A neuronal network hospital data correlation analysis system

Abstract

The application discloses a kind of neuron network hospital data correlation analysis systems, belong to disease prediction technical field.The application is by acquisition module, for obtaining the historical diagnosis and treatment data of the electronic medical record system record of hospital, health correlation state data and the illness data of target patient, the historical diagnosis and treatment data are divided according to disease category, obtain the disease diagnosis and treatment data related to each disease;Construction module is used to generate preset disease correlation model according to health correlation state data, the preset disease correlation model includes: each disease corresponding neuron node and the first connection between neuron node;Optimization module is used to adjust preset disease correlation model according to the disease diagnosis and treatment data, obtain target correlation model;Correlation module is used to determine the associated disease of target patient according to the target correlation model and the illness data, realizes the technical effect of effectively and comprehensively identifying illness.

Description

Technical Field

[0001] This invention relates to the field of disease prediction, and more particularly to a neural network hospital data association analysis system. Background Technology

[0002] Medical resources are the foundation for disease diagnosis, treatment, and rehabilitation. Sufficient medical testing equipment and corresponding examinations can effectively improve the accuracy of diagnosing the cause of a patient's illness. Currently, during the diagnostic process, doctors generally set one or more examinations based on the patient's condition, such as blood and urine tests. However, they typically do not exhaustively test all items, but only need to test certain data. This testing method is based on the doctor's personal judgment of the condition or on the recommendations of treatment guidelines. When the test data is insufficient or incomplete, it is difficult to comprehensively judge the condition and discover other related conditions.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this invention is to provide a neural network hospital data association analysis system, which aims to effectively and comprehensively identify disease conditions. To achieve the above objective, this invention provides a neural network hospital data association analysis system, which includes:

[0005] The acquisition module is used to acquire historical medical data, health-related status data, and target patient condition data recorded by the hospital's electronic medical record system, and to divide the historical medical data according to disease type to obtain disease medical data related to each disease.

[0006] The construction module is used to generate a preset disease association model based on health association status data. The preset disease association model includes: neuron nodes corresponding to each disease and the first connection relationship between neuron nodes.

[0007] The optimization module is used to adjust the preset disease association model based on the disease diagnosis and treatment data to obtain the target association model;

[0008] The association module is used to determine the associated diseases of the target patient based on the target association model and the disease data.

[0009] Furthermore, to achieve the above objectives, the present invention also provides a method for neural network hospital data association analysis, the steps of which include:

[0010] The system acquires historical medical records, health status data, and target patient condition data from the hospital's electronic medical record system. It then divides the historical medical records by disease type to obtain disease medical records related to each disease.

[0011] A preset disease association model is generated based on health association status data. The preset disease association model includes: neuron nodes corresponding to each disease and the first connection relationship between neuron nodes.

[0012] The preset disease association model is adjusted based on the disease diagnosis and treatment data to obtain the target association model;

[0013] The associated diseases of the target patient are determined based on the target association model and the disease data.

[0014] Optionally, the health association status data includes: secondary disease relationships, complication relationships, and disease classification relationships, and the step of generating a preset disease association model based on the health association status data includes:

[0015] Generate neuron nodes corresponding to the diseases present in the health-related status data;

[0016] A first connection relationship between neuronal nodes is generated based on the secondary disease relationship, the complication relationship, and the disease classification relationship.

[0017] Optionally, the disease diagnosis and treatment data includes: laboratory indicator data, and the step of adjusting the preset disease association model based on the disease diagnosis and treatment data to obtain the target association model includes:

[0018] Based on the test index data, determine the standard characteristic data corresponding to each disease;

[0019] Add the standard feature data to the corresponding neuron node;

[0020] The first connection relationship is updated based on the standard feature data, the first connection relationship, and the test index data to obtain the target association model.

[0021] Optionally, the step of updating the first connection relationship based on the standard feature data, the first connection relationship, and the test index data to obtain the target association model includes:

[0022] The target data range of the test index data is determined based on the standard feature data and the first connection relationship.

