Health data management method and system based on big data analysis

Abstract

The invention discloses a health data management method and system based on big data analysis, and belongs to the technical field of health management. Performing primary disease risk estimation through a preset first disease reasoning model to obtain a first estimated primary disease, and removing primary disease information of the patient from the first estimated primary disease to obtain a second estimated primary disease of the patient; whether the disease screening is carried out in the corresponding address area is judged according to the number of patients taking the disease as the primary disease and the number of patients taking the disease as the second estimated primary disease, so that the accurate screening of the disease target and the area target is realized, the singleness of the screening target disease is avoided, the effective range of the screening target disease is expanded, and the screening efficiency is improved. Meanwhile, the screening efficiency and the effective pertinence are improved.

Description

Technical Field

[0001] This invention relates to the field of health management technology, and in particular to health data management methods and systems based on big data analysis. Background Technology

[0002] Disease screening is the process of early detection of potential diseases or risk factors in healthy or asymptomatic individuals through medical testing. The goal is to achieve early diagnosis and treatment and reduce the risk of serious illness. Disease screening programs are based on the analysis of medical data to determine which diseases to screen for and which areas to screen for. Therefore, disease screening is an extremely important part of medical and health data management. Current disease screening relies solely on manually determining screening targets based on diagnosed patient disease data and address data, ignoring potentially hidden undiagnosed disease information. This results in a narrow range of disease targets and insufficient effectiveness, leading to inadequate targeting and efficiency in screening. Summary of the Invention

[0003] To address the technical problems existing in the prior art, this invention provides a health data management method based on big data analysis, comprising the following steps: S1. Obtain patient medical information, including diagnostic information, personal information, symptom information, and indicator information; S2. Obtain information on the primary disease, secondary disease, and concurrent disease of each patient based on the diagnostic information; S3. Based on the patient's secondary disease information, concurrent disease information, and the latest symptom information and indicator information, the primary disease risk is estimated through the preset first disease reasoning model to obtain the first estimated primary disease, and the patient's primary disease information is removed from it to obtain the patient's second estimated primary disease. S4. Divide patients into address regions based on personal information, and count the number of patients with various diseases as primary diseases (A1) and the number of patients with various diseases as secondary estimated primary diseases (A2) in each address region. S5. If the number of patients A1 and the number of patients A2 corresponding to any disease in any address area reach the preset screening warning condition, a screening suggestion signal for that disease in that address area is generated.

[0004] Furthermore, the step of obtaining the first estimated primary disease by estimating the risk of the primary disease through a preset first disease reasoning model specifically includes the following steps: S31. Using a pre-constructed medical knowledge graph, the first estimated primary disease set, the second estimated primary disease set, and the third estimated primary disease set are obtained based on the secondary disease information, concurrent disease information, and the latest symptom information and indicator information of the corresponding patients, respectively. S32. After intersecting the third estimated primary disease set with the first estimated primary disease set and the second estimated primary disease set to obtain the first intersection and the second intersection, perform a union operation on the first intersection and the second intersection to obtain the candidate primary disease set. S33. Obtain the correlation degree G1 between each candidate primary disease and the patient's secondary diseases and the correlation degree G2 between each candidate primary disease and the patient's comorbid diseases. Calculate the first risk value F1 of each candidate primary disease for the corresponding patient based on the correlation degree G1 and the correlation degree G2. S34. Using the patient's candidate primary disease and the latest symptom and indicator information as input, the preset deep learning model outputs the predicted secondary disease and predicted concurrent disease of the corresponding patient when the current candidate primary disease is used as input. The first prediction accuracy Z1 is calculated based on the patient's secondary disease information and the predicted secondary disease, and the second prediction accuracy Z2 is calculated based on the patient's concurrent disease information and the predicted secondary disease. S35. Calculate the second risk value F2 for each candidate primary disease for the corresponding patient based on the first prediction accuracy Z1 and the second prediction accuracy Z2. S36. Calculate the comprehensive risk value F3 of each candidate primary disease for the corresponding patient based on the first risk value F1 and the second risk value F2, and take the candidate primary disease with a comprehensive risk value F3 greater than or equal to the preset risk threshold as the first estimated primary disease.

