Medical record data intelligent acquisition method and system based on smart hospital

Abstract

The invention provides a medical record data intelligent acquisition method and system based on a smart hospital, and the method comprises the following steps: firstly, collecting medical record related multi-source data in the hospital, including basic information of a patient, symptom description, an inspection result and a diagnosis conclusion; and then the data are preprocessed, and data quality evaluation is performed by using an improved Bayesian filtering method, so that a reliable medical record data state can be obtained. And then, inputting the reliable medical record data state into an intelligent classification model trained by considering department categories and disease types, and performing classification to obtain predicted disease categories and corresponding departments. And finally, inputting a prediction result into an acquisition optimization function considering disease characteristics and data association to obtain a data acquisition priority, and generating an acquisition instruction according to the priority to realize intelligent acquisition of medical record data. According to the invention, the accuracy and efficiency of medical record data acquisition can be improved, the configuration of medical resources is optimized, and the diagnosis and treatment quality and management level of hospitals are improved.

Description

Technical Field

[0001] The present invention relates to the field of smart medical care, and more specifically, to a method and system for intelligently collecting medical record data based on a smart hospital. Background Art

[0002] With the rapid development of medical information technology, hospitals have accumulated a large amount of medical record data, which includes basic information of patients, symptom descriptions, test results and diagnosis conclusions. Traditional medical record data collection methods mainly rely on manual entry and simple automation tools, which have problems such as low data collection efficiency, poor accuracy and difficulty in data integration. In addition, medical record data are easily affected by factors such as data quality and data correlation during the collection process, resulting in insufficient data reliability and integrity. In practical applications, how to efficiently and accurately collect and process medical record data to support the hospital's clinical decision-making and scientific research work has become a technical problem that needs to be solved urgently.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: the data collection process lacks intelligence and automation, and cannot effectively cope with situations where the data volume is large and the data type is complex. It is also difficult to dynamically adjust the collection priority according to disease characteristics and data correlation, which affects the collection efficiency and quality of medical record data. Summary of the invention

[0004] The present invention provides a method and system for intelligently collecting medical record data based on a smart hospital.

[0005] In a first aspect of the present invention, a method for intelligently collecting medical record data based on a smart hospital is provided, comprising:

[0006] Collect multi-source data related to medical records in the hospital, including basic information of patients, symptom descriptions, examination and test results, and diagnosis conclusions;

[0007] Preprocess the multi-source data related to medical records, and use the improved Bayesian filtering method to evaluate the data quality of the preprocessed multi-source data related to medical records to obtain the reliable medical record data status;

[0008] Input the reliable medical record data status into the intelligent classification model trained by considering the department category and disease type, and classify the predicted disease category and corresponding department;

[0009] The predicted disease category and the corresponding department are input into the collection optimization function constructed by considering the disease characteristics and data association to obtain the data collection priority;

[0010] Generate corresponding collection instructions according to the data collection priority, and realize intelligent collection of medical record data based on the collection instructions.

[0011] Furthermore, the process of preprocessing the multi-source data related to the medical records is specifically as follows:

[0012] Eliminate erroneous values ​​and abnormal fluctuations in multi-source data related to medical records through data cleaning technology, data smoothing processing and / or outlier detection to obtain first data;

[0013] Eliminating the influence of different data types and value ranges in the first data by normalization processing or regularization processing to obtain second data;

[0014] The missing data items in the second data are supplemented by a data completion algorithm to obtain preprocessed medical record-related multi-source data.

[0015] Furthermore, the department categories are to classify hospital departments into multiple categories including internal medicine, surgery, obstetrics and gynecology, and pediatrics;

[0016] The disease types are considered to take into account the impact of different system diseases on medical record data and the impact of common disease types on collection focus.

[0017] Furthermore, the acquisition optimization function is a first function constructed with the disease characteristics as the main variable and the increment of the data correlation as the secondary variable, and the expression is:

[0018]

[0019] in, Indicates that the solution obtained by the first function is The data collection priority at each moment; represents the main variable function in the first function; represents the subvariable function in the first function; express Predicted disease category at the moment; express Always correspond to the department; represents the first benchmark disease category; represents the constant term in the first function; Represents the sample set for constructing the acquisition optimization function, with size Matrix representation of ; Indicates the total number of samples in the sample set; It represents the number of variables with disease characteristics in the sample set; It indicates the number of variables with data association in the sample set; The disease characteristics are , the data correlation is The data collection priority of the corresponding samples; The disease characteristics are , the data correlation is The data collection priority of the corresponding sample.

[0020] Furthermore, the acquisition optimization function is a second function constructed with data association as the main variable and the increment of disease characteristics as the secondary variable, and the expression is:

[0021]

[0022] in, Indicates that the solution obtained by the second function is The data collection priority at each moment; represents the secondary variable function in the second function; represents the main variable function in the second function; express Predicted disease category at the moment; express Always correspond to the department; Indicates the first benchmark department; represents the constant term in the second function; Represents the sample set for constructing the acquisition optimization function, with size Matrix representation of ; Indicates the total number of samples in the sample set; It represents the number of variables with disease characteristics in the sample set; It indicates the number of variables with data association in the sample set; The disease characteristics are , the data correlation is The data collection priority of the corresponding samples; The disease characteristics are , the data correlation is The data collection priority of the corresponding sample.

