Patient data analysis management system

Abstract

The invention provides a patient data analysis and management system. The system comprises a multi-source data acquisition module, a multi-source data fusion module, a dynamic patient portrait module and a visual display module. The multi-source data acquisition module is used for acquiring basic information and medical history of a patient as well as physiological parameters, genome information and living habit data collected by a medical monitoring instrument; the multi-source data fusion module is used for performing multi-modal fusion on the data acquired by the multi-source data acquisition module; the dynamic patient portrait module is used for constructing a dynamic patient portrait model according to the multi-modal fusion data, the dynamic patient portrait model comprises a health state classification model and a risk assessment model, and results analyzed and output by the dynamic patient portrait model specifically comprise patient health states, risk assessment and personalized suggestions; and the visual display module is used for visually displaying a result analyzed and output by the dynamic patient portrait model.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a patient data analysis and management system. Background Art

[0002] Patient data refers to all medical and health information related to individual patients, usually covering a wide range of areas, including but not limited to the patient's basic information, medical history, diagnosis, treatment records, drug usage, laboratory test results and imaging results. In the modern medical system, the effective management and analysis of patient data plays a vital role in improving the quality of medical services, reducing costs and achieving personalized medicine. It can extract deeper insights from patient data and thus improve medical outcomes and patient experience.

[0003] Patient data analysis and management, through in-depth analysis of patient data, doctors can accurately diagnose diseases, develop treatment plans that are more suitable for patients, reveal trends and potential risks in disease progression, and help doctors make more informed decisions. At the same time, patient data analysis identifies each patient's unique needs and response patterns, thereby customizing personalized treatment plans and improving the targetedness and effectiveness of treatment. Especially when dealing with complex or chronic diseases, patient data analysis and management play a vital role in the modern medical system, not only improving the quality and efficiency of medical services.

[0004] However, the existing patient data analysis and management systems still have problems such as insufficient comprehensiveness of the influencing factors of the analysis and inaccurate judgment of the prediction results. Summary of the invention

[0005] The purpose of the present invention is to provide a patient data analysis and management system to solve the problems raised in the above background technology.

[0006] The present invention is implemented through the following technical solutions: a patient data analysis and management system, the system comprising a multi-source data acquisition module, a multi-source data fusion module, a dynamic patient portrait module, and a visualization display module;

[0007] The multi-source data acquisition module is used to collect basic information of patients, medical history, physiological parameters collected by medical monitoring instruments, genome information and life habit data;

[0008] The multi-source data fusion module is used to perform multi-modal fusion on the data collected by the multi-source data collection module;

[0009] The dynamic patient portrait module is used to construct a dynamic patient portrait model based on multimodal fusion data. The dynamic patient portrait model includes a health status classification model and a risk assessment model. The results of the dynamic patient portrait model analysis output specifically include patient health status, risk assessment, and personalized recommendations.

[0010] The visualization display module is used to visualize the results output by the dynamic patient portrait model analysis.

[0011] Specifically, the multimodal fusion in the multi-source data fusion module also includes diagnostic notes and physician opinion data, which are integrated into the dynamic patient portrait model to enrich the data analysis content.

[0012] Specifically, the specific method of the multimodal fusion is as follows:

[0013] First, define the data field mapping relationship and map the data fields from different sources to a unified dynamic patient portrait model;

[0014] Second, use the data integration platform to integrate the standardized data to generate comprehensive patient information;

[0015] Third, integrate different types of data to construct input data for multi-dimensional patient portraits;

[0016] Fourth, establish a dynamic update mechanism to regularly update medical history and living habits information to achieve comprehensive management of patients and ensure that patient portraits promptly reflect the latest health status.

[0017] Specifically, the construction of the health status classification model specifically includes:

[0018] First, extract and identify key physiological indicators and medical history data features related to health status;

[0019] Second, construct a dataset that contains the patient's features and labeled health status associations, and input the dataset into the trained health status classification model;

[0020] Third, generate a health status label for the patient based on the output of the model, and set a time interval or data trigger based on the data update of the dynamic update mechanism to trigger the model to re-evaluate and update the health status label.

