Intelligent medical information processing and analyzing system based on big data

Abstract

The invention belongs to the technical field of smart medical information, and discloses a smart medical information processing and analyzing system based on big data, which comprises a data acquisition module used for acquiring a medical behavior data set and preprocessing the medical behavior data set, the patient medical behavior data set comprises patient attention data on diseases, patient staying data in a certain disease related diagnosis and treatment link and patient interaction data with medical services or information; the data processing module is used for processing the patient medical behavior data set; according to the method, the patient interest reference value is obtained by analyzing the patient medical behavior data set, the attention tendency and interest preference of the patient for specific diseases or medical services can be deeply informed, and the patient preference classification result and the required adjustment result are obtained by analyzing the patient behavior data. And a medical institution can clearly know the demand popularity and resource utilization efficiency of medical services related to different diseases.

Description

Technical Field

[0001] The present invention relates to the field of smart medical information technology, and more specifically, to a smart medical information processing and analysis system based on big data. Background Art

[0002] With the acceleration of the process of medical informatization, medical institutions have accumulated a huge amount of medical data in their daily operations. These data include basic information of patients, medical records, examination and test results, details of the diagnosis and treatment process, and various interactive information related to medical services. However, at present, the utilization of these rich data resources by medical institutions is still relatively limited.

[0003] On the one hand, traditional medical information management systems mostly focus on data storage and basic query and retrieval functions, lacking effective means for in-depth data mining and analysis. Massive amounts of medical data are scattered in various isolated systems, forming data islands, which are difficult to integrate and correlate and analyze, and cannot give full play to the potential value of data. For example, during the diagnosis and treatment process of patients, it is difficult for doctors to quickly and comprehensively obtain all relevant behavioral data of patients at different stages and in different departments, making it difficult to accurately grasp the patient's cognition of the disease, focus of attention, and real needs for medical services. This may lead to the formulation of diagnosis and treatment plans that are not personalized and accurate enough, affecting the treatment effect and patient satisfaction. On the other hand, in terms of the provision and optimization of medical services, there is a lack of scientific decision-making mechanisms based on data-driven. When planning the allocation of medical resources, designing diagnosis and treatment processes, and determining the focus of publicity and promotion, medical institutions often rely on experience judgment or local and simple data analysis, making it difficult to fully understand the differences in the patient group's interest preferences, demand intensity, and behavior patterns for different diseases and medical services from a macro and micro level. This may result in waste or mismatch of medical resources. For example, some popular departments are short of medical resources, while some relatively unpopular departments with diseases that actually exist in patient demand have idle resources. At the same time, it is not conducive to medical institutions to adjust their service strategies in a timely manner to meet the ever-changing needs of patients, and it is difficult for them to stand out in the fierce competition in the medical market.

[0004] In view of this, the present invention proposes a smart medical information processing and analysis system based on big data to solve the above problems. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: comprising: a data collection module, used to collect and pre-process a medical behavior data set, wherein the patient medical behavior data set includes patient attention data on a disease, patient stay data in a diagnosis and treatment link related to a disease, and patient interaction data with medical services or information;

[0006] The data processing module processes the patient medical behavior data set to obtain the patient interest reference value, classifies the patient interest reference value, and obtains the patient preference classification result;

[0007] The data analysis module analyzes the patient's medical behavior data set to obtain the required adjustment results;

[0008] The optimization suggestion module analyzes the required adjustment results and obtains suggestions based on patient needs.

[0009] Furthermore, the methods for collecting patient medical behavior data sets include:

[0010] Data monitoring points are set up on the medical information platform to count the number of times that the patient's stay time on a specific disease information page exceeds a preset attention time threshold, and obtain the patient's gaze data;

[0011] Through the location tracking equipment inside the hospital, the time from the patient entering to leaving a certain disease-related diagnosis and treatment link is collected to obtain the patient's stay data;

[0012] By accessing the hospital's internal interactive data platform, the number of interactions between patients and medical services or information is counted to obtain patient interaction data. The forms of interaction include clicks on symptom information, consultation questions on treatment plans, or appointments for medical services.

[0013] Preprocessing methods also include data cleaning, removing errors and duplicate data, and data standardization, unifying data from different sources and formats for subsequent processing.