[0005] Furthermore, the first risk value is calculated as follows: ; in, Let be the first risk value of the i-th candidate primary disease for the corresponding patient, and n be the number of secondary diseases in the corresponding patient. Let m be the correlation between the i-th candidate primary disease and the j-th secondary disease of the corresponding patient, and m be the number of concurrent diseases of the corresponding patient. Let be the degree of association between the i-th candidate primary disease and the p-th comorbidity of the corresponding patient; The second risk value is calculated as follows: ; in, Let be the second risk value for the i-th candidate primary disease for the corresponding patient. and These are the first and second prediction accuracy rates for the i-th candidate primary disease, respectively. The calculation method for the comprehensive risk value is as follows: ; in, Let K1 be the comprehensive risk value of the i-th candidate primary disease for the corresponding patient, and K2 be the first preset weight and the second preset weight, respectively.

[0006] Furthermore, the first prediction accuracy rate is the ratio of the number of correctly predicted secondary diseases to the actual number of secondary diseases the patient has; The second prediction accuracy rate is the ratio of the number of correctly predicted comorbidities to the actual number of comorbidities in the patient.

[0007] Furthermore, the step of dividing patients into address regions based on personal information specifically involves: S41. Obtain multi-level administrative regions; S42. After extracting the address information field from the patient's personal information, use spatial indexing technology to perform geocoding and convert the geocoding into latitude and longitude coordinates; S43. Map latitude and longitude coordinates to the corresponding smallest level of administrative region through a geographic information system.

[0008] Furthermore, the preset screening and early warning conditions are specifically: the number of patients A1 corresponding to any disease in any address area is greater than or equal to a first preset threshold, and / or the sum of the number of patients A1 and the number of patients A2 is greater than or equal to a second preset threshold.

[0009] Furthermore, step S3 also includes the following steps: The risk of secondary diseases and the risk of concurrent diseases are estimated by using a pre-set second disease reasoning model to obtain the first estimated secondary disease and the second estimated concurrent disease, respectively. The patient's secondary disease information and concurrent disease information are then removed from the model to obtain the patient's second estimated secondary disease and second estimated concurrent disease. Based on the second estimate of secondary diseases and the second estimate of comorbid diseases, disease precautions and reminder messages are generated and sent to the patient and / or attending physician.

[0010] Furthermore, the process of estimating the risk of secondary diseases and the risk of concurrent diseases through a preset second disease inference model to obtain the first estimated secondary disease and the second estimated concurrent disease, respectively, is as follows: Using a pre-constructed medical knowledge graph, a first estimated set of secondary diseases, a second estimated set of secondary diseases, a first estimated set of concurrent diseases, and a second estimated set of concurrent diseases are obtained based on the primary disease information, the latest symptom information, and indicator information of the corresponding patients. The first estimated set of secondary diseases and the second estimated set of secondary diseases are intersected to obtain the candidate set of secondary diseases. The first estimated set of concurrent diseases and the second estimated set of concurrent diseases are intersected to obtain the candidate set of concurrent diseases. Using the patient's primary disease and the latest symptom and indicator information as input, all of the patient's primary diseases are sequentially input into the deep learning model to obtain all the first predicted secondary diseases and the first predicted concurrent diseases; The first estimated set of secondary diseases is obtained by intersecting all the first predicted secondary diseases with the set of candidate secondary diseases. The first estimated set of concurrent diseases is obtained by intersecting all the first predicted concurrent diseases with the set of candidate concurrent diseases.

[0011] This invention also provides a health data management system based on big data analytics, employing any of the big data analytics methods described above, comprising: The data acquisition module acquires patient medical information, including diagnostic information, personal information, symptom information, and indicator information; The disease acquisition module obtains information on each patient's primary disease, secondary disease, and concurrent disease based on the diagnostic information. The first estimation module, based on the patient's secondary disease information, concurrent disease information, and the latest symptom and indicator information, uses a preset first disease reasoning model to estimate the risk of the primary disease to obtain the first estimated primary disease, and then removes the patient's primary disease information to obtain the patient's second estimated primary disease. The statistics module divides patients into address regions based on personal information, and counts the number of patients with various diseases as primary diseases (A1) and the number of patients with various diseases as secondary estimated primary diseases (A2) in each address region. The early warning module generates a screening suggestion signal for that disease in any address area if the number of patients A1 and A2 corresponding to any disease in any address area reaches the preset screening early warning condition.