[0023] Determine the associated disease statistics results corresponding to the standard feature data based on the target data range;

[0024] A second connection relationship is generated based on the associated disease statistics;

[0025] The second connection relationship is used as the first connection relationship.

[0026] Optionally, the first connection relationship is a link between a first neuron node and a second neuron node, and the step of determining the target data range of the test index data based on the standard feature data and the first connection relationship includes:

[0027] Multiple target data ranges are generated based on the standard feature data corresponding to the first neuron node.

[0028] Optionally, the step of determining the standard feature data corresponding to each disease based on the test index data includes:

[0029] Based on the aforementioned test index data, determine the disease-strongly correlated and weakly correlated indicators;

[0030] Based on the test index data, determine the average incidence rate and incidence rate variance corresponding to the disease strong correlation index.

[0031] Based on the test index data, determine the first value of the weakly correlated index with the highest frequency.

[0032] The average incidence rate index, the incidence rate variance index, and the first value of the weak correlation index are used as the standard feature data.

[0033] Optionally, before the step of determining the standard feature data corresponding to each disease based on the test index data, the method further includes:

[0034] Based on the disease diagnosis and treatment data and clustering algorithm, the centroid distance between each centroid in the cluster space is calculated to obtain multiple centroid distances;

[0035] The adjustment result of the corresponding neuron node is determined based on multiple centroid distances, and the adjustment result includes: maintaining the current number of neuron nodes or dividing the neuron node into more than one sub-neuron node.

[0036] Optionally, the step of determining the associated disease of the target patient based on the target association model and the disease data includes:

[0037] The corresponding target neuron node is determined based on the aforementioned disease data;

[0038] The incidence of associated diseases is determined based on the first connection relationship corresponding to the target neuron node and the disease data.

[0039] Furthermore, to achieve the above objectives, the present invention also provides a neural network hospital data association analysis device, the neural network hospital data association analysis device comprising: a memory, a processor, and a neural network hospital data association analysis program stored in the memory and executable on the processor, the neural network hospital data association analysis program being configured to implement the steps of the neural network hospital data association analysis method described in any of the above claims.

[0040] This invention proposes a neural network-based hospital data association analysis method. This method acquires historical medical data, health association status data, and target patient condition data recorded in the hospital's electronic medical record system. The historical medical data is categorized by disease type to obtain disease-related medical data for each disease. A preset disease association model is generated based on the health association status data. Compared to doctors determining whether a disease is present solely through data and adjusting the preset disease association model based on the disease medical data to obtain a target association model, this method can identify multiple possible associated diseases and their corresponding probabilities. Therefore, it can effectively and comprehensively identify conditions even when detection data is insufficient or incomplete. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the structure of the neural network hospital data association analysis device in the hardware operating environment involved in the embodiments of the present invention;

[0042] Figure 2 This is a flowchart illustrating an embodiment of the neural network hospital data association analysis method of the present invention;

[0043] Figure 3 This is a flowchart illustrating two embodiments of the neural network hospital data association analysis method of the present invention;

[0044] Figure 4 The flowcharts for three embodiments of the neural network hospital data association analysis method of the present invention are shown below;

[0045] Figure 5 The following are flowcharts illustrating four embodiments of the neural network hospital data association analysis method of the present invention;

[0046] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0047] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0048] Reference Figure 1 , Figure 1This is a schematic diagram of the structure of a neural network hospital data association analysis device in the hardware operating environment involved in the embodiments of the present invention.

[0049] like Figure 1 As shown, the neural network hospital data association analysis device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, an interactive device 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The interactive device 1003 may include a display screen or an input unit such as a keyboard. Optionally, the interactive device 1003 may also be connected to the communication bus via standard wired or wireless interfaces. The network interface 1004 may optionally include standard wired or wireless interfaces (such as a Wi-Fi interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0050] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the neural network hospital data association analysis device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0051] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a data storage module, a network communication module, a user interface module, and a neural network hospital data association analysis program.

