Intelligent management system for chronic comorbidities in the elderly

Abstract

The present invention proposes an intelligent management system for chronic comorbidities in the elderly, including a comprehensive elderly assessment module, a multidisciplinary collaborative diagnosis and treatment module, a multiple medication risk prediction management module, a chronic disease intelligent follow-up management module, and a health science knowledge base module. The comprehensive elderly assessment module inputs the patient's individual chronic disease profile into a risk factor prediction model for prediction and outputs a comprehensive assessment result. The multidisciplinary collaborative diagnosis and treatment module is used by the first-visit / admitting physician to coordinate multidisciplinary physicians for collaborative diagnosis and treatment. The multiple medication risk prediction management module is used to provide patients with comprehensive diagnosis, multiple medication management, and risk prediction through the collaborative diagnosis and treatment. The chronic disease intelligent follow-up management module performs follow-up tracking, health risk prediction, and review prompts for patients. The health science knowledge base module matches and pushes corresponding health education knowledge. The present invention improves the accuracy and timeliness of risk assessment, prediction, and management of chronic comorbidities in the elderly, and plays a good auxiliary role in improving the quality of life of patients.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and in particular to an intelligent management system for chronic comorbidities in the elderly. Background Art

[0002] According to census data, the proportion of people aged 60 and above continues to rise. As they age, the elderly experience pathophysiological characteristics such as decreased body function, organ dysfunction, impaired immune and cognitive function, impaired limb mobility, and endocrine and metabolic disorders. When acute changes occur in one organ, other organs are also affected. Consequently, the elderly often suffer from multiple diseases affecting multiple organs. Furthermore, the cumulative effects of various symptoms and injuries become more pronounced, leading to the common phenomenon of multiple diseases in elderly patients, known as "multimorbidity." In 2008, the World Health Organization (WHO) defined multimorbidity as "a condition in which two or more chronic conditions coexist in the same patient." This emphasis on the coexistence of multiple chronic conditions broadens the concept of multimorbidity, shifting from a disease-based approach to a condition-based approach.

[0003] It is estimated that over 75% of older adults have one or more chronic conditions, and approximately 15% of the population has comorbid chronic diseases. Comorbid chronic diseases among older adults have become a prominent global issue, leading to greater healthcare demand, increased healthcare service utilization, and increased costs. Comorbid chronic diseases not only lead to imbalances in individual physiological functions and increase adverse drug events, but also affect disease prognosis and increase the healthcare burden. The prevention and treatment of comorbid chronic diseases has become an urgent issue. Effectively predicting and assessing the future disease risk of patients with comorbid chronic diseases enables physicians to intervene early, reduce the risk of related diseases, and thus prevent them from occurring, which is of great significance.

[0004] At present, there is no unified standard for risk prediction and assessment models and management systems for chronic disease comorbidities in the elderly. Factors such as research subjects, risk factors, model construction methods, and physicians' experience levels will affect the management of chronic disease comorbidities in the elderly. How to comprehensively consider multi-dimensional data and the interactions between diseases and drugs is an urgent problem that needs to be solved to improve the precise management of chronic disease comorbidities in the elderly. Summary of the Invention

[0005] Based on the defects and shortcomings of the existing technology, the present invention provides an intelligent management system for chronic comorbidities in the elderly, which includes multi-dimensional data characteristics of patients, coordinates and brings together doctors from different departments to jointly consider the interactions between diseases and therapeutic drugs, improves the accuracy and timeliness of risk assessment, prediction and management of chronic comorbidities in the elderly, and plays a good auxiliary role in improving the quality of life of patients.

[0006] To achieve the above object, the technical solution of the present invention is as follows:

[0007] An intelligent management system for chronic comorbidities in the elderly, including: a comprehensive elderly assessment module, a multidisciplinary collaborative diagnosis and treatment module, and a multiple medication risk prediction and management module;

[0008] The comprehensive elderly assessment module constructs a personal chronic disease map for elderly patients based on their personal information, diagnostic data, and medication information. This map is then input into the risk factor prediction model to generate predictions and output health risk alerts and comprehensive assessment results for the patient.

