Hypertension target organ damage assessment and adjuvant therapy scheme generation system

Abstract

The invention discloses a hypertension target organ damage assessment and adjuvant therapy scheme generation system, and relates to the technical field of damage assessment, the hypertension target organ damage assessment and adjuvant therapy scheme generation system comprises an assessment module used for collecting medical data of a hypertensive patient, an analysis module connected with the assessment module and used for analyzing the medical data of the hypertensive patient, the analysis module is used for integrating medical data and outputting patient analysis data through a learning model, the scheme generation module is connected with the analysis module and is used for matching a treatment recommendation scheme according to the patient analysis data and generating a personalized recommendation scheme for a patient, and the dynamic monitoring module is connected with the scheme generation module and is used for monitoring the patient after the personalized recommendation scheme is executed. And the medical data of the hypertensive patient is dynamically obtained, and a personalized recommendation scheme is optimized. According to the method, repeated medical data shared among multiple target organs can be integrated, extraction of the medical data is not delayed, the medical data can be lightened, and a good medical data storage and management effect is achieved.

Description

Technical Field

[0001] This invention relates to the field of damage assessment technology, specifically to a system for assessing target organ damage in hypertension and generating adjunctive treatment plans. Background Technology

[0002] Hypertension, as the leading risk factor for cardiovascular disease worldwide, is harmful not only in terms of elevated blood pressure levels, but more importantly, in the insidious and progressive damage it causes to target organs such as the heart, brain, kidneys, eyes, and blood vessels throughout the body over a long period. For newly diagnosed hypertensive patients, target organ damage assessment is crucial. The results can be directly used for hypertension staging and cardiovascular risk stratification, providing key information for the formulation and selection of drug treatment strategies. However, hypertensive patients require long-term physical examinations and monitoring, necessitating the storage of large amounts of data. Furthermore, the storage of identical data for assessments of multiple target organs makes it difficult to achieve both efficient lightweight storage and rapid retrieval of the data, thus hindering the rapid use of medical data to recommend treatment plans for hypertensive patients. Summary of the Invention

[0003] The purpose of this invention is to provide a system for assessing target organ damage in hypertension and generating adjunctive treatment plans, in order to address the shortcomings of the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a system for assessing target organ damage in hypertension and generating adjunctive treatment plans, comprising: The assessment module is used to collect medical data from hypertensive patients. The medical data includes basic clinical data, molecular markers and multimodal imaging data. Based on the medical data, baseline assessment of target organ damage is performed, and individual models are obtained by modeling patient information. The analysis module, connected to the evaluation model, is used to integrate medical data through the learning model and output patient analysis data. The patient analysis data includes stratification results, individualized blood pressure reduction targets, target organ protection priorities, and treatment decision logic. The patient analysis data is labeled in the individual model. The treatment plan generation module, connected to the analysis module, is used to match treatment recommendation plans based on patient analysis data and generate personalized recommendation plans for patients. The dynamic monitoring module, connected to the solution generation module, is used to dynamically acquire the medical data of hypertensive patients after the personalized recommendation plan is executed, and to optimize the personalized recommendation plan based on the dynamically acquired medical data.

[0005] In a preferred embodiment, the evaluation module includes: The registration and data collection unit is used to register patients on the corresponding medical platform and collect basic clinical data, molecular markers, and multimodal imaging data of hypertensive patients as medical data. The assessment unit is used to identify the target organs that need to be assessed for damage, allocate medical data to the target organs that need to be assessed for damage, obtain multiple allocation data, and perform damage assessment on the target organs that need to be assessed based on the allocation data to obtain the assessment results. The construction unit is used to build a standard organ human model corresponding to the patient in the medical platform, and to perform organ annotation to obtain an individual model. The evaluation results of the target organ are labeled and stored in the organ model corresponding to the individual model.

[0006] In a preferred embodiment, the building unit includes: The model building unit is used to construct the target organ that needs to be damaged in three dimensions, and to perform three-dimensional association and annotation according to the human body structure of the target organ to obtain an individual model; The configuration unit is used to configure multiple storage pages for multiple target organs in the individual model in the medical platform. The multiple storage pages are linked sequentially to obtain the organ model. The storage unit is used to store the allocation data and the corresponding evaluation results in the storage page of the corresponding organ model.