[0052] exist Figure 1 In the neural network hospital data association analysis device shown, the network interface 1004 is mainly used for data communication with other devices; the interactive device 1003 is mainly used for data interaction with users; the processor 1001 and memory 1005 in the neural network hospital data association analysis device of the present invention can be set in the neural network hospital data association analysis device, and the neural network hospital data association analysis device calls the neural network hospital data association analysis program stored in the memory 1005 through the processor 1001 and executes the neural network hospital data association analysis method provided in the embodiment of the present invention.

[0053] This invention provides a method for correlation analysis of hospital data using neural networks, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of a neural network hospital data association analysis method according to the present invention.

[0054] In this embodiment, the neural network hospital data association analysis method includes:

[0055] Step S1: Obtain historical medical records, health-related status data, and target patient's condition data recorded by the hospital's electronic medical record system. Divide the historical medical records according to disease types to obtain disease medical records related to each disease.

[0056] In this embodiment, since electronic medical record systems are commonly used in hospitals, connecting to such systems allows for efficient acquisition of historical medical data. Furthermore, it should be noted that this embodiment does not limit the name of the system storing historical medical data; it only needs to be a system that substantially stores the historical medical data, and can have other names. Specifically, historical medical data refers to data related to the patient's diagnosis and treatment obtained up to the current moment. Here, health-related data refers to the correlation data between multiple diseases. This correlation data can include relationships between complications, secondary diseases, sequelae, etc., and also includes: multiple subtypes of the same type of disease, such as different types of diabetes; and relationships between diseases with similar symptoms. Since historical medical data already provides diagnostic results, it is convenient to obtain disease-related medical data for each disease.

[0057] Step S2: Generate a preset disease association model based on health association status data. The preset disease association model includes: neuron nodes corresponding to each disease and the first connection relationship between neuron nodes.

[0058] In this embodiment, a preset disease association model is constructed based on health-related status data. The preset disease association model includes: neuron nodes corresponding to each disease and a first connection relationship between these neuron nodes. Each neuron node can store data related to that neuron node. This first connection relationship is directed and is used to calculate the probability of a connection from a first neuron node to a second neuron node. Through this preset disease association model, the probability of other related diseases can be predicted after identifying a disease.

[0059] Step S3: Adjust the preset disease association model based on the disease diagnosis and treatment data to obtain the target association model;

[0060] Optionally, adjusting the preset disease association model using the disease diagnosis and treatment data can improve the accuracy of predicting associated diseases through a large amount of data. Optionally, adjusting the preset disease association model here can be by adding neuron nodes, adding first connections, or adjusting the first connections, etc.

[0061] Step S4: Determine the associated diseases of the target patient based on the target association model and the disease data.

[0062] Specifically, the disease data is used to determine the corresponding neuron node in the target association model. Based on the first connection relationship between the disease data and the neuron node, the probability of the associated disease is determined. After determining the probability of the associated disease, a decision is made based on the probability of the associated disease and a preset disease probability threshold to determine whether to continue searching for the probability of possible associated diseases until the probability of the associated disease is less than the preset disease probability threshold. The possible associated diseases and their corresponding probabilities are then displayed.

[0063] In this embodiment, historical medical data, health association status data, and target patient condition data recorded by the hospital's electronic medical record system are acquired. The historical medical data is then divided into disease types to obtain disease medical data related to each disease. A preset disease association model is generated based on the health association status data. Compared to doctors judging whether a disease is met solely based on data and adjusting the preset disease association model according to the disease medical data to obtain a target association model, the target disease association model can determine multiple possible associated diseases and their corresponding probabilities. This allows for effective and comprehensive identification of the condition even when the detection data is insufficient or incomplete.

