight parameters in the hidden layer closest to the output layer in the first blood coagulation curve recognition model to generate the second blood coagulation curve recognition model, it is beneficial to shorten the path between the optimized features and the output, while improving the training efficiency, strengthening the correlation between the optimized features and the output.

[0111] In a possible example, the determining the second blood coagulation curve recognition model according to the first feedback information and the first blood coagulation data further includes: obtaining a second user data set, where the second user data set includes third blood coagulation data and its corresponding blood coagulation curve type, and the blood coagulation curve type in the recognition result corresponding to the third blood coagulation data is the same as the blood coagulation curve type in the feedback information corresponding to the third blood coagulation data; adjusting the weight parameters in at least one hidden layer in the first blood coagulation curve recognition model according to the information in the first user data set and the second user data set to generate the second blood coagulation curve recognition model.

[0112] Among them, the second user data set is used to store relevant data when the user confirms that the recognition result of the blood coagulation curve recognition model is correct. Specifically, the second user data set includes at least one set of blood coagulation data and the blood coagulation curve type obtained by the blood coagulation curve recognition model for recognizing the blood coagulation data. Specifically, the information in the second user data set is the data obtained before processing the first blood coagulation data, that is, the third blood coagulation data is the content recognized before the first blood coagulation data.

[0113] In a specific implementation, after the blood coagulation detection device obtains the first feedback information, it can store the currently recognized first blood coagulation data and the second blood coagulation curve type in the first feedback information into the first user data set, and establish a correspondence between the first blood coagulation data and the second blood coagulation curve type. At the same time, obtain the second user data set. And adjust the weight parameters in at least one hidden layer in the first blood coagulation curve recognition model according to the information in the first user data set and the second user data set to generate the second blood coagulation curve recognition model, so as to realize the synchronous update of the first blood coagulation curve recognition model according to the accurately recognized blood coagulation curve type corresponding to the blood coagulation curve recognition model and the blood coagulation curve type after correcting the recognition result of the blood coagulation curve recognition model, so that the second blood coagulation curve recognition model further optimizes the recognition accuracy, improves the recognition effect, and optimizes the user experience on the basis of adapting to the judgment habits, detection reagents used, and operator operation habits in the user's area.

[0114] It can be seen that in this example, by adjusting the weight parameters in at least one hidden layer of the first blood coagulation curve recognition model through the information in the first user data set and the second user data, the recognition accuracy can be further optimized, and the user experience can be optimized.

[0115] In a possible example, during the recognition process of the second coagulation data, the method further includes: obtaining the second coagulation data, and determining a second recognition result corresponding to the second coagulation data according to the second coagulation data and the second coagulation curve recognition model.

[0116] In specific implementation, after updating the first coagulation curve recognition model to the second coagulation curve recognition model, when the second coagulation data to be detected is obtained, the second coagulation data can be compared to the above-mentioned first coagulation data, and the second coagulation curve recognition model can be compared to the above-mentioned first coagulation curve recognition model, so as to determine a second recognition result corresponding to the second coagulation data according to the second coagulation data and the second coagulation curve recognition model, and to determine the type of coagulation curve corresponding to the second coagulation data.

[0117] Specifically, after determining the second recognition result, the second recognition result can be compared to the above-mentioned first recognition result, and steps S430 and S440 are analogously executed, so as to update the model algorithm according to the user's usage process and continuously improve the accuracy of abnormal curve recognition.

[0118] It can be seen that in this example, the second recognition result determined according to the second coagulation data and the second coagulation curve recognition model has high accuracy and can continue to be used for model algorithm update to further improve and optimize the recognition accuracy.

[0119] In a possible example, after determining the second recognition result corresponding to the second coagulation data, the method further includes: obtaining second feedback information of the user based on the second recognition result, where the second feedback information includes the trigger state of the reset function provided by the coagulation detection device, and the trigger state is a triggered state or an untriggered state; when the trigger state is the triggered state, reset the second coagulation curve recognition model to the first coagulation curve recognition model, so as to apply the first coagulation curve recognition model to the recognition process of the fourth coagulation data.

[0120] Among them, the trigger state is used to represent the user's requirement for whether to reset the second coagulation curve recognition model. Among them, the fourth coagulation data is the content for coagulation abnormal curve recognition after the second coagulation data, and specifically, it can be the first or the Pth coagulation data to be detected arranged after the second coagulation data, and P is a positive integer greater than or equal to 2, which is not further limited here.

