A cloud computing-based health record management method and system for the elderly

Abstract

The present invention relates to the field of health management technology, and discloses a cloud computing-based health record management method and system for the elderly, including: performing integrity verification processing on individual identification information to obtain verification identification information, performing standardization processing on the verification identification information to obtain benchmark identification information; performing classification and labeling processing on previous disease records to obtain classification and labeling records; performing quality purification processing on the classification and labeling records to obtain target labeling records, extracting pathological abnormality coupling data from periodic physical examination data, and analyzing the current pathological factors and potential pathological factors corresponding to the elderly; evaluating the health risk level corresponding to the current pathological factors, mining pathological inducing factors from daily behavioral factors, and calculating the inducing contribution of pathological inducing factors to potential pathological factors; and using cloud computing technology to construct health management files corresponding to the elderly. The main purpose of the present invention is to solve the problem of low rationality of health record management for the elderly.

Description

Technical Field

[0001] The present invention relates to a cloud computing-based elderly health record management method and system, belonging to the technical field of health management. Background Art

[0002] With the rapid development of information technology, the way of health record management is also constantly evolving. Traditional health record management of the elderly mostly relies on paper records or local electronic document storage. The data are relatively scattered and cannot reflect the dynamic changes in the health status of the elderly in a timely manner. For example, when the elderly undergo routine physical examinations at community health service centers, if they want to be referred to large hospitals, it often takes a lot of time and energy to transmit and integrate health record information, which seriously affects the efficiency and quality of medical services. In addition, the existing health record management of the elderly only relies on relevant medical test data during the analysis process, and does not combine the elderly's living habits and various test data for health analysis, which leads to a decrease in the accuracy of health analysis. Therefore, a method is needed to improve the rationality of health record management for the elderly. Summary of the invention

[0003] The present invention provides a method and system for managing health records of the elderly based on cloud computing, the main purpose of which is to solve the problem of low rationality in the management of health records of the elderly.

[0004] To achieve the above purpose, the present invention provides a cloud computing-based health record management method for the elderly, comprising:

[0005] Collecting individual identification information corresponding to the elderly, performing integrity verification processing on the individual identification information to obtain verification identification information, and performing standardization processing on the verification identification information to obtain reference identification information;

[0006] Based on the benchmark identification information, the corresponding past disease history of the elderly is collected, and the past disease history is classified and labeled according to the preset disease classification criteria to obtain the classified and labeled history;

[0007] Performing quality purification processing on the classified and labeled resume to obtain a target labeled resume, acquiring periodic physical examination data and daily behavior habit data of the elderly, extracting pathological abnormality coupling data from the periodic physical examination data, and analyzing the current pathological factors and potential pathological factors corresponding to the elderly based on the pathological abnormality coupling data;

[0008] According to the classification and labeling history, the health risk level corresponding to the current pathological factor is evaluated, the daily behavior factors in the daily behavior habit data are analyzed, the pathological inducing factors are mined from the daily behavior factors, and the inducing contribution of the pathological inducing factors to the potential pathological factors is calculated;

[0009] In combination with the current pathological factors, the potential pathological factors, the health risk level, the classification and annotation history and the induced contribution rate, cloud computing technology is used to construct a corresponding health management file for the elderly.

[0010] Optionally, the performing integrity verification processing on the individual identification information to obtain verification identification information includes:

[0011] Checking the information format corresponding to the individual identification information, and performing format verification processing on the individual identification information based on the information format to obtain initial verification identification information;

[0012] According to a preset information database, information verification processing is performed on the initial verification identification information to obtain a verification result;

[0013] Based on the verification result, determining the information to be corrected in the initial verification identification information;

[0014] The information to be corrected is corrected and verified to obtain verification identification information.

[0015] Optionally, the performing standardization processing on the verification identification information to obtain reference identification information includes:

[0016] Identify the identification information characters and the communication number characters in the verification identification information, and extract the rare characters in the identification information characters;

[0017] Performing encoding conversion processing on the rare characters of the information to obtain encoded information characters;

[0018] Performing standard processing on the communication number characters to obtain standard number characters;

[0019] The standard number characters and the rare information characters are used to perform information update processing on the verification identification information to obtain the reference identification information.

[0020] Optionally, the previous disease history is classified and annotated according to a preset disease classification criterion to obtain a classified and annotated history, including:

[0021] Extracting information from the past medical history to obtain detailed disease information;

[0022] Analyzing the correlation and matching degree between the disease detailed information and the disease classification criteria;

[0023] In combination with the correlation matching degree, the disease detailed information and the disease classification criteria, the previous disease history is classified and labeled to obtain a classified and labeled history.

[0024] Optionally, the performing quality purification processing on the classification annotation history to obtain the target annotation history includes:

[0025] Cleaning the classification and labeling history to obtain a cleaned and labeled history;

[0026] Calculating the history information entropy corresponding to the cleaning and marking history;

[0027] Extracting the history representation in the cleaning and marking history, and calculating the representation Gini index corresponding to the history representation;

[0028] The cleaned and labeled records are subjected to quality purification processing in combination with the record information entropy and the characterization Gini index to obtain the target labeled records.

[0029] Optionally, the calculating the Gini impurity corresponding to the resume representation includes:

[0030] Identify the resume sub-features in the resume representation, count the number of sub-features of the resume sub-features in the resume representation, and count the number of representations corresponding to the resume representation;

[0031] Combining the number of sub-features and the number of representations, calculating the distribution probability corresponding to the resume sub-feature;

[0032] Combined with the distribution probability, the Gini impurity corresponding to the resume representation is calculated by the following formula:

[0033]

[0034] Among them, F represents the Gini impurity corresponding to the resume representation, G d represents the distribution probability corresponding to the dth feature in the resume sub-feature, d represents the feature sequence number corresponding to the resume sub-feature, and p represents the number of features corresponding to the resume sub-feature.