[0064] Furthermore, based on the first embodiment, a second embodiment of the present invention for a neural network hospital data association analysis method is proposed. In this embodiment, referring to... Figure 3 The health association status data includes: secondary disease relationships, complication relationships, and disease classification relationships. The step of generating a preset disease association model based on the health association status data includes:

[0065] Step S21: Generate neuron nodes corresponding to the diseases present in the health association status data;

[0066] Since diseases are not entirely independent of each other, the diseases present in the health association status data generally include all common diseases. A neuron node is generated corresponding to each disease present in the health association status data.

[0067] Step S22: Generate a first connection relationship between neuronal nodes based on the secondary disease relationship, the complication relationship, and the disease classification relationship.

[0068] In this embodiment, a corresponding first connection relationship is generated based on different health association status data, thereby improving the completeness of the constructed preset disease association model.

[0069] In this embodiment, neuron nodes are generated corresponding to diseases present in the health association status data; a first connection relationship is generated between neuron nodes based on the secondary disease relationship, the complication relationship, and the disease classification relationship, thereby obtaining a preset disease association model with complete connection relationships.

[0070] Furthermore, based on the first or second embodiment, a third embodiment of the present invention for a neural network hospital data association analysis method is proposed. In this embodiment, reference is made to... Figure 4 The disease diagnosis and treatment data includes: test indicator data. The step of adjusting the preset disease association model based on the disease diagnosis and treatment data to obtain the target association model includes:

[0071] Step S31: Determine the standard characteristic data corresponding to each disease based on the test index data;

[0072] In this embodiment, the type of test indicators is not limited. Therefore, the test indicator data here can be various different types of data. Commonly, they generally include blood and urine tests. By analyzing changes in the components of blood and urine, doctors can help discover disease clues. Specifically, blood tests generally include: detecting components such as white blood cells, red blood cells, hemoglobin, and platelets, as well as blood glucose, blood lipids, and inflammatory markers. Urine tests generally include: routine urinalysis, urine sediment microscopy, and urine protein detection. Optionally, the average value of the test indicator data corresponding to each disease can be calculated. It should be noted that the standard feature data is obtained by extracting features from the test indicator data. The purpose of feature extraction is to reduce the amount of data and improve computational efficiency. Preferably, multiple standard feature data can be extracted from one type of test indicator data. For example, average blood pressure, average blood lipids, etc.

[0073] Step S32: Add the standard feature data to the corresponding neuron node;

[0074] Specifically, the standard feature data is added to the corresponding neuron node and saved.

[0075] Step S33: Update the first connection relationship based on the standard feature data, the first connection relationship, and the test index data to obtain the target association model.

[0076] Optionally, the connection relationship that needs to be updated is determined by the first connection relationship, and the connection relationship that needs to be updated is updated according to the standard feature data and test index data. After the update is completed, the target association model described by Deao can be realized.

[0077] In this embodiment, by generating neuron nodes corresponding to diseases existing in the health association status data, and generating a first connection relationship between neuron nodes based on the secondary disease relationship, the complication relationship, and the disease classification relationship, the preset disease association model can be adjusted to update the target association model.

[0078] Furthermore, based on any of the above embodiments, a fourth embodiment of the present invention for a neural network hospital data association analysis method is proposed, referring to... Figure 5 The step of updating the first connection relationship based on the standard feature data, the first connection relationship, and the test index data to obtain the target association model includes:

[0079] Step S331: Determine the target data range of the test index data based on the standard feature data and the first connection relationship;

[0080] Preferably, the standard feature data includes: the average and standard deviation of the test indicators; based on the average and standard deviation of each test indicator, a target data range is selected from the test indicator data; when the data in the test indicator data reaches a preset standard, the data corresponding to the preset standard is determined as the target data; wherein, the preset standard is that the difference between the data in the test indicator data and the average of the test indicator data is less than the standard deviation. Of course, the preset standards here can be adjusted simultaneously, and multiple preset standards can be set.