[0121] In a specific implementation, based on the second recognition result, the user can determine the reset requirement for the current second blood coagulation curve recognition model according to the actual situation. When there is no reset requirement, the trigger state corresponding to the second feedback information is the untriggered state. At this time, the blood coagulation detection device can use the second blood coagulation curve recognition model for subsequent recognition of the fourth blood coagulation data or for the update of the blood coagulation curve recognition model. When there is a reset, the trigger state corresponding to the second feedback information is the triggered state. At this time, the blood coagulation detection device can reset the second blood coagulation curve recognition model to the first blood coagulation curve recognition model and use the first blood coagulation curve recognition model for subsequent recognition of the fourth blood coagulation data or for the update of the blood coagulation curve recognition model.

[0122] Specifically, the blood coagulation detection device is configured with a reset function. When the blood coagulation detection device detects that the reset function is triggered, it can mark the trigger state as the triggered state and reset the second blood coagulation curve recognition model to the first blood coagulation curve recognition model. Exemplarily, refer to Figure 6 , in the application interface provided by the blood coagulation detection device, a "Reset" control corresponding to the reset function can be configured. When the blood coagulation detection device detects that the "Reset" control is triggered (such as when detecting operations such as single-click, double-click, long-press on the "Reset" control), it is determined that the reset function is triggered. When the blood coagulation detection device detects that the reset function is not triggered, it can automatically determine that the trigger state of the second blood coagulation curve recognition model is the untriggered state.

[0123] It can be seen that in this example, when the trigger state is the triggered state, the second blood coagulation curve recognition model is reset to the first blood coagulation curve recognition model to apply the first blood coagulation curve recognition model to the recognition process of the fourth blood coagulation data. The blood coagulation curve recognition model can be dynamically optimized to meet the differentiated needs of users.

[0124] Consistent with the above-described embodiments, please refer to Figure 9 , Figure 9 is a functional unit composition block diagram of a blood coagulation abnormal curve recognition device provided by an embodiment of the present application. The blood coagulation abnormal curve recognition device is the above-mentioned server or a part of the server. The blood coagulation abnormal curve recognition device 900 includes:

[0125] A first acquisition unit 910, configured to acquire first blood coagulation data to be detected;

[0126] A first determination unit 920, configured to obtain a first recognition result corresponding to the first blood coagulation data according to the first blood coagulation data and the first blood coagulation curve recognition model, where the first recognition result includes a first blood coagulation curve type;

[0127] A second acquisition unit 930, configured to acquire first feedback information of a user based on the first recognition result, where the first feedback information includes an updated second blood coagulation curve type, and the second blood coagulation curve type is different from the first blood coagulation curve type;

[0128] A second determination unit 940, configured to determine a second blood coagulation curve recognition model according to the first feedback information and the first blood coagulation data, so as to apply the second blood coagulation curve recognition model to the recognition process of second blood coagulation data.

[0129] In a possible example, in the aspect of determining a second blood coagulation curve recognition model according to the first feedback information and the first blood coagulation data, and applying the second blood coagulation curve recognition model to the recognition process of second blood coagulation data, the second determination unit is specifically configured to:

[0130] Store the second blood coagulation curve type and the first blood coagulation data in a first user dataset;

[0131] Generate the second blood coagulation curve recognition model according to the information in the first user dataset, so that the second blood coagulation curve recognition model and the first blood coagulation curve recognition model are integrally applied to the recognition process of the second blood coagulation data.

[0132] In a possible example, during the recognition process of the second blood coagulation data, the second determination unit is specifically configured to:

[0133] Acquire the second blood coagulation data, and determine a first output result according to the second blood coagulation data and the first blood coagulation curve recognition model;

[0134] Determine a second output result according to the second blood coagulation data and the second blood coagulation curve recognition model;

[0135] Perform an integration operation on the first output result and the second output result to determine a second recognition result corresponding to the second blood coagulation data.

[0136] In a possible example, in the aspect of performing an integration operation on the first output result and the second output result to determine a second recognition result corresponding to the second blood coagulation data, the second determination unit is specifically configured to:

[0137] Average the first output result and the second output result to determine the second recognition result; or,

[0138] Perform a weighted average on the first output result and the second output result to determine the second recognition result.

[0139] In a possible example, in terms of determining the second blood coagulation curve recognition model according to the first feedback information and the first blood coagulation data, the second determination unit is specifically configured to:

[0140] Store the second blood coagulation curve type and the first blood coagulation data in the first user dataset;

[0141] According to the information in the first user dataset, adjust the weight parameters in at least one hidden layer of the first blood coagulation curve recognition model to generate the second blood coagulation curve recognition model.