[0035] Optionally, extracting the pathological abnormality coupling data from the periodic physical examination data includes:

[0036] Identify the abnormal physical examination index corresponding to the periodic physical examination data and the corresponding abnormal physical examination index data;

[0037] Normalizing the abnormal physical examination index data to obtain physical examination observation values;

[0038] Based on the physical examination observation value, calculating the indicator deviation value corresponding to the abnormal physical examination indicator;

[0039] Analyzing the pathological physical examination indicators among the abnormal physical examination indicators according to the indicator deviation values;

[0040] Based on the pathological physical examination index, pathological abnormality coupling data in the periodic physical examination data is extracted.

[0041] Optionally, evaluating the health risk level corresponding to the current pathological factor according to the classification and labeling history includes:

[0042] Extracting the pathological medication information in the classified and marked resume, wherein the pathological medication information includes: pathological medication name, pathological medication dosage, pathological medication frequency and pathological medication cycle;

[0043] Calculate the drug utilization intensity corresponding to the pathological medication name by combining the pathological medication dosage and the pathological medication frequency;

[0044] In combination with the pathological medication cycle and the medication utilization intensity, analyzing the medication utilization characteristics corresponding to the pathological medication name;

[0045] The health risk level corresponding to the current pathological factor is evaluated according to the drug utilization characteristics.

[0046] Optionally, the calculating the contribution of the pathology-inducing factor to the potential pathology factor comprises:

[0047] Dispatching historical pathological data corresponding to the potential pathological factors, and analyzing the pathological factors inducing the pathological factors;

[0048] Performing correlation matching processing on the pathological factors and the potential pathological factors to obtain a correlation factor set;

[0049] Determining, according to the set of associated factors, the factor weights corresponding to the pathological factors causing the disease;

[0050] Calculating the correlation strength between the pathological factors causing the disease and the potential pathological factors according to the historical pathological data;

[0051] The association strength and the factor weight are combined to calculate the contribution of the pathological factor to the potential pathological factor.

[0052] The cloud computing-based health record management system for the elderly is characterized in that the system includes:

[0053] An information processing module is used to collect individual identification information corresponding to the elderly, perform integrity verification processing on the individual identification information to obtain verification identification information, and perform standardization processing on the verification identification information to obtain reference identification information;

[0054] A disease history processing module is used to collect the corresponding past disease history of the elderly based on the benchmark identification information, and classify and annotate the past disease history according to the preset disease classification criteria to obtain a classified and annotated history;

[0055] The medical record factor analysis module is used to perform quality purification processing on the classified and labeled resumes to obtain target labeled resumes, obtain the periodic physical examination data and daily behavior habit data of the elderly, extract the pathological abnormality coupling data in the periodic physical examination data, and analyze the current pathological factors and potential pathological factors corresponding to the elderly based on the pathological abnormality coupling data;

[0056] An induction contribution calculation module is used to evaluate the health risk level corresponding to the current pathological factor according to the classification and annotation history, analyze the daily behavior factors in the daily behavior habit data, mine the pathological inducing factors from the daily behavior factors, and calculate the induction contribution of the pathological inducing factors to the potential pathological factors;

[0057] The file construction module is used to combine the current pathological factors, the potential pathological factors, the health risk level, the classification and annotation history and the induced contribution rate to construct a corresponding health management file for the elderly using cloud computing technology.

[0058] Compared with the problems described in the background technology, the present invention obtains verification identification information by performing integrity verification processing on the individual identification information, which can accurately screen out erroneous information in the individual identification information and correct it, avoid information mismatch or confusion, and improve the accuracy of the health records subsequently constructed. The present invention classifies and annotates the previous disease history according to preset disease classification criteria to obtain a classified and annotated history, which can make the disease information more orderly and clear, facilitate rapid retrieval of specific disease conditions, and help subsequent medical staff accurately grasp the patient's full medical history, improve diagnostic accuracy and efficiency, provide strong support for the formulation of health management plans for the elderly, and promote the rational allocation and utilization of medical resources. Quality purification processing can accurately eliminate erroneous or ambiguous disease labeling information, making the classification and labeling history more accurate and reliable, which helps medical staff to formulate scientific treatment plans based on high-quality data. The present invention evaluates the health risk level corresponding to the current pathological factor according to the classification and labeling history, and can understand the severity of the current pathological factor through the health risk level, thereby improving the accuracy of the subsequent construction of the health management file corresponding to the elderly. The present invention combines the current pathological factor, the potential pathological factor, the health risk level, the classification and labeling history and the induced contribution rate, and uses cloud computing technology to construct the health management file corresponding to the elderly, which can improve the rationality of the health file management of the elderly. Therefore, the cloud computing-based health file management method and system for the elderly proposed in the present invention can improve the rationality of the health file management of the elderly. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A schematic diagram of a process flow of a cloud computing-based elderly health record management method provided by an embodiment of the present invention;

[0060] Figure 2 A functional module diagram for implementing the cloud computing-based health record management system for the elderly provided in one embodiment of the present invention.

[0061] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0062] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0063] The embodiment of the present application provides a method for managing the health records of the elderly based on cloud computing. The execution subject of the method for managing the health records of the elderly based on cloud computing includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for managing the health records of the elderly based on cloud computing can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0064] Embodiment 1:

[0065] Reference Figure 1 FIG. 1 is a flow chart of a method for managing health records of the elderly based on cloud computing provided by an embodiment of the present invention. In this embodiment, the method for managing health records of the elderly based on cloud computing includes:

[0066] S1. Collect individual identification information corresponding to the elderly, perform integrity verification on the individual identification information to obtain verification identification information, and perform standardization on the verification identification information to obtain reference identification information.