[0009] The multidisciplinary collaborative diagnosis and treatment module is used by the first-visit / receiving doctor to coordinate multidisciplinary doctors to conduct collaborative diagnosis and treatment based on the comprehensive assessment results of the geriatric comprehensive assessment module;

[0010] The polypharmacy risk prediction management module is used by the first-visit / admitting physician to provide comprehensive diagnosis, polypharmacy management and risk prediction to the patient through the collaborative diagnosis and treatment.

[0011] Furthermore, the comprehensive elderly assessment module sets up a personal information database to store the patient's personal information, sets up a diagnostic test database to store the diagnostic data of the patient's visits and hospitalization and all related examinations, and sets up a medication information database to store the patient's daily and hospitalized medication information.

[0012] Furthermore, the comprehensive elderly assessment module also includes: obtaining a scale score based on the patient's evaluation scale, inputting the patient's personal chronic disease map and scale score into the risk factor prediction model for prediction, and outputting health risk prompts and comprehensive assessment results for the patient.

[0013] Furthermore, the comprehensive geriatric assessment module is provided with an assessment scale database to store assessment scales covering physical function, geriatric syndrome, mental and psychological status, and the Charlson comorbidity index.

[0014] Furthermore, the risk factor prediction model is one or more of a multivariate logistic regression model, a Cox regression model, a neural network machine learning model, and a time series disease network model.

[0015] Furthermore, the multidisciplinary collaborative diagnosis and treatment module includes a multidisciplinary integrated triage platform and a multidisciplinary patient precision clinic; the multidisciplinary integrated triage platform includes the information of doctors in each department, and the first-visit / admitting doctor comprehensively selects the required consultation departments and relevant doctors based on the comprehensive evaluation results, and the above-mentioned relevant doctors jointly form a multidisciplinary patient precision clinic; the multidisciplinary patient precision clinic is a specialty related to the diagnosis and treatment of chronic diseases, including geriatrics, endocrinology and metabolism, cardiovascular medicine, respiratory medicine, gastroenterology, nephrology, neurology, neurosurgery, oncology, clinical psychology, clinical nutrition, rehabilitation medicine, pharmacy, and nursing department.

[0016] Furthermore, the multiple medication risk prediction management module includes comprehensive diagnostic testing, multiple medication management, and diet and daily life management functions.

[0017] Furthermore, multiple medication management follows the principles of non-drug treatment priority, benefit principle, five-drug principle, small dose principle, timing principle and medication suspension principle; the multiple medication management also includes AI medication warning, which promptly alerts patients of medication contraindications.

[0018] Furthermore, it also includes a smart follow-up management module for chronic diseases, which includes a follow-up tracking function, a review reminder function and a risk factor reminder function, and performs follow-up tracking, health risk prediction and review reminders for elderly patients with chronic diseases and comorbidities.

[0019] Furthermore, it also includes a health science popularization knowledge base module, which includes a medical health education science popularization knowledge base. The medical health education science popularization knowledge base is dynamically updated, and according to the diagnosis results made by the multidisciplinary patient precision clinic, the corresponding health education knowledge is timely matched and pushed to the patient.

[0020] Compared with the prior art, the present invention has the following advantages:

[0021] 1. This invention integrates all functional modules of the whole process management of chronic diseases in the elderly:

[0022] (1) Comprehensive geriatric assessment: Establish a comprehensive geriatric assessment platform to fully understand the patient's health status through comprehensive assessment and provide a basis for subsequent management.

[0023] (2) Multidisciplinary team integrated management: Establish a multidisciplinary team integrated management team and platform to provide systematic, comprehensive, safe, convenient and effective medical services, and conduct comprehensive assessment, standardized diagnosis and treatment, and health guidance for elderly patients.

[0024] (3) Multiple medication management: Doctors will implement precise individualized medication management by giving priority to non-drug treatment and following the benefit principle, five-drug principle, low-dose principle, timing principle and medication suspension principle.

[0025] (4) Health education and health guidance: including health guidance for the elderly on chronic disease risk factors and vaccination, osteoporosis prevention and fall prevention measures, accidental injuries and self-rescue.

[0026] (5) Chronic disease management: For chronic diseases such as hypertension and diabetes, regular health guidance and full-process follow-up are provided through the platform to improve the effectiveness of disease management and significantly improve the quality of life of patients.