[0023] Furthermore, the acquisition optimization function is a third function constructed with disease characteristics and data correlation as the co-dominant role, and the expression is:

[0024]

[0025] in, Indicates that the solution obtained by the third function is The data collection priority at each moment; It represents the main variable function when constructing the acquisition optimization function with disease characteristics as the main variable and the increment of data correlation as the secondary variable; It represents the secondary variable function when constructing the acquisition optimization function with the disease characteristics as the main variable and the increment of the data correlation as the secondary variable; It represents the secondary variable function when constructing the acquisition optimization function with data correlation as the main variable and the increment of disease characteristics as the secondary variable; It represents the main variable function when constructing the acquisition optimization function with data correlation as the main variable and the increment of disease characteristics as the secondary variable; express Predicted disease category at the moment; express Always correspond to the department; Indicates the third benchmark disease category; Indicates the third benchmark department; It represents the constant term when constructing the acquisition optimization function with disease characteristics as the main variable and the increment of data correlation as the secondary variable; It represents the constant term when constructing the acquisition optimization function with data association as the main variable and the increment of disease characteristics as the secondary variable; Represents the sample set for constructing the acquisition optimization function, with size Matrix representation of ; Indicates the total number of samples in the sample set; It represents the number of variables with disease characteristics in the sample set; It indicates the number of variables with data association in the sample set; The disease characteristics are , the data correlation is The data collection priority of the corresponding samples; The disease characteristics are , the data correlation is The data collection priority of the corresponding sample.

[0026] Furthermore, the generation process of the acquisition instruction is specifically as follows:

[0027] Establish a basic collection task and the basic instructions required to complete the basic collection task;

[0028] The adjustment factor is determined by the ratio of the data collection priority to the basic priority;

[0029] The detail level of the acquisition instruction is determined by the product of the basic instruction and the adjustment coefficient, and the corresponding acquisition instruction is generated.

[0030] In a second aspect of the present invention, a medical record data intelligent collection system based on a smart hospital is provided, comprising:

[0031] The data collection module is used to collect multi-source data related to medical records in the hospital, including basic information of patients, symptom descriptions, examination and test results, and diagnosis conclusions;

[0032] The data evaluation module is used to pre-process the multi-source data related to the medical records, and use the improved Bayesian filtering method to evaluate the data quality of the pre-processed multi-source data related to the medical records to obtain the reliable medical record data status;

[0033] The classification prediction module is used to input the reliable medical record data status into the intelligent classification model trained by considering the department category and disease type, and classify and predict the disease category and the corresponding department;

[0034] The priority solving module is used to input the predicted disease category and the corresponding department into the acquisition optimization function constructed by considering the disease characteristics and data association to obtain the data acquisition priority;

[0035] The instruction execution module is used to generate corresponding collection instructions according to the data collection priority, and realize the intelligent collection of medical record data based on the collection instructions.

[0036] In a third aspect of the present invention, an electronic device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor; wherein when the processor executes the program, the method described in any one of the first aspects is implemented.

[0037] In a fourth aspect of the present invention, a computer-readable storage medium is provided, which includes instructions, and when the instructions are executed on a computer, the computer executes any one of the methods in the first aspect.

[0038] The above-mentioned embodiments of the present invention have at least the following beneficial effects: the present invention can effectively improve the accuracy and reliability of medical record data by collecting multi-source data related to medical records in the hospital, and preprocessing and quality evaluating the data. The improved Bayesian filtering method is used to perform quality evaluation on the preprocessed data, which can identify and eliminate erroneous values ​​and abnormal fluctuations in the data, and ensure the integrity and consistency of the data. At the same time, by inputting the reliable medical record data status into the intelligent classification model, the disease category and the corresponding department can be accurately predicted, providing accurate guidance for subsequent data collection. In addition, the present invention also constructs an acquisition optimization function, comprehensively considers disease characteristics and data correlation, and dynamically adjusts the data acquisition priority, thereby realizing the intelligent collection of medical record data. This method can significantly improve the efficiency and pertinence of data collection, reduce unnecessary data redundancy and repeated collection, and optimize the configuration and utilization of medical resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:

[0040] Figure 1 A flowchart of a method for intelligently collecting medical record data based on a smart hospital provided by an embodiment of the present invention;

[0041] Figure 2 A schematic diagram of the structure of an intelligent medical record data collection system based on a smart hospital provided by an embodiment of the present invention;

[0042] Figure 3 The schematic diagram schematically shows the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0044] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, apparatus, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0045] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0046] Reference below Figure 1 , Figure 1 The following is a flow chart of a method for intelligently collecting medical record data based on a smart hospital according to an embodiment of the present invention. Figure 1 As shown, a method 100 for intelligently collecting medical record data based on a smart hospital includes:

[0047] Step 101, collecting multi-source data related to medical records in the hospital, where the multi-source data related to medical records include basic information of patients, symptom descriptions, examination and test results, and diagnosis conclusions;

[0048] Step 102, preprocessing the medical record related multi-source data, and using an improved Bayesian filtering method to evaluate the data quality of the preprocessed medical record related multi-source data to obtain a reliable medical record data status;

[0049] Step 103, inputting the reliable medical record data status into the intelligent classification model trained by considering the department category and disease type, and classifying to obtain the predicted disease category and the corresponding department;

[0050] Step 104, input the predicted disease category and the corresponding department into the collection optimization function constructed by considering the disease characteristics and data association, and obtain the data collection priority;

[0051] Step 105, generating corresponding collection instructions according to the data collection priority, and realizing intelligent collection of medical record data based on the collection instructions.

