Bee venom allergy possibility auxiliary analysis method and system based on detection data
By combining the basic information of the target patients and the detection data of the bee venom protein, using historical patient data and possibility range evaluation model, the shortcomings of the existing technology of bee venom allergy prediction methods are solved, and a more accurate and personalized analysis of bee venom allergy possibility is achieved.
Patent Information
- Application Number
- CN202510203418.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-24
- Publication Date
- 2025-05-13
AI Technical Summary
The existing bee venom allergy prediction methods cannot comprehensively and accurately evaluate patients' allergic risks, and lack methods to effectively utilize historical patient data, resulting in a single prediction model and room for improvement in accuracy and reliability.
By obtaining basic information and bee venom protein detection data of the target patient, and combining historical patient data, vector similarity calculation and possibility range evaluation models (including decision tree models, knowledge graph models, and neural network models) are used to determine the possibility of bee venom allergic in the target patient.
A more intelligent and accurate analysis of the possibility of bee venom allergy is achieved, which can personalize the specific situation of the target patient and refer to historical patient data to provide more accurate prediction results, providing better support and reference for doctors' diagnostic work.
Smart Images
Figure CN119993542A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of auxiliary diagnosis of bee venom allergy, and in particular to an auxiliary analysis method and system for the possibility of bee venom allergy based on detection data. Background Art
[0002] In bee venom therapy and related medical scenarios, it is crucial to accurately predict the possibility of a patient's allergy to bee venom protein. Currently, there are many difficulties in clinically predicting bee venom allergy.
[0003] On the one hand, existing detection methods mainly focus on the detection of specific bee venom proteins, but relying solely on these test data cannot comprehensively and accurately assess the patient's allergy risk. Because allergic reactions are a complex physiological process, affected by a variety of factors, simply testing bee venom proteins is difficult to cover many factors that have a significant impact on the possibility of allergies, such as age, gender, height, weight, family allergy history, and living environment.
[0004] On the other hand, the use of historical patient data is not sufficient and scientific enough. Although a large amount of historical patient data has been accumulated, including basic data, bee venom protein test data, and allergic conclusions, in actual prediction, there is a lack of effective methods to screen out historical data similar to the current patient and accurately determine its reference degree. Traditional methods are often just simple comparisons, and it is difficult to explore the deep-level connections between data, resulting in the failure to fully utilize the value of historical data.
[0005] Furthermore, the existing prediction models are relatively simple. Whether it is the decision tree model built based on industry standards and expert advice, or other simple prediction methods, they cannot fully consider various complex factors, and there is a large room for improvement in accuracy and reliability. This makes doctors lack sufficiently accurate references when facing whether patients are suitable for bee venom-related treatments, which increases treatment risks and uncertainties. Summary of the invention
[0006] The present application provides a method and system for auxiliary analysis of the possibility of bee venom allergy based on detection data, which can intelligently and accurately analyze and predict the possibility of bee venom allergy in target patients, and provide a reference for doctors' diagnosis work.
[0007] In a first aspect, the present application provides a method for auxiliary analysis of the possibility of bee venom allergy based on detection data. The method comprises: Obtaining basic information of the target patient and test data of multiple types of melittin, and obtaining historical patient data, wherein the historical patient data carries a timestamp and includes basic information of the historical patient, test data of multiple types of melittin, and conclusions on the allergy to bee venom; Determine similar patient data in the historical patient data based on the basic information of the target patient, wherein the similar patient data is the historical patient data and the target patient's basic information has information similarity higher than a first similarity threshold, and the data similarity of the detection data of multiple types of melittin is higher than a second similarity threshold; Substituting the basic information of the target patient and the detection data of various melittin proteins into the pre-acquired possibility range evaluation model to obtain the allergy possibility range of the target patient; The similar patient data are combined to determine the possibility of bee venom allergy in the target patient within the range of allergy possibility.
[0008] By adopting the above technical scheme, the target patient's basic information, bee venom protein detection data and a possibility range assessment model pre-constructed based on the expert experience knowledge base are combined to determine the target patient's allergy possibility range for bee venom allergy, and the target patient's basic information, bee venom protein detection data and historical patient data are combined to analyze and determine the target patient's bee venom allergy possibility within the allergy possibility range, so that the analyzed bee venom allergy possibility is consistent with expert experience and can be personalized to the specific situation of the target patient. The historical patient data is also referred to, which is conducive to making the analyzed and predicted bee venom allergy possibility more accurate, and providing better support and reference for doctors' diagnostic work.