[0007] In a preferred embodiment, the storage unit includes: The marking unit is used to divide multiple allocated data into multiple type data according to type, and to mark the type data that has repetition among the multiple allocated data. The data storage unit is used to store multiple types of data from multiple allocated data into the storage pages of the corresponding organ models respectively, and to enable the storage pages of the organ models in sequence. The image capture unit is used to integrate and capture the type data marked in the storage pages of multiple organ models through the image capture network, which includes a main image capture point and multiple sub-image capture points.

[0008] In a preferred embodiment, the photo-taking unit includes: The corresponding unit is used to configure multiple image retention nets for the corresponding individual model. When there is labeled type data in the storage pages of multiple organ models, one image retention net is activated, and multiple sub-image retention points in the image retention net are respectively matched with the labeled type data. The binding unit is used to delete the type data marked in the organ model, bind the storage location of the deleted type data in the storage page of the organ model to the sub-image point, and store the same type data in the main image point.

[0009] In a preferred embodiment, the analysis module includes: The feedback unit is used to determine the corresponding learning model for each of the multiple target organs. Based on the target organ, the data stored in the master image point is fed back to the storage page of the corresponding organ model to obtain the allocation data of the corresponding organ model and the corresponding evaluation results. The processing unit is used to input the allocated data and the corresponding evaluation results into the learning model of the corresponding target organ and output the output patient analysis data. The data storage unit is used to mark patient analysis data in the storage page of the corresponding organ model in the individual model.

[0010] In a preferred embodiment, the scheme generation module includes: The matching unit is used to generate a treatment plan database, which includes various patient analysis data and corresponding treatment plan databases. Based on the patient analysis data, the treatment plan database is matched to obtain recommended treatment plans. The generation unit is used to provide treatment recommendations to the medical management end (physician) for confirmation and generate a personalized recommendation plan for the corresponding patient.

[0011] In a preferred embodiment, the dynamic monitoring module includes: The dynamic acquisition unit is used to execute personalized recommendation plans and dynamically acquire real-time blood pressure fluctuations, molecular markers and multimodal imaging data of hypertensive patients as dynamic medical data. The optimization unit is used to re-input dynamic medical data into the learning model to obtain dynamic patient analysis data. Based on the dynamic patient analysis data, it matches the treatment plan database to obtain a new personalized recommendation plan, thus completing the optimization of the personalized recommendation plan.

[0012] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention enables paginated storage and management of large amounts of patient data through individual models. It can also correlate medical data from the same period with the corresponding organs being evaluated, integrate duplicate medical data shared by multiple target organs without delaying the retrieval of medical data, and streamline medical data, thus providing a good medical data storage and management function. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0014] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Example 1, please refer to Figure 1 As shown in this embodiment, a system for assessing hypertension target organ damage and generating adjunctive treatment plans includes: The assessment module is used to collect medical data from hypertensive patients. The medical data includes basic clinical data, molecular markers and multimodal imaging data. Based on the medical data, baseline assessment of target organ damage is performed, and individual models are obtained by modeling patient information. The analysis module, connected to the evaluation model, is used to integrate medical data through the learning model and output patient analysis data. The patient analysis data includes stratification results, individualized blood pressure reduction targets, target organ protection priorities, and treatment decision logic. The patient analysis data is labeled in the individual model. The treatment plan generation module, connected to the analysis module, is used to match treatment recommendation plans based on patient analysis data and generate personalized recommendation plans for patients. The dynamic monitoring module, connected to the solution generation module, is used to dynamically acquire the medical data of hypertensive patients after the personalized recommendation plan is executed, and to optimize the personalized recommendation plan based on the dynamically acquired medical data.

[0017] It should be noted that the individual model can store and manage large amounts of patient data in pages, and can also match medical data from the same period with the corresponding organs being evaluated. It can integrate duplicate medical data shared by multiple target organs without delaying the extraction of medical data, and can also streamline medical data, thus having a good role in medical data storage and management.