[0027] 2. This invention realizes the intelligent management of chronic comorbidities in the elderly through artificial intelligence, big data analysis and risk factor prediction models, which helps doctors to accurately intervene in chronic comorbidities in the elderly and implement full-process management. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a schematic diagram of the architecture of the intelligent management system for chronic comorbidities in the elderly according to Example 1 of the present invention.

[0029] Figure 2 This is a flow chart of the intelligent management system for chronic comorbidities in the elderly according to Example 1 of the present invention.

[0030] Figure 3 This is a schematic diagram of the doctor side of the intelligent management system for chronic comorbidities in the elderly according to Example 1 of the present invention.

[0031] Figure 4 This is a patient-side schematic diagram of the intelligent management system for chronic comorbidities in the elderly according to Example 1 of the present invention. DETAILED DESCRIPTION

[0032] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other, and the specific features, structures or characteristics described in conjunction with the embodiments can be included in at least one embodiment of the present application. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art to which this application belongs; the terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit this application; it should be noted that the conventional conditions and methods not described in detail in the embodiments are all carried out in accordance with the experimental conditions and methods conventionally used by those skilled in the art.

[0033] Example 1:

[0034] As shown in Figure 1, the intelligent management system for chronic comorbidities in the elderly includes five system modules: (1) comprehensive assessment module for the elderly, (2) multidisciplinary collaborative diagnosis and treatment module, (3) multiple medication risk prediction management module, (4) intelligent follow-up management module for chronic diseases, and (5) health science knowledge base module; data communication connections are established between each module.

[0035] The comprehensive assessment module for the elderly includes multiple databases. In this embodiment, a patient personal information database, a diagnostic test database, a medication information database, and an evaluation scale database are set up, and relevant information therein can be retrieved. The risk factor prediction model is input based on the patient's previous medical history diagnosis, test and medication information, scale results, etc., and a comprehensive calculation and evaluation is performed to obtain a health value range displayed in various forms such as "mathematical formulas, scoring systems, forest plots, nomograms, etc."

[0036] In the multidisciplinary collaborative diagnosis and treatment module, the first-visit / admitting doctor coordinates multidisciplinary doctors to conduct collaborative diagnosis and treatment based on the initial evaluation results of the comprehensive geriatric assessment module; doctors from different departments jointly form a multidisciplinary patient precision clinic.

[0037] In the polypharmacy risk prediction management module, doctors prescribe comprehensive diagnostic tests for patients through the collaborative diagnosis and treatment, avoiding repeated redundant tests; through joint comprehensive judgment of the test results, they provide patients with comprehensive and systematic comprehensive diagnosis and polypharmacy management and guidance; if necessary, they will also provide patients with dietary and daily life advice and corresponding management.

[0038] After diagnosis and treatment, doctors can use the intelligent follow-up management module for chronic diseases to follow up on patients, predict health risks, and provide reexamination reminders.

[0039] The health science knowledge base module is dynamically updated and can timely match and push corresponding health education knowledge to patients based on the diagnosis results made by the multidisciplinary patient precision clinic. The module also includes common health education knowledge for the elderly, such as vaccination, osteoporosis prevention and fall prevention measures, accidental injuries and self-rescue, to help elderly patients and their families with daily management in a comprehensive, systematic and comprehensive manner.

[0040] Figure 2 shows the workflow of the intelligent management system for chronic comorbidities in the elderly. After entering the system, patients can enter and retrieve data from the database, establish basic data, and, after computational analysis, enter the multidisciplinary integrated triage platform. The initial / receiving physician will coordinate the establishment of a multidisciplinary precision clinic for the patient, which issues diagnostic test notifications, medication management notices, and dietary and lifestyle reminders to manage polypharmacy risks. Each time corresponding diagnostic test information and polypharmacy information is obtained, it is automatically integrated into the patient's basic database (personal information database, diagnostic test database, and medication information database) for use / access the next time the patient logs into the system. After the patient completes their medical consultation, the physician uses the intelligent chronic disease follow-up management module for follow-up tracking, reexamination reminders, and risk factor alerts. The physician can also push relevant information from the health knowledge base module to the patient for learning and self-management.