[0052] It should be noted that the intelligent medical record data collection method involved in the present invention is mainly used in a smart hospital environment to improve the efficiency and accuracy of medical record data collection. Medical record-related multi-source data refers to various data related to patient medical records generated in the daily operation of the hospital, including but not limited to the patient's basic information, symptom description, examination and test results, and diagnostic conclusions. These data are usually stored in the hospital's information system, such as an electronic medical record system, a test information system, etc. Preprocessing refers to a series of operations such as cleaning, smoothing, and outlier detection on the original data to eliminate erroneous values ​​and abnormal fluctuations in the data to ensure the accuracy and consistency of the data. The improved Bayesian filtering method is a data quality assessment technology that evaluates the reliability of data through statistical analysis and probability models, thereby screening out high-quality medical record data.

[0053] Specifically, the collection of multi-source data related to medical records can be achieved through the hospital's information system interface, such as extracting patients' personal information and medical history information from the electronic medical record system, and obtaining patients' examination and test results from the inspection information system. Data cleaning techniques in the preprocessing process can include operations such as removing duplicate data and correcting erroneous data. Data smoothing can use methods such as moving average and exponential smoothing to eliminate noise in the data. Outlier detection can use statistical methods such as standard deviation and box plots to identify and process abnormal data. Normalization processing is to convert data of different data types and value ranges into a unified numerical range to facilitate subsequent data analysis and processing. Common normalization methods include maximum-minimum normalization and Z-score normalization. Regularization processing prevents data overfitting and improves the generalization ability of the model by adding penalty terms. Data completion algorithms can use methods such as mean filling and interpolation filling to supplement missing data items to ensure data integrity.

[0054] Preferably, when performing data quality assessment, appropriate Bayesian filter models and parameters can be selected in combination with the actual situation of the hospital and the characteristics of the data. For example, the prior probability and transition probability in the Bayesian filter model can be adjusted according to the distribution characteristics and correlation of the data to improve the accuracy of the evaluation results. In the training process of the intelligent classification model, machine learning algorithms such as support vector machines, random forests, neural networks, etc. can be used to build an efficient classification model in combination with the department categories and disease type data of the hospital. Department categories can be divided according to the department settings of the hospital, such as internal medicine, surgery, obstetrics and gynecology, pediatrics, etc., and disease types can be divided according to the systematic classification of diseases and common disease types.

[0055] It should be noted that in the traditional Bayesian filtering method, the prior probability is usually set based on fixed statistical laws or assumptions, which is difficult to adapt to the dynamic changes and complexity of medical record data. In the present invention, the setting of the prior probability is dynamically adjusted according to the historical medical record data of the hospital. Specifically, the system will regularly analyze the statistical characteristics of the hospital's past medical record data, such as the distribution of common diseases in different departments, the frequency of occurrence of patient symptoms, etc., as the basis for updating the prior probability. For example, for a specific department (such as cardiology), if historical data shows that the diagnosis frequency of coronary heart disease is high, then when the quality assessment of the medical record data of the department is performed, the prior probability related to coronary heart disease will be increased accordingly. This dynamic adjustment mechanism enables the Bayesian filtering method to better reflect the actual diagnosis and treatment situation of the hospital and improve the accuracy of data quality assessment.

[0056] The calculation of transition probability is used in Bayesian filtering to describe the conversion relationship between data states. When calculating the transition probability, the present invention fully considers the characteristics of hospital medical record data and practical application scenarios. For example, in medical record data, the occurrence of certain symptoms may have a high correlation with specific examination results. The system will dynamically adjust the transition probability according to these correlations to ensure that the evaluation results can more accurately reflect the true situation of the data. Specifically, by analyzing the correlation between symptoms and examination results in historical data, an association matrix can be constructed, and the transition probability can be calculated using the matrix. For example, if a patient has symptoms of chest pain and the electrocardiogram results are abnormal, the system will increase the transition probability of the patient suffering from coronary heart disease based on the high correlation between chest pain and electrocardiogram abnormalities in the association matrix.

[0057] In practical applications, the distribution and characteristics of medical record data may change over time and with changes in hospital business. The present invention also includes regularly analyzing the distribution characteristics of the data, such as mean, variance, etc., and adjusting the prior probability and transition probability in the model based on these characteristics. In addition, the system will dynamically adjust the weight parameters based on the correlation of the data to ensure that the model can adapt to changes in the data and improve the accuracy and reliability of the evaluation results.

[0058] In some embodiments, before training the intelligent classification model, it is necessary to preprocess and extract features from the multi-source data related to the medical records. Preprocessing includes steps such as data cleaning, smoothing, outlier detection, normalization, and data completion to ensure the quality and integrity of the data. Feature extraction includes extracting features related to disease diagnosis from medical record data, such as symptom descriptions, examination and test results, and basic patient information. These features will be used as input variables of the model for training and prediction. The training process of the intelligent classification model is as follows:

[0059] Dataset division step: Divide the preprocessed medical record data into training set, validation set and test set. The training set is used to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the performance of the model.