[0021] Specifically, the risk assessment model construction specifically includes:

[0022] First, extract patient characteristics related to health risks and construct a dataset containing the association between patient characteristics and corresponding risk outcomes;

[0023] Second, a risk assessment model is trained based on the data set and a risk assessment index is calculated, and then the risk assessment index is graded, specifically including three levels: low risk, medium risk and high risk.

[0024] Specifically, after each data update, a dynamic patient portrait is regenerated and a timestamp is added to record when the health status changes, forming a time series to support subsequent analysis, including:

[0025] First, continuously monitor the patient's health information, including physiological indicators, medication use, medical history updates, and lifestyle information;

[0026] Second, trigger data updates at predetermined time intervals or when specific events are monitored;

[0027] Third, each time the data is updated, the current system time is obtained and the current time is combined with the updated data. After each data update and timestamp attachment, the dynamic patient portrait needs to be regenerated. A history record table is created for each patient in the database to save all health status changes and the regenerated patient portrait. The data recorded by timestamp is used to analyze the changing trends of specific health indicators over time, identify potential health risks and generate early warnings.

[0028] Specifically, the dynamic patient portrait module also includes using a natural language processing algorithm to extract entities and their relationships from the collected data: specifically including:

[0029] First, identify medical entities in the text, including diseases, drugs, and symptoms;

[0030] Second, extract the relationships between different entities, including “disease-symptoms” and “drug-treatment disease”;

[0031] Third, extract the attribute information of the entity, including the dosage and indication of the drug;

[0032] Fourth, the extracted knowledge will be cleaned to remove duplicate or inaccurate information to ensure data consistency and accuracy. At the same time, information from different sources will be integrated to eliminate redundancy. The extracted knowledge will then be structured and a graph database will be used to store and represent the knowledge graph, with each entity represented as a node and each relationship as an edge.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The patient data analysis and management system provided by the present invention can provide a more comprehensive health perspective by integrating information from multiple data sources to construct a dynamic patient portrait. The comprehensive information of the system not only affects the daily management and intervention of patients, but also provides strong support for clinicians in decision-making and formulation of treatment plans. Compared with traditional static analysis systems, this system has higher accuracy, timeliness and comprehensiveness, and can meet the precision medicine needs of modern medicine. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only preferred embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 This is a structural diagram of a patient data analysis and management system provided by the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical scheme and advantages of the present invention more obvious, the exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described in the present invention, all other embodiments obtained by those skilled in the art without creative work should fall within the protection scope of the present invention.

[0038] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known in the art are not described.

[0039] It should be understood that the present invention can be implemented in different forms and should not be interpreted as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to make the disclosure thorough and complete and to fully convey the scope of the present invention to those skilled in the art.

[0040] The purpose of the terms used herein is only to describe specific embodiments and is not intended to be limiting of the present invention. When used herein, the singular forms "one", "an" and "said / the" are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "consisting of" and / or "comprising", when used in this specification, determine the presence of the features, integers, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, parts and / or groups. When used herein, the term "and / or" includes any and all combinations of the relevant listed items.

[0041] In order to fully understand the present invention, a detailed structure will be proposed in the following description to illustrate the technical solution proposed by the present invention. The optional embodiments of the present invention are described in detail as follows, but in addition to these detailed descriptions, the present invention may also have other implementations.

[0042] See also Figure 1 , a patient data analysis and management system, the system comprising a multi-source data acquisition module, a multi-source data fusion module, a dynamic patient portrait module, and a visualization display module;

[0043] The multi-source data acquisition module is used to collect basic information of patients, medical history, physiological parameters collected by medical monitoring instruments, genome information and life habit data;

[0044] Exemplarily, the basic information specifically includes:

[0045] Patient’s personal information: such as name, gender, date of birth, and contact information.