[0014] Furthermore, the steps of processing the patient medical behavior data set include:

[0015] Q1: Based on the patient medical behavior data set, by substituting into the calculation formula: Get the average attention time of each disease , Average duration of stay in the diagnosis and treatment phase Number of medical interactions ,in Provide patients with data on their symptoms. Total time spent caring for patients, Keep data for patients during the diagnosis and treatment process. is the total length of stay of the patient, Interactive data for patients, is the total number of patient interactions;

[0016] By substituting into the calculation formula: Get the patient behavior reference value, where , and They are the corresponding weight factors, which can be determined according to factors such as the severity of the disease and the importance of the diagnosis and treatment links;

[0017] Q2: Collect historical patient behavior reference values ​​of group L as the sample set, and divide the sample set into a training set of 70% L, a test set of 15% L, and a validation set of 15% L;

[0018] Q3: Establish a preference classification model based on the training set and obtain the historical patient behavior reference value in the training set; preset the lower data cluster center K, the general data cluster center U and the higher data cluster center P, and randomly select three data points in the training set as the primary cluster center, representing the lower data cluster center K, the general data cluster center U and the higher data cluster center P respectively;

[0019] Q4: By substituting into the calculation formula: Get the distance between data items, calculate the distance between the data items in the training set and the lower data cluster center K, the general data cluster center U and the higher data cluster center P, and assign the data items to the nearest primary cluster center to obtain three intermediate cluster centers, where and is the coordinate value of the data point, and For two data points, The value of the sub-data item, is the number of sub-data items;

[0020] Q5: Calculate the mean of the three intermediate cluster centers respectively and calculate them again as the new primary cluster centers;

[0021] Q6: Repeat Q4 and Q5 until the preset number of iterations is reached to obtain the patient interest classification model;

[0022] Q7: Output the patient behavior reference value to the interest classification model and output the patient interest reference value.

[0023] Further, the methods for classifying the patient interest reference values ​​include:

[0024] A patient preference threshold interval (W1, W2) is preset. When the patient interest reference value is less than W1, a lower attention signal information is generated. When the patient interest reference value is greater than or equal to W1 and less than or equal to W2, a general attention signal information is generated. When the patient interest reference value is greater than W2, a higher attention signal information is generated.

[0025] The lower attention signal information includes a field showing the degree of preference of a group of patients for the disease or medical service in the range of 1 to 3 points, the general attention signal information includes a field showing the degree of preference in the range of 4 to 7 points, and the higher attention signal information includes a field showing the degree of preference in the range of 8 to 10 points, and the total score is 10 points;

[0026] The lower attention signal information, the general attention signal information and the higher attention signal information are packaged to obtain the patient preference classification results.

[0027] Furthermore, the methods for analyzing the patient medical behavior dataset include:

[0028] A gaze threshold interval (D1, D2) is preset. When the patient's attention data to the symptom is less than D1, low attention information is obtained. When the patient's attention data to the symptom is greater than D1 and less than D2, general attention information is obtained. When the patient's attention data to the symptom is greater than D2, high attention information is obtained.

[0029] The preset stay threshold interval (D3, D4) is that when the patient's stay data in the diagnosis and treatment link is less than D3, a lower stay information is obtained; when the patient's stay data in the diagnosis and treatment link is greater than D3 and less than D4, a general stay information is obtained; when the patient's stay data in the diagnosis and treatment link is greater than D4, a higher stay information is obtained;

[0030] Preset interaction threshold interval (D5, D6), when the patient interaction data is less than D5, obtain low interaction information, when the patient interaction data is greater than D5 and less than D6, obtain general interaction information, when the patient interaction data is greater than D6, obtain high interaction information;

[0031] The lower attention information, the general attention information, the higher attention information, the lower stay information, the general stay information, the higher stay information, the lower interaction information, the general interaction information and the higher interaction information are packaged to obtain the desired adjustment result.

[0032] Further, the method of analyzing the required adjustment results includes:

[0033] When the required adjustment result is lower attention information, lower retention information or lower interaction information, an adjustment decision is generated; when the required adjustment result is average attention information, average retention information or average interaction information, an improvement decision is generated; when the required adjustment result is higher attention information, higher retention information or higher interaction information, a recommendation decision is generated;

[0034] Adjustment decisions include reducing the push ratio of medical services or information related to the current disease in real time and marking it as information that needs to be adjusted. Information that needs to be adjusted includes explaining that the medical services or information related to the current disease are less attractive to the current patients and requiring medical personnel to promptly supplement the knowledge and information related to the current disease and introduce more diversified treatment plan information; improvement decisions include explaining that the medical services or information related to the current disease are generally attractive to the current patients and requiring medical personnel to optimize the advantages of the current medical services; recommendation decisions include increasing the push ratio of medical services or information related to the current disease in real time and marking it as information that needs to be recommended. Information that needs to be recommended includes explaining that the medical services or information related to the current disease are more attractive to the current patients and requiring medical personnel to expand the interactive functions of the current medical services;

[0035] Package and adjust decisions, improve decisions, and recommend decisions to get advice on patient needs.