[0060] Algorithm selection step: Select a machine learning algorithm suitable for medical record data classification, such as support vector machine (SVM), random forest (RF), neural network (NN), etc. According to the characteristics of the data and classification requirements, select the optimal algorithm for training.

[0061] Model training steps: Use the training set to train the selected algorithm and optimize the performance of the model by adjusting model parameters (such as learning rate, regularization parameters, etc.). During the training process, methods such as cross-validation can be used to ensure the generalization ability of the model.

[0062] Model validation and optimization steps: Use the validation set to validate the trained model, adjust the model parameters based on the validation results, and optimize the model performance. For example, if the model has low accuracy on the validation set, you can adjust the model complexity or regularization parameters to improve the model's accuracy and generalization ability.

[0063] Furthermore, after the model training is completed, the test set is used to conduct a final evaluation of the model to ensure the accuracy and reliability of the model. In practical applications, the preprocessed medical record data is input into the trained intelligent classification model, and the model will output the predicted disease category and the corresponding department based on the input feature data. For example, if the input medical record data contains features such as chest pain and abnormal electrocardiogram, the model will predict that the patient has coronary heart disease and assign him to the cardiology department.

[0064] Furthermore, when generating collection instructions, the adjustment coefficient can be determined based on the ratio of the data collection priority to the basic priority, and the level of detail and execution order of the collection instructions can be flexibly adjusted based on the specific needs and resource conditions of the hospital to achieve efficient collection and management of medical record data.

[0065] In some embodiments, the process of preprocessing the multi-source data related to the medical records is specifically as follows:

[0066] Eliminate erroneous values ​​and abnormal fluctuations in multi-source data related to medical records through data cleaning technology, data smoothing processing and / or outlier detection to obtain first data;

[0067] Eliminating the influence of different data types and value ranges in the first data by normalization processing or regularization processing to obtain second data;

[0068] The missing data items in the second data are supplemented by a data completion algorithm to obtain preprocessed medical record-related multi-source data.

[0069] It should be noted that the preprocessing process of medical record-related multi-source data in the present invention is a key step to ensure data quality. Data cleaning technology refers to the process of improving data quality by identifying and correcting errors and inconsistencies in data. Data smoothing is a method used to reduce noise and fluctuations in data and make data smoother and more stable. Outlier detection is to identify and process data points that are significantly different from other data in the data set, which may be caused by errors or special circumstances. Through these preprocessing steps, erroneous values ​​and abnormal fluctuations in the data can be effectively eliminated, providing a reliable basis for subsequent data analysis and processing.

[0070] Specifically, data cleaning techniques can include a variety of methods, such as removing duplicate records, correcting spelling errors, filling missing values, etc. Data smoothing can be achieved through methods such as moving average and exponential smoothing, which smooth the data by calculating the average or weighted average of data points. Outlier detection can identify outliers through statistical methods such as standard deviation and box plots, or machine learning methods such as isolation forest and DBSCAN can be used for more complex anomaly detection. Normalization usually uses maximum-minimum normalization or Z-score normalization to convert data to a uniform numerical range to facilitate comparison and analysis between different data. Regularization prevents model overfitting by adding penalty terms. Common regularization methods include L1 regularization and L2 regularization. Data completion algorithms can use methods such as mean filling, interpolation filling, or model-based prediction filling to ensure data integrity.

[0071] Preferably, during data preprocessing, a suitable processing method can be selected according to the specific characteristics of the data and the application scenario. For example, when processing inspection results with time series characteristics, a time series smoothing method such as exponential smoothing can be used to better reflect the changing trend of the data over time. For outlier detection, a reasonable threshold can be set in combination with domain knowledge to avoid misjudging normal data as outliers. In the normalization process, if the distribution of the data is close to the normal distribution, Z-score normalization can be selected; if the distribution range of the data is wide and uneven, maximum-minimum normalization can be selected.

[0072] Furthermore, you can also choose a suitable data completion algorithm based on the missing data and relevance. For example, when there is a lot of missing data, you can use a predictive filling method based on machine learning to improve the accuracy of data completion.

[0073] In some embodiments, the department categories are to classify hospital departments into multiple categories such as internal medicine, surgery, obstetrics and gynecology, and pediatrics;

[0074] The disease types are considered to take into account the impact of different system diseases on medical record data and the impact of common disease types on collection focus.

[0075] It should be noted that the division of department categories and disease types is the basis for the construction of the intelligent classification model in the present invention. Department categories refer to different departments in the hospital divided according to treatment fields or specialties, such as internal medicine, surgery, obstetrics and gynecology, pediatrics, etc. Disease types are divided according to the systematic classification of diseases and common disease types, such as cardiovascular diseases, digestive system diseases, respiratory system diseases, etc. This classification helps to more accurately predict the patient's disease category and corresponding department during the intelligent collection of medical record data, thereby improving the pertinence and efficiency of data collection.

[0076] Specifically, the classification of department categories can be based on the hospital's department settings and professional characteristics. For example, internal medicine can be further divided into cardiology, gastroenterology, respiratory medicine, etc. The classification of disease types can be based on the International Classification of Diseases (ICD) or the hospital's own disease classification system to ensure the scientificity and consistency of the classification. In the training process of the intelligent classification model, the department category and disease type can be used as one of the input features of the model, combined with other medical record-related data, such as the patient's symptom description, examination and test results, etc., to train the model so that it can accurately predict the patient's disease category and the corresponding department.