[0018] In one embodiment, the evaluation module includes: The registration and data collection unit is used to register patients on the corresponding medical platform and collect basic clinical data, molecular markers, and multimodal imaging data of hypertensive patients as medical data. The assessment unit is used to identify the target organs that need to be assessed for damage, allocate medical data to the target organs that need to be assessed for damage, obtain multiple allocation data, and perform damage assessment on the target organs that need to be assessed based on the allocation data to obtain the assessment results. The construction unit is used to build a standard organ human model corresponding to the patient in the medical platform, and to perform organ annotation to obtain an individual model. The evaluation results of the target organ are labeled and stored in the organ model corresponding to the individual model.

[0019] In one embodiment, the building unit includes: The model building unit is used to construct the target organ that needs to be damaged in three dimensions, and to perform three-dimensional association and annotation according to the human body structure of the target organ to obtain an individual model; The configuration unit is used to configure multiple storage pages for multiple target organs in the individual model in the medical platform. The multiple storage pages are linked sequentially to obtain the organ model. The storage unit is used to store the allocation data and the corresponding evaluation results in the storage page of the corresponding organ model.

[0020] In one embodiment, the storage unit includes: The marking unit is used to divide multiple allocated data into multiple type data according to type, and to mark the type data that has repetition among the multiple allocated data. The data storage unit is used to store multiple types of data from multiple allocated data into the storage pages of the corresponding organ models respectively, and to enable the storage pages of the organ models in sequence. The image capture unit is used to integrate and capture the type data marked in the storage pages of multiple organ models through the image capture network, wherein the image capture network includes a main image capture point and multiple sub-image capture points; In one embodiment, the photo-taking unit includes: The corresponding unit is used to configure multiple image retention nets for the corresponding individual model. When there is labeled type data in the storage pages of multiple organ models, one image retention net is activated, and multiple sub-image retention points in the image retention net are respectively matched with the labeled type data. The binding unit is used to delete the type data marked in the organ model, bind the storage location of the deleted type data in the storage page of the organ model to the sub-image point, and store the same type data in the main image point.