[0012] Furthermore, the system also includes: The second estimation module estimates the risk of secondary diseases and the risk of concurrent diseases through a preset second disease reasoning model to obtain the first estimated secondary disease and the second estimated concurrent disease, respectively. Then, the second estimated secondary disease and the second estimated concurrent disease of the patient are obtained by removing the patient's secondary disease information and concurrent disease information from them. The early warning module also includes generating disease precautions and reminder messages based on the second estimated secondary disease and the second estimated concurrent disease, which are then sent to the patient and / or the attending physician.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention estimates the primary disease based on a patient's secondary disease information, concurrent disease information, and the latest symptom and indicator information using a pre-set first disease reasoning model. After removing the patient's primary disease information from this estimate, a second estimated primary disease is obtained. Then, based on the number of patients with the primary disease and the number of patients with the second estimated primary disease, the invention determines whether screening for that disease should be conducted in the corresponding address area. This achieves precise screening of both disease and regional targets, avoiding the limitation of screening for a single disease, increasing the effective scope of disease screening, and simultaneously improving screening efficiency and effectiveness. Attached Figure Description

[0014] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of a health data management method based on big data analysis according to the present invention; Figure 2 This is a structural block diagram of a health data management system based on big data analysis according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0019] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" and "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed by this invention.

[0020] Example 1 See Figure 1 As shown, the present invention provides a health data management method based on big data analysis, which specifically includes the following steps: S1. Obtain patient medical information, including diagnostic information, personal information, symptom information, and indicator information; S2. Obtain information on the primary disease, secondary disease, and concurrent disease of each patient based on the diagnostic information; S3. Based on the patient's secondary disease information, concurrent disease information, and the latest symptom information and indicator information, the primary disease risk is estimated through the preset first disease reasoning model to obtain the first estimated primary disease, and the patient's primary disease information is removed from it to obtain the patient's second estimated primary disease. S4. Divide patients into address regions based on personal information, and count the number of patients with various diseases as primary diseases (A1) and the number of patients with various diseases as secondary estimated primary diseases (A2) in each address region. S5. If the number of patients A1 and the number of patients A2 corresponding to any disease in any address area reach the preset screening warning condition, a screening suggestion signal for that disease in that address area is generated.

[0021] In step S1, the symptom information and indicator information include, but are not limited to, various symptom data and indicator data in the patient's various test reports, various symptom data and indicator data obtained through various medical wearable devices, etc.

[0022] In step S1, the patient medical information is information data obtained after data preprocessing of the raw data. The data preprocessing includes data cleaning, data integration and aggregation, etc. Data cleaning includes operations such as deduplication, null value imputation, outlier handling, and field filtering, such as removing redundant records and filling missing values; data integration and aggregation includes integrating data from multiple sources, associating patient information through a master index, eliminating redundancy and conflicts, and unifying different terms with the same meaning into standard vocabulary through field mapping and union.

[0023] In step S2, a primary disease typically refers to an independent disease that occurs first, usually related to genetics, environment, or lifestyle; a secondary disease is usually a health problem directly caused by a definite underlying disease or factor; and a concurrent disease typically refers to another disease that is directly or indirectly caused by a particular disease during its development, and the two may coexist or appear sequentially. It should be noted that in this protocol, diseases are not fixed as primary, secondary, or concurrent diseases; the type of disease for each patient is determined based on their diagnostic information.

[0024] In step S3, the process of obtaining the first estimated primary disease by estimating the risk of the primary disease through a preset first disease reasoning model specifically includes the following steps: S31. Using a pre-constructed medical knowledge graph, the first estimated primary disease set, the second estimated primary disease set, and the third estimated primary disease set are obtained based on the secondary disease information, concurrent disease information, and the latest symptom information and indicator information of the corresponding patients, respectively. S32. After intersecting the third estimated primary disease set with the first estimated primary disease set and the second estimated primary disease set to obtain the first intersection and the second intersection, perform a union operation on the first intersection and the second intersection to obtain the candidate primary disease set. S33. Obtain the correlation G1 between each candidate primary disease and the patient's secondary diseases, and the correlation G2 between each candidate primary disease and the patient's comorbidities. Calculate the first risk value F1 for each candidate primary disease for the corresponding patient based on the correlation G1 and correlation G2. ;