[0077] Preferably, in the process of dividing department categories and disease types, flexible adjustments can be made according to the actual situation of the hospital and the characteristics of the data. For example, for some comprehensive hospitals, the department division may be more detailed and professional, and a more detailed classification can be made according to the department settings and professional characteristics of the hospital. In the division of disease types, the disease classification can be optimized and adjusted according to the common diseases and diagnosis and treatment needs of the hospital to better meet the clinical and scientific research needs of the hospital.

[0078] Furthermore, the hospital's historical medical record data and expert knowledge can be combined to verify and optimize the classification of department categories and disease types to improve the accuracy and reliability of the intelligent classification model.

[0079] In some embodiments, the acquisition optimization function is a first function constructed with disease characteristics as the main variable and the increment of data association as the secondary variable, and the expression is:

[0080]

[0081] in, Indicates that the solution obtained by the first function is The data collection priority at each moment; represents the main variable function in the first function; represents the subvariable function in the first function; express Predicted disease category at the moment; express Always correspond to the department; represents the first benchmark disease category; represents the constant term in the first function; Represents the sample set for constructing the acquisition optimization function, with size Matrix representation of ; Indicates the total number of samples in the sample set; It represents the number of variables with disease characteristics in the sample set; It indicates the number of variables with data association in the sample set; The disease characteristics are , the data correlation is The data collection priority of the corresponding samples; The disease characteristics are , the data correlation is The data collection priority of the corresponding sample.

[0082] It should be noted that the collection optimization function is a key tool for determining the priority of medical record data collection in the present invention. This function is constructed with disease characteristics as the main variable and the increment of data correlation as the secondary variable, aiming to optimize the data collection process through quantitative analysis. Disease characteristics refer to various attributes and indicators related to the disease, such as symptoms, test results, etc., while data correlation reflects the correlation and dependency between different data. By constructing the collection optimization function, the priority of data collection can be dynamically adjusted to ensure that important and relevant data can be collected and processed first.

[0083] Specifically, the construction of the acquisition optimization function requires determining the specific form and parameters of the main variable function and the secondary variable function. The main variable function can use a linear or nonlinear function to represent the impact of disease characteristics on data acquisition priority. For example, the weighted summation method can be used to multiply the weights of different disease characteristics by their corresponding values ​​and then sum them. The secondary variable function is used to represent the impact of data correlation on priority. The correlation between data can be quantified by constructing a correlation matrix or using a correlation analysis method. In addition, the specific parameters of the constant term and the sample set need to be determined. The constant term can be used to adjust the overall level of the function, while the sample set is used to train and verify the effectiveness of the function.

[0084] Preferably, when constructing the acquisition optimization function, it can be adjusted and optimized according to the specific needs of the hospital and the characteristics of the data. For example, for departments where certain disease characteristics are particularly important, the weights of these characteristics in the primary variable function can be appropriately increased to increase their importance in the data acquisition priority. When determining the secondary variable function, more complex association analysis methods can be used, such as association network analysis based on graph theory, to more accurately reflect the complex association relationships between data. In addition, the acquisition optimization function can be verified and optimized in combination with the hospital's historical data and expert experience to improve its accuracy and effectiveness in practical applications.

[0085] In some embodiments, the acquisition optimization function is a second function constructed with data association as the main variable and the increment of disease characteristics as the secondary variable, and the expression is:

[0086]

[0087] in, Indicates that the solution obtained by the second function is The data collection priority at each moment; represents the secondary variable function in the second function; represents the main variable function in the second function; express Predicted disease category at the moment; express Always correspond to the department; Indicates the first benchmark department; represents the constant term in the second function; Represents the sample set for constructing the acquisition optimization function, with size Matrix representation of ; Indicates the total number of samples in the sample set; It represents the number of variables with disease characteristics in the sample set; It indicates the number of variables with data association in the sample set; The disease characteristics are , the data correlation is The data collection priority of the corresponding samples; The disease characteristics are , the data correlation is The data collection priority of the corresponding sample.

[0088] It should be noted that another way to construct the collection optimization function is to use data correlation as the main variable and the increment of disease characteristics as the secondary variable. Data correlation refers to the correlation and dependence between different data, while the increment of disease characteristics indicates the marginal impact of disease characteristics on the data collection priority during the data collection process. The collection optimization function constructed in this way can emphasize the correlation between data more, so that data with high correlation with existing data can be given priority during the data collection process to improve the integration and consistency of data.

[0089] Specifically, the secondary variable function can use a linear or nonlinear function to represent the impact of the disease feature increment on the data collection priority. For example, a weighted summation method can be used to multiply the increments of different disease features by their corresponding weights and then sum them. The primary variable function is used to represent the impact of data correlation on priority. The correlation between data can be quantified by constructing a correlation matrix or using correlation analysis methods. In addition, the specific parameters of the constant term and the sample set need to be determined. The constant term can be used to adjust the overall level of the function, while the sample set is used to train and verify the effectiveness of the function.

