CN112669967A - Active health medical decision-making assisting method and equipment - Google Patents
Active health medical decision-making assisting method and equipment Download PDFInfo
- Publication number
- CN112669967A CN112669967A CN202011549296.4A CN202011549296A CN112669967A CN 112669967 A CN112669967 A CN 112669967A CN 202011549296 A CN202011549296 A CN 202011549296A CN 112669967 A CN112669967 A CN 112669967A
- Authority
- CN
- China
- Prior art keywords
- health
- user
- data
- diseases
- medical decision
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention relates to an active health medical decision auxiliary method and equipment, wherein the method comprises the following steps: acquiring individual characteristics of ill people with various health problems and diseases as training samples, and training the training samples by a machine learning method or a deep learning method to generate ill risk assessment models corresponding to the health problems or the diseases; pre-compiling health medical decision auxiliary schemes corresponding to various diseases, and storing the compiled health medical decision auxiliary schemes into a health medical decision auxiliary database; individual data of a user are crawled from a multi-source heterogeneous health medical platform and are preprocessed and cleaned, so that individual characteristics of the user are obtained; inputting individual characteristics of the user into the health assessment model, and outputting health problems or diseases possibly suffered by the user; and acquiring a health medical decision auxiliary scheme corresponding to the health problems or diseases from a health medical intervention database according to the diseases possibly suffered by the user, and sending the health medical decision auxiliary scheme to the user after the doctor performs cooperative processing.
Description
Technical Field
The invention relates to an active health medical decision-making auxiliary method and equipment, belonging to the technical field of health medical treatment and data processing.
Background
At present, the concrete manifestations of the medical model of the major countries in the world are very similar and different in nature, and passive medical treatment is the dominant, so far, the home-based medical model has been continuous for thousands of years. The system is characterized in that a patient blindly surrounds a doctor and an operation subject, the system belongs to passive consumption, the doctor also provides services passively, and the served objects are random and have no pertinence; the information is asymmetric and the patient is in a passive consumption state. Prevention is the most economical and effective health strategy, and the main health problems are solved from the 'source management'. Health issues mainly include hygiene prevention and treatment; due to limited medical resources, people cannot quickly know the health state of themselves, people often cannot know the health risk of themselves when many diseases are in sub-health or sprouting states, and health prevention intervention measures are not taken; most of major diseases are usually discovered and treated by patients in middle and late stages, so that the treatment difficulty and cost are greatly increased; patients with minor diseases or chronic diseases lack the systematic guidance of doctors, and many patients basically take care of the development of the patients without early intervention, and the patients do not pay attention until the patients develop large diseases, and all the patients are passively intervened; moreover, many patients with big diseases and difficult diseases do not know how to select medical institutions and do not understand the treatment directions of the patients, and no systematic, real-time and reliable healthy medical system can help the patients to make decision assistance in healthy and unhealthy states; moreover, when a medical institution diagnoses, a doctor does not have comprehensive understanding on a patient in advance, the difficulty in making a diagnosis decision within 5 minutes is high, the data detected only by the instrument and equipment at a time point is one-sided, the cause of a disease cannot be comprehensively and accurately reflected, the diagnosis cost is high, and the doctor provides services passively. From the perspective of a medical health service system, current health prevention and clinical treatment are in a fracture state, and a service chain is discontinuous.
Disclosure of Invention
In order to solve the problems in the prior art, the present invention provides an active health medical decision-making assisting method and device, which can evaluate the risk of suffering from a disease according to the health characteristics of a user, and provide a professional health medical diagnosis and decision-making assisting scheme to help the user to prevent and treat the disease.
The technical scheme of the invention is as follows:
the first technical scheme is as follows:
an active health medical decision assistance method comprises the following steps:
establishing a health assessment model, acquiring individual characteristics of ill people with various health problems and diseases as training samples, and training the training samples by a machine learning method or a deep learning method to generate an ill risk assessment model corresponding to the health problems or the diseases;
establishing a health medical intervention database, and pre-compiling a health medical decision auxiliary scheme corresponding to various diseases, wherein the health medical decision auxiliary scheme at least comprises a health education scheme, life diet intervention, Chinese medicine health maintenance conditioning, exercise physical therapy, self medication, psychological intervention, treatment department selection, a treatment route and a treatment scheme; writing the compiled health medical decision auxiliary scheme into a health medical decision auxiliary database;
acquiring health data, crawling individual data of a user from a multi-source heterogeneous health medical platform, and preprocessing and cleaning the multi-source heterogeneous individual data to obtain individual characteristics of the user;
estimating the disease risk, namely inputting the individual characteristics of the user into a health evaluation model, and outputting the health problems or diseases possibly suffered by the user and the probability of suffering from the corresponding health problems or diseases by the health evaluation model according to the individual characteristics of the user;
and providing health medical decision assistance, acquiring a health medical decision assistance scheme corresponding to health problems or diseases from a health medical intervention database according to diseases possibly suffered by the user, and sending the acquired health medical decision assistance scheme to the user after cooperative processing by a doctor.