[0025] in, Let be the first risk value of the i-th candidate primary disease for the corresponding patient, and n be the number of secondary diseases in the corresponding patient. Let m be the correlation between the i-th candidate primary disease and the j-th secondary disease of the corresponding patient, and m be the number of concurrent diseases of the corresponding patient. Let be the degree of association between the i-th candidate primary disease and the p-th comorbidity of the corresponding patient; S34. Using the patient's candidate primary disease and the latest symptom and indicator information as input, the preset deep learning model outputs the predicted secondary disease and predicted concurrent disease of the corresponding patient when the current candidate primary disease is used as input. The first prediction accuracy Z1 is calculated based on the patient's secondary disease information and the predicted secondary disease, and the second prediction accuracy Z2 is calculated based on the patient's concurrent disease information and the predicted secondary disease. S35. Calculate the second risk value F2 for each candidate primary disease for the corresponding patient based on the first prediction accuracy Z1 and the second prediction accuracy Z2: ; in, Let be the second risk value for the i-th candidate primary disease for the corresponding patient. and These are the first and second prediction accuracy rates for the i-th candidate primary disease, respectively. S36. Calculate the comprehensive risk value F3 for each candidate primary disease for the corresponding patient based on the first risk value F1 and the second risk value F2. Select candidate primary diseases whose comprehensive risk value F3 is greater than or equal to a preset risk threshold as the first estimated primary disease. ; in, Let K1 be the comprehensive risk value of the i-th candidate primary disease for the corresponding patient, and K2 be the first preset weight and the second preset weight, respectively.

[0026] In step S31, the medical knowledge graph is constructed by extracting medical knowledge from various medical knowledge bases and clinical guidelines. The triplet form includes (primary disease, possible concurrent diseases, concurrent diseases), (primary disease, possible secondary diseases, secondary diseases), (primary disease, corresponding symptoms, detailed symptoms), and (primary disease, indicator, indicator threshold or range), etc. In this scheme, all corresponding primary disease head entities are extracted from the tail entities of secondary disease information, concurrent disease information, and the latest symptom and indicator information. The first estimated primary disease set is the union of the primary disease set found based on symptom information and the primary disease set found based on indicator information.

[0027] In step S33, the range of correlation is [0,1]. The correlation between the primary disease and the secondary disease, as well as the correlation between the primary disease and the comorbid disease, can be obtained by a pre-trained learning model or by domain experts based on medical knowledge such as various medical knowledge bases and clinical guidelines, as well as a large number of patients' historical medical records. The correlation is recorded in the corresponding triplet description in the pre-constructed medical knowledge graph.

[0028] In step S34, the deep learning model is trained using a large number of patients' historical primary diseases and corresponding historical secondary diseases, historical complications, historical symptom information and historical indicator information as training sample groups.

[0029] In step S35, the first prediction accuracy rate is the ratio of the number of correctly predicted secondary diseases to the actual number of secondary diseases. For example, if a patient actually has two secondary diseases, disease a and disease b, and the predicted secondary diseases obtained based on the candidate primary disease c and the patient's latest symptom and indicator information are disease a and disease d, then the first prediction accuracy rate corresponding to the candidate primary disease c is 0.5. Similarly, the second prediction accuracy rate is the ratio of the number of correctly predicted concurrent diseases to the actual number of concurrent diseases of the patient.

[0030] In step S4, the step of dividing the patient's address region based on personal information specifically involves: S41. Obtain multi-level administrative regions; S42. After extracting the address information field from the patient's personal information, use spatial indexing technology to perform geocoding and convert the geocoding into latitude and longitude coordinates; S43. Map latitude and longitude coordinates to the corresponding smallest level of administrative region through a geographic information system.

[0031] The multi-level administrative regions in step S41 are obtained based on the officially defined administrative regions, such as (province-city-district-street).

[0032] In step S5, the preset screening and early warning conditions are specifically as follows: The number of patients A1 corresponding to any disease in any address region is greater than or equal to a first preset threshold, and / or the sum of the number of patients A1 and the number of patients A2 is greater than or equal to a second preset threshold.

