Method, device, electronic device and medium for predicting the disease development trend of rare diseases

Abstract

The present invention relates to the field of Internet technologies, and particularly to a method, device, electronic device, and medium for predicting the development trend of rare diseases. The method includes: dividing all physical examination data according to the data characteristics of all physical examination data within the current time period to obtain physical examination data within multiple target time periods; respectively inputting the physical examination data within each target time period into a disease index determination model corresponding to the rare disease type to obtain the disease index corresponding to each target time period; generating a target trend line graph with the target time period as the abscissa and the corresponding disease index as the ordinate; calculating the similarity between the target trend line graph and each control trend line graph; and determining the development trend of the rare disease patient for the rare disease type according to the similarity. Through the present application, the development trend of rare diseases can be accurately predicted.

Description

Technical Field

[0001] The present invention relates to the field of Internet technologies, and in particular, to a method, device, electronic device, and medium for predicting the development trend of rare diseases. Background Art

[0002] Currently, for rare diseases, clinicians receive a relatively small number of related patients, and the accumulation of experience is limited. For clinicians with insufficient experience, it is difficult to accurately predict the condition of patients, and it is impossible to take treatment measures for patients in advance. Many patients miss the best treatment time, increasing the treatment time and cycle.

[0003] Therefore, how to predict the development trend of rare diseases has become an urgent problem in this field. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, device, electronic device, and medium for predicting the development trend of rare diseases, which can accurately predict the development trend of rare diseases.

[0005] In a first aspect, an embodiment of this application provides a method for predicting the development trend of rare diseases, and the method includes:

[0006] Obtain the rare disease type corresponding to the target rare disease patient and the physical examination data within the current time period;

[0007] For each time point within the current time period, determine whether there is a missing physical examination item corresponding to this time point according to the physical examination data at this time point and the set of physical examination items corresponding to the rare disease type; a missing physical examination item refers to a physical examination item without physical examination data;

[0008] If there is a missing physical examination item corresponding to this time point, then based on the association relationship of physical examination items corresponding to the rare disease type and the physical examination data at this time point, determine the supplementary weight corresponding to each missing physical examination item at this time point; the association relationship of physical examination items includes other physical examination items associated with each physical examination item;

[0009] Determine the missing physical examination item with the largest supplementary weight as the target missing physical examination item at the time point;

[0010] In the physical examination data at this time point, input the physical examination data corresponding to other physical examination items associated with the target missing physical examination item at this time point into the supplementary model corresponding to the rare disease type and the target missing physical examination item, and obtain the physical examination data corresponding to the target missing physical examination item at this time point; and jump to determine whether there is a missing physical examination item corresponding to this time point according to the physical examination data at this time point and the set of physical examination items corresponding to the rare disease type to continue execution;

[0011] Divide all the physical examination data according to the data characteristics of all the physical examination data in the current time period to obtain the physical examination data in multiple target time periods;

[0012] Input the physical examination data in each target time period into the disease index determination model corresponding to the rare disease type respectively to obtain the disease index corresponding to each target time period; the disease index determination model is trained based on the physical examination data samples and the corresponding disease index labels;

[0013] Generate a target trend line graph with the target time period as the abscissa and the corresponding disease index as the ordinate;

[0014] Calculate the similarity between the target trend line graph and each control trend line graph; the control trend line graph is generated based on the disease indices corresponding to the patients of this rare disease type in multiple time periods;

[0015] Determine the disease development trend of the rare disease patient for this rare disease type according to the similarity.

[0016] In a possible implementation manner, determine the association relationship of the physical examination items corresponding to this rare disease type through the following steps:

[0017] Obtain the historical physical examination data corresponding to each physical examination item in the physical examination item set for multiple patients corresponding to this rare disease type at each historical moment;

[0018] Calculate the initial correlation coefficient between every two physical examination items corresponding to each patient according to all the historical physical examination data corresponding to each patient;

[0019] For every two physical examination items, calculate the average value of the initial correlation coefficients between the two physical examination items corresponding to all patients to obtain the target correlation coefficient between the two physical examination items;

[0020] If the target correlation coefficient between two physical examination items is greater than or equal to the preset correlation coefficient, there is an association relationship between the two physical examination items.

[0021] In a possible implementation manner, calculating the initial correlation coefficient between every two physical examination items corresponding to each patient according to all the historical physical examination data corresponding to each patient includes:

[0022] Substitute the historical physical examination data of every two physical examination items corresponding to this patient into the following formula to calculate the initial correlation coefficient between every two physical examination items corresponding to this patient:

[0023] ;

[0024] Wherein, is the initial correlation coefficient between the historical physical examination data of the first physical examination item X and the historical physical examination data of the second physical examination item Y among the two physical examination items corresponding to the patient. is the distance covariance between the historical physical examination data of the first physical examination item X and the historical physical examination data of the second physical examination item Y corresponding to the patient. is the distance variance of the historical physical examination data of the first physical examination item X corresponding to the patient. is the distance variance of the historical physical examination data of the second physical examination item Y corresponding to the patient.

[0025] In a possible implementation manner, all physical examination data are divided according to the data characteristics of all physical examination data within the current time period, and the physical examination data within multiple target time periods are obtained, including:

[0026] All physical examination data are divided according to a preset division duration to obtain the physical examination data within multiple initial time periods;

[0027] For each initial time period, if the variance of the physical examination data corresponding to any physical examination item within the initial time period is greater than the preset variance, the physical examination data within the initial time period are divided according to the physical examination data corresponding to the physical examination item with the largest variance within the initial time period to obtain the physical examination data within the target time period;

[0028] If the variance of the physical examination data corresponding to all physical examination items within the initial time period is less than or equal to the preset variance, the physical examination data within the initial time period are determined as the physical examination data within the target time period.

[0029] In a possible implementation manner, the physical examination data within the initial time period are divided according to the physical examination data corresponding to the physical examination item with the largest variance within the initial time period to obtain the physical examination data within the target time period, including:

[0030] If the time point where the maximum value in the physical examination data corresponding to the physical examination item with the largest variance is located is not the latest time point within the initial time period and is not the earliest time point within the initial time period, the time point where the maximum value is located is determined as the split time point;

[0031] If the time point where the minimum value in the physical examination data corresponding to the physical examination item with the largest variance is located is not the latest time point within the initial time period and is not the earliest time point within the initial time period, the time point where the minimum value is located is determined as the split time point;

[0032] The physical examination data within the initial time period are divided at the split time point to obtain the physical examination data within the target time period;

[0033] If the time point corresponding to the maximum value in the physical examination data of the physical examination item with the largest variance is the latest time point or the earliest time point within the initial time period, and the time point corresponding to the minimum value in the physical examination data of the physical examination item with the largest variance is the latest time point or the earliest time point within the initial time period, then the physical examination data within the initial time period is evenly split to obtain the physical examination data within two target time periods.

[0034] In a possible implementation manner, determining the disease development trend of the rare disease patient for the rare disease type according to the similarity includes:

[0035] If the maximum value of the similarity is greater than or equal to the preset similarity, then determine the actual disease development trend corresponding to the control trend line with the largest similarity as the disease development trend corresponding to the rare disease patient;

[0036] If the maximum value of the similarity is less than the preset similarity and there is a target point with a tangent slope of 0 in the target trend line, then calculate the difference between the disease index corresponding to the target point with the latest time and the disease index of the point with the latest time in the target trend line;

[0037] If the difference is greater than 0, then the disease development trend corresponding to the rare disease patient is a trend of improvement;

[0038] If the difference is less than 0, then the disease development trend corresponding to the rare disease patient is a trend of deterioration.

[0039] In a possible implementation manner, if the maximum value of the similarity is less than the preset similarity, then the method further includes:

[0040] Determine the target trend line as the control trend line.

[0041] In a second aspect, an embodiment of the present application further provides a device for predicting the disease development trend of a rare disease, and the device includes:

[0042] An acquisition module, configured to acquire the rare disease type corresponding to the target rare disease patient and the physical examination data within the current time period;

[0043] A judgment module, configured to, for each time point within the current time period, judge whether there is a missing physical examination item corresponding to the time point according to the physical examination data at the time point and the set of physical examination items corresponding to the rare disease type; a missing physical examination item refers to a physical examination item without physical examination data;

[0044] A determination module, configured to, if there is a missing physical examination item corresponding to the time point, determine the supplementary weight corresponding to each missing physical examination item at the time point based on the association relationship of physical examination items corresponding to the rare disease type and the physical examination data at the time point; the association relationship of physical examination items includes other physical examination items associated with each physical examination item;

[0045] The determination module is further configured to determine the missing physical examination item with the largest supplementary weight as the target missing physical examination item at this time point;

[0046] The input module is configured to input, within the physical examination data at this time point, the physical examination data corresponding to other physical examination items associated with the target missing physical examination item at this time point into the supplementary model corresponding to the rare disease type and the target missing physical examination item, to obtain the physical examination data corresponding to the target missing physical examination item at this time point; and jump to determine whether there is a missing physical examination item corresponding to the time point according to the physical examination data at this time point and the set of physical examination items corresponding to the rare disease type to continue execution;

[0047] The partitioning module is configured to partition all the physical examination data according to the data characteristics of all the physical examination data within the current time period, to obtain the physical examination data within multiple target time periods;

[0048] The input module is further configured to input the physical examination data within each target time period into the disease index determination model corresponding to the rare disease type respectively, to obtain the disease index corresponding to each target time period; the disease index determination model is trained based on the physical examination data samples and the corresponding disease index labels;

[0049] The generation module is configured to generate a target trend line graph with the target time period as the abscissa and the corresponding disease index as the ordinate;

[0050] The calculation module is configured to calculate the similarity between the target trend line graph and each control trend line graph; the control trend line graph is generated based on the disease indices corresponding to the patients with the rare disease type in multiple time periods;

[0051] The determination module is further configured to determine the disease development trend of the rare disease patient for the rare disease type according to the similarity.

[0052] In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of the method for predicting the disease development trend of any rare disease in the first aspect.

[0053] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, it performs the steps of the method for predicting the disease development trend of any rare disease in the first aspect.

[0054] The embodiments of the present application provide a method, device, electronic device and medium for predicting the disease development trend of rare diseases. The method includes: dividing all physical examination data according to the data characteristics of all physical examination data in the current time period to obtain physical examination data in multiple target time periods; respectively inputting the physical examination data in each target time period into a disease index determination model corresponding to the rare disease type to obtain the disease index corresponding to each target time period; generating a target trend line graph with the target time period as the abscissa and the corresponding disease index as the ordinate; calculating the similarity between the target trend line graph and each control trend line graph; and determining the disease development trend of the rare disease patient for the rare disease type according to the similarity. Through the present application, the disease development trend of rare diseases can be accurately predicted. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0056] Figure 1 The flowchart showing a method for predicting the disease development trend of a rare disease provided by an embodiment of the present application;

[0057] Figure 2 The flowchart showing the division of physical examination data provided by an embodiment of the present application;

[0058] Figure 3 The structural schematic diagram showing a device for predicting the disease development trend of a rare disease provided by an embodiment of the present application;

[0059] Figure 4 The structural schematic diagram showing an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of illustration and description, and are not used to limit the protection scope of the present application. In addition, it should be understood that the schematic drawings are not drawn to actual scale. The flowcharts used in the present application show the operations implemented according to some embodiments of the present application. It should be understood that the operations in the flowchart may not be implemented in sequence, and steps without logical context relationships may be reversed or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart or remove one or more operations from the flowchart under the guidance of the content of the present application.

[0061] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all embodiments. The components of the embodiments of the present application described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application.

[0062] In order to enable those skilled in the art to use the content of the present application, the following implementation manners are given in combination with a specific application scenario, the "Internet technology field". For those skilled in the art, without departing from the spirit and scope of the present application, the general principles defined here can be applied to other embodiments and application scenarios. Although the present application is mainly described around the "Internet technology field", it should be understood that this is only an exemplary embodiment.

[0063] It should be noted that the term "including" will be used in the embodiments of the present application to indicate the existence of the features stated thereafter, but does not exclude adding other features.

[0064] The following provides a detailed description of a method for predicting the disease development trend of a rare disease provided by an embodiment of the present application.

[0065] Refer to Figure 1 As shown, it is a schematic flowchart of a method for a target rare disease patient provided by an embodiment of the present application. The following describes each step of the embodiment of the present application exemplarily:

[0066] S101. Obtain the rare disease type corresponding to the target rare disease patient and the physical examination data within the current time period.

[0067] In the implementation manner of the present application, the target rare disease patient refers to a patient for whom the disease development trend of a rare disease needs to be predicted. The rare disease type corresponding to the target rare disease patient refers to the rare disease type suffered by the target rare disease patient (such as hemophilia, myasthenia gravis, neuromyelitis optica, etc.). The duration corresponding to the current time period can be one week, half a month, etc., and can be determined according to the actual situation. The physical examination data can be uploaded by the target rare disease patient to the physical examination database of the medical decision-making assistance system during the doctor's follow-up, or can be uploaded by the doctor to the physical examination database of the medical decision-making assistance system during the follow-up of the target rare disease patient. Different rare disease types correspond to different sets of physical examination items. The physical examination data within the current time period should be the physical examination values of the physical examination items (such as blood pressure, blood sugar, etc.) in the set of physical examination items corresponding to the rare disease type.

[0068] Among them, the medical decision-making assistance system is a software development system that can be used to execute the rare disease condition development trend prediction method provided by the embodiments of the present application. Multiple data sets from different sources are stored in the medical decision-making assistance system (the data can include medical information, papers, rare disease treatment data of patients, etc.), with multiple data types (text data, tabular data, images, videos, etc.), multiple data sources (medical devices, user input, etc.), multiple data owners (patients, rare disease medical experts, etc.), multiple time points, multiple regions, etc. Therefore, this system is a medical decision-making assistance system based on multi-source data.

[0069] S102. For each time point within the current time period, determine whether there is a missing physical examination item corresponding to this time point according to the physical examination data at this time point and the set of physical examination items corresponding to the rare disease type; a missing physical examination item refers to a physical examination item without physical examination data.

[0070] In the implementation manner of the present application, a physical examination item without corresponding physical examination data in the set of physical examination items corresponding to the rare disease type is a missing physical examination item. For example, the set of physical examination items corresponding to the rare disease type includes physical examination item 1, physical examination item 2, and physical examination item 3, and the physical examination data corresponding to this time point includes the physical examination data of physical examination item 1 and the physical examination data of physical examination item 2. Therefore, the missing physical examination item corresponding to the physical examination data at this time point is physical examination item 3.

[0071] S103. If there is a missing physical examination item corresponding to this time point, then based on the association relationship of physical examination items corresponding to the rare disease type and the physical examination data at this time point, determine the supplementary weight corresponding to each missing physical examination item at this time point.

[0072] In the implementation manner of the present application, different rare disease types correspond to different association relationships of physical examination items. The association relationship of physical examination items contains other physical examination items associated with each physical examination item, and the number of other physical examination items associated with each physical examination item can be one or more. Specifically, for each missing physical examination item, count the number of other physical examination items associated with this missing physical examination item and corresponding to physical examination data at this time point according to the physical examination data at this time point, and obtain the supplementary weight corresponding to this missing physical examination item; among them, the physical examination data of other physical examination items can be obtained from the physical examination database in step S101, or can be supplemented and obtained through a supplementary model; the larger the value corresponding to the supplementary weight, the more accurate the physical examination data supplemented and obtained through the supplementary model.

[0073] In addition, the association relationship of physical examination items corresponding to the rare disease type is determined through the following steps:

[0074] Step 1. Obtain the historical physical examination data corresponding to each physical examination item in the set of physical examination items for multiple patients corresponding to the rare disease type at each historical moment.

[0075] Example: The set of physical examination items corresponding to this rare disease type includes physical examination item 1, physical examination item 2, and physical examination item 3. Obtain the physical examination data of multiple patients with this rare disease type at multiple historical moments from the physical examination database. The physical examination data of each patient at each historical moment includes the physical examination value corresponding to physical examination item 1 for this patient at this historical moment, the physical examination value corresponding to physical examination item 2 for this patient at this historical moment, and the physical examination value corresponding to physical examination item 3 for this patient at this historical moment.

[0076] Step 2: Calculate the initial correlation coefficient between every two physical examination items corresponding to each patient based on all the historical physical examination data corresponding to each patient.

[0077] In the embodiment of the present application, for each patient, it is necessary to calculate the initial correlation coefficient between every two physical examination items corresponding to this patient based on the historical physical examination data of this patient. The greater the initial correlation coefficient, the stronger the correlation between the two physical examination items corresponding to this patient.

[0078] Example: There are patient A and patient B. The set of physical examination items includes physical examination item 1, physical examination item 2, and physical examination item 3. Then, for patient A, calculate the initial correlation coefficient between physical examination item 1 and physical examination item 2 corresponding to patient A based on the historical physical examination data of patient A. Calculate the initial correlation coefficient between physical examination item 2 and physical examination item 3 corresponding to patient A based on the historical physical examination data of patient A. Calculate the initial correlation coefficient between physical examination item 1 and physical examination item 3 corresponding to patient A based on the historical physical examination data of patient A. For patient B, calculate the initial correlation coefficient between physical examination item 1 and physical examination item 2 corresponding to patient B based on the historical physical examination data of patient B. Calculate the initial correlation coefficient between physical examination item 2 and physical examination item 3 corresponding to patient B based on the historical physical examination data of patient B. Calculate the initial correlation coefficient between physical examination item 1 and physical examination item 3 corresponding to patient B based on the historical physical examination data of patient B.

[0079] Specifically, substitute the historical physical examination data of every two physical examination items corresponding to the patient into the following formula to calculate the initial correlation coefficient between every two physical examination items corresponding to the patient:

[0080] ;

[0081] ;

[0082] ;

[0083] ;

[0084] ; ;

[0085] ; ;

[0086] ;

[0087] ;

[0088] Wherein, is the initial correlation coefficient between the historical physical examination data of the first physical examination item X and the historical physical examination data of the second physical examination item Y among the two physical examination items corresponding to the patient, is the distance covariance between the historical physical examination data of the first physical examination item X and the historical physical examination data of the second physical examination item Y corresponding to the patient, is the distance variance of the historical physical examination data of the first physical examination item X corresponding to the patient, is the distance variance of the historical physical examination data of the second physical examination item Y corresponding to the patient; is the value corresponding to the i-th row and j-th column in the double-centered distance matrix of the historical physical examination data of the first physical examination item X corresponding to the patient; is the value corresponding to the i-th row and j-th column in the double-centered distance matrix of the historical physical examination data of the second physical examination item Y corresponding to the patient; is the value corresponding to the i-th row and j-th column in the distance matrix of the historical physical examination data of the first physical examination item X corresponding to the patient, is the value corresponding to the i-th row and j-th column in the distance matrix of the historical physical examination data of the second physical examination item Y corresponding to the patient; is the average value of the values corresponding to the i-th row in the distance matrix of the historical physical examination data of the first physical examination item X corresponding to the patient, is the average value of the values corresponding to the i-th row in the distance matrix of the historical physical examination data of the second physical examination item Y corresponding to the patient, is the average value of the values corresponding to the j-th column in the distance matrix of the historical physical examination data of the first physical examination item X corresponding to the patient, is the average value of the values corresponding to the j-th column in the distance matrix of the historical physical examination data of the second physical examination item Y corresponding to the patient, is the i-th historical physical examination data of the first physical examination item X corresponding to the patient, is the j-th historical physical examination data of the first physical examination item X corresponding to the patient, The i-th historical physical examination data of the second physical examination item Y corresponding to the patient, is the j-th historical physical examination data of the second physical examination item Y corresponding to the patient, and n is the number of elements in the distance matrix of the historical physical examination data of the first physical examination item X or the number of elements in the distance matrix of the historical physical examination data of the second physical examination item Y.

[0089] Step 3: For every two physical examination items, calculate the average value of the initial correlation coefficients between the two physical examination items corresponding to all patients, and obtain the target correlation coefficient between the two physical examination items.

[0090] For example, there are patient A and patient B. For physical examination item 1 and physical examination item 2, add the initial correlation coefficient between physical examination item 1 and physical examination item 2 corresponding to patient A and the initial correlation coefficient between physical examination item 1 and physical examination item 2 corresponding to patient B to obtain a sum value, and divide the sum value by 2 to obtain the target correlation coefficient between physical examination item 1 and physical examination item 2.

[0091] Step 4: If the target correlation coefficient between the two physical examination items is greater than or equal to the preset correlation coefficient, there is an association relationship between the two physical examination items.

[0092] Step 5: If the target correlation coefficient between the two physical examination items is less than the preset correlation coefficient, there is no association relationship between the two physical examination items.

[0093] S104: Determine the missing physical examination item with the largest supplementary weight as the target missing physical examination item at this time point.

[0094] S105: In the physical examination data at this time point, input the physical examination data corresponding to other physical examination items associated with the target missing physical examination item at this time point into the supplementary model corresponding to the rare disease type and the target missing physical examination item to obtain the physical examination data corresponding to the target missing physical examination item at this time point; and jump to determine whether there is a missing physical examination item corresponding to this time point according to the physical examination data at this time point and the set of physical examination items corresponding to the rare disease type to continue execution.

[0095] In the embodiment of the present application, different rare disease types and target missing physical examination items correspond to different supplementary models, and the supplementary model is used to supplement the physical examination data corresponding to the target missing physical examination item. The supplementary model is a deep learning model pre-trained through the physical examination data samples corresponding to other physical examination items associated with the missing physical examination item at the same time of the same patient and the corresponding physical examination data labels. The input is the physical examination data corresponding to other physical examination items associated with the missing physical examination item, and the output is the physical examination data corresponding to the missing physical examination item.

[0096] Here, to avoid inaccurate prediction of the subsequent disease development trend due to the possible incompleteness of the physical examination data stored in the physical examination database, the present application continues to supplement the physical examination data. In addition, after the present application supplements the physical examination data corresponding to each missing physical examination item, it re-determines the supplementary weights of other missing physical examination items to continue supplementing the missing physical examination item with the largest supplementary weight, improving the accuracy of the supplemented physical examination data.

[0097] S106. Divide all the physical examination data according to the data characteristics of all the physical examination data in the current time period to obtain the physical examination data in multiple target time periods.

[0098] In the embodiment of the present application, all the physical examination data in the current time period includes the data obtained from the physical examination database in step S101 and supplemented by the supplementary model. The numerical dispersion degree of the physical examination data in the target time period should meet the preset dispersion degree condition, and the specific preset dispersion degree condition can be determined according to the actual situation, such as being set based on methods such as average, variance or average difference, so that the disease index corresponding to the subsequent determined target time period can represent the overall condition of the target time period.

[0099] Specifically, referring to Figure 2 As shown, it is a flow chart of dividing physical examination data provided by the embodiment of the present application. The following will explain each step of the embodiment of the present application exemplarily:

[0100] S201. Divide all the physical examination data according to the preset division duration to obtain the physical examination data in multiple initial time periods.

[0101] In the embodiment of the present application, the preset duration can be determined according to the actual situation, such as one day, two days, etc. By dividing according to the preset duration, the numerical dispersion degree of the physical examination data in each time period can be effectively reduced.

[0102] For example, if the physical examination data in the current time period is the physical examination data within one week, then dividing according to the preset duration of one day, the physical examination data from the 0th day to the 1st day, the physical examination data from the 1st day to the 2nd day, the physical examination data from the 2nd day to the 3rd day, the physical examination data from the 3rd day to the 4th day, the physical examination data from the 4th day to the 5th day, the physical examination data from the 5th day to the 6th day, and the physical examination data from the 6th day to the 7th day can be obtained.

[0103] S202. For each initial time period, if the variance of the physical examination data corresponding to any physical examination item in the initial time period is greater than the preset variance, then divide the physical examination data in the initial time period according to the physical examination data corresponding to the physical examination item with the largest variance in the initial time period to obtain the physical examination data in the target time period.

[0104] Specifically, divide the physical examination data in the initial time period according to the physical examination data corresponding to the physical examination item with the largest variance in the initial time period through the following steps:

[0105] a. If the time point where the maximum value in the physical examination data corresponding to the physical examination item with the largest variance is located is not the latest time point in the initial time period and is not the earliest time point in the initial time period, then determine the time point where the maximum value is located as the split time point.

[0106] Example: The physical examination item with the largest variance in the initial time period is physical examination item 2. The physical examination data in the initial time period includes the physical examination data at 3:00, 4:00, 7:00, 8:00, 9:00, and 12:00. The maximum value in the physical examination data corresponding to physical examination item 2 is the physical examination data at 4:00. Since 4:00 is neither the latest time 12:00 nor the earliest time 3:00 corresponding to the physical examination data in the initial time period, the time point where the maximum value is located is determined as the split time point.

[0107] b. If the time point where the minimum value in the physical examination data corresponding to the physical examination item with the largest variance is located is neither the latest time point nor the earliest time point in the initial time period, then the time point where the minimum value is located is determined as the split time point.

[0108] Example: The physical examination item with the largest variance in the initial time period is physical examination item 2. The physical examination data in the initial time period includes the physical examination data at 3:00, 4:00, 7:00, 8:00, 9:00, and 12:00. The minimum value in the physical examination data corresponding to physical examination item 2 is the physical examination data at 8:00. Since 8:00 is neither the latest time 12:00 nor the earliest time 3:00 corresponding to the physical examination data in the initial time period, the time point where the minimum value is located is determined as the split time point.

[0109] c. Divide the physical examination data in the initial time period at the split time point to obtain the physical examination data in the target time period.

[0110] Example: The physical examination data in the initial time period includes the physical examination data at 3:00, 4:00, 7:00, 8:00, 9:00, and 12:00. The split time points are 4:00 and 8:00. Then the obtained target time periods include 3:00 to 4:00, 4:00 to 8:00, and 8:00 to 12:00.

[0111] d. If the time point where the maximum value in the physical examination data corresponding to the physical examination item with the largest variance is located is the latest time point or the earliest time point in the initial time period, and the time point where the minimum value in the physical examination data corresponding to the physical examination item with the largest variance is located is the latest time point or the earliest time point in the initial time period, then the physical examination data in the initial time period is evenly split to obtain the physical examination data in two target time periods.

[0112] Example: The medical examination item with the largest variance in the initial time period is medical examination item 2. The medical examination data in the initial time period includes the medical examination data at 3:00, 4:00, 7:00, 8:00, 9:00, and 12:00. The maximum value of the medical examination data corresponding to medical examination item 2 is the medical examination data at 3:00, and the minimum value of the medical examination data corresponding to medical examination item 2 is the medical examination data at 12:00. Then the target time periods are from 3:00 to 7:00 and from 8:00 to 12:00.

[0113] S203: If the variance of the medical examination data corresponding to all medical examination items in the initial time period is less than or equal to the preset variance, then determine the medical examination data in the initial time period as the medical examination data in the target time period.

[0114] S107: Input the medical examination data in each target time period into the disease index determination model corresponding to the rare disease type to obtain the disease index corresponding to each target time period; the disease index determination model is trained based on the medical examination data samples and the corresponding disease index labels.

[0115] In the embodiment of the present application, each rare disease type corresponds to a disease index determination model. The disease index determination model is a deep learning model pre-trained through the medical examination data samples of the patients corresponding to the rare disease type in the target time period sample and the corresponding disease index labels. The input is the medical examination data sample, and the output is the disease index. The higher the disease index, the deeper the degree of the patient's illness.

[0116] S108: Generate a target trend line graph with the target time period as the abscissa and the corresponding disease index as the ordinate.

[0117] S109: Calculate the similarity between the target trend line graph and each control trend line graph; the control trend line graph is generated based on the disease indexes corresponding to the patients of the rare disease type in multiple time periods.

[0118] In the embodiment of the present application, each control trend line graph corresponds to an actual disease development trend; the disease development trend includes a trend of improvement, a trend of deterioration, and a trend of stability; and the trend of improvement also includes a slow improvement trend and a rapid improvement trend; the trend of deterioration also includes a slow deterioration trend and a rapid deterioration trend. The control trend line graph is generated based on the disease indexes corresponding to the patients of the rare disease type in each historical time period; the actual disease development trend corresponding to the control trend line graph is determined manually based on the actual disease index development of the patients within the preset time period after each time period.

[0119] Calculate the similarity between the target trend line graph and each control trend line graph through the following steps:

[0120] Step 1. If the length of the target trend line is greater than that of the control trend line, cut off the front part of the curve of the target trend line so that the remaining target trend line has the same length as the control trend line.

[0121] Step 2. If the length of the target trend line is less than that of the control trend line, cut off the front part of the curve of the control trend line so that the remaining target trend line has the same length as the control trend line.

[0122] Step 3. According to the following formula, calculate the similarity between the target trend line and each control trend line based on the target trend line and the control trend line after cutting.

[0123] ;

[0124] where is the similarity between the target trend line and the control trend line, is the length of the target trend line, is the disease index of the point with abscissa i in the target trend line after cutting, is the disease index of the point with abscissa i in the control trend line.

[0125] Step 4. If the length of the target trend line is equal to that of the control trend line, calculate the similarity between the target trend line and the control trend line according to the following formula.

[0126] ;

[0127] where is the similarity between the target trend line and the control trend line, is the length of the target trend line, is the disease index of the point with abscissa i in the target trend line, is the disease index of the point with abscissa i in the control trend line.

[0128] S110. Determine the disease development trend corresponding to the rare disease patient according to the similarity.

[0129] In the embodiment of the present application, the disease development trend corresponding to the rare disease patient refers to the disease development trend within a preset time period in the future.

[0130] Specifically, the disease development trend corresponding to the rare disease patient is determined through the following steps:

[0131] Step 1. If the maximum value of the similarity is greater than or equal to the preset similarity, determine the actual disease development trend corresponding to the control trend line with the maximum similarity as the disease development trend corresponding to the rare disease patient.

[0132] Step 2: If the maximum value of the similarity is less than the preset similarity and there is a target point with a tangent slope of 0 in the target trend line, then calculate the difference between the disease index corresponding to the target point with the latest time and the disease index of the point with the latest time in the target trend line.

[0133] Here, if the maximum value of the similarity is less than the preset similarity, the target trend line is determined as the control trend line, which can increase the number of control trend lines and improve the prediction accuracy.

[0134] Step 3: If the difference is greater than 0, the disease development trend corresponding to the rare disease patient is a trend of improvement.

[0135] In the embodiment of the present application, if the difference is greater than the first preset difference, the disease development trend corresponding to the rare disease patient is a rapid improvement trend; if the difference is greater than 0 and less than the first preset difference, the disease development trend corresponding to the rare disease patient is a slow improvement trend.

[0136] Step 4: If the difference is less than 0, the disease development trend corresponding to the rare disease patient is a trend of deterioration.

[0137] In the embodiment of the present application, if the difference is less than the second preset difference, the disease development trend corresponding to the rare disease patient is a rapid deterioration trend; if the difference is less than 0 and greater than the second preset difference, the disease development trend corresponding to the rare disease patient is a slow deterioration trend.

[0138] Step 5: If the difference is equal to 0, the disease development trend corresponding to the rare disease patient is a trend of stable condition.

[0139] The embodiment of the present application provides a method for predicting the disease development trend of rare diseases. The method includes: dividing all physical examination data according to the data characteristics of all physical examination data in the current time period to obtain the physical examination data in multiple target time periods; respectively inputting the physical examination data in each target time period into the disease index determination model corresponding to the rare disease type to obtain the disease index corresponding to each target time period; using the target time period as the abscissa and the corresponding disease index as the ordinate to generate a target trend line; calculating the similarity between the target trend line and each control trend line; and determining the disease development trend of the rare disease patient for the rare disease type according to the similarity. Through the present application, the disease development trend of rare diseases can be accurately predicted.

[0140] Based on the same inventive concept, in the embodiments of the present application, a device for predicting the disease development trend of rare diseases corresponding to the method for predicting the disease development trend of rare diseases is further provided. Since the principle of solving problems by the device in the embodiments of the present application is similar to that of the above-mentioned method for predicting the disease development trend of rare diseases in the embodiments of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0141] Referring to Figure 3 As shown in the figure, it is a schematic diagram of a device for predicting the disease development trend of rare diseases provided by an embodiment of the present application. The device for predicting the disease development trend of rare diseases includes:

[0142] An acquisition module 301, configured to acquire the rare disease type corresponding to the target rare disease patient and the physical examination data within the current time period;

[0143] A judgment module 302, configured to, for each time point within the current time period, judge whether there is a missing physical examination item corresponding to this time point according to the physical examination data at this time point and the set of physical examination items corresponding to the rare disease type; a missing physical examination item refers to a physical examination item without physical examination data;

[0144] A determination module 303, configured to, if there is a missing physical examination item corresponding to this time point, determine the supplementary weight corresponding to each missing physical examination item at this time point based on the physical examination item association relationship corresponding to the rare disease type and the physical examination data at this time point; the physical examination item association relationship includes other physical examination items associated with each physical examination item;

[0145] The determination module 303 is further configured to determine the missing physical examination item with the largest supplementary weight as the target missing physical examination item at this time point;

[0146] An input module 304, configured to input the physical examination data corresponding to other physical examination items associated with the target missing physical examination item at this time point into the supplementary model corresponding to the rare disease type and the target missing physical examination item within the physical examination data at this time point, to obtain the physical examination data corresponding to the target missing physical examination item at this time point; and jump to judge whether there is a missing physical examination item corresponding to the time point according to the physical examination data at this time point and the set of physical examination items corresponding to the rare disease type to continue execution;

[0147] A partitioning module 305, configured to partition all the physical examination data according to the data characteristics of all the physical examination data within the current time period, to obtain the physical examination data within multiple target time periods;

[0148] The input module 304 is further configured to input the physical examination data within each target time period into the disease index determination model corresponding to the rare disease type, to obtain the disease index corresponding to each target time period; the disease index determination model is trained based on the physical examination data samples and the corresponding disease index labels;

[0149] A generating module 306, configured to generate a target trend line graph with the target time period as the abscissa and the corresponding disease index as the ordinate;

[0150] A calculating module 307, configured to calculate the similarity between the target trend line graph and each control trend line graph; the control trend line graph is generated based on the disease indices corresponding to patients of a rare disease type in multiple time periods;

[0151] A determining module 303, further configured to determine the disease development trend of the rare disease patient for the rare disease type according to the similarity.

[0152] In a possible implementation manner, the determining module 303 is specifically configured to obtain the historical physical examination data corresponding to each physical examination item in the physical examination item set for multiple patients corresponding to the rare disease type at each historical moment; calculate the initial correlation coefficient between every two physical examination items corresponding to each patient according to all the historical physical examination data corresponding to each patient; for every two physical examination items, calculate the average value of the initial correlation coefficients between the two physical examination items corresponding to all patients to obtain the target correlation coefficient between the two physical examination items; if the target correlation coefficient between the two physical examination items is greater than or equal to the preset correlation coefficient, there is an association relationship between the two physical examination items.

[0153] In a possible implementation manner, the determining module 303 is specifically configured to substitute the historical physical examination data of every two physical examination items corresponding to the patient into the following formula to calculate the initial correlation coefficient between every two physical examination items corresponding to the patient:

[0154] ;

[0155] wherein, is the initial correlation term coefficient between the historical physical examination data of the first physical examination item X and the historical physical examination data of the second physical examination item Y among the two physical examination items corresponding to the patient, is the distance covariance between the historical physical examination data of the first physical examination item X and the historical physical examination data of the second physical examination item Y corresponding to the patient, is the distance variance of the historical physical examination data of the first physical examination item X corresponding to the patient, is the distance variance of the historical physical examination data of the second physical examination item Y corresponding to the patient.

[0156] In a possible implementation, the partitioning module 305 is specifically configured to partition all physical examination data according to a preset partitioning duration to obtain physical examination data within multiple initial time periods; for each initial time period, if the variance of the physical examination data corresponding to any physical examination item within the initial time period is greater than a preset variance, then partition the physical examination data within the initial time period according to the physical examination data corresponding to the physical examination item with the largest variance within the initial time period to obtain physical examination data within a target time period; if the variance of the physical examination data corresponding to all physical examination items within the initial time period is less than or equal to the preset variance, then determine the physical examination data within the initial time period as the physical examination data within the target time period.

[0157] In a possible implementation, the partitioning module 305 is specifically configured to: if the time point where the maximum value in the physical examination data corresponding to the physical examination item with the largest variance is located is not the latest time point within the initial time period and is not the earliest time point within the initial time period, then determine the time point where the maximum value is located as the splitting time point; if the time point where the minimum value in the physical examination data corresponding to the physical examination item with the largest variance is located is not the latest time point within the initial time period and is not the earliest time point within the initial time period, then determine the time point where the minimum value is located as the splitting time point; partition the physical examination data within the initial time period at the splitting time point to obtain physical examination data within a target time period; if the time point where the maximum value in the physical examination data corresponding to the physical examination item with the largest variance is the latest time point within the initial time period or the earliest time point within the initial time period, and the time point where the minimum value in the physical examination data corresponding to the physical examination item with the largest variance is the latest time point within the initial time period or the earliest time point within the initial time period, then evenly split the physical examination data within the initial time period to obtain physical examination data within two target time periods.

[0158] In a possible implementation, the determination module 303 is specifically configured to: if the maximum value of the similarity is greater than or equal to a preset similarity, then determine the actual disease development trend corresponding to the control trend line with the largest similarity as the disease development trend corresponding to the rare disease patient; if the maximum value of the similarity is less than the preset similarity and there is a target point with a tangent slope of 0 in the target trend line, then calculate the difference between the disease index corresponding to the target point with the latest time and the disease index of the point with the latest time in the target trend line; if the difference is greater than 0, then the disease development trend corresponding to the rare disease patient is a trend of improvement; if the difference is less than 0, then the disease development trend corresponding to the rare disease patient is a trend of deterioration.

[0159] In a possible implementation, if the maximum value of the similarity is less than the preset similarity, then the determination module 303 is further configured to: determine the target trend line as the control trend line.

[0160] The embodiments of the present application provide a device for predicting the development trend of rare diseases. Through the device of the present application, the development trend of rare diseases can be accurately predicted.

[0161] As Figure 4 shown, an electronic device 400 provided by an embodiment of the present application includes: a processor 401, a memory 402, and a bus. The memory 402 stores machine-readable instructions executable by the processor 401. When the electronic device runs, communication between the processor 401 and the memory 402 is carried out through the bus, and the processor 401 executes the machine-readable instructions to perform the steps of the method for predicting the development trend of rare diseases as described above.

[0162] Specifically, the above-mentioned memory 402 and processor 401 can be general-purpose memory and processor, which are not specifically limited here. When the processor 401 runs the computer program stored in the memory 402, it can execute the method for predicting the development trend of rare diseases as described above.

[0163] Corresponding to the above method for predicting the development trend of rare diseases, an embodiment of the present application also provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, it executes the steps of the method for predicting the development trend of rare diseases as described above.

[0164] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the method embodiments, which will not be elaborated in the present application. In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there can be other division methods in actual implementation. Also, for example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the couplings or direct couplings or communication connections shown or discussed with each other can be through some communication interfaces. The indirect couplings or communication connections of the devices or modules can be in electrical, mechanical, or other forms.

[0165] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they can be located in one place, or can be 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.

[0166] In addition, in each embodiment of the present application, each functional unit may be integrated into one processing unit, may exist physically alone for each unit, or two or more units may be integrated into one unit.

[0167] If the described function is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, may be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the information processing method described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0168] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for predicting the disease progression trend of a rare disease, characterized in that, The method includes: Obtaining the rare disease type corresponding to the target rare disease patient and the physical examination data within the current time period; For each time point within the current time period, determining whether there is a missing physical examination item corresponding to the time point according to the physical examination data at the time point and the set of physical examination items corresponding to the rare disease type; the missing physical examination item refers to a physical examination item without physical examination data; If there is a missing physical examination item corresponding to the time point, based on the association relationship of physical examination items corresponding to the rare disease type and the physical examination data at the time point, determining the supplementary weight corresponding to each missing physical examination item at the time point; the association relationship of physical examination items includes other physical examination items associated with each physical examination item; Determining the missing physical examination item with the largest supplementary weight as the target missing physical examination item at the time point; Among the physical examination data at the time point, inputting the physical examination data corresponding to other physical examination items associated with the target missing physical examination item at the time point into the supplementary model corresponding to the rare disease type and the target missing physical examination item to obtain the physical examination data corresponding to the target missing physical examination item at the time point; and jumping to determining whether there is a missing physical examination item corresponding to the time point according to the physical examination data at the time point and the set of physical examination items corresponding to the rare disease type to continue execution; Dividing all the physical examination data according to the data characteristics of all the physical examination data within the current time period to obtain the physical examination data within multiple target time periods; Inputting the physical examination data within each target time period into the disease index determination model corresponding to the rare disease type respectively to obtain the disease index corresponding to each target time period; the disease index determination model is trained based on physical examination data samples and corresponding disease index labels; Generating a target trend line graph with the target time period as the abscissa and the corresponding disease index as the ordinate; Calculating the similarity between the target trend line graph and each control trend line graph; the control trend line graph is generated based on the disease indices corresponding to the patients with the rare disease type in multiple time periods; Determining the disease development trend of the rare disease patient for the rare disease type according to the similarity.

2. The method for predicting the disease progression trend of rare diseases according to claim 1, wherein Determining the association relationship of physical examination items corresponding to the rare disease type through the following steps: Obtaining the historical physical examination data corresponding to each physical examination item in the set of physical examination items for multiple patients corresponding to the rare disease type at each historical moment; Calculating the initial correlation coefficient between every two physical examination items corresponding to each patient according to all the historical physical examination data corresponding to each patient; For every two physical examination items, calculating the average value of the initial correlation coefficients between the two physical examination items corresponding to all patients to obtain the target correlation coefficient between the two physical examination items; If the target correlation coefficient between the two physical examination items is greater than or equal to the preset correlation coefficient, there is an association relationship between the two physical examination items.

3. The method for predicting the disease progression trend of rare diseases according to claim 2, wherein The calculating the initial correlation coefficient between every two physical examination items corresponding to each patient according to all the historical physical examination data corresponding to each patient includes: Substituting the historical physical examination data of every two physical examination items corresponding to the patient into the following formula to calculate the initial correlation coefficient between every two physical examination items corresponding to the patient: ； wherein, is the initial correlation coefficient between the historical physical examination data of the first physical examination item X and the historical physical examination data of the second physical examination item Y among the two physical examination items corresponding to the patient, is the distance covariance between the historical physical examination data of the first physical examination item X and the historical physical examination data of the second physical examination item Y corresponding to the patient, is the distance variance of the historical physical examination data of the first physical examination item X corresponding to the patient, is the distance variance of the historical physical examination data of the second physical examination item Y corresponding to the patient.

4. The method for predicting the disease progression trend of rare diseases according to claim 1, characterized in that Dividing all the physical examination data according to the data characteristics of all the physical examination data within the current time period to obtain the physical examination data within multiple target time periods, including: Dividing all the physical examination data according to a preset division duration to obtain the physical examination data within multiple initial time periods; For each of the initial time periods, if the variance of the physical examination data corresponding to any physical examination item within the initial time period is greater than a preset variance, then dividing the physical examination data within the initial time period according to the physical examination data corresponding to the physical examination item with the largest variance within the initial time period to obtain the physical examination data within the target time period; If the variance of the physical examination data corresponding to all physical examination items within the initial time period is less than or equal to the preset variance, then determining the physical examination data within the initial time period as the physical examination data within the target time period.

5. The method for predicting the disease development trend of rare diseases according to claim 4, characterized in that, The dividing the physical examination data within the initial time period according to the physical examination data corresponding to the physical examination item with the largest variance within the initial time period to obtain the physical examination data within the target time period includes: If the time point where the maximum value in the physical examination data corresponding to the physical examination item with the largest variance is located is not the latest time point within the initial time period and is not the earliest time point within the initial time period, then determining the time point where the maximum value is located as the splitting time point; If the time point where the minimum value in the physical examination data corresponding to the physical examination item with the largest variance is located is not the latest time point within the initial time period and is not the earliest time point within the initial time period, then determining the time point where the minimum value is located as the splitting time point; Dividing the physical examination data within the initial time period at the splitting time point to obtain the physical examination data within the target time period; If the time point where the maximum value in the physical examination data corresponding to the physical examination item with the largest variance is the latest time point within the initial time period or the earliest time point within the initial time period, and the time point where the minimum value in the physical examination data corresponding to the physical examination item with the largest variance is the latest time point within the initial time period or the earliest time point within the initial time period, then evenly splitting the physical examination data within the initial time period to obtain the physical examination data within two target time periods.

6. The method for predicting the disease development trend of rare diseases according to claim 5, wherein The determining the disease development trend of the rare disease patient for the rare disease type according to the similarity includes: If the maximum value of the similarity is greater than or equal to a preset similarity, then determining the actual disease development trend corresponding to the control trend line with the largest similarity as the disease development trend corresponding to the target rare disease patient; If the maximum value of the similarity is less than the preset similarity and there is a target point with a tangent slope of 0 in the target trend line, then calculating the difference between the disease index corresponding to the target point with the latest time and the disease index of the point with the latest time in the target trend line; If the difference is greater than 0, then the disease development trend corresponding to the rare disease patient is a trend of improvement; If the difference is less than 0, then the disease development trend corresponding to the rare disease patient is a trend of deterioration.

7. The method for predicting the disease development trend of rare diseases according to claim 6, characterized in that If the maximum value of the similarity is less than the preset similarity, then the method further includes: Determine the target trend broken line as the control trend broken line.

8. A device for predicting the disease progression trend of a rare disease, characterized in that, The device includes: An acquisition module, configured to acquire the rare disease type corresponding to the target rare disease patient and the physical examination data within the current time period; A judgment module, configured to, for each time point within the current time period, determine whether there is a missing physical examination item corresponding to the time point according to the physical examination data at the time point and the set of physical examination items corresponding to the rare disease type; the missing physical examination item refers to a physical examination item without physical examination data; A determination module, configured to, if there is a missing physical examination item corresponding to the time point, determine the supplementary weight corresponding to each missing physical examination item at the time point based on the physical examination item association relationship corresponding to the rare disease type and the physical examination data at the time point; the physical examination item association relationship includes other physical examination items associated with each physical examination item; The determination module is further configured to determine the missing physical examination item with the largest supplementary weight as the target missing physical examination item at the time point; An input module, configured to input the physical examination data corresponding to other physical examination items associated with the target missing physical examination item at the time point into the supplementary model corresponding to the rare disease type and the target missing physical examination item within the physical examination data at the time point, to obtain the physical examination data corresponding to the target missing physical examination item at the time point; and jump to determining whether there is a missing physical examination item corresponding to the time point according to the physical examination data at the time point and the set of physical examination items corresponding to the rare disease type to continue execution; A division module, configured to divide all the physical examination data according to the data characteristics of all the physical examination data within the current time period, to obtain the physical examination data within multiple target time periods; The input module is further configured to input the physical examination data within each target time period into the disease index determination model corresponding to the rare disease type respectively, to obtain the disease index corresponding to each target time period; the disease index determination model is trained based on physical examination data samples and corresponding disease index labels; A generation module, configured to generate a target trend broken line with the target time period as the abscissa and the corresponding disease index as the ordinate; A calculation module, configured to calculate the similarity between the target trend broken line and each control trend broken line; the control trend broken line is generated based on the disease indexes corresponding to the patients corresponding to the rare disease type in multiple time periods; The determination module is further configured to determine the disease development trend of the rare disease patient for the rare disease type according to the similarity.

9. An electronic device, characterized in that, It includes: A processor, a storage medium and a bus, the storage medium stores machine-readable instructions executable by the processor, when the electronic device runs, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to execute the steps of the method for predicting the disease development trend of a rare disease according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, it executes the steps of the method for predicting the disease development trend of a rare disease according to any one of claims 1 to 7.

Images