Further, the specific steps of obtaining individual characteristics of the ill population of various health problems and diseases as training samples, training the training samples by a machine learning method or a deep learning method, and generating the ill risk assessment model corresponding to the diseases are as follows:
for each health problem or disease, acquiring a group of individual characteristics of a diseased population corresponding to the health problem or disease as training samples, wherein the diseased population at least comprises confirmed patients, highly suspected persons and slightly suspected persons, and the individual characteristics comprise characteristic items and characteristic values corresponding to the characteristic items; adding labels to each training sample, wherein the labels comprise confirmed diagnosis, high suspicion and mild suspicion, and dividing the training samples added with the labels into a training set, a verification set and a test set;
constructing a deep learning neural network, training the deep learning neural network by using a training set, determining the accuracy of the deep learning neural network through a verification set and a test set, and outputting a disease risk evaluation model corresponding to a health problem or a disease after the accuracy of the deep learning neural network meets a preset condition;
wherein, the characteristic items at least comprise life data, physiological sign data, permanent address and environment data, medical history and medical history data and health medical consumption data.
Furthermore, a prediction module of the mass-triggered paroxysmal epidemic, a disease complication analysis module, a medication analysis module and a diagnosis and treatment item association relation analysis module are added in the step of establishing the health assessment model; the prediction module of the mass outbreak epidemic disease is used for capturing real-time information of the mass outbreak epidemic disease from the Internet and outputting risk data that the epidemic disease is possibly infected at the location of the user; the disease complication analysis module is used for acquiring health problems or diseases possibly suffered by the user and output by the disease risk assessment model, and analyzing and outputting complications possibly caused by the corresponding health problems or diseases; the medication analysis module is used for acquiring health problems or diseases possibly suffered by the user, analyzing and outputting available medicines and corresponding medicine information corresponding to the health problems or diseases, available food for diet therapy and corresponding food information, and physical therapy items and corresponding physical therapy information; the diagnosis and treatment item association relation analysis module is used for acquiring health problems or diseases possibly suffered by the user, analyzing and outputting diagnosis and treatment items and diagnosis and treatment item information which are associated with the corresponding health problems or diseases.
Further, the specific steps of crawling the individual data of the user from the multi-source heterogeneous health medical platform, preprocessing and cleaning the multi-source heterogeneous health data, and obtaining the individual characteristics of the cleaned user are as follows:
data preprocessing: traversing the user health information page of each data source, acquiring all data items and characteristic values thereof in the page, merging all the acquired data items and characteristic values thereof, putting the merged data items and characteristic values into a data set, merging all the data items with inconsistent names but same actual meanings in the data set, and acquiring a non-duplicated data item set;
data cleaning: and removing the data items which are not related to the health information in the non-repeated data item set to obtain the individual characteristics of the user comprising a plurality of characteristic items related to the health of the user.
Further, when the individual characteristics of the user are input into the health assessment model, required characteristic items and corresponding characteristic values are captured from the individual characteristics of the user and input into the disease risk assessment model corresponding to the health problems or diseases.
Further, the health medical intervention database is further added with regional problem data, and the step of obtaining the regional data specifically comprises:
setting a plurality of monitoring areas;
adding the user and the individual data of the user into the corresponding monitoring area according to the address information of the user;
monitoring each monitoring area, setting a threshold T, recording the health problem as area problem data and storing the area problem data into a health medical intervention database when the number of users who generate the same health problem in one monitoring area exceeds the threshold T.
Further, the method also comprises the step of actively providing regional group medical auxiliary service, and specifically comprises the following steps:
selecting a monitoring area to be served, and acquiring area data corresponding to the monitoring area from the health medical intervention database; performing cluster analysis and layering on the regional data to obtain homotopic group data suffering from the same health problem or disease and corresponding individual data of users, wherein the homotopic group data at least comprises group population, a user list and coverage rate of corresponding health or disease;
and acquiring a health medical decision auxiliary scheme corresponding to health problems or diseases from the health medical intervention database, calling corresponding individual data of the user according to a user list, sending the health medical decision auxiliary scheme and the individual data of the user to a doctor team, customizing the corresponding individual health medical decision auxiliary scheme according to the health medical decision auxiliary scheme and the individual data of the user, storing the customized individual health medical decision auxiliary scheme in a memory, and distributing the customized individual health medical decision auxiliary scheme to the corresponding user.
Further, when the disease risk estimation step is carried out, the input individual characteristics of the user, the output health problems or diseases possibly suffered by the user and the probability of suffering from the corresponding health problems or diseases are sent to a doctor team, the doctor team evaluates and corrects the output result according to the input individual characteristics of the user, the corrected result is fed back to the disease risk estimation model, and the disease risk estimation model is further trained.
Further, the health medical decision-making auxiliary scheme is sent to the user or the guardian of the user through a mail box, a short message or a third-party communication client.
The second technical scheme is as follows:
an active health medical decision assistance device comprising a memory and a processor, the memory storing instructions adapted to be loaded by the processor and to perform an active health medical decision assistance method as set forth in claim one.
The invention has the following beneficial effects:
1. the invention relates to an active health medical decision auxiliary method and equipment, which help an individual user to change from passive medical treatment to active health, actively provide the risk probability of illness for the individual user, carry out early diagnosis and preventive treatment, realize accurate prevention and accurate treatment, provide the user with the details of disease diagnosis and treatment in all aspects and assist the user in making decisions by writing a health medical decision auxiliary scheme in advance and providing the user with the treatment route, treatment cost, disease complication analysis, medication pattern analysis and the like of diseases.