[0142] In a possible example, the first blood coagulation curve recognition model includes an input layer, a plurality of hidden layers, and an output layer;

[0143] In terms of adjusting the weight parameters in at least one hidden layer of the first blood coagulation curve recognition model according to the information in the first user dataset to generate the second blood coagulation curve recognition model, the second determination unit is specifically configured to:

[0144] According to the information in the first user dataset, at least adjust the weight parameters in the hidden layer closest to the output layer of the first blood coagulation curve recognition model to generate the second blood coagulation curve recognition model.

[0145] In a possible example, in terms of determining the second blood coagulation curve recognition model according to the first feedback information and the first blood coagulation data, the second determination unit is specifically configured to:

[0146] Obtain a second user dataset, where the second user dataset includes third blood coagulation data and its corresponding blood coagulation curve type, and the blood coagulation curve type in the recognition result corresponding to the third blood coagulation data is the same as the blood coagulation curve type in the feedback information corresponding to the third blood coagulation data;

[0147] According to the information in the first user dataset and the second user dataset, adjust the weight parameters in at least one hidden layer of the first blood coagulation curve recognition model to generate the second blood coagulation curve recognition model.

[0148] In a possible example, in terms of the recognition process of the second blood coagulation data, the second determination unit is specifically further configured to:

[0149] Obtain the second blood coagulation data, and according to the second blood coagulation data and the second blood coagulation curve recognition model, determine the second recognition result corresponding to the second blood coagulation data.

[0150] In a possible example, the second determination unit is further specifically configured to: after determining the second recognition result corresponding to the second coagulation data, obtain second feedback information of the user based on the second recognition result, where the second feedback information includes a trigger state of a reset function provided by the coagulation detection device, and the trigger state is a triggered state or an untriggered state; when the trigger state is the triggered state, reset the second coagulation curve recognition model to the first coagulation curve recognition model, so as to apply the first coagulation curve recognition model to the recognition process of the fourth coagulation data.

[0151] It can be understood that since the method embodiment and the device embodiment are different presentation forms of the same technical concept, therefore, the content of the method embodiment part in this application should be synchronously adapted to the device embodiment part, and will not be elaborated here.

[0152] In the case of adopting an integrated unit, the functional unit composition block diagram of another coagulation abnormality curve recognition device provided in the embodiments of the present application is as Figure 10 shown. In Figure 10 , the coagulation abnormality curve recognition device 900 includes: a processing module 1020 and a communication module 1010. The processing module 1020 is used to control and manage the actions of the above-mentioned coagulation abnormality curve recognition device. For example, the steps executed by the first acquisition unit 910, the first determination unit 920, the second acquisition unit 930, and the second determination unit 940, and / or used to execute other processes of the technologies described herein. The communication module 1010 is used to support the interaction between the coagulation abnormality curve recognition device 900 and other devices. As Figure 10 shown, the coagulation abnormality curve recognition device 900 may further include a storage module 1030, and the storage module 1030 is used to store the program code and data of the coagulation abnormality curve recognition device 900.

[0153] Among them, the processing module 1020 can be a processor or a controller. For example, it can be a Central Processing Unit (CPU), a general-purpose processor, a Digital Signal Processor (DSP), an Application-Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in connection with the disclosed content of the embodiments of the present application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and so on. The communication module 1010 can be a transceiver, an RF circuit, a communication interface, etc. The storage module 1030 can be a memory.

[0154] Among them, the storage module 1030 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), or a flash memory. The volatile memory can be a Random Access Memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory are available, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), SynchLink DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).

[0155] Among them, all relevant contents of each scenario involved in the above method embodiments can be cited in the function descriptions of the corresponding functional modules, and will not be elaborated here. All of the above coagulation abnormality curve recognition devices can execute the above Figure 4 coagulation abnormality curve recognition method shown.

[0156] This application can divide the functional units of the electronic device according to the above method examples. For example, each functional unit can be divided corresponding 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 a software functional unit. It should be noted that the division of units in the embodiments of this application is illustrative, and is only a logical function division. There may be other division methods in actual implementation.

[0157] The embodiments of this application also provide a computer-readable storage medium. Among them, this computer-readable storage medium stores a computer program for electronic data exchange, and this computer program enables the computer to execute some or all of the steps of any of the methods described in the above method embodiments. The above computer includes a server.

[0158] The embodiments of this application also provide a computer program product. The above computer program product includes a non-transitory computer-readable storage medium storing a computer program. The above computer program can operate to enable the computer to execute some or all of the steps of any of the information push methods described in the above method embodiments. This computer program product can be a software installation package.

[0159] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, some steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0160] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0161] In several embodiments provided by 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 merely illustrative. For example, the above division of units is only a logical function division. In actual implementation, there may be other division methods. For example, 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 displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in electrical or other forms.