[0033] In some embodiments, step S3 further includes the following steps: The risk of secondary diseases and the risk of concurrent diseases are estimated by using a pre-set second disease reasoning model to obtain the first estimated secondary disease and the second estimated concurrent disease, respectively. The patient's secondary disease information and concurrent disease information are then removed from the model to obtain the patient's second estimated secondary disease and second estimated concurrent disease. Based on the second estimate of secondary diseases and the second estimate of comorbid diseases, disease precautions and reminder messages are generated and sent to the patient and / or attending physician.

[0034] The process of estimating the risk of secondary diseases and the risk of concurrent diseases using a preset second disease reasoning model yields a first estimated secondary disease and a second estimated concurrent disease, respectively. Specifically: Using a pre-constructed medical knowledge graph, a first estimated set of secondary diseases, a second estimated set of secondary diseases, a first estimated set of concurrent diseases, and a second estimated set of concurrent diseases are obtained based on the primary disease information, the latest symptom information, and indicator information of the corresponding patients. The first estimated set of secondary diseases and the second estimated set of secondary diseases are intersected to obtain the candidate set of secondary diseases. The first estimated set of concurrent diseases and the second estimated set of concurrent diseases are intersected to obtain the candidate set of concurrent diseases. Using the patient's primary disease and the latest symptom and indicator information as input, all of the patient's primary diseases are sequentially input into the deep learning model to obtain all the first predicted secondary diseases and the first predicted concurrent diseases; The first estimated set of secondary diseases is obtained by intersecting all the first predicted secondary diseases with the set of candidate secondary diseases. The first estimated set of concurrent diseases is obtained by intersecting all the first predicted concurrent diseases with the set of candidate concurrent diseases.

[0035] The system generates disease-related notices and reminders to send to patients and / or attending physicians via relevant apps or mini-programs to their accounts.

[0036] Example 2 See Figure 2 As shown, the present invention also provides a health data management system for big data analysis, wherein the system applies the method described in Embodiment 1, and wherein the system specifically includes: The central processing unit, and the data acquisition module, disease acquisition module, first estimation module, statistics module and early warning module that are connected in communication with the central processing unit; The data acquisition module acquires patient medical information, including diagnostic information, personal information, symptom information, and indicator information; The disease acquisition module obtains information on each patient's primary disease, secondary disease, and concurrent disease based on the diagnostic information. The first estimation module, based on the patient's secondary disease information, concurrent disease information, and the latest symptom and indicator information, uses a preset first disease reasoning model to estimate the risk of the primary disease to obtain the first estimated primary disease, and then removes the patient's primary disease information to obtain the patient's second estimated primary disease. The statistics module divides patients into address regions based on personal information, and counts the number of patients with various diseases as primary diseases (A1) and the number of patients with various diseases as secondary estimated primary diseases (A2) in each address region. The early warning module generates a screening suggestion signal for that disease in any address area if the number of patients A1 and A2 corresponding to any disease in any address area reaches the preset screening early warning condition.

[0037] In some embodiments, the system further includes a second estimation module that is communicatively connected to a central processing unit, used to perform secondary disease risk estimation and concurrent disease risk estimation through a preset second disease inference model to obtain a first estimated secondary disease and a second estimated concurrent disease, and to remove the patient's secondary disease information and concurrent disease information from them to obtain the patient's second estimated secondary disease and second estimated concurrent disease. The early warning module also includes generating disease precautions and reminder messages based on the second estimated secondary disease and the second estimated concurrent disease, which are then sent to the patient and / or the attending physician.

[0038] Example 3 The present invention also provides an electronic device, including: a processor, a transmitting device, an input device, an output device, and a memory. The processor may be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit, or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory may be implemented using a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), and is used to store computer program code. The computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes a method as described in any of the above possible implementation methods.

[0039] Example 4 The present invention also provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor of an electronic device, cause the processor to perform a method as described in any of the above possible implementations.