[0081] Step S332: Determine the associated disease statistics results corresponding to the standard feature data based on the target data range;

[0082] The data within the target data range definitely corresponds to a disease. However, it's necessary to statistically determine, within the target data range, the presence of disease-related statistical results for the neuron nodes pointed to by the first connection relationship, in addition to the originally corresponding disease. This is known as the associated disease statistical results. Furthermore, since the standard feature data has different types of detected data, there are multiple associated disease statistical results, forming a data table. The table includes the associated disease statistical results, i.e., probabilities. A specific example is shown below:

[0083]

[0084] Step S333: Generate a second connection relationship based on the associated disease statistics results;

[0085] Specifically, the second connection here can be related to the statistical results of the disease.

[0086] Step S334: Use the second connection relationship as the first connection relationship.

[0087] In this embodiment, the target data range of the test index data is determined by the standard feature data and the first connection relationship, thereby enabling the determination of associated disease statistics based on the target data range. This generates an accurate second connection relationship, and the first connection relationship is then generated based on the second connection relationship.

[0088] Furthermore, the first connection relationship is a link between a first neuron node and a second neuron node. The step of determining the target data range of the test index data based on the standard feature data and the first connection relationship includes:

[0089] Multiple target data ranges are generated based on the standard feature data corresponding to the first neuron node.

[0090] In this embodiment, multiple target data ranges are generated using the standard feature data corresponding to the first neuron node. Each target data range can obtain corresponding associated disease statistics, thereby obtaining multiple associated disease statistics. All associated disease statistics are used as the second connection relationship.

[0091] Furthermore, based on any of the above embodiments, a fifth embodiment of the present invention for a neural network hospital data association analysis method is proposed, wherein the step of determining the standard feature data corresponding to each disease based on the test index data includes:

[0092] Based on the aforementioned test index data, determine the disease-strongly correlated and weakly correlated indicators;

[0093] Based on the test index data, determine the average incidence rate and incidence rate variance corresponding to the disease strong correlation index.

[0094] In this embodiment, the average incidence rate index refers to the average value of test indicator data within the range of data for the current disease, and the incidence rate variance index refers to the variance of test indicator data within the range of data for the current disease. In some embodiments, the incidence rate variance index may also be replaced by the incidence rate standard deviation index. Specifically, the average incidence rate index here can be the average blood glucose level or the average white blood cell count. The corresponding test indicator data may be different for different types of diseases.

[0095] Based on the test index data, determine the first value of the weakly correlated index with the highest frequency.

[0096] In this embodiment, the frequency of occurrence of the values ​​in the weakly correlated indicators is statistically analyzed, and the value of the weakly correlated indicator with the highest frequency is selected as the first value of the weakly correlated indicator.

[0097] The average incidence rate index, the incidence rate variance index, and the first value of the weak correlation index are used as the standard feature data.

[0098] In other embodiments, different disease-related strong and weak correlation indicators may be selected than those in this embodiment.

[0099] Furthermore, before the step of determining the standard characteristic data corresponding to each disease based on the test index data, the method further includes:

[0100] Based on the disease diagnosis and treatment data and clustering algorithm, the centroid distance between each centroid in the clustering space is calculated to obtain multiple centroid distances;

[0101] The adjustment result of the corresponding neuron node is determined based on multiple centroid distances, and the adjustment result includes: maintaining the current number of neuron nodes or dividing the neuron node into more than one sub-neuron node.

[0102] Specifically, based on multiple centroid distances, it is determined whether two or more abnormal categories of data centers exist. When all centroid distances are less than a preset distance, it is determined that no two or more abnormal categories of data centers exist, thus maintaining the current neuron node. When there are centroid distances greater than or equal to the preset distance, the number of centroid distances greater than or equal to the preset distance is used to determine the number of sub-neuron nodes to be divided into. In other embodiments, other methods can be selected to determine whether to divide a neuron node into more than one sub-neuron node. For example, based on known disease subtype classifications, types of infecting viruses, etc. Specifically, it needs to be determined according to the actual situation. In addition, for a type of disease, since there will be significant differences between early and late-stage data, the neuron node can be divided into more than one sub-neuron node according to the disease progression stage.