[0067] The present invention performs integrity verification on the individual identification information to obtain verification identification information, which can accurately screen out erroneous information in the individual identification information and correct it, avoid information mismatch or confusion, and improve the accuracy of the health records subsequently constructed. It should be explained that the individual identification information is the individual basic information of the elderly, such as name and age, and the verification identification information is the information obtained after the erroneous information in the individual identification information is corrected. The collection of the individual identification information corresponding to the elderly can be achieved through a questionnaire survey.

[0068] Specifically, the performing integrity verification processing on the individual identification information to obtain verification identification information includes:

[0069] Checking the information format corresponding to the individual identification information, and performing format verification processing on the individual identification information based on the information format to obtain initial verification identification information;

[0070] According to a preset information database, information verification processing is performed on the initial verification identification information to obtain a verification result;

[0071] Based on the verification result, determining the information to be corrected in the initial verification identification information;

[0072] The information to be corrected is corrected and verified to obtain verification identification information.

[0073] It should be explained that the information format is the information form corresponding to the individual identification information, the initial verification identification information is the information obtained after the information format of the individual identification information is verified and corrected, the preset information database is a pre-built data storage library, the verification result is the result obtained by comparing and analyzing the initial verification identification information according to the preset information database, and the information to be corrected is the information with errors in the initial verification identification information.

[0074] Furthermore, the check of the information format corresponding to the individual identification information can be achieved through a specially developed format detection algorithm, and the format detection algorithm is compiled by a programming language; based on the information format, the individual identification information can be format-checked by compiling a corresponding format matching verification program to obtain initial verification identification information; according to a preset information database, the initial verification identification information can be information-checked by a data comparison and verification module to obtain a verification result, and the data comparison and verification module is composed of a rule engine unit for defining and managing data comparison rules; based on the verification result, the information to be corrected in the initial verification identification information can be determined through a difference analysis program; the information to be corrected can be corrected and verified by a correction execution and secondary verification tool to obtain verification identification information.

[0075] The present invention obtains benchmark identification information by standardizing the verification identification information, which can enhance data universality and standardization, facilitate subsequent data interaction and sharing between different systems, and effectively reduce recognition errors caused by data differences, etc., so that health analysis based on this information, medical service docking and other operations are more accurate and efficient, thereby improving the overall quality and coordination of health management for the elderly. It should be explained that the benchmark representation information is a standardized presentation form of the verification identification information.

[0076] In detail, the standardizing the verification identification information to obtain the reference identification information includes:

[0077] Identify the identification information characters and the communication number characters in the verification identification information, and extract the rare characters in the identification information characters;

[0078] Performing encoding conversion processing on the rare characters of the information to obtain encoded information characters;

[0079] Performing standard processing on the communication number characters to obtain standard number characters;

[0080] The standard number characters and the rare information characters are used to perform information update processing on the verification identification information to obtain the reference identification information.

[0081] It should be explained that the identification information characters are the character content in the verification identification information used to reflect identity and other related identification attributes, the communication number characters are the character content in the verification identification information related to telephone numbers and other communications, the obscure information characters are those obscure and uncommon character parts in the identification information characters, the coded information characters are the character forms formed after the obscure information characters are converted according to specific coding rules, and the standard number characters are the character forms obtained after the communication number characters are normalized by removing non-numeric characters and arranging them according to standard formats. Furthermore, the identification information characters and communication number characters in the verification identification information can be identified by OCR recognition technology, and the obscure information characters in the identification information characters can be extracted by an extraction function, and the extraction function is compiled by a scripting language, such as JS scripting language; the obscure information characters can be encoded and converted according to the obscure character encoding conversion table to obtain the coded information characters, and the table stores the correspondence between obscure characters and general encodings (such as Unicode encoding); the communication number characters can be processed by string processing functions to obtain standard number characters, and spaces may appear in different positions of the telephone number, and string processing functions can be used to delete spaces.

[0082] S2. Based on the benchmark identification information, collect the corresponding past disease history of the elderly, and classify and annotate the past disease history according to preset disease classification criteria to obtain a classified and annotated history.

[0083] The present invention classifies and annotates the previous disease history according to preset disease classification criteria to obtain a classified and annotated history, which can make the disease information more orderly and clear, facilitate rapid retrieval of specific disease conditions, and help subsequent medical staff to accurately grasp the patient's full medical history, improve diagnostic accuracy and efficiency, provide strong support for the formulation of health management plans for the elderly, and promote the rational allocation and utilization of medical resources.

[0084] It should be explained that the past medical history is a record summary of various diseases suffered by the elderly in the past. The preset disease classification criteria are standards and rules for dividing disease types based on medical expertise and actual application needs. The classified and labeled history is a more organized and easy-to-view and analyze disease information presentation form formed by classifying the past disease history according to the preset disease classification criteria and adding corresponding annotations. Furthermore, the collection of the corresponding past disease history of the elderly can be obtained through different medical institutions.

[0085] Specifically, the previous disease history is classified and labeled according to the preset disease classification criteria to obtain the classified and labeled history, including:

[0086] Extracting information from the past medical history to obtain detailed disease information;

[0087] Analyzing the correlation and matching degree between the disease detailed information and the disease classification criteria;

[0088] In combination with the correlation matching degree, the disease detailed information and the disease classification criteria, the previous disease history is classified and labeled to obtain a classified and labeled history.