[0009] Further, the determining similar patient data in the historical patient data based on the basic information of the target patient includes: The basic information is constructed as a first vector, and the detection data of multiple types of melittin are constructed as a second vector; Determine that the vector similarity between the first vectors of the historical patient and the target patient is the information similarity, and the vector similarity between the second vectors is the data similarity; The historical patient data whose vector similarity is higher than a first similarity threshold and whose data similarity is higher than a second similarity threshold is determined as the similar patient data.
[0010] Furthermore, the possibility range assessment model includes one or more of a decision tree model, a knowledge graph model, and a neural network model.
[0011] Further, the step of determining the possibility of bee venom allergy of the target patient within the range of allergy possibility by combining the similar patient data includes: Analyzing the data reference of each similar patient data, wherein the data reference is associated with one or more of a timestamp carried by the similar patient data, information similarity with basic information of the target patient, and data similarity with multiple melittin detection data of the target patient; determining a data base value for each similar patient data, the data base value being associated with a bee venom allergy conclusion; Calculate the likelihood influence coefficient based on the data reference degree, the data base value, and the number of similar data of similar patient data; The allergy possibility of the target patient is determined within the allergy possibility range according to the possibility influence coefficient.
[0012] Furthermore, the analysis of the data reference of each similar patient data includes: Assume that the timestamp carried by the i-th similar patient data is , and the information similarity with the basic information of the target patient is , and the data similarity with the target patient's various bee venom protein detection data is , the data reference is , the current time is , the first similarity threshold is , the second similarity threshold is ,but , where All are preset calculation coefficients greater than zero.
[0013] Further, determining the data base value of each similar patient data includes: The conclusion of bee venom allergy of similar patient data is bee venom allergy or non-bee venom allergy. The bee venom allergy also carries allergy degree data. Let the data base value of the i-th similar patient data be 、Allergy degree data is If the conclusion of bee venom allergy is non-bee venom allergy, then ,otherwise .
[0014] Furthermore, the calculating of the possibility influence coefficient according to the data reference degree, the data base value and the number of similar data of similar patient data includes: Assume that there are n similar patient data and the probability influence coefficient is k, then ; ; ; ; ; In the formula, are all preset calculation coefficients greater than zero and , To preset the allergy level, is the preset reference degree, for The standard deviation of for The standard deviation of .
[0015] Further, determining the allergy possibility of the target patient within the allergy possibility range according to the possibility influence coefficient includes: Assume that the range of allergy possibility is , the probability influence coefficient is k and , the target patient's probability of bee venom allergy, then .
[0016] In a second aspect, the present application provides a system for auxiliary analysis of the possibility of bee venom allergy based on detection data. The system applies any one of the methods described in the first aspect above.
[0017] In summary, this application at least has the following beneficial effects: A method and system for auxiliary analysis of the possibility of bee venom allergy based on test data are provided, which can intelligently analyze the possibility of bee venom allergy of the target patient by comprehensively considering the specific situation of the target patient, historical patient data and expert treatment experience, which is conducive to providing more accurate analysis results of the possibility of bee venom allergy and providing better support and reference for doctors' diagnostic work.
[0018] It should be understood that the contents described in the Summary of the Invention are not intended to limit the key or important features of the embodiments of the present application, nor are they intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein: Figure 1 A flow chart of a method for auxiliary analysis of the possibility of bee venom allergy based on detection data in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are 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.
[0021] In addition, the term "and / or" in this article is only a description of the association relationship between the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.
[0022] The present application provides a method and system for auxiliary analysis of the possibility of bee venom allergy based on test data, which can combine expert experience, historical treatment data and the patient's personal situation to intelligently and accurately analyze the patient's possibility of bee venom allergy, thereby providing a reference for doctors' treatment.
[0023] In the first aspect, the embodiment of the present application discloses a method for auxiliary analysis of the possibility of bee venom allergy based on test data. The method can be executed by a server or medical system (such as an HIS system, a CIS system, etc.) of a hospital to determine the possibility of bee venom allergy of a target patient, and the possibility of bee venom allergy can be provided to doctors to provide reference and support for the doctors' diagnosis and treatment work.
[0024] Figure 1 A flow chart of a method for auxiliary analysis of the possibility of bee venom allergy based on detection data in an embodiment of the present application is shown.
[0025] Reference Figure 1 The method specifically comprises the following steps: S110: Obtain basic information of target patients and test data of various bee venom proteins, and obtain historical patient data.
[0026] The basic information of the target patient is multi-dimensional information, which can theoretically contain any relevant information of the patient that can be obtained, for example, through doctor-patient communication or medical information collection. Specific information dimensions include gender, age, height, weight, lifestyle, allergy history, past medical history, family allergy history, etc.
[0027] When judging the possibility of bee venom allergy, it is necessary to detect multiple bee venom proteins and obtain multiple test data, which can be collected.
[0028] Historical patient data or historical treatment data represent the medical records of bee venom allergy-related visits received by the hospital in the past. The historical patient data carries a timestamp and includes basic information of historical patients, test data of multiple bee venom proteins, and conclusions on bee venom allergy. The meaning of the basic information of historical patients is the same as that of the target patients, but the specific information in different dimensions is different. The quantity and type of bee venom proteins in the bee venom protein test data are also the same as those of the target patients, but the specific test data are different. The historical patient data contains clear conclusions on bee venom allergy, and the conclusions on bee venom allergy are bee venom allergy or non-bee venom allergy. The conclusions on bee venom allergy also carry allergy degree data, and the allergy degree data can be a quantitative allergy degree score, such as a score between 1 and 10 points, or a semi-quantitative allergy degree assessment conclusion, such as mild allergy, moderate allergy, and severe allergy.
[0029] S120: Determine similar patient data in the historical patient data based on the basic information of the target patient.
[0030] The similar patient data is the information similarity between the basic information of the historical patient and the target patient is higher than the first similarity threshold, and the data similarity of the detection data of multiple melittin proteins is higher than the second similarity threshold. In the method of this step, in order to facilitate the comparison of the information similarity between the basic information of the target patient and the historical patient, it is generally necessary to quantify the basic information, and the detection data of melittin is generally a quantitative concentration value, so there is no need to quantify it. Regarding the quantification of basic information of each dimension, it can be configured based on treatment experience and data experience. For non-quantitative dimensional information such as gender, lifestyle, allergy history, past medical history, family allergy history, etc., a situation comparison table or situation mapping can be used to determine the dimensional value corresponding to each specific dimensional information. For example, a clear dimensional value can be configured for males and females respectively to ensure that the difference in their dimensional values is reasonable. For another specific example, the dimensional value of the family allergy history dimension is determined based on the presence or absence, number of family allergy history, and the relationship with historical patients. For dimensional information that is quantitative in itself, such as height, age, and weight, the specific values of height, age, and weight can be directly used as the corresponding dimensional values. Alternatively, an activation function can be used based on treatment experience and data experience to adjust the specific value of the dimensional value when the dimensional information is in different value ranges, change the way of mapping the dimensional value to the actual dimensional information, change the speed at which the dimensional value changes with the different values of the actual dimensional information, etc., so that it is more in line with the needs of bee venom allergy medical profession.
[0031] In one example, the method of this step specifically includes: constructing the basic information into a first vector, and constructing the detection data of multiple bee venom proteins into a second vector; determining the vector similarity between the first vectors of the historical patient and the target patient as the information similarity, and the vector similarity between the second vectors as the data similarity; determining the historical patient data whose vector similarity is higher than a first similarity threshold and whose data similarity is higher than a second similarity threshold as the similar patient data.
[0032] In another example, the absolute value of the difference of the dimension value of each basic information can be calculated, and the inverse of the absolute value of the difference can be calculated as the dimension similarity. The weighted sum of all dimensional similarities is calculated as the comprehensive similarity, and the comprehensive similarity is normalized to be between 0 and 1 as the final information similarity result. Data similarity can also be calculated using the same idea.
[0033] Other methods of calculating information similarity and data similarity are not listed one by one. It is only necessary to be able to calculate information similarity and data similarity.
[0034] S130: Substituting the basic information of the target patient and the detection data of various melittin proteins into the pre-acquired possibility range evaluation model to obtain the allergy possibility range of the target patient.
[0035] The possibility range assessment model is pre-built based on an expert experience knowledge base; the possibility range assessment model includes one or more of a decision tree model, a knowledge graph model, and a neural network model.
[0036] In a specific embodiment, the possibility range assessment model is specifically selected as a decision tree model. The decision tree model is an interpretable model. It is possible to trace how the possibility range of allergies is determined based on basic information and detection data of multiple bee venom proteins, thereby facilitating doctors and experts to adjust the model structure and parameters based on professional knowledge and experience.
[0037] The specific means of constructing and training the possibility range assessment model based on the expert experience knowledge base can be determined based on the content of the feasibility range assessment model of the specific configuration. The construction and training methods of decision tree models, knowledge graph models, neural network models, etc. are all known technologies and are not disclosed in detail here. It is only necessary to be able to input the basic information of the target patient and the detection data of various bee venom proteins, and output the possibility range of allergy of bee venom allergy that conforms to the expert experience. Of course, the possibility range assessment model can also be characterized as other models, or a fusion model composed of multiple models, which will not be listed here one by one.
[0038] S140: Determine the possibility of bee venom allergy in the target patient within the range of allergy possibility by combining the similar patient data.
[0039] In one example, the method of this step specifically includes: analyzing the data reference degree of each similar patient data, the data reference degree is associated with one or more of the timestamp carried by the similar patient data, the information similarity with the basic information of the target patient, and the data similarity with multiple bee venom protein detection data of the target patient; determining the data base value of each similar patient data, the data base value is associated with the bee venom allergy conclusion; calculating the possibility influence coefficient based on the data reference degree, the data base value and the number of similar data of the similar patient data; determining the allergy possibility of the target patient within the allergy possibility range based on the possibility influence coefficient.
[0040] The analysis of the data reference of each similar patient data includes: assuming that the timestamp carried by the i-th similar patient data is , and the information similarity with the basic information of the target patient is , and the data similarity with the target patient's various bee venom protein detection data is , the data reference is , the current time is , the first similarity threshold is , the second similarity threshold is ,but , where All are preset calculation coefficients greater than zero. In the expression model of data reference, the overall data reference value range is between 0 and 1, and it is negatively correlated with the time length from the timestamp carried by similar patient data to the current moment, and positively correlated with the ratio of information similarity exceeding the first similarity threshold and data similarity exceeding the second similarity threshold. The value range is adjusted using the pre-configured calculation coefficient to cover as much as possible between 0 and 1.
[0041] The data base value of each similar patient data is determined as follows: the conclusion of bee venom allergy of the similar patient data is bee venom allergy or non-bee venom allergy, and the bee venom allergy also carries allergy degree data. Suppose the data base value of the i-th similar patient data is 、Allergy degree data is If the conclusion of bee venom allergy is non-bee venom allergy, then ,otherwise Based on this, the conclusion of bee venom allergy can be quantified as a data base value, so that it can truly reflect the bee venom allergy situation of similar patient data. The allergy degree data can be directly a quantitative allergy degree score, or a semi-quantitative allergy degree mapping value.
[0042] The calculation of the possibility influence coefficient according to the data reference degree, the data base value and the number of similar data of similar patient data includes: assuming that there are n similar patient data and the possibility influence coefficient is k, then ; ; ; ; ; In the formula, are all preset calculation coefficients greater than zero and , To preset the allergy level, is the preset reference degree, for The standard deviation of for The model for calculating the likelihood influence coefficient fully considers the data base value and data reference degree in similar patient data, and takes into account the proportion of bee venom allergy in similar patient data to the total number, the proportion of bee venom allergy in similar patient data with a higher data reference degree to the total number of this part, and considers the specific data distribution to analyze the credibility of the likelihood influence coefficient, and finally comprehensively determines a reasonable likelihood influence coefficient.
[0043] The calculation method of the aforementioned possibility influence coefficient fully considers a variety of specific parameter characteristics, and it can of course also consider only a few of the parameter characteristics, such as , , or change the fusion calculation method of the parameter features, for example , That is, under the inventive concept of the present application, the calculation method of the possibility influence coefficient can be implemented as a variety of examples, which will not be listed one by one here.
[0044] Determining the allergy possibility of the target patient within the allergy possibility range according to the possibility influence coefficient includes: assuming that the allergy possibility range is , the probability influence coefficient is k and , the target patient's probability of bee venom allergy, then . Again, only one adaptation is shown here The method of mapping the possibility influence coefficient to a specific bee venom allergy possibility within the allergy possibility range can adaptively adjust the calculation method of this part when the possibility influence coefficient is represented by other calculation methods and other value ranges. The specific methods are not listed one by one here.
[0045] It should be understood that the calculation coefficients in the method model can be configured or trained based on professional knowledge, data experience and expert experience, so that the method model can more accurately analyze and determine the possibility of bee venom allergy in the target patient based on basic information, detection data of multiple bee venom proteins and expert experience knowledge base.
[0046] In summary, this method can comprehensively consider the patient's personalized data, expert experience, and historical treatment experience to intelligently and accurately analyze the possibility of bee venom allergy, so as to provide better reference and support for doctors' diagnosis and treatment work.
[0047] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity of description, 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 embodiments of 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 optional embodiments, and the actions and modules involved are not necessarily required by the present application.
[0048] In a second aspect, the present application discloses a system for auxiliary analysis of the possibility of bee venom allergy based on detection data. The system is used to perform the method disclosed in the first aspect above, and the system may include a server or medical system of a hospital (such as an HIS system, a CIS system, etc.), or be implemented as a server or medical system of a hospital (such as an HIS system, a CIS system, etc.).
[0049] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described device can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.
[0050] In summary, this application at least has the following beneficial effects: A method and system for auxiliary analysis of the possibility of bee venom allergy based on test data are provided, which can intelligently analyze the possibility of bee venom allergy of the target patient by comprehensively considering the specific situation of the target patient, historical patient data and expert treatment experience, which is conducive to providing more accurate analysis results of the possibility of bee venom allergy and providing better support and reference for doctors' diagnostic work.
[0051] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.
Claims
1. A method for auxiliary analysis of the possibility of bee venom allergy based on detection data, characterized in that: include: Obtaining basic information of the target patient and test data of multiple types of melittin, and obtaining historical patient data, wherein the historical patient data carries a timestamp and includes basic information of the historical patient, test data of multiple types of melittin, and conclusions on the allergy to bee venom; Determine similar patient data in the historical patient data based on the basic information of the target patient, wherein the similar patient data is the historical patient data and the target patient's basic information has information similarity higher than a first similarity threshold, and the data similarity of the detection data of multiple types of melittin has data similarity higher than a second similarity threshold; Substituting the basic information of the target patient and the detection data of various melittin proteins into a pre-acquired possibility range assessment model to obtain the possibility range of allergy of the target patient, wherein the possibility range assessment model is pre-constructed based on an expert experience knowledge base; The likelihood of bee venom allergy in the target patient is determined within the range of allergy likelihood by combining the similar patient data.
2. The method according to claim 1, characterized in that Determining similar patient data in the historical patient data based on the basic information of the target patient includes: The basic information is constructed as a first vector, and the detection data of multiple types of melittin are constructed as a second vector; Determine that the vector similarity between the first vectors of the historical patient and the target patient is the information similarity, and the vector similarity between the second vectors is the data similarity; The historical patient data whose vector similarity is higher than a first similarity threshold and whose data similarity is higher than a second similarity threshold is determined as the similar patient data.
3. The method according to claim 1, characterized in that The possibility range evaluation model includes one or more of a decision tree model, a knowledge graph model and a neural network model.
4. The method according to any one of claims 1 to 3, characterized in that: The step of combining the similar patient data to determine the possibility of bee venom allergy in the target patient within the range of allergy possibility comprises: Analyzing the data reference of each similar patient data, wherein the data reference is associated with one or more of a timestamp carried by the similar patient data, information similarity with basic information of the target patient, and data similarity with multiple melittin detection data of the target patient; Determining a data base value for each similar patient data, the data base value being associated with a bee venom allergy conclusion; Calculate the likelihood influence coefficient based on the data reference degree, the data base value, and the number of similar data of similar patient data; The allergy possibility of the target patient is determined within the allergy possibility range according to the possibility influence coefficient.
5. The method according to claim 4, characterized in that The data reference of analyzing each similar patient data includes: Assume that the timestamp carried by the i-th similar patient data is , and the information similarity with the basic information of the target patient is , and the data similarity with the target patient's various bee venom protein detection data is , the data reference is , the current time is , the first similarity threshold is , the second similarity threshold is ,but , where All are preset calculation coefficients greater than zero.
6. The method according to claim 4, characterized in that Determining the data base value of each similar patient data includes: The conclusion of bee venom allergy of similar patient data is bee venom allergy or non-bee venom allergy. The bee venom allergy also carries allergy degree data. Let the data base value of the i-th similar patient data be 、Allergy degree data is If the conclusion of bee venom allergy is non-bee venom allergy, then ,otherwise .
7. The method according to claim 4, characterized in that The calculation of the possibility influence coefficient according to the data reference degree, the data base value and the number of similar data of similar patient data includes: Assume that there are n similar patient data and the probability influence coefficient is k, then ； ； ； ； ； In the formula, are all preset calculation coefficients greater than zero and , To preset the allergy level, is the preset reference degree, for The standard deviation of for The standard deviation of .
8. The method according to claim 4, characterized in that Determining the allergy possibility of the target patient within the allergy possibility range according to the possibility influence coefficient includes: Assume that the range of allergy possibility is , the probability influence coefficient is k and , the target patient's probability of bee venom allergy, then .
9. A system for assisting analysis of the possibility of bee venom allergy based on detection data, characterized in that: Application of the method as claimed in any one of claims 1 to 8.