[0046] Socio-demographic information: such as occupation, address, and family structure.

[0047] The medical history data includes:

[0048] Past medical history: including past diagnoses, hospitalization records, surgical procedures, and chronic disease information.

[0049] Medication history: including all prescriptions, allergic reactions and adverse reactions.

[0050] Vaccination record: vaccination status.

[0051] The physiological parameters collected by the medical monitoring instrument include:

[0052] Physiological parameters collected from wearable devices (such as smart watches) and medical monitors (such as blood glucose monitors and heart rate monitors), including heart rate, blood pressure, blood glucose, and oxygen saturation information.

[0053] The genome information includes:

[0054] Genome sequencing results: extract gene mutation and single nucleotide polymorphism (SNP) information.

[0055] Family history: The history of genetic diseases in the family to assess genetic risk.

[0056] The living habit data includes:

[0057] Dietary records: include the type, amount and frequency of daily food intake.

[0058] Sleep records: sleep time and sleep quality information.

[0059] Exercise records: exercise frequency, duration and type.

[0060] The multi-source data fusion module is used to perform multi-modal fusion on the data collected by the multi-source data collection module;

[0061] The dynamic patient portrait module is used to construct a dynamic patient portrait model based on multimodal fusion data. The dynamic patient portrait model includes a health status classification model and a risk assessment model. The results of the dynamic patient portrait model analysis output specifically include patient health status, risk assessment, and personalized recommendations.

[0062] The visualization display module is used to visualize the results output by the dynamic patient portrait model analysis.

[0063] The following is an example of a patient portrait display:

[0064] Patient portrait:

[0065] Name: Zhang San

[0066] Age: 65

[0067] Health status label: High risk (heart disease)

[0068] Risk Assessment Index: 0.75 (High Risk)

[0069] Key physiological indicators:

[0070] Blood pressure: 130 / 85 mmHg

[0071] Blood sugar: 150mg / dl

[0072] BMI: 28

[0073] Personalized recommendations:

[0074] Monitor blood pressure and blood sugar regularly, and a heart health assessment is recommended.

[0075] Increase your exercise; at least 150 minutes of moderate-intensity aerobic activity per week is recommended.

[0076] Through the above steps, the patient portrait constructed will be more real-time and targeted, not only providing timely risk assessment for medical personnel, but also providing corresponding personalized management suggestions based on the patient's health status. This comprehensive data processing and analysis method is particularly important in modern medicine and provides strong support for precision medicine.

[0077] Specifically, the multimodal fusion in the multi-source data fusion module also includes diagnostic notes and physician opinion data, which are integrated into the dynamic patient portrait model to enrich the data analysis content.

[0078] Specifically, the specific method of the multimodal fusion is as follows:

[0079] First, define the data field mapping relationship to map data fields from different sources (such as hospital information systems, real-time monitoring devices, and patient self-reporting applications) to a unified dynamic patient portrait model;

[0080] Second, use data integration platforms (such as Apache NiFi, Pentaho, etc.) to integrate standardized data to generate comprehensive patient information;

[0081] Third, integrate different types of data (structured, semi-structured, and unstructured) to construct input data for multi-dimensional patient portraits;

[0082] Fourth, establish a dynamic update mechanism to regularly update medical history and living habits information to achieve comprehensive management of patients and ensure that patient portraits promptly reflect the latest health status.

[0083] Specifically, the construction of the health status classification model specifically includes:

[0084] First, extract and identify key physiological indicators and medical history data features related to health status, such as blood pressure, blood sugar, body mass index (BMI), cholesterol level, previous medical history, family medical history, etc.;

[0085] Second, construct a dataset that contains the patient's features and labeled health status, including normal, high-risk, and chronic disease states, and input the dataset into a trained health status classification model;

[0086] Third, generate a health status label for the patient based on the output of the model, and set a time interval or data trigger based on the data update of the dynamic update mechanism (for example, when a physiological indicator exceeds a critical value) to trigger the model to re-evaluate and update the health status label.

[0087] Specifically, the risk assessment model construction specifically includes:

[0088] First, extract patient characteristics related to health risks, such as age, gender, weight, existing health conditions (chronic disease history), lifestyle habits, and other information, and construct a data set that contains the association between patient characteristics and corresponding risk outcomes, such as records of whether a specific disease has occurred;

[0089] Second, a risk assessment model is trained based on the data set and a risk assessment index is calculated, and then the risk assessment index is graded, specifically including three levels: low risk, medium risk and high risk.

[0090] For example, the risk assessment index is a probability value that indicates the risk of a patient experiencing a health event (such as cardiovascular disease, diabetes) within a specific period of time. The output is a value between 0 and 1, and the closer to 1, the higher the risk. According to different levels, corresponding health management plans and intervention measures are formulated. For example: a risk index between 0-0.2 is low risk, 0.2-0.5 is medium risk, and above 0.5 is high risk.

[0091] Specifically, after each data update, a dynamic patient portrait is regenerated and a timestamp is added to record when the health status changes, forming a time series to support subsequent analysis, including:

[0092] First, continuously monitor the patient's health information, including physiological indicators, medication use, medical history updates, and lifestyle information;

[0093] Second, trigger data updates at predetermined time intervals or when specific events are monitored;

[0094] Third, each time the data is updated, the current system time is obtained and the current time is combined with the updated data. After each data update and timestamp attachment, the dynamic patient portrait needs to be regenerated. A history record table is created for each patient in the database to save all health status changes and the regenerated patient portrait. The data recorded by timestamp is used to analyze the changing trends of specific health indicators over time, identify potential health risks and generate early warnings.

[0095] For example, through the above steps, the introduction of timestamp records in the generation and update process of dynamic patient portraits can effectively track changes in the patient's health status. This time series data not only enables medical staff to understand the patient's current status in real time, but also provides important information support for subsequent long-term health management.

[0096] Specifically, the dynamic patient portrait module also includes using a natural language processing algorithm to extract entities and their relationships from the collected data: specifically including:

[0097] First, identify the medical entities in the text, specifically including diseases, drugs, and symptoms;

[0098] Second, extract the relationships between different entities, specifically including "disease - symptom" and "drug - treating disease";

[0099] Third, extract the attribute information of the entities, including the dosage and indications of the drugs;

[0100] Fourth, clean the extracted knowledge, remove duplicate or inaccurate information, ensure the consistency and accuracy of the data. At the same time, integrate the information from different sources, eliminate redundancy, and then structure the extracted knowledge. Use a graph database to store and represent the knowledge graph, where each entity is represented as a node and each relationship is represented as an edge.

[0101] Exemplarily, the specific process of extracting entities and their relationships from the collected data using natural language processing algorithms is as follows:

[0102] Text data collection: Collect unstructured data in medical records, such as doctors' diagnostic remarks, physician opinions, surgical records, etc.

[0103] Data cleaning: Process typos, special characters, language noise, etc., and standardize the text format.

[0104] Text preprocessing:

[0105] Word segmentation: Use word segmentation tools (such as NLTK, Spacy) to split the text into words.

[0106] Part - of - speech tagging: Tag the part - of - speech of each word for subsequent analysis.

[0107] Stop - word filtering: Remove meaningless stop words (such as "de", "shi" in Chinese, which are translated as "of", "is" in English).

[0108] Word vector representation: Represent the words as vectors to capture the semantic relationships between words.

[0109] Information extraction:

[0110] Entity recognition: Identify the key entities in the text (such as disease names, drugs, symptoms).

[0111] Relationship extraction: Extract the relationships between entities (such as "The patient has diabetes", "The drug is used to treat hypertension").

[0112] Sentiment analysis: Analyze the sentiment tendency of the text and identify the doctor's attitude towards the condition (such as "mild", "severe").

[0113] Data integration: Integrate the extracted information into the patient profile to form a rich patient dataset.

[0114] Finally, a medical knowledge graph is constructed to represent entities such as diseases, symptoms, and drugs and their relationships to facilitate further analysis and reasoning.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A patient data analysis and management system, characterized in that: The system includes a multi-source data acquisition module, a multi-source data fusion module, a dynamic patient portrait module, and a visualization display module; The multi-source data acquisition module is used to collect basic information of patients, medical history, physiological parameters collected by medical monitoring instruments, genome information and life habit data; The multi-source data fusion module is used to perform multi-modal fusion on the data collected by the multi-source data collection module; The dynamic patient portrait module is used to construct a dynamic patient portrait model based on multimodal fusion data. The dynamic patient portrait model includes a health status classification model and a risk assessment model. The results of the dynamic patient portrait model analysis output specifically include patient health status, risk assessment, and personalized recommendations. The visualization display module is used to visualize the results output by the dynamic patient portrait model analysis.

2. A patient data analysis and management system according to claim 1, characterized in that: The multimodal fusion in the multi-source data fusion module also includes diagnostic notes and physician opinion data, which are integrated into the dynamic patient portrait model to enrich the data analysis content.

3. A patient data analysis and management system according to claim 2, characterized in that: The specific method of the multimodal fusion is as follows: First, define the data field mapping relationship and map the data fields from different sources to a unified dynamic patient portrait model; Second, use the data integration platform to integrate the standardized data to generate comprehensive patient information; Third, integrate different types of data to construct input data for multi-dimensional patient portraits; Fourth, establish a dynamic update mechanism to regularly update medical history and living habits information to achieve comprehensive management of patients and ensure that patient portraits promptly reflect the latest health status.

4. A patient data analysis and management system according to claim 3, characterized in that: The construction of the health status classification model specifically includes: First, extract and identify key physiological indicators and medical history data features related to health status; Second, construct a dataset that contains the patient's features and labeled health status associations, and input the dataset into the trained health status classification model; Third, generate a health status label for the patient based on the output of the model, and set a time interval or data trigger based on the data update of the dynamic update mechanism to trigger the model to re-evaluate and update the health status label.

5. A patient data analysis and management system according to claim 4, characterized in that: The risk assessment model construction specifically includes: First, extract patient characteristics related to health risks and construct a dataset containing the association between patient characteristics and corresponding risk outcomes; Second, a risk assessment model is trained based on the data set and a risk assessment index is calculated, and then the risk assessment index is graded, specifically including three levels: low risk, medium risk and high risk.

6. A patient data analysis and management system according to claim 5, characterized in that: After each data update, a dynamic patient portrait is regenerated and a timestamp is added to record when the health status changes, forming a time series to support subsequent analysis, including: First, continuously monitor the patient's health information, including physiological indicators, medication use, medical history updates, and lifestyle information; Second, trigger data updates at predetermined time intervals or when specific events are monitored; Third, each time the data is updated, the current system time is obtained and the current time is combined with the updated data. After each data update and timestamp attachment, the dynamic patient portrait needs to be regenerated. A history record table is created for each patient in the database to save all health status changes and the regenerated patient portrait. The data recorded by timestamp is used to analyze the changing trends of specific health indicators over time, identify potential health risks and generate early warnings.

7. A patient data analysis and management system according to claim 6, characterized in that: The dynamic patient portrait module also includes extracting entities and their relationships from the collected data using a natural language processing algorithm: specifically, First, identify medical entities in the text, including diseases, drugs, and symptoms; Second, extract the relationships between different entities, including "disease-symptoms" and "drug-treatment disease"; Third, extract the attribute information of the entity, including the dosage and indication of the drug; Fourth, the extracted knowledge will be cleaned to remove duplicate or inaccurate information to ensure data consistency and accuracy. At the same time, information from different sources will be integrated to eliminate redundancy. The extracted knowledge will then be structured and a graph database will be used to store and represent the knowledge graph, with each entity represented as a node and each relationship as an edge.

Images