[0041] Figure 3 shows an example of the physician-side visual interface of the intelligent management system for chronic comorbidities in the elderly. The first interface contains database ports, accessible data, and comprehensive evaluation results. This interface is visible to both doctors and patients. In the second interface, the initial / receiving physician can view the physician information and schedules of various hospital departments. They can select the appropriate department and physician for a coordinated consultation. This interface is visible only to the physician and not to the patient. In the third interface, the initial / receiving physician selects relevant physicians to form a multidisciplinary patient precision clinic (e.g., cardiologist 1, endocrinologist 2, nephrologist 3, etc.). This interface is visible to both doctors and patients, allowing for dialogue and information exchange. In the fourth interface, multidisciplinary physicians diagnose the patient's condition, select appropriate tests, and send them to the patient. This interface is visible to both doctors and patients. In the fifth interface, multidisciplinary physicians discuss and exchange opinions, ultimately developing a comprehensive diagnosis and corresponding medication instructions. This interface is visible only to the physician and not to the patient. The sixth interface is visible to both doctors and patients, providing them with dietary and daily life information, medication guidance, review recommendations, and health education information. The seventh interface is the follow-up tracking interface, where doctors can issue medication reminders, review reminders, risk factor prompts, etc., and provide timely health education guidance based on the patient's medication or physical sign data feedback. The risk factor prompts are based on the health value range obtained by the comprehensive elderly assessment module to distinguish whether they are risk factors that require attention.

[0042] Figure 4 shows an example of the patient-side visual interface of the intelligent management system for chronic comorbidities in the elderly. The first interface contains database portals, accessible data, and comprehensive evaluation results. This interface is visible to both doctors and patients, allowing them to enter, modify, and confirm information. In the second interface, a multidisciplinary precision clinic tailored to the patient's condition has been established, allowing both the patient and the multidisciplinary medical team to view and communicate. In the third interface, patients can view test notifications from the medical team, along with relevant results and images of the effects. In the fourth interface, patients can view dietary and daily routines, medication guidance, follow-up recommendations, and health education information provided by the medical team. In the fifth interface, patients receive medication and follow-up reminders from doctors and the system after their diagnosis and treatment, and can keep records of these reminders. For special circumstances, patients can also consult with doctors in the internet hospital through this platform and make appointments using the hospital's integrated online platform. This intelligent management system for chronic comorbidities in the elderly can also be loaded into an app or other mini-programs on computers, mobile phones, or watches, providing timely alerts about potential risk factors and offering convenient medical service options.

[0043] Example 2:

[0044] This embodiment proposes a further detailed solution for the comprehensive assessment module for the elderly.

[0045] In this embodiment, the comprehensive elderly assessment module is equipped with multiple databases and computing tools, which are described in detail as follows:

[0046] The patient personal information database stores the patient's personal information, such as the patient's age, gender, education level, marital status, living habits, smoking, drinking, body mass index, sleep conditions, past medical history, health self-assessment, medical insurance reimbursement method, etc.

[0047] The diagnostic test database stores records of tests and all related examinations performed on patients during their visits and hospitalizations; related examination records include data such as the patient's diagnosis time, test reports, images, and diagnostic results for each time; the diagnostic test database may also include the patient's genetic test data. For example, the following imaging, biochemical and genetic testing indicators are collected to preliminarily construct a diagnostic test database: blood pressure, electrocardiogram, echocardiogram, cardiac computed tomography, cardiac magnetic resonance imaging, coronary angiography, and cardiac angiography; biomarkers (B-type natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP), troponin (cTn), matrix protein 2 (ST2) and soluble ST2 (sST2), galectin 3 (Gal-3), growth differentiation factor 15 (GDF-15), serum cystatin C (Cys C), neutrophil gelatinase transporter (NGAL)); five-item complete blood test including white blood cell count, red blood cell count, hemoglobin concentration, platelet count, clotting time, blood sugar, blood lipids, liver function, and kidney function; and genetic testing data related to cardiovascular disease pharmacogenomics.