[0103] In this embodiment, the centroid distances between centroids in the clustering space are calculated using the disease diagnosis and treatment data and a clustering algorithm to obtain multiple centroid distances. Based on the multiple centroid distances, the adjustment results of the corresponding neuron nodes are determined. The adjustment results include: maintaining the current number of neuron nodes or dividing the neuron nodes into more than one sub-neuron nodes, thereby distinguishing the same disease in different situations, and different neuron nodes corresponding to different first connection relationships, thereby improving the accuracy of discovering other related conditions.

[0104] Furthermore, the step of determining the associated disease of the target patient based on the target association model and the disease data includes:

[0105] The corresponding target neuron node is determined based on the aforementioned disease data;

[0106] The incidence of associated diseases is determined based on the first connection relationship corresponding to the target neuron node and the disease data.

[0107] In this embodiment, a pathfinding algorithm can be used to determine other neuron nodes besides the target neuron node, and the probability of reaching other neuron nodes can be calculated based on the associated disease statistics of the first connection relationship. It should be noted that the biggest advantage of this method is that as long as a disease has been identified in the current target patient's disease data, and when the disease data does not include the disease data corresponding to all the standard feature data of the neuron node, the standard feature data can be directly used as the disease data. Commonly, if the patient has only completed a portion of the blood test, the standard feature data can be chosen as the disease data.

[0108] In addition, optionally, when the disease data includes data corresponding to all standard feature data, a target data range in which the disease data falls is determined, and the associated disease statistics are determined based on a table of the target data range.

[0109] Furthermore, this invention also proposes a neural network hospital data association analysis system, which includes:

[0110] The acquisition module is used to acquire historical medical data, health-related status data, and target patient condition data recorded by the hospital's electronic medical record system, and to divide the historical medical data according to disease type to obtain disease medical data related to each disease.

[0111] The construction module is used to generate a preset disease association model based on health association status data. The preset disease association model includes: neuron nodes corresponding to each disease and the first connection relationship between neuron nodes.

[0112] The optimization module is used to adjust the preset disease association model based on the disease diagnosis and treatment data to obtain the target association model;

[0113] The association module is used to determine the associated diseases of the target patient based on the target association model and the disease data.

[0114] The neural network hospital data association analysis system can implement the steps of any of the embodiments of the neural network hospital data association analysis method described above.

[0115] In addition, the neural network hospital data association analysis system also includes a display module, which displays the associated diseases of the target patient and their corresponding probabilities through a display device.

[0116] Furthermore, this invention also proposes a neural network hospital data association analysis device, which includes: a memory, a processor, and a neural network hospital data association analysis program stored in the memory and executable on the processor. The neural network hospital data association analysis program is configured to implement the steps of an embodiment of the neural network hospital data association analysis method described above.

[0117] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0118] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0119] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0120] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for correlation analysis of hospital data using neural networks, characterized in that, The steps of the neural network hospital data association analysis method include: The system acquires historical medical records, health status data, and target patient condition data from the hospital's electronic medical record system. It then divides the historical medical records by disease type to obtain disease medical records related to each disease. A preset disease association model is generated based on health association status data. The preset disease association model includes: neuron nodes corresponding to each disease and the first connection relationship between neuron nodes. The health association status data includes: secondary disease relationships, complication relationships, and disease classification relationships. The step of generating the preset disease association model based on the health association status data includes: Generate neuron nodes corresponding to the diseases present in the health-related status data; A first connection relationship between neuronal nodes is generated based on the secondary disease relationship, the complication relationship, and the disease classification relationship, wherein the first connection relationship is directed and is used to calculate the probability of realization from the first neuronal node to the second neuronal node. The preset disease association model is adjusted based on the disease diagnosis and treatment data to obtain the target association model; The associated diseases of the target patient are determined based on the target association model and the disease data; The disease diagnosis and treatment data includes: laboratory indicator data. The step of adjusting the preset disease association model based on the disease diagnosis and treatment data to obtain the target association model includes: Based on the test index data, determine the standard characteristic data corresponding to each disease; Add the standard feature data to the corresponding neuron node; The first connection relationship is updated based on the standard feature data, the first connection relationship, and the test index data to obtain the target association model; The step of updating the first connection relationship based on the standard feature data, the first connection relationship, and the test index data to obtain the target association model includes: The target data range of the test index data is determined based on the standard feature data and the first connection relationship. Determine the associated disease statistics results corresponding to the standard feature data based on the target data range; A second connection relationship is generated based on the associated disease statistics; The second connection relationship is used as the first connection relationship.