[0036] The technical effects and advantages of the big data-based intelligent medical information processing and analysis system of the present invention are as follows:

[0037] The present invention analyzes the patient's medical behavior data set to obtain the patient's interest reference value, which can deeply understand the patient's attention tendency and interest preference for specific diseases or medical services. With the help of the patient preference classification results and the required adjustment results obtained by analyzing the patient's behavior data, medical institutions can clearly understand the demand for medical services related to different diseases and the efficiency of resource utilization. Based on this, medical resources such as manpower, material resources and financial resources can be reasonably allocated; this smart medical information processing and analysis system based on big data can effectively solve many problems such as the current insufficient utilization of medical data, the lack of personalization and accuracy of medical services, and the unreasonable allocation of medical resources. It has extremely important significance and broad application prospects for promoting the development of smart medicine. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a schematic diagram of a big data-based intelligent medical information processing and analysis system of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] See also Figure 1As shown, the smart medical information processing and analysis system based on big data described in this embodiment includes: a data acquisition module, which is used to collect and pre-process a medical behavior data set, wherein the patient medical behavior data set includes patient attention data on symptoms, patient stay data in a diagnosis and treatment link related to a certain symptom, and patient interaction data with medical services or information;

[0041] Ways to collect patient medical behavior data sets include:

[0042] Data monitoring points are set up on the medical information platform to count the number of times that the patient's stay time on a specific disease information page exceeds a preset attention time threshold, and obtain the patient's gaze data;

[0043] Through the location tracking equipment inside the hospital, the time from the patient entering to leaving a certain disease-related diagnosis and treatment link is collected to obtain the patient's stay data;

[0044] By accessing the hospital's internal interactive data platform, the number of interactions between patients and medical services or information is counted to obtain patient interaction data. The forms of interaction include clicks on symptom information, consultation questions on treatment plans, or appointments for medical services.

[0045] Preprocessing methods also include data cleaning, removing errors and duplicate data, and data standardization, unifying data from different sources and formats for subsequent processing.

[0046] The data processing module processes the patient medical behavior data set to obtain the patient interest reference value, classifies the patient interest reference value, and obtains the patient preference classification result;

[0047] The steps for processing the patient medical behavior dataset include:

[0048] Q1: Based on the patient medical behavior data set, by substituting into the calculation formula: Get the average attention time of each disease , Average duration of stay in the diagnosis and treatment phase Number of medical interactions ,in Provide patients with data on their symptoms. Total time spent caring for patients, Keep data for patients during the diagnosis and treatment process. is the total length of stay of the patient, Interactive data for patients, is the total number of patient interactions;

[0049] By substituting into the calculation formula: Get the patient behavior reference value, where , and They are the corresponding weight factors, which can be determined according to factors such as the severity of the disease and the importance of the diagnosis and treatment links;

[0050] Q2: Collect historical patient behavior reference values ​​of group L as the sample set, and divide the sample set into a training set of 70% L, a test set of 15% L, and a validation set of 15% L;

[0051] Q3: Establish a preference classification model based on the training set and obtain the historical patient behavior reference value in the training set; preset the lower data cluster center K, the general data cluster center U and the higher data cluster center P, and randomly select three data points in the training set as the primary cluster center, representing the lower data cluster center K, the general data cluster center U and the higher data cluster center P respectively;

[0052] Q4: By substituting into the calculation formula: Get the distance between data items, calculate the distance between the data items in the training set and the lower data cluster center K, the general data cluster center U and the higher data cluster center P, and assign the data items to the nearest primary cluster center to obtain three intermediate cluster centers, where and is the coordinate value of the data point, and For two data points, The value of the sub-data item, is the number of sub-data items;

[0053] Q5: Calculate the mean of the three intermediate cluster centers respectively and calculate them again as the new primary cluster centers;

[0054] Q6: Repeat Q4 and Q5 until the preset number of iterations is reached to obtain the patient interest classification model;

[0055] Q7: Output the patient behavior reference value to the interest classification model and output the patient interest reference value.

[0056] The data analysis module analyzes the patient's medical behavior data set to obtain the required adjustment results;

[0057] Methods for analyzing patient medical behavior datasets include:

[0058] A gaze threshold interval (D1, D2) is preset. When the patient's attention data to the symptom is less than D1, low attention information is obtained. When the patient's attention data to the symptom is greater than D1 and less than D2, general attention information is obtained. When the patient's attention data to the symptom is greater than D2, high attention information is obtained.