[0089] It should be explained that the disease detailed information is a summary of the detailed contents such as the specific symptoms, onset, treatment process, etc. of each disease after information extraction of the previous disease history, and the correlation matching degree indicates the degree of correlation between the disease detailed information and the disease classification criteria.

[0090] Furthermore, natural language processing technology can be used to extract information from the past medical history to obtain detailed disease information; the past medical history is classified and labeled in combination with the correlation matching degree, the disease detailed information and the disease classification criteria to obtain a classified and labeled history. For example, if an elderly person has symptoms such as cough, low fever, fatigue, etc. in his past medical history, information entropy selects cough and fatigue as key detailed information. After analysis, it is found that the correlation matching degree with the respiratory disease classification criteria is relatively high, so the past medical history is labeled as a respiratory disease category to form a classified and labeled history, which is convenient for quickly understanding the elderly person's medical history tendencies in subsequent medical services.

[0091] Further, as an optional embodiment of the present invention, the analyzing the correlation matching degree between the disease detailed information and the disease classification criteria includes:

[0092] Vectorizing the disease detailed information and the disease classification criteria respectively to obtain a detailed information vector and a classification criteria vector;

[0093] Analyze the information dimensions corresponding to the key detailed information, and calculate the dimension weights corresponding to the information dimensions;

[0094] Combining the detailed information vector, the classification criteria vector and the dimension weight, the correlation coefficient between the disease detailed information and the disease classification criteria is obtained by the following formula:

[0095]

[0096] Among them, A represents the correlation coefficient between disease details and disease classification criteria, B a represents the dimension weight corresponding to the ath information in the disease details, D arepresents the detailed information vector corresponding to the ath information in the disease detailed information, a represents the sequence number corresponding to the disease detailed information, t represents the number of information in the disease detailed information, E b represents the classification criterion vector corresponding to the bth criterion in the disease classification criterion, b represents the serial number of the disease classification criterion, and r represents the number of criteria in the disease classification criterion;

[0097] Combined with the correlation coefficient, the correlation matching degree between the disease detailed information and the disease classification criteria is analyzed.

[0098] It should be explained that the detailed information vector and the classification criteria vector are respectively the vector expressions corresponding to the key detailed information and the disease classification criteria, and the information dimension is the nature of the key detailed information in different aspects. For example, in the key detailed information of the disease, the information dimension may include the symptom manifestation dimension (such as fever, cough and other different symptoms), the frequency of onset dimension (such as daily attacks, monthly attacks and other frequency situations), the disease severity dimension (mild, moderate, severe, etc.), etc. The dimension weight represents the importance corresponding to the information dimension. Furthermore, the vectorization processing of the key detailed information and the disease classification criteria can be achieved through the word2vec algorithm; the information dimension corresponding to the key detailed information can be analyzed by principal component analysis, and the dimension weight corresponding to the information dimension can be calculated by hierarchical analysis; according to the numerical value of the correlation coefficient, the correlation matching degree between the disease detailed information and the disease classification criteria is analyzed.

[0099] S3. Perform quality purification processing on the classified and labeled records to obtain target labeled records, acquire the periodic physical examination data and daily behavioral habit data of the elderly, extract the pathological abnormality coupling data in the periodic physical examination data, and analyze the current pathological factors and potential pathological factors corresponding to the elderly based on the pathological abnormality coupling data.

[0100] The present invention can accurately eliminate erroneous or ambiguous disease labeling information by performing quality purification processing on the classified labeling resume, making the classified labeling resume more accurate and reliable, and helping medical staff to formulate scientific treatment plans based on high-quality data. It should be explained that the target labeling resume is the resume obtained after removing redundant and unclear information from the classified labeling resume.

[0101] Specifically, the quality purification process of the classification labeling history to obtain the target labeling history includes:

[0102] Cleaning the classification and labeling history to obtain a cleaned and labeled history;

[0103] Calculating the history information entropy corresponding to the cleaning and marking history;

[0104] Extracting the history representation in the cleaning and marking history, and calculating the representation Gini index corresponding to the history representation;

[0105] The cleaned and labeled records are subjected to quality purification processing in combination with the record information entropy and the characterization Gini index to obtain the target labeled records.

[0106] It should be explained that the cleaned and labeled history is the history obtained after removing the repeated description information in the classified and labeled history, the history information entropy represents the complexity and uncertainty corresponding to the cleaned and labeled history, the history representation is the key element in the cleaned and labeled history that can reflect the disease-related characteristics, attributes and various aspects of manifestation, and the representation Gini index represents the representation impurity corresponding to the history representation.

[0107] Furthermore, the classified annotation history can be cleaned by a box plot method to obtain a cleaned annotation history; the history information entropy corresponding to the cleaned annotation history can be calculated by a Shannon entropy algorithm; the history representation in the cleaned annotation history can be extracted by a keyword extraction algorithm; the Gini impurity corresponding to the history representation is calculated; the cleaned annotation history is quality purified by combining the history information entropy and the Gini impurity to obtain a target annotation history;.

[0108] Further, as an optional embodiment of the present invention, the calculating the Gini impurity corresponding to the resume representation includes:

[0109] Identify the resume sub-features in the resume representation, count the number of sub-features of the resume sub-features in the resume representation, and count the number of representations corresponding to the resume representation;

[0110] Combining the number of sub-features and the number of representations, calculating the distribution probability corresponding to the resume sub-feature;

[0111] Combined with the distribution probability, the Gini impurity corresponding to the resume representation is calculated by the following formula:

[0112]

[0113] Among them, F represents the Gini impurity corresponding to the resume representation, G d represents the distribution probability corresponding to the dth feature in the resume sub-feature, d represents the feature sequence number corresponding to the resume sub-feature, and p represents the number of features corresponding to the resume sub-feature.