2. The neural network hospital data association analysis method as described in claim 1, characterized in that, The first connection relationship is a link between a first neuron node and a second neuron node. The step of determining the target data range of the test index data based on the standard feature data and the first connection relationship includes: Multiple target data ranges are generated based on the standard feature data corresponding to the first neuron node.

3. The neural network hospital data association analysis method as described in claim 1, characterized in that, The step of determining the standard characteristic data corresponding to each disease based on the test index data includes: Based on the aforementioned test index data, determine the disease-strongly correlated and weakly correlated indicators; Based on the test index data, determine the average incidence rate and incidence rate variance corresponding to the disease strong correlation index. Based on the test index data, determine the first value of the weakly correlated index with the highest frequency. The average incidence rate index, the incidence rate variance index, and the first value of the weak correlation index are used as the standard feature data.

4. The neural network hospital data association analysis method as described in claim 1, characterized in that, Before the step of determining the standard characteristic data corresponding to each disease based on the test index data, the method further includes: Based on the disease diagnosis and treatment data and clustering algorithm, the centroid distance between each centroid in the cluster space is calculated to obtain multiple centroid distances; The adjustment result of the corresponding neuron node is determined based on multiple centroid distances, and the adjustment result includes: maintaining the current number of neuron nodes or dividing the neuron node into more than one sub-neuron node.

5. The neural network hospital data association analysis method according to any one of claims 2 to 4, characterized in that, The step of determining the associated disease of the target patient based on the target association model and the disease data includes: The corresponding target neuron node is determined based on the aforementioned disease data; The incidence of associated diseases is determined based on the first connection relationship corresponding to the target neuron node and the disease data.

6. A neural network-based hospital data association analysis system, characterized in that, The neural network hospital data association analysis system is used to perform the steps of the neural network hospital data association analysis method as described in claim 1, and the neural network hospital data association analysis system includes: The acquisition module is used to acquire historical medical data, health-related status data, and target patient condition data recorded by the hospital's electronic medical record system, and to divide the historical medical data according to disease type to obtain disease medical data related to each disease. A construction module is used to generate a preset disease association model based on health association status data. The preset disease association model includes: neuron nodes corresponding to each disease and the first connection relationship between neuron nodes. The health association status data includes: secondary disease relationships, complication relationships, and disease classification relationships. The step of generating the preset disease association model based on the health association status data includes: Generate neuron nodes corresponding to the diseases present in the health-related status data; A first connection relationship between neuronal nodes is generated based on the secondary disease relationship, the complication relationship, and the disease classification relationship, wherein the first connection relationship is directed and is used to calculate the probability of realization from the first neuronal node to the second neuronal node. The optimization module is used to adjust the preset disease association model based on the disease diagnosis and treatment data to obtain the target association model; The association module is used to determine the associated diseases of the target patient based on the target association model and the disease data.

7. A neural network hospital data association analysis device, characterized in that, The neural network hospital data association analysis device includes: a memory, a processor, and a neural network hospital data association analysis program stored in the memory and executable on the processor, wherein the neural network hospital data association analysis program is configured to implement the steps of the neural network hospital data association analysis method as described in any one of claims 1 to 5.

Images