[0059] The preset stay threshold interval (D3, D4) is that when the patient's stay data in the diagnosis and treatment link is less than D3, a lower stay information is obtained; when the patient's stay data in the diagnosis and treatment link is greater than D3 and less than D4, a general stay information is obtained; when the patient's stay data in the diagnosis and treatment link is greater than D4, a higher stay information is obtained;

[0060] Preset interaction threshold interval (D5, D6), when the patient interaction data is less than D5, obtain low interaction information, when the patient interaction data is greater than D5 and less than D6, obtain general interaction information, when the patient interaction data is greater than D6, obtain high interaction information;

[0061] The lower attention information, the general attention information, the higher attention information, the lower stay information, the general stay information, the higher stay information, the lower interaction information, the general interaction information and the higher interaction information are packaged to obtain the desired adjustment result.

[0062] The optimization suggestion module analyzes the required adjustment results and obtains suggestions on patient needs;

[0063] Ways to analyze the results of the required adjustments include:

[0064] When the required adjustment result is lower attention information, lower retention information or lower interaction information, an adjustment decision is generated; when the required adjustment result is average attention information, average retention information or average interaction information, an improvement decision is generated; when the required adjustment result is higher attention information, higher retention information or higher interaction information, a recommendation decision is generated;

[0065] Adjustment decisions include reducing the push ratio of medical services or information related to the current disease in real time and marking it as information that needs to be adjusted. Information that needs to be adjusted includes explaining that the medical services or information related to the current disease are less attractive to the current patients and requiring medical personnel to promptly supplement the knowledge and information related to the current disease and introduce more diversified treatment plan information; improvement decisions include explaining that the medical services or information related to the current disease are generally attractive to the current patients and requiring medical personnel to optimize the advantages of the current medical services; recommendation decisions include increasing the push ratio of medical services or information related to the current disease in real time and marking it as information that needs to be recommended. Information that needs to be recommended includes explaining that the medical services or information related to the current disease are more attractive to the current patients and requiring medical personnel to expand the interactive functions of the current medical services;

[0066] Package and adjust decisions, improve decisions, and recommend decisions to get advice on patient needs.

[0067] Since the electronic device introduced in this embodiment is an electronic device used to implement the exhibition display method based on virtual reality technology in the embodiment of this application, based on the exhibition display method based on virtual reality technology introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not described in detail here. As long as the technical personnel of this field implement the electronic device used in the exhibition display method based on virtual reality technology in the embodiment of this application, it belongs to the scope of protection of this application.

[0068] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0069] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical patients in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A smart medical information processing and analysis system based on big data, characterized in that: include: The data collection module is used to collect and preprocess the medical behavior data set. The patient medical behavior data set includes the patient's attention to the disease, the patient's stay in a certain disease diagnosis and treatment link, and the patient's interaction data with medical services or information; The data processing module processes the patient medical behavior data set to obtain the patient interest reference value, classifies the patient interest reference value, and obtains the patient preference classification result; The data analysis module analyzes the patient's medical behavior data set to obtain the required adjustment results; The optimization suggestion module analyzes the required adjustment results and obtains suggestions based on patient needs.

2. According to claim 1, a smart medical information processing and analysis system based on big data is characterized in that: Ways to collect patient medical behavior data sets include: Data monitoring points are set up on the medical information platform to count the number of times that the patient's stay time on a specific disease information page exceeds a preset attention time threshold, and obtain the patient's gaze data; Through the location tracking equipment inside the hospital, the time from when the patient enters to when they leave a certain disease diagnosis and treatment link is collected to obtain the patient's stay data; By accessing the hospital's internal interactive data platform, the number of interactions between patients and medical services or information is counted to obtain patient interaction data. The forms of interaction include clicks on symptom information, consultation questions on treatment plans, or appointments for medical services. Preprocessing methods also include data cleaning, removing errors and duplicate data, and data standardization, unifying data from different sources and formats for subsequent processing.