[0114] It should be explained that the resume sub-feature is a component of the resume representation, the number of sub-features is the number of occurrences of the resume sub-feature in each representation in the resume representation, and the distribution probability represents the corresponding occurrence probability of the resume sub-feature. Furthermore, the resume sub-features in the resume representation can be identified by a machine learning model, such as a recursive feature elimination machine learning algorithm; the distribution probability corresponding to the resume sub-feature can be obtained by calculating the quantitative ratio between the number of sub-features and the number of representations.

[0115] The present invention can obtain the data related to pathology in the periodic physical examination data by extracting the pathological abnormality coupling data in the periodic physical examination data, laying a foundation for the subsequent analysis of the current pathological factors and potential pathological factors corresponding to the elderly. It should be explained that the periodic physical examination data and the daily behavior habit data are respectively the regular physical examination data and data in daily life of the elderly, and the pathological abnormality coupling data are specific data in the periodic physical examination data that can reflect the mutual correlation or coordinated occurrence of multiple pathological abnormalities. Furthermore, the acquisition of the periodic physical examination data and daily behavior habit data of the elderly can be achieved through medical examination institutions.

[0116] In detail, the step of extracting the pathological abnormality coupling data from the periodic physical examination data includes:

[0117] Identify the abnormal physical examination index corresponding to the periodic physical examination data and the corresponding abnormal physical examination index data;

[0118] Normalizing the abnormal physical examination index data to obtain physical examination observation values;

[0119] Based on the physical examination observation value, calculating the indicator deviation value corresponding to the abnormal physical examination indicator;

[0120] Analyzing the pathological physical examination indicators among the abnormal physical examination indicators according to the indicator deviation values;

[0121] Based on the pathological physical examination index, pathological abnormality coupling data in the periodic physical examination data is extracted.

[0122] It should be explained that the abnormal physical examination indicators are physical examination items with abnormalities in the periodic physical examination data, such as blood routine tests. The abnormal physical examination indicator data are detailed test data corresponding to the abnormal physical examination indicators. The physical examination observation values ​​are quantitative expressions corresponding to the abnormal indicator physical examination data. The indicator deviation values ​​represent the differences between the abnormal physical examination indicators and the normal reference values. The pathological physical examination indicators are indicators in the abnormal physical examination indicators that reflect pathology. Furthermore, the abnormal physical examination indicators corresponding to the periodic physical examination data and their corresponding abnormal physical examination indicator data can be identified through preset abnormality detection rules. The abnormality detection rules are pre-established medical detection standards. The abnormal physical examination indicator data can be normalized through a maximum-minimum normalization algorithm to obtain physical examination observation values. The difference between the physical examination observation values ​​and the reference observation values ​​is calculated to obtain the indicator deviation values ​​corresponding to the abnormal physical examination indicators. According to the indicator deviation values, the pathological physical examination indicators in the abnormal physical examination indicators can be analyzed through medical clinical experience. The data corresponding to the pathological physical examination indicators are extracted to obtain the pathological abnormality coupling data in the periodic physical examination data.

[0123] The present invention analyzes the current pathological factors and potential pathological factors corresponding to the elderly based on the pathological abnormality coupling data, and can obtain the current pathological factors and the pathological factors that may occur in the elderly, thereby providing a basis for the subsequent evaluation of the health risk level and the calculation of the induced contribution. It should be explained that the current pathological factors and the potential pathological factors are respectively the pathological factors that have occurred in the elderly and the pathological factors that have not yet been triggered but may cause the occurrence of the disease. Furthermore, based on the pathological abnormality coupling data, the current pathological factors and potential pathological factors corresponding to the elderly are analyzed. For example, if the pathological abnormality coupling data shows that the blood pressure is long-term high and the heart structure has changed, hypertension is the current pathological factor. If only dyslipidemia is present but there are no obvious symptoms, dyslipidemia is the potential pathological factor.

[0124] S4. According to the classification and labeling history, the health risk level corresponding to the current pathological factor is evaluated, the daily behavior factors in the daily behavior habit data are analyzed, the pathological inducing factors are mined from the daily behavior factors, and the inducing contribution of the pathological inducing factors to the potential pathological factors is calculated.

[0125] The present invention evaluates the health risk level corresponding to the current pathological factor based on the classification and labeling history, and the severity corresponding to the current pathological factor can be understood through the health risk level, thereby improving the accuracy of subsequent construction of health management files corresponding to the elderly. It should be explained that the health risk level represents the severity of the pathology corresponding to the current pathological factor.

[0126] In detail, the step of evaluating the health risk level corresponding to the current pathological factor according to the classification and labeling history includes:

[0127] Extracting the pathological medication information in the classified and marked resume, wherein the pathological medication information includes: pathological medication name, pathological medication dosage, pathological medication frequency and pathological medication cycle;

[0128] Calculate the drug utilization intensity corresponding to the pathological medication name by combining the pathological medication dosage and the pathological medication frequency;

[0129] In combination with the pathological medication cycle and the medication utilization intensity, analyzing the medication utilization characteristics corresponding to the pathological medication name;

[0130] The health risk level corresponding to the current pathological factor is evaluated according to the drug utilization characteristics.

[0131] It should be explained that the pathological medication information is the medication use record information recorded in the classification and annotation history, the medication utilization intensity indicates the quantitative degree of medication use per unit time corresponding to the pathological medication name, taking into account factors such as medication dosage and frequency, and the medication utilization characteristics are the comprehensive characteristics corresponding to the pathological medication name, including medication use intensity, duration of use, dynamic changes in use, etc., which are used to comprehensively describe the application of drugs in the treatment process.