[0040] The beneficial effects of this invention are as follows: This invention estimates the primary disease risk based on a patient's secondary disease information, concurrent disease information, and the latest symptom and indicator information using a pre-set first disease reasoning model. After removing the patient's primary disease information from this estimate, a second estimated primary disease is obtained. Finally, based on the number of patients with the primary disease and the number of patients with the second estimated primary disease, the invention determines whether to conduct disease screening in the corresponding address area. This achieves precise screening of both disease and region targets, avoiding the limitation of screening for a single disease, increasing the effective scope of disease screening, and simultaneously improving screening efficiency and effectiveness. The present invention also uses a preset second disease reasoning model to estimate the risk of secondary diseases and the risk of concurrent diseases to obtain a first estimated secondary disease and a second estimated concurrent disease, thereby generating disease precautions and reminder messages to be sent to patients and / or attending physicians, so as to promptly remind patients of the estimated hidden secondary and concurrent diseases and ensure timely medical treatment.

[0041] In the description of this specification, the references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0042] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0043] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A health data management method based on big data analysis, characterized in that, Includes the following steps: S1. Obtain patient medical information, including diagnostic information, personal information, symptom information, and indicator information; S2. Obtain information on the primary disease, secondary disease, and concurrent disease of each patient based on the diagnostic information; S3. Based on the patient's secondary disease information, concurrent disease information, and the latest symptom information and indicator information, the primary disease risk is estimated through the preset first disease reasoning model to obtain the first estimated primary disease, and the patient's primary disease information is removed from it to obtain the patient's second estimated primary disease. S4. Divide patients into address regions based on personal information, and count the number of patients with various diseases as primary diseases (A1) and the number of patients with various diseases as secondary estimated primary diseases (A2) in each address region. S5. If the number of patients A1 and the number of patients A2 corresponding to any disease in any address area reach the preset screening warning condition, a screening suggestion signal for that disease in that address area is generated.

2. The health data management method based on big data analysis according to claim 1, characterized in that, The step of obtaining the first estimated primary disease by estimating the risk of the primary disease through a preset first disease reasoning model specifically includes the following steps: S31. Using a pre-constructed medical knowledge graph, the first estimated primary disease set, the second estimated primary disease set, and the third estimated primary disease set are obtained based on the secondary disease information, concurrent disease information, and the latest symptom information and indicator information of the corresponding patients, respectively. S32. After intersecting the third estimated primary disease set with the first estimated primary disease set and the second estimated primary disease set to obtain the first intersection and the second intersection, perform a union operation on the first intersection and the second intersection to obtain the candidate primary disease set. S33. Obtain the correlation degree G1 between each candidate primary disease and the patient's secondary diseases and the correlation degree G2 between each candidate primary disease and the patient's comorbid diseases. Calculate the first risk value F1 of each candidate primary disease for the corresponding patient based on the correlation degree G1 and the correlation degree G2. S34. Using the patient's candidate primary disease and the latest symptom and indicator information as input, the preset deep learning model outputs the predicted secondary disease and predicted concurrent disease of the corresponding patient when the current candidate primary disease is used as input. The first prediction accuracy Z1 is calculated based on the patient's secondary disease information and the predicted secondary disease, and the second prediction accuracy Z2 is calculated based on the patient's concurrent disease information and the predicted secondary disease. S35. Calculate the second risk value F2 for each candidate primary disease for the corresponding patient based on the first prediction accuracy Z1 and the second prediction accuracy Z2. S36. Calculate the comprehensive risk value F3 of each candidate primary disease for the corresponding patient based on the first risk value F1 and the second risk value F2, and take the candidate primary disease with a comprehensive risk value F3 greater than or equal to the preset risk threshold as the first estimated primary disease.

3. The health data management method based on big data analysis according to claim 2, characterized in that, The first risk value is calculated as follows: ; in, Let be the first risk value of the i-th candidate primary disease for the corresponding patient, and n be the number of secondary diseases in the corresponding patient. Let m be the correlation between the i-th candidate primary disease and the j-th secondary disease of the corresponding patient, and m be the number of concurrent diseases of the corresponding patient. Let be the degree of association between the i-th candidate primary disease and the p-th comorbidity of the corresponding patient; The second risk value is calculated as follows: ; in, Let be the second risk value for the i-th candidate primary disease for the corresponding patient. and These are the first and second prediction accuracy rates for the i-th candidate primary disease, respectively. The calculation method for the comprehensive risk value is as follows: ; in, Let K1 be the comprehensive risk value of the i-th candidate primary disease for the corresponding patient, and K2 be the first preset weight and the second preset weight, respectively.

4. The health data management method based on big data analysis according to claim 2, characterized in that, The first prediction accuracy rate is the ratio of the number of correctly predicted secondary diseases to the actual number of secondary diseases in the patient; The second prediction accuracy rate is the ratio of the number of correctly predicted comorbidities to the actual number of comorbidities in the patient.

5. The health data management method based on big data analysis according to claim 1, characterized in that, The process of dividing patients into address regions based on personal information is as follows: S41. Obtain multi-level administrative regions; S42. After extracting the address information field from the patient's personal information, use spatial indexing technology to perform geocoding and convert the geocoding into latitude and longitude coordinates; S43. Map latitude and longitude coordinates to the corresponding smallest level of administrative region through a geographic information system.

6. The health data management method based on big data analysis according to claim 1, characterized in that, The preset screening and early warning conditions are specifically: the number of patients A1 corresponding to any disease in any address area is greater than or equal to the first preset threshold, and / or the sum of the number of patients A1 and the number of patients A2 is greater than or equal to the second preset threshold.

7. The health data management method based on big data analysis according to claim 2, characterized in that, Step S3 also includes the following steps: The risk of secondary diseases and the risk of concurrent diseases are estimated by using a pre-set second disease reasoning model to obtain the first estimated secondary disease and the second estimated concurrent disease, respectively. The patient's secondary disease information and concurrent disease information are then removed from the model to obtain the patient's second estimated secondary disease and second estimated concurrent disease. Based on the second estimate of secondary diseases and the second estimate of comorbid diseases, disease precautions and reminder messages are generated and sent to the patient and / or attending physician.

8. The health data management method based on big data analysis according to claim 7, characterized in that, The process of estimating the risk of secondary diseases and the risk of concurrent diseases using a preset second disease reasoning model yields a first estimated secondary disease and a second estimated concurrent disease, respectively. Specifically: Using a pre-constructed medical knowledge graph, a first estimated set of secondary diseases, a second estimated set of secondary diseases, a first estimated set of concurrent diseases, and a second estimated set of concurrent diseases are obtained based on the primary disease information, the latest symptom information, and indicator information of the corresponding patients. The first estimated set of secondary diseases and the second estimated set of secondary diseases are intersected to obtain the candidate set of secondary diseases. The first estimated set of concurrent diseases and the second estimated set of concurrent diseases are intersected to obtain the candidate set of concurrent diseases. Using the patient's primary disease and the latest symptom and indicator information as input, all of the patient's primary diseases are sequentially input into the deep learning model to obtain all the first predicted secondary diseases and the first predicted concurrent diseases; The first estimated set of secondary diseases is obtained by intersecting all the first predicted secondary diseases with the set of candidate secondary diseases. The first estimated set of concurrent diseases is obtained by intersecting all the first predicted concurrent diseases with the set of candidate concurrent diseases.

9. A health data management system based on big data analytics, employing the health data management method based on big data analytics as described in any one of claims 1 to 8, characterized in that, include: The data acquisition module acquires patient medical information, including diagnostic information, personal information, symptom information, and indicator information; The disease acquisition module obtains information on each patient's primary disease, secondary disease, and concurrent disease based on the diagnostic information. The first estimation module, based on the patient's secondary disease information, concurrent disease information, and the latest symptom and indicator information, uses a preset first disease reasoning model to estimate the risk of the primary disease to obtain the first estimated primary disease, and then removes the patient's primary disease information to obtain the patient's second estimated primary disease. The statistics module divides patients into address regions based on personal information, and counts the number of patients with various diseases as primary diseases (A1) and the number of patients with various diseases as secondary estimated primary diseases (A2) in each address region. The early warning module generates a screening suggestion signal for that disease in any address area if the number of patients A1 and A2 corresponding to any disease in any address area reaches the preset screening early warning condition.

10. The health data management system for big data analysis according to claim 9, characterized in that, The system also includes: The second estimation module estimates the risk of secondary diseases and the risk of concurrent diseases through a preset second disease reasoning model to obtain the first estimated secondary disease and the second estimated concurrent disease, respectively. Then, the second estimated secondary disease and the second estimated concurrent disease of the patient are obtained by removing the patient's secondary disease information and concurrent disease information from them. The early warning module also includes generating disease precautions and reminder messages based on the second estimated secondary disease and the second estimated concurrent disease, which are then sent to the patient and / or the attending physician.

Images