3. The intelligent medical information processing and analysis system based on big data according to claim 2 is characterized in that: The steps for processing the patient medical behavior dataset include: Q1: Based on the patient medical behavior data set, by substituting into the calculation formula: Get the average attention time of each disease , Average duration of stay in the diagnosis and treatment phase Number of medical interactions ,in Provide patients with data on their symptoms. Total time spent caring for patients, Keep data for patients during the diagnosis and treatment process. is the total length of stay of the patient, Interactive data for patients, is the total number of patient interactions; By substituting into the calculation formula: Get the patient behavior reference value, where , and They are the corresponding weight factors, which can be determined according to the severity of the disease and the importance of the diagnosis and treatment links; Q2: Collect historical patient behavior reference values ​​of group L as the sample set, and divide the sample set into a training set of 70% L, a test set of 15% L, and a validation set of 15% L; Q3: Establish a preference classification model based on the training set and obtain the historical patient behavior reference value in the training set; preset the lower data cluster center K, the general data cluster center U and the higher data cluster center P, and randomly select three data points in the training set as the primary cluster center, representing the lower data cluster center K, the general data cluster center U and the higher data cluster center P respectively; Q4: By substituting into the calculation formula: Get the distance between data items, calculate the distance between the data items in the training set and the lower data cluster center K, the general data cluster center U and the higher data cluster center P, and assign the data items to the nearest primary cluster center to obtain three intermediate cluster centers, where and is the coordinate value of the data point, and For two data points, The value of the sub-data item, is the number of sub-data items; Q5: Calculate the mean of the three intermediate cluster centers respectively and calculate them again as the new primary cluster centers; Q6: Repeat Q4 and Q5 until the preset number of iterations is reached to obtain the patient interest classification model; Q7: Output the patient behavior reference value to the interest classification model and output the patient interest reference value.

4. The intelligent medical information processing and analysis system based on big data according to claim 3 is characterized in that: Ways to categorize patient interest references include: A patient preference threshold interval (W1, W2) is preset. When the patient interest reference value is less than W1, a lower attention signal information is generated. When the patient interest reference value is greater than or equal to W1 and less than or equal to W2, a general attention signal information is generated. When the patient interest reference value is greater than W2, a higher attention signal information is generated. The lower attention signal information includes a field showing the degree of preference of a group of patients for the disease or medical service in the range of 1 to 3 points, the general attention signal information includes a field showing the degree of preference in the range of 4 to 7 points, and the higher attention signal information includes a field showing the degree of preference in the range of 8 to 10 points, and the total score is 10 points; The lower attention signal information, the general attention signal information and the higher attention signal information are packaged to obtain the patient preference classification results.

5. The intelligent medical information processing and analysis system based on big data according to claim 4 is characterized in that: Methods for analyzing patient medical behavior datasets include: A gaze threshold interval (D1, D2) is preset. When the patient's attention data to the symptom is less than D1, low attention information is obtained. When the patient's attention data to the symptom is greater than D1 and less than D2, general attention information is obtained. When the patient's attention data to the symptom is greater than D2, high attention information is obtained. The preset stay threshold interval (D3, D4) is that when the patient's stay data in the diagnosis and treatment link is less than D3, a lower stay information is obtained; when the patient's stay data in the diagnosis and treatment link is greater than D3 and less than D4, a general stay information is obtained; when the patient's stay data in the diagnosis and treatment link is greater than D4, a higher stay information is obtained; Preset interaction threshold interval (D5, D6), when the patient interaction data is less than D5, obtain low interaction information, when the patient interaction data is greater than D5 and less than D6, obtain general interaction information, when the patient interaction data is greater than D6, obtain high interaction information; The lower attention information, the general attention information, the higher attention information, the lower stay information, the general stay information, the higher stay information, the lower interaction information, the general interaction information and the higher interaction information are packaged to obtain the desired adjustment result.

6. The intelligent medical information processing and analysis system based on big data according to claim 5 is characterized in that: Ways to analyze the results of the required adjustments include: When the required adjustment result is lower attention information, lower retention information or lower interaction information, an adjustment decision is generated; when the required adjustment result is average attention information, average retention information or average interaction information, an improvement decision is generated; when the required adjustment result is higher attention information, higher retention information or higher interaction information, a recommendation decision is generated; Adjustment decisions include reducing the push ratio of medical services or information for current symptoms in real time and marking it as information that needs to be adjusted. Information that needs to be adjusted includes explaining that the medical services or information for current symptoms are less attractive to current patients and requiring medical personnel to promptly supplement the current disease knowledge information and introduce more diversified treatment plan information; improvement decisions include explaining that the medical services or information for current symptoms are generally attractive to current patients and requiring medical personnel to optimize the advantages of current medical services; recommendation decisions include increasing the push ratio of medical services or information for current symptoms in real time and marking it as information that needs to be recommended. Information that needs to be recommended includes explaining that the medical services or information for current symptoms are more attractive to current patients and requiring medical personnel to expand the interactive functions of current medical services; Package and adjust decisions, improve decisions, and recommend decisions to get advice on patient needs.

Images