[0090] Preferably, when constructing the acquisition optimization function, it can be adjusted and optimized according to the specific needs of the hospital and the characteristics of the data. For example, for those departments with particularly high data correlation, the weight of the data correlation in the main variable function can be appropriately increased to increase its importance in the data acquisition priority. When determining the secondary variable function, more complex disease feature analysis methods can be used, such as feature selection algorithms based on machine learning, to more accurately reflect the impact of the increment of disease characteristics on the data acquisition priority. In addition, the acquisition optimization function can be verified and optimized in combination with the hospital's historical data and expert experience to improve its accuracy and effectiveness in practical applications.

[0091] In some embodiments, the acquisition optimization function is a third function constructed with disease characteristics and data correlation as the co-dominant role, and the expression is:

[0092]

[0093] in, Indicates that the solution obtained by the third function is The data collection priority at each moment; It represents the main variable function when constructing the acquisition optimization function with disease characteristics as the main variable and the increment of data correlation as the secondary variable; It represents the secondary variable function when constructing the acquisition optimization function with the disease characteristics as the main variable and the increment of the data correlation as the secondary variable; It represents the secondary variable function when constructing the acquisition optimization function with data correlation as the main variable and the increment of disease characteristics as the secondary variable; It represents the main variable function when constructing the acquisition optimization function with data correlation as the main variable and the increment of disease characteristics as the secondary variable; express Predicted disease category at the moment; express Always correspond to the department; Indicates the third benchmark disease category; Indicates the third benchmark department; It represents the constant term when constructing the acquisition optimization function with disease characteristics as the main variable and the increment of data correlation as the secondary variable; It represents the constant term when constructing the acquisition optimization function with data association as the main variable and the increment of disease characteristics as the secondary variable; Represents the sample set for constructing the acquisition optimization function, with size Matrix representation of ; Indicates the total number of samples in the sample set; It represents the number of variables with disease characteristics in the sample set; It indicates the number of variables with data association in the sample set; The disease characteristics are , the data correlation is The data collection priority of the corresponding samples; The disease characteristics are , the data correlation is The data collection priority of the corresponding sample.

[0094] It should be noted that the collection optimization function can also be constructed by considering both disease characteristics and data relevance as co-dominant factors. This construction method aims to comprehensively consider the impact of the two factors on data collection priorities to achieve a more comprehensive and balanced data collection strategy. Disease characteristics refer to various attributes and indicators related to the disease, such as symptoms, test results, etc., while data relevance reflects the correlation and dependency between different data. In this way, it can be ensured that in the data collection process, both the importance of disease characteristics and the correlation between data are emphasized, thereby improving the accuracy and completeness of data collection.

[0095] Specifically, the construction of the acquisition optimization function can be in the form of a combination of multiple functions, each of which represents the impact of disease characteristics and data correlation on the data acquisition priority. For example, the weighted summation method can be used to multiply the weights of different disease characteristics by their corresponding values ​​and then sum them up to obtain the contribution value of the disease characteristics to the priority; similarly, the weights of different data correlations are multiplied by their corresponding values ​​and then summed to obtain the contribution value of the data correlation to the priority. Then, the two contribution values ​​are weighted and summed to obtain the final data acquisition priority. In addition, it is necessary to determine the specific parameters in each function, such as weight coefficients, constant terms, etc. These parameters can be adjusted and optimized according to the actual situation of the hospital and the characteristics of the data.

[0096] Preferably, when constructing the acquisition optimization function, more detailed adjustments and optimizations can be made according to the specific needs of the hospital and the characteristics of the data. For example, a machine learning algorithm can be used to automatically optimize the weights of disease characteristics and data associations to improve the accuracy and adaptability of the function. When determining the weight coefficient, the importance of different disease characteristics and data associations can be evaluated in combination with the hospital's historical data and expert experience, so as to reasonably allocate weights. In addition, other relevant factors, such as the timeliness of data and the scarcity of data, can be introduced to further enrich the construction of the acquisition optimization function to achieve a more accurate and efficient data collection strategy.

[0097] In some embodiments, the generation process of the acquisition instruction is specifically as follows:

[0098] Establish a basic collection task and the basic instructions required to complete the basic collection task;

[0099] The adjustment factor is determined by the ratio of the data collection priority to the basic priority;

[0100] The detail level of the acquisition instruction is determined by the product of the basic instruction and the adjustment coefficient, and the corresponding acquisition instruction is generated.

[0101] It should be noted that the generation process of collection instructions is a key step in the medical record data intelligent collection system, which determines the specific operations and execution order of data collection. Basic collection tasks refer to the basic tasks that must be completed in the process of medical record data collection, such as collecting basic information of patients, diagnosis results, etc. Basic instructions are the basic operation instructions required to complete these basic tasks, such as data reading instructions, data storage instructions, etc. The adjustment coefficient is calculated based on the ratio of data collection priority to basic priority, and is used to adjust the level of detail of the collection instructions to ensure the pertinence and efficiency of data collection.

[0102] Specifically, basic collection tasks can be defined according to the hospital's business processes and data requirements, such as collecting basic information of patients when they are admitted to the hospital, and collecting examination and test results after the examination and test are completed. Basic instructions need to be set according to the hospital information system and data storage method used, such as using SQL statements to read data from the database, using file I / O operations to write data to files, etc. The calculation of the adjustment coefficient can be based on the specific value of the data collection priority and the set value of the basic priority. For example, if the data collection priority is 10 and the basic priority is 5, the adjustment coefficient is 2, indicating that the collection instruction needs to be executed in more detail than the basic instruction. The level of detail of the collection instruction can be achieved by adding parameter settings and execution conditions of the instruction. For example, when collecting examination and test results, detailed analysis and recording instructions for the results can be added according to the priority.