2. According to the active health medical decision-making auxiliary method and device, after the individual health characteristics of the user are input into the health evaluation model, the corresponding disease risk evaluation model only obtains the individual characteristics related to the corresponding disease, and the processing efficiency of the model is improved.
3. According to the active health medical decision-making auxiliary method and equipment, the health medical intervention database is also added with regional problem data, so that a user can be assisted to know the related health problems of the region through the regional problem data, and a health disease monitoring mechanism can carry out health monitoring, disease prevention and control, epidemic situation early warning treatment and the like.
4. According to the active health medical decision-making auxiliary method and equipment, the result output by the disease risk assessment model is analyzed by a doctor team and then is assessed and corrected, the corrected data is used as a training sample to train the disease risk assessment model again, and the above processes are circulated, so that the output result of the disease risk assessment model is more accurate.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and the specific embodiments.
The first embodiment is as follows:
referring to fig. 1, an active health medical decision assistance method includes the following steps:
establishing a health assessment model, and acquiring individual characteristics of the sick population of various health problems and diseases as training samples, such as: the individual characteristics of the sick people suffering from coronary heart disease, hypertension, diabetes and the like are introduced, training is carried out through a machine learning method or a deep learning method, a coronary heart disease risk assessment model and a diabetes risk assessment model corresponding to the coronary heart disease, the hypertension and the diabetes are generated, the corresponding sick risk assessment model can output the sick probability corresponding to diseases or health problems according to the individual characteristics of a user, and the sick probability can be expressed by numerical values and also can be expressed by severe and mild equivalent-level vocabularies;
establishing a health medical intervention database, and pre-compiling a health medical decision auxiliary scheme corresponding to various diseases, for example: the decision-making auxiliary scheme of the coronary heart disease, the decision-making auxiliary scheme of the hypertension and the decision-making auxiliary scheme of the diabetes are respectively compiled in advance by expert doctors of the coronary heart disease, the hypertension and the diabetes, the expert doctors are not limited to one, and a plurality of people can be requested to compile and adjust the decision-making auxiliary schemes; the health medical decision-making auxiliary scheme at least comprises a health education scheme, life diet intervention, traditional Chinese medicine health maintenance conditioning, exercise physical therapy, self-medication, psychological intervention, treatment department selection, treatment routes and treatment schemes, except emergency treatment, residents firstly provide services through a new health mode center when having health problems, and the health medical decision-making auxiliary scheme is suitable for self-medication, food therapy, physical therapy and small diseases and chronic diseases of medical means such as injection, operation and hospitalization without reaching medical institutions; the diagnosis and treatment are guided before the admission of the big and difficult diseases, diagnosis and treatment routes, treatment schemes, expense estimation and the like are provided for the patients to select, so that the patients can clearly determine proper medical institutions and departments to treat, and the accurate medical seeking is realized; writing the compiled health medical decision auxiliary scheme into a health medical decision auxiliary database;
acquiring health data, crawling individual data of a user from a multi-source heterogeneous health medical platform (such as an electronic medical record system of a hospital, a national health medical big data center, a medical insurance consumption platform, a third-party internet health medical management platform and the like) under the authorization of the user, preprocessing and cleaning the multi-source heterogeneous individual data to obtain individual characteristics of the cleaned user, wherein the individual characteristics comprise historical illness information, age, weight, height, various item values of recent physical examination, physical examination interval time, address information and the like, and acquiring all data related to diseases or health problems as far as possible;
estimating the disease risk, namely inputting the individual characteristics of the user into a health evaluation model, and outputting the health problems or diseases possibly suffered by the user and the probability of suffering from the corresponding health problems or diseases by the health evaluation model according to the individual characteristics of the user;
providing health medical decision assistance, and acquiring a health medical decision assistance scheme corresponding to a disease from a health medical intervention database according to the disease possibly suffered by a user, for example: the health assessment model outputs a health medical decision auxiliary scheme which is correspondingly obtained if the user is likely to suffer from hypertension and heart diseases, the risk of suffering from hypertension is high, and the probability of suffering from heart diseases is low, the obtained health medical decision auxiliary scheme is cooperatively processed by a doctor or a doctor team and then provided for the patient, the doctor or the doctor team reviews the health medical decision auxiliary scheme according to the individual characteristics of the user, if the health medical decision auxiliary scheme is accurate, the health medical decision auxiliary scheme is directly provided for the patient, if the health medical decision auxiliary scheme is wrong, the health medical decision auxiliary scheme is corrected and then sent to the patient, and the patient can automatically determine a medical route according to the risk probability of suffering from hypertension and the health medical decision auxiliary scheme.