[0132] Furthermore, the pathological medication information in the classified and labeled history can be extracted through text mining technology; the pathological medication dosage and the pathological medication frequency can be combined to calculate the drug utilization intensity corresponding to the pathological medication name, and the pathological medication dosage (such as the number of tablets, milliliters, etc. used each time) can be multiplied by the pathological medication frequency (the number of times the medication is used per day or week, etc.) to obtain the amount of medication used per unit time (such as per day), thereby representing the drug utilization intensity. For example, if 3 tablets of medicine are used each time and 2 times a day, the drug utilization intensity is 6 tablets a day; in combination with the pathological medication cycle and the drug utilization intensity, the drug utilization characteristics corresponding to the pathological medication name are analyzed, such as first determining the length of the pathological medication cycle based on the start and end time of the medication, and then combining the calculated drug utilization intensity. If the medication cycle is long and the intensity is high, it may indicate that the disease is chronic or difficult to cure; if the cycle is short and the intensity is high, it may be a strong intervention for acute symptoms; if the cycle is long and the intensity is low, it may be maintenance treatment; based on the drug utilization characteristics, the health risk level corresponding to the current pathological factor is evaluated. Based on the drug utilization characteristics, if the drug utilization intensity is high and the medication cycle is long, it may imply that the current pathological factor is more stubborn and the health risk level is higher. If the intensity is low and the cycle is short, the pathological condition may be relatively mild and the risk level is lower. At the same time, the risk level is comprehensively judged based on the type of drug and common adverse reactions and complications.

[0133] The present invention calculates the inducing contribution of the pathological inducing factors to the potential pathological factors, so as to understand the extent of the role played by the medical history inducing factors in the appearance of the potential pathological factors, accurately identify the key risk factors of the pathological inducing factors, and intervene in advance to effectively reduce the possibility of the potential pathological factors developing into diseases. It should be explained that the daily behavior factors are the habitual elements in the daily behavior habit data, such as diet, physical exercise, etc. The pathological inducing factors are the factors in the daily behavior factors that can induce pathology. The inducing contribution degree indicates the extent of the role played by the pathological inducing factors in the appearance of the potential pathological factors. Furthermore, the daily behavior factors in the daily behavior habit data can be analyzed through a data mining algorithm, and the pathological inducing factors can be mined from the daily behavior factors through a medical knowledge base.

[0134] In detail, the calculation of the contribution of the pathological inducing factor to the potential pathological factor includes:

[0135] Dispatching historical pathological data corresponding to the potential pathological factors, and analyzing the pathological factors inducing the pathological factors;

[0136] Performing correlation matching processing on the pathological factors and the potential pathological factors to obtain a correlation factor set;

[0137] Determining, according to the set of associated factors, the factor weights corresponding to the pathological factors causing the disease;

[0138] Calculating the correlation strength between the pathological factors causing the disease and the potential pathological factors according to the historical pathological data;

[0139] The association strength and the factor weight are combined to calculate the contribution of the pathological factor to the potential pathological factor.

[0140] It should be explained that the historical pathological data is a collection of various types of information reflecting the occurrence and development of the disease, such as medical records, disease records, diagnosis and treatment data related to the patient group that has been diagnosed with the pathological condition in the past corresponding to the potential pathological factor; the inducing pathological factors are various factors that have been analyzed and summarized in the historical pathological data and have been proven to cause the potential pathological factor to develop into an overt disease; the associated factor set is a set obtained by associating and matching the inducing pathological factors and the pathological inducing factors; the factor weight represents the corresponding importance of the pathological inducing factors; and the association strength represents the degree of association between the pathological inducing factors and the potential pathological factors.

[0141] Furthermore, the historical pathological data corresponding to the potential pathological factors can be scheduled through a medical data management system or a database query interface, and the inducing pathological factors in the historical pathological data can be analyzed through data mining and statistical analysis algorithms; the inducing pathological factors and the pathological inducing factors can be associated and matched by establishing a feature matching model and a similarity calculation method to obtain a set of associated factors; based on the set of associated factors, the factor weights corresponding to the pathological inducing factors can be determined through expert experience evaluation and hierarchical analysis method; based on the historical pathological data, the association strength between the pathological inducing factors and the potential pathological factors can be calculated by constructing an association rule model and calculating conditional probability; the association strength is multiplied by the corresponding factor weight to obtain the contribution value corresponding to the inducing pathological factor, and each contribution value is divided by the sum of the contribution values ​​to obtain the inducing contribution degree corresponding to the inducing pathological factor.

[0142] S5. Combining the current pathological factors, the potential pathological factors, the health risk level, the classification and annotation history and the induced contribution rate, a corresponding health management file for the elderly is constructed using cloud computing technology.

[0143] The present invention combines the current pathological factors, the potential pathological factors, the health risk level, the classification and annotation history and the induction contribution rate, and uses cloud computing technology to construct a health management file corresponding to the elderly, which can improve the rationality of the health file management of the elderly. It should be explained that the health management file is an integration of the pathological factors currently present in the elderly, the potential pathological factors that have not yet occurred but have the risk of disease, the health risk level obtained through evaluation, the classification and annotation history containing information such as medication, and the induction contribution rate of pathological inducing factors to potential pathological factors, etc., which is used to comprehensively and systematically reflect the health status of the elderly and provide personalized health information. The information record set that provides the basis for health management is further combined with the current pathological factors, the potential pathological factors, the health risk level, the classification and annotation history and the induction contribution rate to construct the corresponding health management file for the elderly using cloud computing technology. For example, the elderly's high blood pressure (current pathological factors), abnormal physical indicators in the early stage that may develop into diabetes (potential pathological factors), the higher health risk level evaluated due to a comprehensive assessment of multiple factors, the classification and annotation history of past disease diagnosis and treatment and medication, the high induction contribution rate of high-sugar diet to diabetes and other data are integrated and constructed on the cloud platform with the help of the powerful storage and computing capabilities of cloud computing.