[0103] Preferably, when generating collection instructions, more flexible adjustments and optimizations can be made according to the actual situation of the hospital and the characteristics of the data. For example, for some data collection tasks with higher priority, more data verification and validation instructions can be added to ensure the accuracy and completeness of the data. When setting the adjustment coefficient, the adjustment coefficient of tasks of different priorities can be dynamically adjusted in combination with the hospital's historical data and expert experience to adapt to changes in hospital business and data collection needs.

[0104] Furthermore, an intelligent scheduling algorithm can be introduced to dynamically adjust the execution order and priority of collection instructions according to the real-time data collection priority and system resource status to achieve a more efficient and intelligent data collection process.

[0105] The above-mentioned embodiments of the present invention have the following beneficial effects: The intelligent medical record data collection method based on smart hospitals described in the present invention can realize the efficient and intelligent collection of hospital medical record data. By collecting multi-source data and performing preprocessing, the method can effectively eliminate erroneous values ​​and abnormal fluctuations in the data to ensure the accuracy and integrity of the data. The reliability of the data can be further improved by using an improved Bayesian filtering method to assess the quality of the data. In addition, by predicting the disease category and department of reliable medical record data through an intelligent classification model, more accurate guidance can be provided for data collection to ensure the pertinence and effectiveness of data collection. This method can also dynamically adjust the data collection priority according to disease characteristics and data correlation, thereby optimizing the data collection process and improving collection efficiency.

[0106] At the same time, the implementation of this method can significantly improve the hospital's medical record management level and medical service quality. By intelligently collecting medical record data, hospitals can obtain and process patient information more quickly, providing timely and accurate data support for clinical decision-making. This not only helps to improve diagnosis and treatment efficiency, but also promotes the rational allocation and utilization of medical resources and reduces medical costs. In addition, this method can also provide rich data resources for the hospital's scientific research work and promote the development and innovation of medical research.

[0107] like Figure 2 As shown, in some embodiments, a medical record data intelligent collection system 200 based on a smart hospital includes:

[0108] The data collection module 201 is used to collect multi-source data related to medical records in the hospital, and the multi-source data related to medical records includes basic information of patients, symptom descriptions, examination and test results, and diagnosis conclusions;

[0109] The data evaluation module 202 is used to pre-process the medical record related multi-source data, and use the improved Bayesian filtering method to evaluate the data quality of the pre-processed medical record related multi-source data to obtain the reliable medical record data status;

[0110] The classification prediction module 203 is used to input the reliable medical record data status into the intelligent classification model trained by considering the department category and disease type, and classify and obtain the predicted disease category and the corresponding department;

[0111] The priority solving module 204 is used to input the predicted disease category and the corresponding department into the acquisition optimization function constructed by considering the disease characteristics and data association to obtain the data acquisition priority;

[0112] The instruction execution module 205 is used to generate corresponding collection instructions according to the data collection priority, and realize the intelligent collection of medical record data based on the collection instructions.

[0113] It is understandable that the modules recorded in the intelligent medical record data collection system 200 based on the smart hospital are similar to the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the intelligent collection method for medical record data based on a smart hospital are also applicable to the intelligent collection system 200 for medical record data based on a smart hospital and the modules contained therein, and will not be repeated here.

[0114] Reference below Figure 3, which shows a schematic diagram of the structure of an electronic device 300 suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include but are not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0115] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 to a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0116] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0117] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0118] It should be noted that the medical record data involved in this application, including but not limited to basic patient information, symptom descriptions, test results and diagnostic conclusions, are obtained with the full informed consent and explicit authorization of the patient or his or her legal guardian. In the process of data collection, storage, use and processing, the relevant national and regional laws and regulations as well as industry standards are strictly followed to ensure the legality, security and confidentiality of the data. At the same time, this technical solution fully respects the privacy rights of patients, takes necessary technical and management measures to prevent risks such as data leakage and abuse, and protects the security of patients' personal information.

[0119] The above descriptions are only some preferred embodiments of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A method for intelligently collecting medical record data based on a smart hospital, characterized in that: The following steps are involved: Collect multi-source data related to medical records in the hospital, including basic information of patients, symptom descriptions, examination and test results, and diagnosis conclusions; Preprocess the multi-source data related to medical records, and use the improved Bayesian filtering method to evaluate the data quality of the preprocessed multi-source data related to medical records to obtain the reliable medical record data status; Input the reliable medical record data status into the intelligent classification model trained by considering the department category and disease type, and classify the predicted disease category and corresponding department; The predicted disease category and the corresponding department are input into the collection optimization function constructed by considering the disease characteristics and data association to obtain the data collection priority; Generate corresponding collection instructions according to the data collection priority, and realize intelligent collection of medical record data based on the collection instructions.

2. According to claim 1, the intelligent medical record data collection method based on smart hospital is characterized in that: The process of preprocessing the multi-source data related to medical records is specifically as follows: Eliminate erroneous values ​​and abnormal fluctuations in multi-source data related to medical records through data cleaning technology, data smoothing processing and / or outlier detection to obtain first data; Eliminating the influence of different data types and value ranges in the first data by normalization processing or regularization processing to obtain second data; The missing data items in the second data are supplemented by a data completion algorithm to obtain preprocessed medical record-related multi-source data.