At present, generally, a user finds a doctor when the disease condition is displayed or the disease condition is serious, and the intervention and diagnosis time is judged by the user subjectively, so that the time is missed and the blindness is realized, the traditional medical seeking mode is changed by the aid of the decision-making auxiliary method, the current mode that the patient finds the doctor is changed into the mode that the patient finds the doctor, namely, the active service of continuous monitoring is changed, the condition of the disease is found in advance, the tracking and observation of undifferentiated diseases are facilitated, the early intervention is carried out, the individual user is helped to change from passive medical treatment to active medical treatment, the individual data of the user is obtained from the big data, the risk probability of the disease is actively provided for the individual user, the early diagnosis and the preventive treatment are carried out, and the; the health medical decision-making auxiliary scheme is written in advance, so that the user can know the treatment route, treatment cost, disease complication analysis, medication pattern analysis and the like of the disease, the comprehensive details of disease diagnosis and treatment are provided for the user, and the decision making of the user is assisted.
Meanwhile, the decision-making auxiliary scheme enables individuals faced by the user to face medical teams 'general practitioners' including traditional Chinese medical doctors, general practitioners, specialized doctors, public health doctors and the like instead of departments classified according to human organs, tissues, diseases and the like, enables clinicians to actively participate in health services, and has irreplaceable advantages.
Further, the specific steps of obtaining individual characteristics of the ill population of various health problems and diseases as training samples, training the training samples by a machine learning method or a deep learning method, and generating the ill risk assessment model corresponding to the diseases are as follows:
for each health problem or disease, a group of individual characteristics of a sick people corresponding to the health problem or disease are collected as training samples, for example, health data of a sick people suffering from heart disease is collected as training samples, wherein the sick people at least comprise patients with confirmed heart disease, people with high suspected heart disease and people with mild suspected heart disease, and the health data comprise physiological characteristics and corresponding characteristic values, such as height: 175cm, blood pressure: 107, and so on; adding labels to training samples, wherein the labels comprise confirmed diagnosis, high suspicion and mild suspicion, and dividing the training samples added with the labels into a training set, a verification set and a test set;
constructing a deep learning neural network, training the deep learning neural network by using a training set, determining the accuracy of the deep learning neural network through a verification set and a test set, continuously performing iterative training on the deep learning neural network, saving parameters of the deep learning neural network when the accuracy of the deep learning neural network meets the preset requirement, and outputting a disease risk evaluation model corresponding to a health problem or a disease;
wherein, the characteristic items at least comprise life data, physiological sign data, permanent address and environment data, medical history and medical history data and health medical consumption data.
Furthermore, a prediction module of the mass-triggered paroxysmal epidemic, a disease complication analysis module, a medication analysis module and a diagnosis and treatment item association relation analysis module are added in the step of establishing the health assessment model to provide other services for the user; the prediction module of the mass outbreak epidemic disease is used for capturing real-time information of the mass outbreak epidemic disease from the Internet and outputting risk data that the epidemic disease is possibly infected at the location of the user; the disease complication analysis module is used for acquiring health problems or diseases possibly suffered by the user and output by the disease risk assessment model, and analyzing and outputting complications possibly caused by the corresponding health problems or diseases; the medication analysis module is used for acquiring health problems or diseases possibly suffered by the user, analyzing and outputting medicines and corresponding medicine information which can be used corresponding to the health problems or the diseases, food and corresponding food information which can be used for food therapy, physical therapy items and corresponding physical therapy information, for example, informing the user of information such as prices, dosage, production places and instruction manuals of the medicines A and the medicines A, or informing the user of information such as functions and purchase channels of the foods which can be used for food therapy, such as black beans and black beans; the diagnosis and treatment item association relation analysis module is used for acquiring health problems or diseases possibly suffered by the user, analyzing and outputting diagnosis and treatment items and diagnosis and treatment item information which are associated with the corresponding health problems or diseases.
Further, the specific steps of crawling the individual data of the user from the multi-source heterogeneous health medical platform, preprocessing and cleaning the multi-source heterogeneous health data, and obtaining the individual characteristics of the cleaned user are as follows:
data preprocessing: traversing the user health information page of each data source, acquiring all data items and characteristic values thereof in the page, merging all the acquired data items and characteristic values thereof, putting the merged data items and characteristic values into a data set, merging all the data items with inconsistent names but same actual meanings in the data set, and acquiring a non-duplicated data item set;
data cleaning: and removing data items which are irrelevant to the physiological health in the data set without the repeated items, such as personal identification card information, web page addresses and the like, and obtaining the individual characteristics of the user comprising a plurality of characteristic items relevant to the health of the user.
Further, when the individual characteristics of the user are input into the health assessment model, required characteristic items and corresponding characteristic values are captured from the individual characteristics of the user and input into a disease risk assessment model corresponding to health problems or diseases, when a disease risk assessment model is established and a sample is labeled, the label is added by an expert doctor, a disease is usually diagnosed by a plurality of fixed physiological characteristics, for example, the physiological characteristics required by hypertension are blood pressure, pulse rate, body mass index, electrocardiogram, urine and the like, the hypertension risk assessment model only collects the relevant physiological characteristics when the model is built, after the model is established, after the individual health characteristics of the user are input into the health evaluation model, the hypertension risk evaluation model only obtains the relevant individual characteristics, and the processing efficiency of the model is improved.
When active health medical decision assistance is carried out, the closed-loop operation of the health management data of the whole life cycle (except emergency treatment) of the user is carried out by methods such as dynamic health assessment, continuous monitoring of chronic diseases, intervention effect evaluation and the like, and practical data are continuously provided for a health assessment model.