[0144] Compared with the problems described in the background technology, the present invention obtains verification identification information by performing integrity verification processing on the individual identification information, which can accurately screen out erroneous information in the individual identification information and correct it, avoid information mismatch or confusion, and improve the accuracy of the health records subsequently constructed. The present invention classifies and annotates the previous disease history according to preset disease classification criteria to obtain a classified and annotated history, which can make the disease information more orderly and clear, facilitate rapid retrieval of specific disease conditions, and help subsequent medical staff accurately grasp the patient's full medical history, improve diagnostic accuracy and efficiency, provide strong support for the formulation of health management plans for the elderly, and promote the rational allocation and utilization of medical resources. Quality purification can accurately eliminate erroneous or ambiguous disease labeling information, making the classification and labeling history more accurate and reliable, which helps medical staff to formulate scientific treatment plans based on high-quality data. The present invention evaluates the health risk level corresponding to the current pathological factor according to the classification and labeling history, and can understand the severity of the current pathological factor through the health risk level, thereby improving the accuracy of the subsequent construction of the health management file corresponding to the elderly. The present invention combines the current pathological factor, the potential pathological factor, the health risk level, the classification and labeling history and the induced contribution rate, and uses cloud computing technology to construct the health management file corresponding to the elderly, which can improve the rationality of the health file management of the elderly. Therefore, the cloud computing-based health file management method for the elderly proposed in the present invention can improve the rationality of the health file management of the elderly.

[0145] Embodiment 2:

[0146] like Figure 2 The figure shows a functional module diagram of a cloud computing-based health record management system for the elderly provided by one embodiment of the present invention.

[0147] The cloud computing-based health record management system 100 for the elderly described in the present invention can be installed in an electronic device. According to the functions implemented, the cloud computing-based health record management system 100 for the elderly can include an information processing module 101, a disease history processing module 102, a medical history factor analysis module 103, an induced contribution calculation module 104 and a file construction module 105. The module described in the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0148] In this embodiment, the functions of each module / unit are as follows:

[0149] The information processing module 101 is used to collect individual identification information corresponding to the elderly, perform integrity verification processing on the individual identification information to obtain verification identification information, and perform standardization processing on the verification identification information to obtain reference identification information;

[0150] The disease history processing module 102 is used to collect the corresponding past disease history of the elderly based on the benchmark identification information, and classify and annotate the past disease history according to the preset disease classification criteria to obtain the classified and annotated history;

[0151] The medical record factor analysis module 103 is used to perform quality purification processing on the classified and labeled resume to obtain the target labeled resume, obtain the periodic physical examination data and daily behavior habit data of the elderly, extract the pathological abnormality coupling data in the periodic physical examination data, and analyze the current pathological factors and potential pathological factors corresponding to the elderly based on the pathological abnormality coupling data;

[0152] The induction contribution calculation module 104 is used to evaluate the health risk level corresponding to the current pathological factor according to the classification and annotation history, analyze the daily behavior factors in the daily behavior habit data, mine the pathological inducing factors from the daily behavior factors, and calculate the induction contribution of the pathological inducing factors to the potential pathological factors;

[0153] The file construction module 105 is used to combine the current pathological factors, the potential pathological factors, the health risk level, the classification and annotation history and the induced contribution rate to construct a corresponding health management file for the elderly using cloud computing technology.

[0154] In detail, each module described in the cloud computing-based elderly health record management system 100 described in the embodiment of the present application is used in the same manner as described above. Figure 1 The technical means described in the article are the same as the method for managing health records of the elderly based on cloud computing, and can produce the same technical effects, so I will not go into details here.

[0155] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for managing health records of the elderly based on cloud computing, characterized in that: The method comprises: Collecting individual identification information corresponding to the elderly, performing integrity verification processing on the individual identification information to obtain verification identification information, and performing standardization processing on the verification identification information to obtain reference identification information; Based on the benchmark identification information, the corresponding past disease history of the elderly is collected, and the past disease history is classified and labeled according to the preset disease classification criteria to obtain the classified and labeled history; Performing quality purification processing on the classified and labeled resume to obtain a target labeled resume, acquiring periodic physical examination data and daily behavior habit data of the elderly, extracting pathological abnormality coupling data from the periodic physical examination data, and analyzing the current pathological factors and potential pathological factors corresponding to the elderly based on the pathological abnormality coupling data; According to the classification and labeling history, the health risk level corresponding to the current pathological factor is evaluated, the daily behavior factors in the daily behavior habit data are analyzed, the pathological inducing factors are mined from the daily behavior factors, and the inducing contribution of the pathological inducing factors to the potential pathological factors is calculated; In combination with the current pathological factors, the potential pathological factors, the health risk level, the classification and annotation history and the induced contribution rate, cloud computing technology is used to construct a corresponding health management file for the elderly.

2. The cloud computing-based health record management method for the elderly as claimed in claim 1, characterized in that: The performing integrity verification processing on the individual identification information to obtain verification identification information includes: Checking the information format corresponding to the individual identification information, and performing format verification processing on the individual identification information based on the information format to obtain initial verification identification information; According to a preset information database, information verification processing is performed on the initial verification identification information to obtain a verification result; Based on the verification result, determining the information to be corrected in the initial verification identification information; The information to be corrected is corrected and verified to obtain verification identification information.