3. According to claim 1, the intelligent medical record data collection method based on smart hospital is characterized in that: The department categories are to classify hospital departments into multiple categories including internal medicine, surgery, obstetrics and gynecology, and pediatrics; The disease types are considered to take into account the impact of different system diseases on medical record data and the impact of common disease types on collection focus.

4. According to claim 1, the intelligent medical record data collection method based on smart hospital is characterized in that: The acquisition optimization function is a first function constructed with disease characteristics as the main variable and the increment of data association as the secondary variable, and the expression is: ； in, Indicates that the solution obtained by the first function is The data collection priority at each moment; represents the main variable function in the first function; represents the subvariable function in the first function; express Predicted disease category at the moment; express Always correspond to the department; represents the first benchmark disease category; represents the constant term in the first function; Represents the sample set for constructing the acquisition optimization function, with size Matrix representation of ; Indicates the total number of samples in the sample set; It represents the number of variables with disease characteristics in the sample set; It indicates the number of variables with data association in the sample set; The disease characteristics are , the data correlation is The data collection priority of the corresponding samples; The disease characteristics are , the data correlation is The data collection priority of the corresponding sample.

5. According to claim 1, the intelligent medical record data collection method based on smart hospital is characterized in that: The acquisition optimization function is a second function constructed with data association as the main variable and the increment of disease characteristics as the secondary variable, and the expression is: ； in, Indicates that the solution obtained by the second function is The data collection priority at each moment; represents the secondary variable function in the second function; represents the main variable function in the second function; express Predicted disease category at the moment; express Always correspond to the department; Indicates the first benchmark department; represents the constant term in the second function; Represents the sample set for constructing the acquisition optimization function, with size Matrix representation of ; Indicates the total number of samples in the sample set; It represents the number of variables with disease characteristics in the sample set; It indicates the number of variables with data association in the sample set; The disease characteristics are , the data correlation is The data collection priority of the corresponding samples; The disease characteristics are , the data correlation is The data collection priority of the corresponding sample.

6. According to claim 1, the intelligent medical record data collection method based on smart hospital is characterized in that: The acquisition optimization function is a third function constructed with disease characteristics and data correlation as the co-dominant role, and the expression is: ； in, Indicates that the solution obtained by the third function is The data collection priority at each moment; It represents the main variable function when constructing the acquisition optimization function with disease characteristics as the main variable and the increment of data correlation as the secondary variable; It represents the secondary variable function when constructing the acquisition optimization function with the disease characteristics as the main variable and the increment of the data correlation as the secondary variable; It represents the secondary variable function when constructing the acquisition optimization function with data correlation as the main variable and the increment of disease characteristics as the secondary variable; It represents the main variable function when constructing the acquisition optimization function with data correlation as the main variable and the increment of disease characteristics as the secondary variable; express Predicted disease category at the moment; express Always correspond to the department; Indicates the third benchmark disease category; Indicates the third benchmark department; It represents the constant term when constructing the acquisition optimization function with disease characteristics as the main variable and the increment of data correlation as the secondary variable; It represents the constant term when constructing the acquisition optimization function with data association as the main variable and the increment of disease characteristics as the secondary variable; Represents the sample set for constructing the acquisition optimization function, with size Matrix representation of ; Indicates the total number of samples in the sample set; It represents the number of variables with disease characteristics in the sample set; It indicates the number of variables with data association in the sample set; The disease characteristics are , the data correlation is The data collection priority of the corresponding samples; The disease characteristics are , the data correlation is The data collection priority of the corresponding sample.

7. The intelligent medical record data collection method based on a smart hospital according to claim 1 is characterized in that: The generation process of the acquisition instruction is specifically as follows: Establish a basic collection task and the basic instructions required to complete the basic collection task; The adjustment factor is determined by the ratio of the data collection priority to the basic priority; The detail level of the acquisition instruction is determined by the product of the basic instruction and the adjustment coefficient, and the corresponding acquisition instruction is generated.

8. An intelligent medical record data collection system based on a smart hospital, characterized by: The system is used to implement a method for intelligently collecting medical record data based on a smart hospital as described in any one of claims 1 to 7, comprising: The data collection module is used to collect multi-source data related to medical records in the hospital, including basic information of patients, symptom descriptions, examination and test results, and diagnosis conclusions; The data evaluation module is used to pre-process the multi-source data related to the medical records, and use the improved Bayesian filtering method to evaluate the data quality of the pre-processed multi-source data related to the medical records to obtain the reliable medical record data status; The classification prediction module is used to input the reliable medical record data status into the intelligent classification model trained by considering the department category and disease type, and classify and predict the disease category and the corresponding department; The priority solving module is used to input the predicted disease category and the corresponding department into the acquisition optimization function constructed by considering the disease characteristics and data association to obtain the data acquisition priority; The instruction execution module is used to generate corresponding collection instructions according to the data collection priority, and realize the intelligent collection of medical record data based on the collection instructions.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, it implements an intelligent medical record data collection method based on a smart hospital as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement an intelligent medical record data collection method based on a smart hospital as described in any one of claims 1 to 7.

Images