Further, the health medical intervention database is further added with regional problem data, and the step of obtaining the regional data specifically comprises:
setting a plurality of monitoring areas; the monitoring area can be a province, a city, a county and a community, and can also be an area arbitrarily selected on an electronic map;
adding the user and the individual data of the user into the corresponding monitoring area according to the address information of the user;
monitoring each monitoring area, setting a threshold T, recording the health problem as area problem data and storing the area problem data into a health medical intervention database when the number of users who generate the same health problem in one monitoring area exceeds the threshold T. The regional problem data can be provided for users to be available at any time in life health care, medical treatment, health care, old age and the like; the system is provided for a health disease monitoring mechanism to carry out health monitoring, disease prevention and control, epidemic situation early warning treatment and the like.
Further, the method also comprises the step of actively providing regional group medical auxiliary service, and specifically comprises the following steps:
(1) selecting a monitoring area to be served, and acquiring area data corresponding to the monitoring area from the health medical intervention database; performing cluster analysis and layering on the regional data to obtain homotopic group data suffering from the same health problem or disease and corresponding individual data of users, wherein the homotopic group data at least comprises group population, a user list and coverage rate of corresponding health or disease;
(2) and acquiring a health medical decision auxiliary scheme corresponding to health problems or diseases from the health medical intervention database, calling corresponding individual data of the user according to a user list, sending the health medical decision auxiliary scheme and the individual data of the user to a doctor team, carrying out early diagnosis by the doctor through an epidemiology analysis method, carrying out real-time monitoring and evaluation, disease early warning, chronic disease screening and the like, extracting a preventive treatment scheme or an intervention suggestion from the health medical intervention database, and providing individualized and accurate service data. The doctor can serve the regional 'same-symptom' group in a centralized way according to the data, or the doctor can serve the individual independently, so that the doctor can serve the service in a store, remotely or wait for a tour in a centralized way according to objective conditions; can serve individual patients by a plurality of doctors simultaneously, and provides continuous, comprehensive and coordinated health management and disease management services for the patients.
(1) And (2) the steps are a bidirectional circulation process, and the regional problem data and the individual data are fused, and the health clinical intervention and the medical clinical intervention are fused.
Further, when the disease risk estimation step is carried out, the input individual characteristics of the user, the output health problems or diseases possibly suffered by the user and the probability of suffering from the corresponding health problems or diseases are sent to a doctor team, the doctor team evaluates and corrects the output result according to the input individual characteristics of the user, the corrected result is fed back to the disease risk estimation model, and the disease risk estimation model is further trained. A working mode of combining real-time dynamic analysis and timing analysis and combining a health assessment model and man-machine interaction of a doctor team is adopted; when a certain element of the individual data is subjected to biochemistry, the health assessment model is immediately analyzed and processed when multi-source dynamic factors such as disease symptoms, physical sign indexes, age, environment, psychology and the like change, real-time updated individual data is formed, a service request is triggered to a doctor team when the individual data of a user changes, the doctor team actively evaluates and corrects health problems or diseases possibly suffered by the user and the probability of suffering from the corresponding health problems or diseases after analyzing the individual data, the corrected data is used as a training sample again to train the disease risk assessment model, and the processes are circulated, so that the output result of the disease risk assessment model is more accurate.
Furthermore, the health medical decision-making auxiliary scheme is sent to the user or the guardian of the user through a mail box, a short message or a third-party communication client, and the individual user or the family member of the individual user can be endowed with the whole-process decision-making capability as the guardian of the user. The output health medical decision-making auxiliary scheme can be further analyzed by a doctor team, and then the most effective and most beneficial individualized intervention method or medical scheme is determined by combining traditional Chinese medicine, preventive medicine and clinical medicine thought methods, and then is stored in a national health medical big data center in real time or is sent to a user and a guardian of the user according to the setting.
Example two:
an active health medical decision assistance apparatus comprising a memory and a processor, the memory storing instructions adapted to be loaded by the processor and to perform an active health medical decision assistance method as described in the first embodiment.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes performed by the present specification and drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. An active health medical decision assistance method, comprising the steps of:
establishing a health assessment model, acquiring individual characteristics of ill people with various health problems and diseases as training samples, and training the training samples by a machine learning method or a deep learning method to generate an ill risk assessment model corresponding to the health problems or the diseases;
establishing a health medical intervention database, and pre-compiling a health medical decision auxiliary scheme corresponding to various diseases, wherein the health medical decision auxiliary scheme at least comprises a health education scheme, life diet intervention, Chinese medicine health maintenance conditioning, exercise physical therapy, self medication, psychological intervention, treatment department selection, a treatment route and a treatment scheme; writing the compiled health medical decision auxiliary scheme into a health medical decision auxiliary database;
acquiring health data, crawling individual data of a user from a multi-source heterogeneous health medical platform, and preprocessing and cleaning the multi-source heterogeneous individual data to obtain individual characteristics of the user;
estimating the disease risk, namely inputting the individual characteristics of the user into a health evaluation model, and outputting the health problems or diseases possibly suffered by the user and the probability of suffering from the corresponding health problems or diseases by the health evaluation model according to the individual characteristics of the user;
and providing health medical decision assistance, acquiring a health medical decision assistance scheme corresponding to health problems or diseases from a health medical intervention database according to diseases possibly suffered by the user, and sending the acquired health medical decision assistance scheme to the user after cooperative processing by a doctor.