3. The cloud computing-based health record management method for the elderly as claimed in claim 1, characterized in that: The step of performing standardization processing on the verification identification information to obtain reference identification information includes: Identify the identification information characters and the communication number characters in the verification identification information, and extract the rare characters in the identification information characters; Performing encoding conversion processing on the rare characters of the information to obtain encoded information characters; Performing standard processing on the communication number characters to obtain standard number characters; The standard number characters and the rare information characters are used to perform information update processing on the verification identification information to obtain the reference identification information.

4. The cloud computing-based health record management method for the elderly as claimed in claim 1, characterized in that: The previous disease history is classified and annotated according to the preset disease classification criteria to obtain the classified and annotated history, including: Extracting information from the past medical history to obtain detailed disease information; Analyzing the correlation and matching degree between the disease detailed information and the disease classification criteria; In combination with the correlation matching degree, the disease detailed information and the disease classification criteria, the previous disease history is classified and labeled to obtain a classified and labeled history.

5. The cloud computing-based health record management method for the elderly as claimed in claim 1, characterized in that: The step of performing quality purification processing on the classification labeling history to obtain the target labeling history includes: Cleaning the classification and labeling history to obtain a cleaned and labeled history; Calculating the history information entropy corresponding to the cleaning and marking history; Extracting the history representation in the cleaning and marking history, and calculating the representation Gini index corresponding to the history representation; The cleaned and labeled records are subjected to quality purification processing in combination with the record information entropy and the characterization Gini index to obtain the target labeled records.

6. The cloud computing-based health record management method for the elderly as claimed in claim 5, characterized in that: The calculating the Gini impurity corresponding to the resume representation includes: Identify the resume sub-features in the resume representation, count the number of sub-features of the resume sub-features in the resume representation, and count the number of representations corresponding to the resume representation; Combining the number of sub-features and the number of representations, calculating the distribution probability corresponding to the resume sub-feature; Combined with the distribution probability, the Gini impurity corresponding to the resume representation is calculated by the following formula: Among them, F represents the Gini impurity corresponding to the resume representation, G d represents the distribution probability corresponding to the dth feature in the resume sub-feature, d represents the feature sequence number corresponding to the resume sub-feature, and p represents the number of features corresponding to the resume sub-feature.

7. The cloud computing-based health record management method for the elderly as claimed in claim 1, characterized in that: The step of extracting the pathological abnormality coupling data from the periodic physical examination data includes: Identify the abnormal physical examination index corresponding to the periodic physical examination data and the corresponding abnormal physical examination index data; Normalizing the abnormal physical examination index data to obtain physical examination observation values; Based on the physical examination observation value, calculating the indicator deviation value corresponding to the abnormal physical examination indicator; Analyzing the pathological physical examination indicators among the abnormal physical examination indicators according to the indicator deviation values; Based on the pathological physical examination index, pathological abnormality coupling data in the periodic physical examination data is extracted.

8. The cloud computing-based health record management method for the elderly as claimed in claim 1, characterized in that: The step of evaluating the health risk level corresponding to the current pathological factor according to the classification and labeling history includes: Extracting the pathological medication information in the classified and marked resume, wherein the pathological medication information includes: pathological medication name, pathological medication dosage, pathological medication frequency and pathological medication cycle; Calculate the drug utilization intensity corresponding to the pathological medication name by combining the pathological medication dosage and the pathological medication frequency; In combination with the pathological medication cycle and the medication utilization intensity, analyzing the medication utilization characteristics corresponding to the pathological medication name; The health risk level corresponding to the current pathological factor is evaluated according to the drug utilization characteristics.

9. The cloud computing-based health record management method for the elderly as claimed in claim 1, characterized in that: The calculating the contribution of the pathological inducing factor to the potential pathological factor comprises: Dispatching historical pathological data corresponding to the potential pathological factors, and analyzing the pathological factors inducing the pathological factors; Performing correlation matching processing on the pathological factors and the potential pathological factors to obtain a correlation factor set; Determining, according to the set of associated factors, the factor weights corresponding to the pathological factors causing the disease; Calculating the correlation strength between the pathological factors causing the disease and the potential pathological factors according to the historical pathological data; The association strength and the factor weight are combined to calculate the contribution of the pathological factor to the potential pathological factor.

10. The cloud computing-based health record management system for the elderly is characterized by: The system comprises: An information processing module is used to collect individual identification information corresponding to the elderly, perform integrity verification processing on the individual identification information to obtain verification identification information, and perform standardization processing on the verification identification information to obtain reference identification information; A disease history processing module is used to collect the corresponding past disease history of the elderly based on the benchmark identification information, and classify and annotate the past disease history according to the preset disease classification criteria to obtain a classified and annotated history; The medical record factor analysis module is used to perform quality purification processing on the classified and labeled resumes to obtain target labeled resumes, obtain the periodic physical examination data and daily behavior habit data of the elderly, extract the pathological abnormality coupling data in the periodic physical examination data, and analyze the current pathological factors and potential pathological factors corresponding to the elderly based on the pathological abnormality coupling data; An induction contribution calculation module is used to evaluate the health risk level corresponding to the current pathological factor according to the classification and annotation history, analyze the daily behavior factors in the daily behavior habit data, mine the pathological inducing factors from the daily behavior factors, and calculate the induction contribution of the pathological inducing factors to the potential pathological factors; The file construction module is used to combine the current pathological factors, the potential pathological factors, the health risk level, the classification and annotation history and the induced contribution rate to construct a corresponding health management file for the elderly using cloud computing technology.

Images