[0048] The medication information database stores a patient's daily and hospitalized medication usage, including prescription and over-the-counter medications, but excluding health supplements. For example, medication usage for cardiovascular diseases (coronary heart disease (CHD), hypertension, arrhythmias, valvular heart disease, myocarditis / primary cardiomyopathy, congenital heart disease, pericardial disease, etc.), non-cardiovascular diseases (dyslipidemia, anemia, chronic kidney disease (CKD), type 2 diabetes, cerebrovascular disease (CVD), obesity, peripheral vascular disease (PVD), malignant tumors (including solid tumors, leukemia, lymphoma, etc.), chronic obstructive pulmonary disease (COPD), hyperthyroidism, hypothyroidism, etc.) can be collected.

[0049] The evaluation scale database includes comprehensive geriatric evaluation scales such as Activity of daily living (ADL), Frail Scale (FS), Pittsburgh Sleep Quality Index (PSQI), and Charlson co-morbidity index (CCI).

[0050] The calculation program included in the comprehensive assessment module for the elderly includes a patient's personal chronic disease map and a risk factor prediction model. After obtaining the patient's personal information and patient case and medication records from the personal information database, the diagnostic test database, and the medication information database, the calculation program sequentially numbers the chronic diseases suffered by the patient based on the sorting results; based on the patient's personal characteristic information and the sequential numbering results, a patient's personal chronic disease map is constructed. The map includes a node set and an edge set. The node set includes patient nodes and chronic disease nodes. The edges in the edge set represent the chronic diseases diagnosed by the patient, and the order of diagnosis is coded as n, where n is a positive integer greater than or equal to 1. The patient's personal chronic disease map is constructed and input into the risk factor prediction model for prediction, outputting health risk prompts and comprehensive assessment results for the patient.

[0051] Furthermore, the algorithm described in this embodiment can also obtain scale scores from the evaluation scale database as needed. These scale scores include the patient's comorbidity index, daily living ability assessment scale score, and patient frailty level scale score. The algorithm then inputs the constructed patient's personal chronic disease profile and scale scores into the risk factor prediction model to generate predictions, outputting health risk indicators and comprehensive assessment results for the patient. These scale scores provide a real-time assessment of the patient's condition, supplementing and revising the patient's personal chronic disease profile, and enabling a more accurate comprehensive assessment.

[0052] Risk factor prediction models include a variety of models, such as multivariate logistic regression and Cox regression, neural network machine learning models, and time series disease network models. Constructed risk prediction models can also be presented in various ways, such as mathematical formulas, scoring systems, forest plots, and nomograms.

[0053] Example 1 (Multivariate Logistic Regression Analysis): Univariate and multivariate logistic regression analyses were performed using in-hospital mortality as the dependent variable and 10 chronic diseases as independent variables. Four comorbidities, namely chronic ischemic heart disease, hypertension, diabetes, and cerebrovascular disease, were identified as independent risk factors for in-hospital mortality in elderly patients with heart failure. Furthermore, age stratification, New York Heart Association (NYHA) class III or higher, and N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels >3000 ng / L, three indicators associated with death or heart failure severity, were added as independent variables. A binary logistic regression model was then performed with in-hospital mortality as the dependent variable. The big data-trained model generated risk factor indicators corresponding to specific indicators and a forest plot was constructed. The trained model was then fed with the patient's individual chronic disease profile data to output a comprehensive assessment of the patient's risk factors.

[0054] Example 2 (neural network machine learning): The patient node and the chronic disease node (including diagnostic test information data and medication information data) of the patient's personal chronic disease map are input into the fully connected layer network to obtain updated embedding vector representations of the patient node and the chronic disease (chronic ischemic heart disease, hypertension, diabetes and cerebrovascular disease) nodes; the updated embedding vector representations of the patient node and the chronic disease node are input into the decoding prediction module to obtain the prediction results of chronic disease comorbidity risk factors, wherein the decoding prediction module includes a bilinear decoder, which is used to decode the embedding vector representations of the patient node and the chronic disease node, and evaluate the probability of occurrence of patient risk factors based on the decoding results.

[0055] Validation of risk factor prediction models is mainly divided into internal validation and external validation. The commonly used internal validation in this field is to use the data of the modeling set for validation, including random split method, cross-validation method and Bootstrap sampling method. The random split method is to randomly divide the data set into a training set and a validation set according to a certain ratio, build a prediction model with the training set data, and verify the prediction performance of the model with the validation set data. Since the two data sets are very similar, more ideal results are often obtained. Compared with the random sampling method, when the data is limited, the cross-validation method and the bootstrap sampling method are more desirable, which can reduce the similarity between the training set and the validation set. For the two methods mentioned above, the modeling set can be used for internal validation by resampling 500 times through the Bootstrap method; 300 cases of validation set data can be used for external validation to confirm the high accuracy of risk occurrence judgment.

[0056] External validation, a common method in this field, involves evaluating the model using data from a different set than the modeling dataset. Based on the data source, it can be categorized into temporal validation, regional validation, and strong external validation. Temporal validation assesses the subsequent predictive effectiveness of the risk prediction model for the modeled subject; regional validation uses additional subjects from the same region as the modeled subject; and strong external validation uses subjects from other research centers completely separate from the modeling dataset. Strong external validation is considered the most ideal external validation method. The main evaluation metrics for risk factor prediction models are discrimination and calibration. Discrimination refers to the model's ability to correctly predict the occurrence of an event and is typically assessed using the area under the receiver operating characteristic (ROC) curve (AUC). An AUC of 0.5-0.7 indicates low predictive value, an AUC of 0.7-0.9 indicates moderate predictive value, and an AUC above 0.9 indicates high predictive value. Values closer to 1.0 indicate better model discrimination. Calibration reflects the degree to which the model correctly estimates absolute risk. It is often evaluated using the HL goodness-of-fit test, which mainly compares whether the difference between the expected probability and the actual probability of an event is statistically significant. P>0.05 indicates that the model has a good fit.

[0057] The above-described embodiments are merely preferred implementations of the present invention and are intended to help understand the method and core concepts of the present application. The scope of protection of the present invention is not limited to the above-described embodiments. All technical solutions within the scope of protection of the present invention are within the scope of protection of the present invention. It should be noted that for those skilled in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. An intelligent management system for chronic comorbidities in the elderly, characterized by: It integrates the full-process management modules for elderly chronic diseases, including: comprehensive elderly assessment module, multidisciplinary collaborative diagnosis and treatment module, multiple medication risk prediction and management module, intelligent chronic disease follow-up management module, and health science knowledge base module; The comprehensive elderly assessment module constructs a personal chronic disease map for elderly patients based on their personal information, diagnostic data, and medication information. It obtains a scale score based on the patient's evaluation scale. The scale score includes the patient's comorbidity index, which is a real-time assessment of the patient's status. It supplements and amends the patient's personal chronic disease map. The patient's personal chronic disease map and the scale score for the real-time assessment of the patient's status are input into the risk factor prediction model for prediction, and outputs health risk prompts and comprehensive assessment results for the patient. The risk factor prediction model includes: inputting the patient node and chronic disease node of the patient's personal chronic disease map into a fully connected layer network, wherein the chronic disease node includes diagnostic test information data and medication information data; obtaining updated embedding vector representations of the patient node and the chronic disease node; inputting the updated embedding vector representations of the patient node and the chronic disease node into a decoding prediction module to obtain a prediction result of chronic disease comorbidity risk factors, wherein the decoding prediction module includes a bilinear decoder, which is used to decode the embedding vector representations of the patient node and the chronic disease node, and evaluate the probability of occurrence of the patient risk factor based on the decoding result; The multidisciplinary collaborative diagnosis and treatment module is used by the first-visit / receiving doctor to coordinate multidisciplinary doctors to conduct collaborative diagnosis and treatment based on the comprehensive assessment results of the geriatric comprehensive assessment module; the multidisciplinary collaborative diagnosis and treatment module includes a multidisciplinary integrated triage platform and a multidisciplinary patient precision clinic; the multidisciplinary integrated triage platform contains the doctor information of each department, and the first-visit / receiving doctor comprehensively selects the required consultation departments and relevant doctors based on the comprehensive assessment results, and the above-mentioned relevant doctors jointly form a multidisciplinary patient precision clinic; The polypharmacy risk prediction management module is used by the first / receiving doctor to provide comprehensive diagnosis, polypharmacy management and risk prediction to the patient through the collaborative diagnosis and treatment. The polypharmacy risk prediction management module includes comprehensive diagnostic testing, polypharmacy management, and diet and daily life management functions; The intelligent chronic disease follow-up management module includes follow-up tracking functions, review reminder functions and risk factor reminder functions, and provides follow-up tracking, health risk prediction and review reminders for elderly patients with chronic diseases and comorbidities; The health science knowledge base module includes a medical health education science knowledge base, which is dynamically updated and timely matches and pushes corresponding health education knowledge to patients based on the diagnosis results made by the multidisciplinary patient precision clinic; The first interface of the doctor-side visual interface of the intelligent management system for chronic comorbidities in the elderly contains various database ports, callable data and comprehensive evaluation results, which are visible to both doctors and patients; in the second interface, the first-visit / admitting doctor can see the doctor information and schedule of each department in the hospital, and select the corresponding department and doctor according to the needs for overall consultation. This interface is visible only to doctors, not patients; in the third interface, the first-visit / admitting physician selects relevant doctors to form a multidisciplinary patient precision clinic. In this interface, doctors and patients can see it together, conduct dialogues and exchange information; in the fourth interface, multidisciplinary doctors choose to prescribe corresponding examinations for the patient's condition and push them to the patient. The interface is visible to both doctors and patients; in the fifth interface, doctors from multiple disciplines discuss and exchange opinions, and ultimately form a comprehensive diagnosis result and corresponding medication guidance. This interface is visible only to doctors and not to patients; in the sixth interface, both doctors and patients can see it, and provide patients with diet, daily life, medication guidance, review suggestions and health education knowledge push; the seventh interface is the follow-up tracking interface, in which doctors issue medication reminders, review reminders, and risk factor prompts, and provide timely health education guidance based on the patient's medication or physical sign data feedback; the risk factor prompts are based on the health value range obtained by the comprehensive elderly assessment module to distinguish whether they are risk factors that need attention; The first interface of the patient-side visual interface of the intelligent management system for chronic comorbidities in the elderly includes various database ports, callable data and comprehensive evaluation results. This interface is visible to both doctors and patients, and patients can enter, revise and confirm the information; in the second interface, a multidisciplinary patient precision clinic tailored to the patient's condition has been established, which is visible to both patients and the multidisciplinary medical team, and they can have dialogues and exchanges; in the third interface, patients can see the test push and relevant results and impact pictures given by the doctor team; in the fourth interface, patients can see the patient's diet, daily life, medication guidance, review suggestions and health education knowledge push given by the doctor team; in the fifth interface, patients receive medication or review reminders pushed by doctors and the system after diagnosis and treatment, and keep records of medication and review; in special circumstances, they can consult with doctors in the Internet hospital and use the online platform connected to the hospital to make registrations and appointments; The intelligent management system for chronic comorbidities in the elderly can also be loaded into a small program on the watch to promptly remind possible risk factors and provide convenient medical service channels for possible patients to choose.

2. The intelligent management system for chronic comorbidities in the elderly according to claim 1 is characterized by: The comprehensive elderly assessment module sets up a personal information database to store the patient's personal information, sets up a diagnostic test database to store the patient's diagnostic data of tests and all related examinations during the patient's visit and hospitalization, and sets up a medication information database to store the patient's daily and hospitalized medication information.

3. The intelligent management system for chronic comorbidities in the elderly according to claim 1 is characterized by: The comprehensive geriatric assessment module is provided with an assessment scale database, which stores assessment scales covering physical function, geriatric syndrome, mental and psychological status, and the Charlson comorbidity index.

4. The intelligent management system for chronic comorbidities in the elderly according to claim 1 is characterized in that: The risk factor prediction model is one or more of a multi-factor logistic regression model, a Cox regression model, a neural network machine learning model, and a time series disease network model.

5. The intelligent management system for chronic comorbidities in the elderly according to claim 1 is characterized in that: Multiple medication management follows the principles of non-drug treatment priority, benefit principle, five-drug principle, small dose principle, timing principle and medication suspension principle; the multiple medication management also includes AI medication warning, which provides timely alarms for patients' medication contraindications.

Images