2. The active health medical decision-making assistance method according to claim 1, wherein the specific steps of obtaining individual features of the ill population of various health problems and diseases as training samples, training the training samples by a machine learning method or a deep learning method, and generating an ill risk assessment model corresponding to the diseases are as follows:
for each health problem or disease, acquiring a group of individual characteristics of a diseased population corresponding to the health problem or disease as training samples, wherein the diseased population at least comprises confirmed patients, highly suspected persons and slightly suspected persons, and the individual characteristics comprise characteristic items and characteristic values corresponding to the characteristic items; adding labels to each training sample, wherein the labels comprise confirmed diagnosis, high suspicion and mild suspicion, and dividing the training samples added with the labels into a training set, a verification set and a test set;
constructing a deep learning neural network, training the deep learning neural network by using a training set, determining the accuracy of the deep learning neural network through a verification set and a test set, and outputting a disease risk evaluation model corresponding to a health problem or a disease after the accuracy of the deep learning neural network meets a preset condition;
wherein, the characteristic items at least comprise life data, physiological sign data, permanent address and environment data, medical history and medical history data and health medical consumption data.
3. The active health medical decision assistance method of claim 2, wherein: in the step of establishing the health assessment model, a prediction module of the mass-triggered paroxysmal epidemic, a disease complication analysis module, a medication analysis module and a diagnosis and treatment item association relation analysis module are also added; the prediction module of the mass outbreak epidemic disease is used for capturing real-time information of the mass outbreak epidemic disease from the Internet and outputting risk data that the epidemic disease is possibly infected at the location of the user; the disease complication analysis module is used for acquiring health problems or diseases possibly suffered by the user and output by the disease risk assessment model, and analyzing and outputting complications possibly caused by the corresponding health problems or diseases; the medication analysis module is used for acquiring health problems or diseases possibly suffered by the user, analyzing and outputting available medicines and corresponding medicine information corresponding to the health problems or diseases, available food for diet therapy and corresponding food information, and physical therapy items and corresponding physical therapy information; the diagnosis and treatment item association relation analysis module is used for acquiring health problems or diseases possibly suffered by the user, analyzing and outputting diagnosis and treatment items and diagnosis and treatment item information which are associated with the corresponding health problems or diseases.
4. The active health medical decision assistance method according to claim 2, wherein the specific steps of crawling individual data of the user from the multi-source heterogeneous health medical platform, preprocessing and cleaning the multi-source heterogeneous health data, and obtaining the cleaned individual characteristics of the user are as follows:
data preprocessing: traversing the user health information page of each data source, acquiring all data items and characteristic values thereof in the page, merging all the acquired data items and characteristic values thereof, putting the merged data items and characteristic values into a data set, merging all the data items with inconsistent names but same actual meanings in the data set, and acquiring a non-duplicated data item set;
data cleaning: and removing the data items which are not related to the health information in the non-repeated data item set to obtain the individual characteristics of the user comprising a plurality of characteristic items related to the health of the user.
5. The active health medical decision assistance method of claim 4, wherein: when the individual characteristics of the user are input into the health assessment model, required characteristic items and corresponding characteristic values are captured from the individual characteristics of the user and input into a disease risk assessment model corresponding to health problems or diseases.
6. The active health medical decision assistance method according to claim 1, wherein regional problem data is further added to the health medical intervention database, and the step of obtaining the regional data specifically comprises:
setting a plurality of monitoring areas;
adding the user and the individual data of the user into the corresponding monitoring area according to the address information of the user;
monitoring each monitoring area, setting a threshold T, recording the health problem as area problem data and storing the area problem data into a health medical intervention database when the number of users who generate the same health problem in one monitoring area exceeds the threshold T.
7. The active health medical decision assistance method according to claim 6, further comprising the step of actively providing regional group medical assistance services, specifically:
selecting a monitoring area to be served, and acquiring area data corresponding to the monitoring area from the health medical intervention database; performing cluster analysis and layering on the regional data to obtain homotopic group data suffering from the same health problem or disease and corresponding individual data of users, wherein the homotopic group data at least comprises group population, a user list and coverage rate of corresponding health or disease;
and acquiring a health medical decision auxiliary scheme corresponding to health problems or diseases from the health medical intervention database, calling corresponding individual data of the user according to a user list, sending the health medical decision auxiliary scheme and the individual data of the user to a doctor team, customizing the corresponding individual health medical decision auxiliary scheme according to the health medical decision auxiliary scheme and the individual data of the user, storing the customized individual health medical decision auxiliary scheme in a memory, and distributing the customized individual health medical decision auxiliary scheme to the corresponding user.
8. The active health medical decision assistance method of claim 1, wherein: when the disease risk estimation step is carried out, the input individual characteristics of the user, the output health problems or diseases possibly suffered by the user and the probability of suffering from the corresponding health problems or diseases are sent to a doctor team, the doctor team evaluates and corrects the output result according to the input individual characteristics of the user, the corrected result is fed back to a disease risk evaluation model, and the disease risk evaluation model is further trained.