[0021] It should be noted that after being diagnosed with hypertension, it is necessary to regularly monitor and collect blood pressure and organ data for periodic evaluation. The medical data collected includes basic clinical data, molecular markers, and multimodal imaging data. The basic clinical data includes: gender, age, height, weight, BMI, waist circumference, and hip circumference; medical history: duration and grade of hypertension, previous medication history, comorbidities (diabetes, coronary heart disease, etc.), smoking / drinking history, and family history; blood pressure data: clinic blood pressure, 24-hour ambulatory blood pressure, and home blood pressure. Molecular markers include: routine biochemical tests: complete blood count, urinalysis, stool analysis, liver and kidney function tests, fasting blood glucose / HbA1c, blood lipids, thyroid function tests, and electrolytes; markers related to cardiac target organ damage: BNP / NT-proBNP, myocardial enzyme profile (troponin cTnI / cTnT, creatine kinase CK / CKMB); and markers related to kidney damage: urine protein, serum creatinine, eGFR, 24-hour urine protein quantification, urine microalbumin / creatinine ratio (UACR), and serum cystatin C. Multimodal imaging data: Cardiac target organ damage-related examinations: electrocardiogram (heart rate / rhythm, P wave duration, Sokolow-Lyon voltage, Cornell product), echocardiography (left atrial diameter / volume index, left ventricular end-diastolic diameter / volume, left ventricular end-systolic diameter / volume, interventricular septal thickness, left ventricular posterior wall thickness, left ventricular mass index (LVMI), left ventricular ejection fraction (LVEF), E / e', tricuspid regurgitation velocity); Kidney damage-related examinations: renal and renal artery ultrasound (renal cortex thickness, kidney volume, renal artery resistance index); Brain damage-related examinations: transcranial Doppler ultrasound (intracranial and extracranial large artery blood flow), cranial MRI / MRA (white matter lesions, lacunar infarction, cerebral microbleeds); Vascular damage-related examinations: carotid ultrasound (IMT thickness, plaque characteristics), vascular ultrasound (ascending aorta, abdominal aorta diameter), ankle-brachial index (ABI), pulse wave velocity (PWV); Retinal damage-related examinations: fundus examination. Among these, there are five types of target organ damage universal core indicators and medical data: age, BMI, waist circumference, hip circumference, duration of hypertension, and blood pressure data; and five common risk factors for target organ damage: diabetes / blood glucose / HbA1c, hyperlipidemia / blood lipids (LDL-C, TC, TG), and hyperuricemia / serum uric acid. Using these data as medical data, the target organs requiring damage assessment are identified, in this case: heart, kidneys, brain, blood vessels, and retina. The collected medical data is then assigned to the corresponding target organs, resulting in multiple assigned data sets. Damage assessment is then performed on each target organ according to the assigned data. Specifically, cardiac target organ damage assessment includes: left atrial enlargement: ECG P wave duration ≥120 ms; echocardiogram left atrial anteroposterior diameter >40 mm for men, left atrial anteroposterior diameter >38 mm for women, and left atrial volume index >34 mL / m³. 2Left ventricular hypertrophy: ECG Sokolow-Lyon voltage > 3.8 mV or Cornell product > 244 mV·ms; Echocardiography interventricular septal thickness ≥ 11 mm, left ventricular posterior wall thickness ≥ 11 mm, LVMI in males > 109 g / m². 2 Women >105 g / m 2 Left ventricular enlargement: Left ventricular end-diastolic diameter ≥55 mm in men and ≥50 mm in women; Left ventricular end-diastolic volume >74 mL / m³ in men. 2 Women > 61 mL / m 2 Left ventricular diastolic function: mean E / e' ≥ 14, interventricular septum e' < 7 cm / s or lateral wall e' < 10 cm / s, left atrial volume index > 34 mL / m³ 2 Tricuspid regurgitation velocity > 2.8 m / s. Left ventricular systolic function: LVEF < 50%. Increased cardiac volume and pressure load: BNP > 35 ng / L, NT-proBNP ≥ 75 ng / L (< 50 years old), ≥ 150 ng / L (50–74 years old), ≥ 300 ng / L (≥ 75 years old). Each of the above lesions (left atrial enlargement, left ventricular enlargement, left ventricular hypertrophy, left ventricular diastolic function) is scored as 1 point. Left ventricular systolic function and BNP > 400 ng / L or NT-proBNP ≥ 450 ng / L (< 50 years old), ≥ 900 ng / L (50–74 years old), ≥ 1800 ng / L (≥ 75 years old) are each scored as 2 points. The final assessment result is obtained. Individual indicators are compared for the kidneys, brain, blood vessels, and retina until the assessment results for all target organs are obtained.

[0022] Next, a 3D model is constructed based on the standard of the target organ and according to the location of the target organ to obtain an individual model. The initial individual models are all constructed according to the standard human body model. Then, multiple storage pages are configured for each target organ in the individual model. The storage page is a storage space that is allocated in the medical platform and bound to the target organ. Multiple storage pages are multiple storage spaces, which are linked sequentially to form a series of "pageable" data surfaces. The formed storage pages can dynamically display the dynamic changes of the target organ in the individual model, which has more time dynamic change and better understanding of the changes of the target organ. In the individual model, each organ configured with multiple storage pages is an organ model, which can store relevant data on the organ's response. The allocated data and the corresponding evaluation results are stored in the storage pages of the corresponding organ model. Multiple allocation data are divided into multiple types according to type. The duplicate type data in the multiple allocation data are marked so that they can be identified and confirmed by the image network in the future. Multiple image networks are configured for the corresponding individual model in the medical platform. When there is marked type data in the storage page of multiple organ models, an image network is activated, and the multiple sub-image points in the image network are matched one by one with the marked type data. The system removes the labeled type data from the organ model, binds the storage location of the deleted type data in the organ model's storage page to the sub-image points, and stores the same type of data individually in the main image point. Here, the image network is a cloud server, and the multiple sub-image points are virtual signal carriers with small data volumes, occupying space for storing small amounts of data. All the sub-image points are communicatively connected to the main image point, which can integrate a large amount of identical data, reducing the burden of storing medical data on the medical platform. At the same time, it also has a good activation effect on the data and will not affect the use of the data. Through the individual model, a large amount of patient data can be stored and managed in pages. It can also match medical data from the same period with the corresponding evaluated organs, integrate duplicate medical data shared by multiple target organs, without delaying the extraction of medical data, and also lightweighting medical data, which has a good medical data storage and management function.

[0023] In one embodiment, the analysis module includes: The feedback unit is used to determine the corresponding learning model for each of the multiple target organs. Based on the target organ, the data stored in the master image point is fed back to the storage page of the corresponding organ model to obtain the allocation data of the corresponding organ model and the corresponding evaluation results. The processing unit is used to input the allocated data and the corresponding evaluation results into the learning model of the corresponding target organ and output the output patient analysis data. The data storage unit is used to mark patient analysis data in the storage page of the corresponding organ model in the individual model.

[0024] It should be noted that the learning models here are trained using allocation data from multiple target organs, corresponding assessment results, and patient analysis data. For example, the heart, brain, blood vessels, kidneys, and retina each have their own allocation data and corresponding assessment results. Five separate learning models are trained on each target organ to obtain a learning model specifically for analyzing patient analysis data for that organ. The patient analysis data includes: stratification results, individualized blood pressure targets, target organ protection priorities, and treatment decision logic. The allocation data and corresponding assessment results are input into the learning model for the corresponding target organ, outputting the patient analysis data. This patient analysis data is then marked in the storage page of the corresponding organ model within the individual model. The patient analysis data is stored as single-organ data and does not require the use of a retrieval network.

[0025] In one embodiment, the scheme generation module includes: The matching unit is used to generate a treatment plan database, which includes various patient analysis data and corresponding treatment plan databases. Based on the patient analysis data, the treatment plan database is matched to obtain recommended treatment plans. The generation unit is used to provide treatment recommendations to the medical management end (physician) for confirmation and generate a personalized recommendation plan for the corresponding patient.

[0026] It should be noted that a treatment plan database needs to be built, which stores various patient analysis data and corresponding treatment plan databases. After the subsequent patient analysis data is obtained, it can be matched with the treatment plan database to obtain a treatment recommendation plan. This treatment recommendation plan will be provided to the doctor in charge of the patient. After the doctor confirms it, it will be implemented to obtain a personalized recommendation plan for the corresponding patient. Here, the recommendation plan is the plan after confirmation by the doctor and matching with the corresponding patient.

[0027] In one embodiment, the dynamic monitoring module includes: The dynamic acquisition unit is used to execute personalized recommendation plans and dynamically acquire real-time blood pressure fluctuations, molecular markers and multimodal imaging data of hypertensive patients as dynamic medical data. The optimization unit is used to re-input dynamic medical data into the learning model to obtain dynamic patient analysis data. Based on the dynamic patient analysis data, it matches the treatment plan database to obtain a new personalized recommendation plan, thus completing the optimization of the personalized recommendation plan.

[0028] It should be noted that after the personalized recommendation plan is implemented for the patient, real-time blood pressure fluctuations, molecular markers, and multimodal imaging data are collected as dynamic medical data and re-entered into the analysis module for analysis to obtain dynamic patient analysis data. Based on the dynamic patient analysis data, the treatment plan database is matched to obtain a new personalized recommendation plan, and the personalized recommendation plan is optimized until the patient returns to normal physical parameters.

[0029] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A system for assessing target organ damage in hypertension and generating adjunctive treatment plans, characterized in that, include: The assessment module is used to collect medical data from hypertensive patients. The medical data includes basic clinical data, molecular markers and multimodal imaging data. Based on the medical data, baseline assessment of target organ damage is performed, and individual models are obtained by modeling patient information. The analysis module, connected to the evaluation model, is used to integrate medical data through the learning model and output patient analysis data. The patient analysis data includes stratification results, individualized blood pressure reduction targets, target organ protection priorities, and treatment decision logic. The patient analysis data is labeled in the individual model. The treatment plan generation module, connected to the analysis module, is used to match treatment recommendation plans based on patient analysis data and generate personalized recommendation plans for patients. The dynamic monitoring module, connected to the solution generation module, is used to dynamically acquire the medical data of hypertensive patients after the personalized recommendation plan is executed, and to optimize the personalized recommendation plan based on the dynamically acquired medical data.

2. The system for assessing and generating adjunctive treatment plans for hypertension target organ damage according to claim 1, characterized in that, The evaluation module includes: The registration and data collection unit is used to register patients on the corresponding medical platform and collect basic clinical data, molecular markers, and multimodal imaging data of hypertensive patients as medical data. The assessment unit is used to identify the target organs that need to be assessed for damage, allocate medical data to the target organs that need to be assessed for damage, obtain multiple allocation data, and perform damage assessment on the target organs that need to be assessed based on the allocation data to obtain the assessment results. The construction unit is used to build a standard organ human model corresponding to the patient in the medical platform, and to perform organ annotation to obtain an individual model. The evaluation results of the target organ are labeled and stored in the organ model corresponding to the individual model.

3. The system for assessing and generating adjunctive treatment plans for hypertension target organ damage according to claim 2, characterized in that, The building unit includes: The model building unit is used to construct the target organ that needs to be damaged in three dimensions, and to perform three-dimensional association and annotation according to the human body structure of the target organ to obtain an individual model; The configuration unit is used to configure multiple storage pages for multiple target organs in the individual model in the medical platform. The multiple storage pages are linked sequentially to obtain the organ model. The storage unit is used to store the allocation data and the corresponding evaluation results in the storage page of the corresponding organ model.

4. The system for assessing and generating adjunctive treatment plans for hypertension target organ damage according to claim 3, characterized in that, The storage unit includes: The marking unit is used to divide multiple allocated data into multiple type data according to type, and to mark the type data that has repetition among the multiple allocated data. The data storage unit is used to store multiple types of data from multiple allocated data into the storage pages of the corresponding organ models respectively, and to enable the storage pages of the organ models in sequence. The image capture unit is used to integrate and capture the type data marked in the storage pages of multiple organ models through the image capture network, which includes a main image capture point and multiple sub-image capture points.

5. The system for assessing and generating adjunctive treatment plans for hypertension target organ damage according to claim 4, characterized in that, The image capture unit includes: The corresponding unit is used to configure multiple image retention nets for the corresponding individual model. When there is labeled type data in the storage pages of multiple organ models, one image retention net is activated, and multiple sub-image retention points in the image retention net are respectively matched with the labeled type data. The binding unit is used to delete the type data marked in the organ model, bind the storage location of the deleted type data in the storage page of the organ model to the sub-image point, and store the same type data in the main image point.

6. The system for assessing and generating adjunctive treatment plans for hypertension target organ damage according to claim 1, characterized in that, The analysis module includes: The feedback unit is used to determine the corresponding learning model for each of the multiple target organs. Based on the target organ, the data stored in the master image point is fed back to the storage page of the corresponding organ model to obtain the allocation data of the corresponding organ model and the corresponding evaluation results. The processing unit is used to input the allocated data and the corresponding evaluation results into the learning model of the corresponding target organ and output the output patient analysis data. The data storage unit is used to mark patient analysis data in the storage page of the corresponding organ model in the individual model.

7. The system for assessing and generating adjunctive treatment plans for hypertension target organ damage according to claim 1, characterized in that, The scheme generation module includes: The matching unit is used to generate a treatment plan database, which includes various patient analysis data and corresponding treatment plan databases. Based on the patient analysis data, the treatment plan database is matched to obtain recommended treatment plans. The generation unit is used to provide treatment recommendations to the medical management system for confirmation and to generate personalized recommendations for the corresponding patients.

8. The system for assessing and generating adjunctive treatment plans for hypertension target organ damage according to claim 1, characterized in that, The dynamic monitoring module includes: The dynamic acquisition unit is used to execute personalized recommendation plans and dynamically acquire real-time blood pressure fluctuations, molecular markers and multimodal imaging data of hypertensive patients as dynamic medical data. The optimization unit is used to re-input dynamic medical data into the learning model to obtain dynamic patient analysis data. Based on the dynamic patient analysis data, it matches the treatment plan database to obtain a new personalized recommendation plan, thus completing the optimization of the personalized recommendation plan.

Images