[0162] The units described above as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0163] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0164] If the above integrated unit is implemented in the form of a software functional unit 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, in essence, 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. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the above methods in various embodiments of the present application. The foregoing memory includes: USB flash drive, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk, or optical disc, etc., which can store program codes.

[0165] Those 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 relevant hardware through a program. This program can be stored in a computer-readable memory. The memory can include: flash drive, ROM, RAM, magnetic disk, or optical disc, etc.

[0166] The above has introduced the embodiments of the present application in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for identifying coagulation abnormality curves, characterized in that, Applied to a coagulation detection device, the method includes: Obtain first coagulation data to be identified and detected; According to the first coagulation data and a first coagulation curve recognition model, obtain a first recognition result corresponding to the first coagulation data, where the first recognition result includes a first coagulation curve type; Obtain first feedback information of the user based on the first recognition result, where the first feedback information includes an updated second coagulation curve type, and the second coagulation curve type is different from the first coagulation curve type; According to the first feedback information and the first coagulation data, determine a second coagulation curve recognition model to apply the second coagulation curve recognition model to the recognition process of second coagulation data.

2. The method according to claim 1, wherein The step of determining a second coagulation curve recognition model according to the first feedback information and the first coagulation data to apply the second coagulation curve recognition model to the recognition process of second coagulation data includes: Store the second coagulation curve type and the first coagulation data in a first user dataset; Generate the second coagulation curve recognition model according to the information in the first user dataset, so that the second coagulation curve recognition model and the first coagulation curve recognition model are integrally applied to the recognition process of the second coagulation data.

3. The method according to claim 2, characterized in that, In the recognition process of the second coagulation data, the method further includes: Obtain the second coagulation data, and determine a first output result according to the second coagulation data and the first coagulation curve recognition model; Determine a second output result according to the second coagulation data and the second coagulation curve recognition model; Perform an integration operation on the first output result and the second output result to determine a second recognition result corresponding to the second coagulation data.

4. The method according to claim 3, wherein The step of performing an integration operation on the first output result and the second output result to determine a second recognition result corresponding to the second coagulation data includes: Average the first output result and the second output result to determine the second recognition result; or, Perform a weighted average on the first output result and the second output result to determine the second recognition result.

5. The method according to claim 1, wherein The step of determining a second coagulation curve recognition model according to the first feedback information and the first coagulation data includes: Store the second coagulation curve type and the first coagulation data in a first user dataset; Adjust the weight parameters in at least one hidden layer of the first coagulation curve recognition model according to the information in the first user dataset to generate the second coagulation curve recognition model.

6. The method according to claim 5, characterized in that, The first coagulation curve recognition model includes an input layer, multiple hidden layers, and an output layer; The step of adjusting the weight parameters in at least one hidden layer of the first coagulation curve recognition model according to the information in the first user dataset to generate the second coagulation curve recognition model includes: Adjust at least the weight parameters in the hidden layer closest to the output layer of the first coagulation curve recognition model according to the information in the first user dataset to generate the second coagulation curve recognition model.

7. The method according to claim 5, wherein Determining the second blood coagulation curve recognition model according to the first feedback information and the first blood coagulation data further includes: Obtaining a second user dataset, where the second user dataset includes third blood coagulation data and its corresponding blood coagulation curve type, and the blood coagulation curve type in the recognition result corresponding to the third blood coagulation data is the same as the blood coagulation curve type in the feedback information corresponding to the third blood coagulation data; Adjusting the weight parameters in at least one hidden layer in the first blood coagulation curve recognition model according to the information in the first user dataset and the second user dataset to generate the second blood coagulation curve recognition model.

8. The method according to claim 5, characterized in that, During the recognition process of the second blood coagulation data, the method further includes: Obtaining the second blood coagulation data, and determining a second recognition result corresponding to the second blood coagulation data according to the second blood coagulation data and the second blood coagulation curve recognition model.

9. The method according to claim 4 or 8, characterized in that, After determining the second recognition result corresponding to the second blood coagulation data, the method further includes: Obtaining second feedback information of the user based on the second recognition result, where the second feedback information includes the trigger status of the reset function provided by the blood coagulation detection device, and the trigger status is a triggered status or an untriggered status; When the trigger status is the triggered status, resetting the second blood coagulation curve recognition model to the first blood coagulation curve recognition model to apply the first blood coagulation curve recognition model to the recognition process of the fourth blood coagulation data.

10. A coagulation detection device, characterized in that, Including a processor, a memory, a communication interface, and one or more programs, the one or more programs are stored in the memory and are configured to be executed by the processor, and the programs include instructions for performing the steps in the method according to any one of claims 1-9.

Images