9. The active health medical decision assistance method of claim 1, wherein: the health medical decision-making auxiliary scheme is sent to the user or the guardian of the user through a mail box, a short message or a third-party communication client.
10. An active health medical decision assistance device, characterized by: comprising a memory and a processor, the memory storing instructions adapted to be loaded by the processor and to perform an active health medical decision assistance method according to any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011549296.4A CN112669967B (en) | 2020-12-24 | 2020-12-24 | Active health medical decision-making assisting method and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011549296.4A CN112669967B (en) | 2020-12-24 | 2020-12-24 | Active health medical decision-making assisting method and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112669967A true CN112669967A (en) | 2021-04-16 |
| CN112669967B CN112669967B (en) | 2022-09-16 |
Family
ID=75409928
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011549296.4A Active CN112669967B (en) | 2020-12-24 | 2020-12-24 | Active health medical decision-making assisting method and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112669967B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113205890A (en) * | 2021-05-21 | 2021-08-03 | 中国人民大学 | Control method and system based on infectious disease propagation path |
| CN113223713A (en) * | 2021-05-08 | 2021-08-06 | 中国福利会国际和平妇幼保健院 | Dynamic monitoring method, system and storage medium based on risk assessment |
| CN113539480A (en) * | 2021-07-20 | 2021-10-22 | 武汉情智感知科技有限公司 | Novel mental health intervention interactive system |
| CN114864088A (en) * | 2022-04-26 | 2022-08-05 | 福建福寿康宁科技有限公司 | Medical health-based digital twin establishing method and device and storage medium |
| CN115083601A (en) * | 2022-07-25 | 2022-09-20 | 四川省医学科学院·四川省人民医院 | Type 2diabetes auxiliary decision making system based on machine learning |
| CN115274079A (en) * | 2022-07-25 | 2022-11-01 | 成都市第六人民医院 | Medical and health management system based on big data |
| CN115966278A (en) * | 2022-12-07 | 2023-04-14 | 北京健康在线技术开发有限公司 | Cardiovascular metabolic disease prevention and treatment method and system based on digital therapy |
| CN117292829A (en) * | 2023-10-13 | 2023-12-26 | 黄恺 | Graded diagnosis and treatment information system for coronary heart disease |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102110192A (en) * | 2011-04-02 | 2011-06-29 | 中国医学科学院医学信息研究所 | Auxiliary disease judgment method based on diagnostic element data association |
| CN107526938A (en) * | 2017-10-10 | 2017-12-29 | 福建镓康医疗科技有限公司 | A kind of primary care Health Delivery System |
| CN110111884A (en) * | 2019-04-30 | 2019-08-09 | 杭州电子科技大学 | A kind of man-machine coordination intelligent medical treatment aid decision-making system based on CMKMC |
| CN110189823A (en) * | 2019-05-29 | 2019-08-30 | 深圳市国联健康数据科技有限责任公司 | A kind of homologous health medical treatment of intelligence and the medical model of prevention |
| US20200035363A1 (en) * | 2018-07-26 | 2020-01-30 | Siemens Healthcare Gmbh | System to predict health outcomes |
| US10553319B1 (en) * | 2019-03-14 | 2020-02-04 | Kpn Innovations, Llc | Artificial intelligence systems and methods for vibrant constitutional guidance |
| CN111125061A (en) * | 2019-12-18 | 2020-05-08 | 甘肃省卫生健康统计信息中心(西北人口信息中心) | Method for standardizing and promoting health medical big data |
| CN111986757A (en) * | 2020-08-27 | 2020-11-24 | 北京天骥空间科技有限公司 | Parallel human health information system |
-
2020
- 2020-12-24 CN CN202011549296.4A patent/CN112669967B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102110192A (en) * | 2011-04-02 | 2011-06-29 | 中国医学科学院医学信息研究所 | Auxiliary disease judgment method based on diagnostic element data association |
| CN107526938A (en) * | 2017-10-10 | 2017-12-29 | 福建镓康医疗科技有限公司 | A kind of primary care Health Delivery System |
| US20200035363A1 (en) * | 2018-07-26 | 2020-01-30 | Siemens Healthcare Gmbh | System to predict health outcomes |
| US10553319B1 (en) * | 2019-03-14 | 2020-02-04 | Kpn Innovations, Llc | Artificial intelligence systems and methods for vibrant constitutional guidance |
| CN110111884A (en) * | 2019-04-30 | 2019-08-09 | 杭州电子科技大学 | A kind of man-machine coordination intelligent medical treatment aid decision-making system based on CMKMC |
| CN110189823A (en) * | 2019-05-29 | 2019-08-30 | 深圳市国联健康数据科技有限责任公司 | A kind of homologous health medical treatment of intelligence and the medical model of prevention |
| CN111125061A (en) * | 2019-12-18 | 2020-05-08 | 甘肃省卫生健康统计信息中心(西北人口信息中心) | Method for standardizing and promoting health medical big data |
| CN111986757A (en) * | 2020-08-27 | 2020-11-24 | 北京天骥空间科技有限公司 | Parallel human health information system |
Non-Patent Citations (4)
| Title |
|---|
| GEORGY LEBEDEV 等: "Technology of Supporting Medical Decision-Making Using Evidence-Based Medicine and Artificial Intelligence", 《KNOWLEDGE-BASED AND INTELLIGENT INFORMATION & ENGINEERING SYSTEMS: PROCEEDINGS OF THE 24TH INTERNATIONAL CONFERENCE KES2020》 * |
| YAOMING LAI 等: "2019 Novel Coronavirus-Infected Pneumonia on CT: A Feasibility Study of Few-Shot Learning for Computerized Diagnosis of Emergency Diseases", 《IEEE ACCESS》 * |
| 李炳辉: "基于商业智能的医疗卫生资源决策支持系统分析与设计", 《中国优秀博硕士学位论文全文数据库(硕士)医药卫生科技辑》 * |
| 梁书彤 等: "基于机器学习的医疗决策支持系统综述", 《计算机工程与应用》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113223713A (en) * | 2021-05-08 | 2021-08-06 | 中国福利会国际和平妇幼保健院 | Dynamic monitoring method, system and storage medium based on risk assessment |
| CN113205890A (en) * | 2021-05-21 | 2021-08-03 | 中国人民大学 | Control method and system based on infectious disease propagation path |
| CN113539480A (en) * | 2021-07-20 | 2021-10-22 | 武汉情智感知科技有限公司 | Novel mental health intervention interactive system |
| CN114864088A (en) * | 2022-04-26 | 2022-08-05 | 福建福寿康宁科技有限公司 | Medical health-based digital twin establishing method and device and storage medium |
| CN114864088B (en) * | 2022-04-26 | 2023-04-14 | 福建福寿康宁科技有限公司 | Digital twin establishing method and device based on medical health and storage medium |
| WO2023207795A1 (en) * | 2022-04-26 | 2023-11-02 | 吴运良 | Medical health-based digital twin establishing method and device, and storage medium |
| CN115083601A (en) * | 2022-07-25 | 2022-09-20 | 四川省医学科学院·四川省人民医院 | Type 2diabetes auxiliary decision making system based on machine learning |
| CN115274079A (en) * | 2022-07-25 | 2022-11-01 | 成都市第六人民医院 | Medical and health management system based on big data |
| CN115966278A (en) * | 2022-12-07 | 2023-04-14 | 北京健康在线技术开发有限公司 | Cardiovascular metabolic disease prevention and treatment method and system based on digital therapy |
| CN117292829A (en) * | 2023-10-13 | 2023-12-26 | 黄恺 | Graded diagnosis and treatment information system for coronary heart disease |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112669967B (en) | 2022-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112669967B (en) | Active health medical decision-making assisting method and equipment | |
| CN105260588B (en) | A kind of health guards robot system and its data processing method | |
| Bartko | The intraclass correlation coefficient as a measure of reliability | |
| US20170357760A1 (en) | Clinical decision supporting ensemble system and clinical decision supporting method using the same | |
| CN108780663A (en) | Digital personalized medicine platform and system | |
| KR20170140757A (en) | A clinical decision support ensemble system and the clinical decision support method by using the same | |
| KR20170061222A (en) | The method for prediction health data value through generation of health data pattern and the apparatus thereof | |
| KR20190132290A (en) | Method, server and program of learning a patient diagnosis | |
| CN109065162A (en) | A kind of comprehensive intelligent diagnostic system | |
| Richard et al. | Testing a New Proposal for Somatization Disorder | |
| CN110875087B (en) | Chronic pulmonary disease management system | |
| WO2019223508A1 (en) | Method for establishing alzheimer's disease stage assessment model, and computer device | |
| Junaid et al. | Data science and its application in heart disease prediction | |
| CN107707606A (en) | Cerebral apoplexy first aid examination supervision doctor's networked system and its application process | |
| Corbett et al. | Cognition and understanding of neuropathy of inpatients admitted to a specialized tertiary diabetic foot unit with diabetes-related foot ulcers | |
| Tong et al. | Tracking fatigue and health state in multiple sclerosis patients using connnected wellness devices | |
| Dahalan et al. | Heart rate events classification via explainable fuzzy logic systems | |
| CN113674823A (en) | Internet intelligent physical examination management system and application thereof | |
| Shakhmametova et al. | Clinical decision support system for the respiratory diseases diagnosis | |
| Khan et al. | Severe analysis of cardiac disease detection using the wearable device by artificial intelligence | |
| Nakamizo et al. | Radiation effects on atherosclerosis in atomic bomb survivors: a cross‐sectional study using structural equation modeling | |
| RU2752707C1 (en) | Medical system for remote monitoring, analysis and prediction of condition of patient by sequence of electrocardiograms of heart from first deflection and computerised method for monitoring, analysis and prediction of condition of patient | |
| Pandey et al. | The future of AI-enhanced ECG interpretation for valvular heart disease screening | |
| KR20230097332A (en) | System for early diagnosis of dementia and Method for early diagnosis of dementia using the same | |
| Amira et al. | Stress level classification using heart rate variability |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |