Medical follow-up visit method and system

Abstract

The invention discloses a medical follow-up visit method and system, relates to the technical field of medical information systems, and solves the problems that an existing follow-up visit system cannot realize full-process automatic recording of follow-up visit data and cannot realize early screening of diseases in combination with data resources. Converting and storing the input data into follow-up visit data; receiving a follow-up request of a doctor end, creating audio and video communication with a patient end according to the follow-up request, and converting and storing data in the audio and video communication into follow-up data; receiving a follow-up scheme of the doctor side, pushing the follow-up scale to the patient side pushing the group, and converting and storing the follow-up scale returned by the patient side into follow-up data; performing early screening analysis by considering core factors, family genetic factors, internal constitution factors and living environment factors according to follow-up visit data; and generating an early warning signal based on an early screening analysis result, and performing early warning on the doctor terminal and the patient terminal according to a built-in early warning mode, thereby solving the problem.

Description

Technical Field

[0001] The present invention relates to the technical field of medical information systems, and more specifically, to a medical follow-up method and system. Background Art

[0002] Currently, there are many medical follow-up systems on the market, most of which take the disease as the main line, adopt the form of active reporting combined with passive follow-up, and collect important node data in the disease process (disease occurrence, diagnosis, and development). The existing follow-up systems on the market have the following problems and shortcomings.

[0003] First, the existing follow-up system only simply collects disease-related data generated by patients during their life cycle and is unable to fully utilize existing data resources to achieve early disease screening.

[0004] Second, the existing follow-up system fails to combine various technical capabilities to achieve automated follow-up data processing. Current technology applications are often relatively isolated. Although technologies such as OCR recognition, speech-to-text, and AI outbound calls have been applied to a certain extent in their respective fields, in the medical field, there is no complete set of automated collection processes for follow-up data, especially the standardized processing of unstructured data is still insufficient, which greatly reduces the utilization rate of follow-up data and makes it difficult to improve the effect of early disease screening. Summary of the invention

[0005] The purpose of this application is to provide a medical follow-up method and system to solve the problem that the existing follow-up system fails to realize automatic recording of the entire process of follow-up data and cannot combine data resources to realize early screening of diseases; to collect, transform, and organize various types of data generated throughout the follow-up process to generate follow-up data, realize automatic collection of follow-up data, and avoid loss of follow-up data; to consider core factors, family genetic factors, intrinsic physical factors and living environment factors to conduct early screening analysis of follow-up data, and to warn of possible abnormal data, which is conducive to quickly discovering abnormal situations during patient follow-up.

[0006] The present application first provides a medical follow-up method, which is executed based on a capability platform, including: receiving input data from a patient side, the input data including: form data, image data and audio and video data, and converting and storing the input data into follow-up data; receiving a follow-up request from a doctor side, creating an audio and video communication with the patient side according to the follow-up request, and converting and storing the data in the audio and video communication into follow-up data; receiving a follow-up plan from a doctor side, the follow-up plan including: a follow-up scale, a push time and a push group, pushing the follow-up scale to the patient side of the push group according to the push time and the push group, and converting and storing the follow-up scale sent back by the patient side into follow-up data; performing early screening analysis based on the follow-up data, considering core factors, family genetic factors, intrinsic physical factors and living environment factors; generating a warning signal based on the early screening analysis results, and issuing a warning to the doctor side and the patient side according to the built-in warning mode.

[0007] In one possible implementation, early screening analysis is performed based on follow-up data taking into account core factors, family genetic factors, intrinsic physical factors and living environment factors; including: extracting answers to core questions, family genetic questions, intrinsic physical factors and living environment questions from the follow-up data and quantifying the answers into numerical values ​​to obtain quantified answer values; calculating weight values ​​for core questions, family genetic questions, intrinsic physical factors and living environment questions; obtaining an early screening warning coefficient based on the weighted summation of the quantified values ​​of the answers to core questions, family genetic questions, intrinsic physical factors and living environment questions and the corresponding weight values; comparing the early screening warning coefficient with a selected threshold to obtain an early screening analysis result.

[0008] In a possible implementation, the weight values ​​of core issues, family genetic issues, intrinsic physical issues, and living environment issues are calculated; including: obtaining follow-up data of patients and non-patients, extracting quantitative values ​​of patients' answers to various questions and quantitative values ​​of non-patients' answers from the follow-up data; for each core issue, dividing the difference between the mean of the quantitative values ​​of all patients' answers to the core issue and the mean of the quantitative values ​​of all non-patients' answers to the core issue by the difference between the upper and lower limits of the quantitative values ​​of the answers to the core issue, to obtain the weight value of the core issue; for any of family genetic issues, intrinsic physical issues, and living environment issues, the weight value of the core issue is calculated. A question, based on all the quantitative values ​​of each patient's answers to this question, calculate the quantitative mean of each patient's answer to this question, based on the quantitative mean of each patient's answer to this question, calculate the quantitative mean of all patients' answers to this question, and divide the difference between the quantitative mean of all patients' answers to this question and the quantitative mean of all non-sick people's answers to this question by the difference between the upper and lower limits of the quantitative values ​​of the answers to this question to obtain the weight value of this question; calculate the dynamic adjustment factor based on the sum of the weight values ​​of all core questions, and adjust the weight values ​​of family genetic problems, intrinsic physical problems and living environment problems based on the dynamic adjustment factor.

[0009] In a possible implementation, the early screening warning coefficient is compared with a selected threshold to obtain an early screening analysis result; including: calculating the early screening warning coefficients of all patients based on the follow-up data of all patients, and sorting the early screening warning coefficients of all patients according to the size of the coefficients; selecting the early screening warning coefficients in order from small to large, calculating the difference between the selected early screening warning coefficient and the average value of the early screening warning coefficients of all patients and the difference between the selected early screening warning coefficient and the adjacent larger early screening warning coefficient, until both differences are less than or equal to the set threshold, at which time the selected early screening warning coefficient is the selected threshold; comparing the early screening warning coefficient calculated from the follow-up data of patients who have not been determined to be ill with the selected threshold, when the early screening warning coefficient is greater than or equal to the selected threshold, it indicates that there is an abnormal risk, and when the early screening warning coefficient is less than the selected threshold, it indicates that there is no abnormal risk.

[0010] In a possible implementation, it also includes: receiving an education plan from the doctor side, the education plan including: education content, push time and push group, and pushing the education content to the patient side of the corresponding push group according to the push time and push group.

[0011] The present application also provides a medical follow-up system, including: a patient side, for patients to enter data, the entered data including: form data, image data and audio and video data; a doctor side, for grouping patients, initiating follow-up requests and creating follow-up plans, the follow-up requests including: video requests and audio requests, and the follow-up plans including: follow-up scales, push time and push groups; a capability platform side, for receiving the entered data from the patient side and converting and storing it into follow-up data, creating audio and video communication with the patient side according to the follow-up request from the doctor side and converting and storing the data in the audio and video communication into follow-up data, pushing the follow-up scale to the patient side in the push group according to the follow-up plan of the doctor side and converting and storing the returned follow-up scale into follow-up data, performing early screening analysis based on the follow-up data considering core factors, family genetic factors, intrinsic physical factors and living environment factors, generating early warning signals based on the early screening analysis results, and issuing warnings to the doctor side and the patient side according to the built-in warning mode; a management side, for managing the data of the patient side and the doctor side by department.

[0012] In one possible implementation, the capability platform includes: an early warning answer quantification module, which is used to extract answers to core questions, family genetic questions, intrinsic physical problems and living environment questions from follow-up data and quantify the answers into numerical values ​​to obtain answer quantification values; an early warning weight calculation module, which is used to calculate the weight values ​​of core questions, family genetic questions, intrinsic physical problems and living environment questions; an early warning coefficient calculation module, which is used to obtain an early screening early warning coefficient based on the weighted sum of the quantified values ​​of the answers to core questions, family genetic questions, intrinsic physical problems and living environment questions and the corresponding weight values; an early screening early warning analysis module, which is used to compare the early screening early warning coefficient with a selected threshold to obtain an early screening analysis result.

[0013] In a possible implementation, the warning weight calculation module on the capability platform is specifically used to: obtain follow-up data of patients and non-patients, and extract the quantitative values ​​of the patients' answers and the quantitative values ​​of the non-patients' answers to various questions from the follow-up data; for each core question, divide the difference between the mean of the quantitative values ​​of all patients' answers to the core question and the mean of the quantitative values ​​of all non-patients' answers to the core question by the difference between the upper and lower limits of the quantitative values ​​of the answers to the core question to obtain the weight value of the core question; for any one of family genetic problems, intrinsic physical problems and living environment problems, calculate the quantitative mean of each patient's answer to the problem based on all the quantitative values ​​of each patient's answer to the problem, calculate the quantitative mean of all patients' answers to the problem based on the quantitative mean of each patient's answer to the problem, and obtain the weight value of the problem based on the difference between the quantitative mean of all patients' answers to the problem and the quantitative mean of all non-patients' answers to the problem divided by the difference between the upper and lower limits of the quantitative values ​​of the answers to the problem; calculate the dynamic adjustment factor based on the sum of the weight values ​​of all core problems, and adjust the weight values ​​of family genetic problems, intrinsic physical problems and living environment problems based on the dynamic adjustment factor.

[0014] In one possible implementation, the early screening warning analysis module is specifically used to: calculate the early screening warning coefficients of all patients based on the follow-up data of all patients, and sort the early screening warning coefficients of all patients according to the size of the coefficients; select the early screening warning coefficients in order from small to large, calculate the difference between the selected early screening warning coefficient and the average value of the early screening warning coefficients of all patients and the difference between the selected early screening warning coefficient and the adjacent larger early screening warning coefficient, until both differences are less than or equal to the set threshold, at which time the selected early screening warning coefficient is the selected threshold; compare the early screening warning coefficient calculated from the follow-up data of patients who have not been determined to be ill with the selected threshold, when the early screening warning coefficient is greater than or equal to the selected threshold, it indicates that there is an abnormal risk, and when the early screening warning coefficient is less than the selected threshold, it indicates that there is no abnormal risk.

[0015] In a possible implementation, in the system: the doctor side is also used to create an education plan, which includes: education content, push time and push group; the capability platform side is also used to push the education content to the patient side of the corresponding push group according to the push time and push group.

[0016] Compared with the prior art, the present application has the following beneficial effects: on the one hand, it realizes the standardized data collection of the whole follow-up process: the form data, picture data and audio and video data in the follow-up process are automatically collected, and converted into standardized follow-up data for subsequent early disease screening; on the other hand, it realizes early disease screening: based on the collected follow-up data, early screening analysis is carried out by jointly considering core factors, family genetic factors, intrinsic physical factors and living environment factors, and early screening of the disease is carried out. The collected data covers the whole follow-up process to improve the accuracy and efficiency of early disease screening, and can quickly identify potential disease risks, provide doctors with more accurate diagnostic basis, and facilitate early detection and treatment of the disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0018] Figure 1 A flowchart of a medical follow-up method provided in an embodiment of the present application;

[0019] Figure 2 A structural diagram of a medical follow-up system provided in an embodiment of the present application;

[0020] Figure 3 This is the application architecture diagram of the medical follow-up system;

[0021] Figure 4 This is the business architecture diagram of the medical follow-up system;

[0022] Figure 5 This is the technical architecture diagram of the medical follow-up system;

[0023] Figure 6 Flowchart for follow-up data collection;

[0024] Figure 7 Flow chart for statistical analysis of follow-up data;

[0025] Figure 8 Flowchart for early screening of the disease. DETAILED DESCRIPTION

[0026] Hereinafter, the term "include" or "may include" used in various embodiments of the present application indicates the presence of the function, operation or element applied for, and does not limit the increase of one or more functions, operations or elements. In addition, as used in various embodiments of the present application, the terms "include", "have" and their cognates are intended only to indicate specific features, numbers, steps, operations, elements, components or a combination of the foregoing items, and should not be understood as first excluding the presence of one or more other features, numbers, steps, operations, elements, components or a combination of the foregoing items or the possibility of increasing one or more features, numbers, steps, operations, elements, components or a combination of the foregoing items.

[0027] The terms used in the various embodiments of the application are only used to describe the purpose of specific embodiments and are not intended to limit the various embodiments of the application. As used herein, the singular form is intended to also include the plural form, unless the context clearly indicates otherwise. Unless otherwise limited, all terms used here (including technical terms and scientific terms) have the same meaning as the meanings commonly understood by ordinary technicians in the field of the various embodiments of the application. Terms (such as terms defined in generally used dictionaries) will be interpreted as having the same meaning as the contextual meaning in the relevant technical field and will not be interpreted as having idealized meanings or too formal meanings, unless clearly defined in the various embodiments of the application.

[0028] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with examples and drawings. The illustrative implementation scheme of the present application and its description are only used to explain the present application and are not intended to limit the present application.

[0029] See also Figure 1 As shown, Figure 1 A flow chart of a medical follow-up method provided in an embodiment of the present application. The method is executed based on a capability platform and includes: receiving input data from a patient end, the input data includes: form data, image data, and audio and video data, and converting and storing the input data into follow-up data; receiving a follow-up request from a doctor end, creating an audio and video communication with the patient end according to the follow-up request, and converting and storing the data in the audio and video communication into follow-up data; receiving a follow-up plan from a doctor end, the follow-up plan includes: a follow-up scale, a push time, and a push group, pushing the follow-up scale to the patient end of the push group according to the push time and the push group, and converting and storing the follow-up scale sent back by the patient end into follow-up data; performing early screening analysis based on the follow-up data by considering core factors, family genetic factors, intrinsic physical factors, and living environment factors; generating an early warning signal based on the results of the early screening analysis, and issuing an early warning to the doctor end and the patient end according to the built-in early warning mode.

[0030] The improvements of this application are that, first, various types of data generated during the entire follow-up process are collected, converted, and organized to generate follow-up data, so as to realize the automatic collection of follow-up data and avoid the loss of follow-up data; second, core factors, family genetic factors, intrinsic physical factors and living environment factors are considered to conduct early screening analysis of follow-up data, and early warning is issued for possible abnormal data, which is conducive to the rapid discovery of abnormal situations during patient follow-up.

[0031] In one possible implementation, early screening analysis is performed based on follow-up data taking into account core factors, family genetic factors, intrinsic physical factors and living environment factors; including: extracting answers to core questions, family genetic questions, intrinsic physical factors and living environment questions from the follow-up data and quantifying the answers into numerical values ​​to obtain quantified answer values; calculating weight values ​​for core questions, family genetic questions, intrinsic physical factors and living environment questions; obtaining an early screening warning coefficient based on the weighted summation of the quantified values ​​of the answers to core questions, family genetic questions, intrinsic physical factors and living environment questions and the corresponding weight values; comparing the early screening warning coefficient with a selected threshold to obtain an early screening analysis result.

[0032] Specifically, the early screening warning coefficient C is as follows:

[0033]

[0034] Among them, C is the early screening warning coefficient; i is the core problem number variable; n is the number of core problems; h i is the weight of each core question; f i (W i ) is the conversion function of the answer to the core question, and different conversion models can be selected according to the actual situation; j is the number of living environment problems; m is the number of living environment problems; g 1 is the weight value of living environment issues; σ is the dynamic adjustment factor, which is a constant; f j (W j ) is the conversion function of the quantitative value of the answers to the living environment questions; k is the variable of the number of internal physical problems; p is the number of internal physical problems; g 2 The weight of the internal constitution problem; f k (W k ) is the conversion function of the quantitative value of the answer to the internal constitution question; l is the variable of the number of family genetic problems; g 3 The weight of family genetic problems; f l (W l ) is the conversion function for the quantitative value of the answer to the family genetic question.

[0035] Among them, the answer quantization value conversion function f x (W x) is to convert the answers in the patient follow-up data into a unified numerical value that is easy to calculate. Different models can be selected for conversion according to the answers. For example, if the answer is a number, it is scaled and unified according to the proportion. If the answer is non-number, it is recognized by the natural language model and converted into a numerical value.

[0036] Furthermore, the weight values ​​of core issues, family genetic issues, intrinsic physical issues and living environment issues are calculated; including: obtaining follow-up data of patients and non-patients, extracting the quantitative values ​​of patients' answers and non-patient answers to various questions from the follow-up data; for each core issue, the weight value of the core issue is obtained by dividing the difference between the mean of the quantitative values ​​of all patients' answers to the core issue and the mean of the quantitative values ​​of all non-patients' answers to the core issue by the difference between the upper and lower limits of the quantitative values ​​of the answers to the core issue; for any of the family genetic issues, intrinsic physical issues and living environment issues , calculate the quantitative mean of each patient's answer to this question based on all the quantitative values ​​of each patient's answers to this question, calculate the quantitative mean of all patients' answers to this question based on the quantitative mean of each patient's answers to this question, and obtain the weight value of this question by dividing the difference between the quantitative mean of all patients' answers to this question and the quantitative mean of all non-sick people's answers to this question by the difference between the upper and lower limits of the quantitative values ​​of the answers to this question; calculate the dynamic adjustment factor based on the sum of the weight values ​​of all core issues, and adjust the weight values ​​of family genetic problems, intrinsic physical problems and living environment problems based on the dynamic adjustment factor.

[0037] Specifically, h i , g 1 , g 2 , g 3 These weights can be expressed by the formula Get. Calculate the weight value h of the i-th core question i , where u is the variable of patients; where U is the number of patients; s u is the quantitative value of the answers of the patients to the core question; v is the variable of the non-sick people; V is the number of non-sick people; s v The quantitative value of the answer to this core question for people without the disease; f max The upper limit of the quantitative value of the answer to this core question; f min The lower limit of the quantitative value of the answer to this core question, that is, the difference between the mean of the quantitative value of the answer to this core question by all patients and the mean of the quantitative value of the answer to this core question by all non-sick people divided by the difference between the upper and lower limits of the quantitative value of the answer to this core question.

[0038] When calculating the weights of the three types of questions, namely, family genetic problems, internal physical problems, and living environment problems, each type of question contains multiple questions. It is necessary to first calculate the quantitative mean of each patient's answer to a certain question, then calculate the quantitative mean of all patients' answers to that question, and similarly calculate the quantitative mean of all non-sick people's answers to that question. Finally, the difference between the quantitative mean of all patients' answers to that question and the quantitative mean of all non-sick people's answers to that question is divided by the difference between the upper and lower limits of the quantitative value of the answer to that question.

[0039] σ is the dynamic adjustment factor h i Assign a weight to each core question.

[0040] Furthermore, the early screening warning coefficient is compared with the selected threshold to obtain the early screening analysis result; including: calculating the early screening warning coefficients of all patients based on the follow-up data of all patients, and sorting the early screening warning coefficients of all patients according to the size of the coefficients; selecting the early screening warning coefficients in order from small to large, calculating the difference between the selected early screening warning coefficient and the average value of the early screening warning coefficients of all patients and the difference between the selected early screening warning coefficient and the adjacent larger early screening warning coefficient, until both differences are less than or equal to the set threshold, at which time the selected early screening warning coefficient is the selected threshold; comparing the early screening warning coefficient calculated from the follow-up data of patients who have not been determined to be ill with the selected threshold, when the early screening warning coefficient is greater than or equal to the selected threshold, it indicates that there is an abnormal risk, and when the early screening warning coefficient is less than the selected threshold, it indicates that there is no abnormal risk.

[0041] Specifically, the follow-up data of all patients are put into the calculation formula of early screening warning coefficient to obtain the distribution of early screening warning coefficients of patients, and the early screening warning coefficients are arranged from large to small, and the smallest early screening warning coefficient C is min Compared with the average value of the early screening warning coefficient and the next larger value, if the difference is too large, that is, greater than the set threshold, it means that C min It is accidental; select the next early screening warning coefficient and repeat the above comparison until both gaps are less than the set threshold. At this time, the selected threshold C is obtained. 阈值 Finally, the normal screening warning coefficient C and C 阈值 Comparison is performed to obtain the early screening analysis results. For follow-up data with an early screening warning coefficient greater than or equal to the selected threshold, it indicates that the patient to whom the follow-up data belongs has an abnormal risk and has a data tendency that is similar to the disease. At this time, the doctor or patient is reminded to conduct further screening to achieve early screening of the disease during the follow-up process.

[0042] In a possible implementation, it also includes: receiving an education plan from the doctor side, the education plan including: education content, push time and push group, and pushing the education content to the patient side of the corresponding push group according to the push time and push group.

[0043] It can be understood that the medical follow-up method provided in this application, on the one hand, realizes the standardized data collection of the entire follow-up process: the form data, image data and audio and video data in the follow-up process are automatically collected, and converted into standardized follow-up data for subsequent early disease screening; on the other hand, it realizes early disease screening: based on the collected follow-up data, early screening analysis is carried out by jointly considering core factors, family genetic factors, intrinsic physical factors and living environment factors, and early screening of the disease is carried out. The collected data covers the entire follow-up process to improve the accuracy and efficiency of early disease screening, can quickly identify potential disease risks, and provide doctors with more accurate diagnostic basis, which is conducive to early detection and treatment of the disease.

[0044] See also Figure 2 As shown, Figure 2 A structural diagram of a medical follow-up system provided for an embodiment of the present application. The system includes: a patient end, which is used for patients to enter data, and the entered data includes: form data, image data, and audio and video data; a doctor end, which is used to group patients, initiate follow-up requests, and create follow-up plans, and the follow-up requests include: video requests and audio requests, and the follow-up plans include: follow-up scales, push time, and push groups; a capability platform end, which is used to receive the entered data of the patient end and convert and store it into follow-up data, create audio and video communication with the patient end according to the follow-up request of the doctor end, and convert and store the data in the audio and video communication into follow-up data, push the follow-up scale to the patient end in the push group according to the follow-up plan of the doctor end, and convert and store the returned follow-up scale into follow-up data, and conduct early screening analysis based on the follow-up data considering core factors, family genetic factors, intrinsic physical factors, and living environment factors, generate early warning signals based on the results of the early screening analysis, and warn the doctor end and the patient end according to the built-in early warning mode; a management end, which is used to manage the data of the patient end and the doctor end by department.

[0045] In one possible implementation, the capability platform includes: an early warning answer quantification module, which is used to extract answers to core questions, family genetic questions, intrinsic physical problems and living environment questions from follow-up data and quantify the answers into numerical values ​​to obtain answer quantification values; an early warning weight calculation module, which is used to calculate the weight values ​​of core questions, family genetic questions, intrinsic physical problems and living environment questions; an early warning coefficient calculation module, which is used to obtain an early screening early warning coefficient based on the weighted sum of the quantified values ​​of the answers to core questions, family genetic questions, intrinsic physical problems and living environment questions and the corresponding weight values; an early screening early warning analysis module, which is used to compare the early screening early warning coefficient with a selected threshold to obtain an early screening analysis result.

[0046] Furthermore, the warning weight calculation module on the capability platform is specifically used to: obtain follow-up data of patients and non-patients, extract the quantitative values ​​of patients' answers and non-patient answers to various questions from the follow-up data; for each core question, divide the difference between the mean of the quantitative values ​​of all patients' answers to the core question and the mean of the quantitative values ​​of all non-patients' answers to the core question by the difference between the upper and lower limits of the quantitative values ​​of the answers to the core question; for any one of family genetic problems, intrinsic physical problems and living environment problems, calculate the quantitative mean of each patient's answer to the problem based on all the quantitative values ​​of each patient's answers to the problem, calculate the quantitative mean of all patients' answers to the problem based on the quantitative mean of each patient's answers to the problem, and obtain the weight of the problem based on the difference between the quantitative mean of all patients' answers to the problem and the quantitative mean of all non-patients' answers to the problem divided by the difference between the upper and lower limits of the quantitative values ​​of the answers to the problem; calculate the dynamic adjustment factor based on the sum of the weight values ​​of all core problems, and adjust the weight values ​​of family genetic problems, intrinsic physical problems and living environment problems based on the dynamic adjustment factor.

[0047] Furthermore, the early screening warning analysis module is specifically used to: calculate the early screening warning coefficients of all patients based on the follow-up data of all patients, and sort the early screening warning coefficients of all patients according to the size of the coefficients; select the early screening warning coefficients in order from small to large, calculate the difference between the selected early screening warning coefficient and the average value of the early screening warning coefficients of all patients and the difference between the selected early screening warning coefficient and the adjacent larger early screening warning coefficient, until both differences are less than or equal to the set threshold, at which time the selected early screening warning coefficient is the selected threshold; compare the early screening warning coefficient calculated from the follow-up data of patients who have not been determined to be ill with the selected threshold, when the early screening warning coefficient is greater than or equal to the selected threshold, it indicates that there is an abnormal risk, and when the early screening warning coefficient is less than the selected threshold, it indicates that there is no abnormal risk.

[0048] In a possible implementation, in the system: the doctor side is also used to create an education plan, which includes: education content, push time and push group; the capability platform side is also used to push the education content to the patient side of the corresponding push group according to the push time and push group.

[0049] It should be noted that the medical follow-up system provided in this application is used to implement the above-mentioned medical follow-up method and has corresponding technical effects, so it will not be elaborated on here.

[0050] See also Figure 3 , 4 As shown in Figures 5 and 5 , the present application also provides an exemplary medical follow-up system. Figure 3 This is the application architecture diagram of the medical follow-up system. Figure 4This is the business architecture diagram of the medical follow-up system. Figure 5 This is the technical architecture diagram of the medical follow-up system.

[0051] An exemplary medical follow-up system may include: patient side, doctor side, function access standard module, default component library, and data standard association model. Among them, the patient side is used for patients to enter data and interact with the doctor side, and the data format supports dynamic forms, pictures, audio and video; the doctor side is used to configure follow-up plans and the annotation, query and export of various types of data; the function access standard module is used to integrate various data collection and conversion components, and supports component expansion and dynamic configuration; the default component library includes dynamic forms, SMS, intelligent voice outbound calls, video communications, OCR, ASR, and realizes the basic capabilities of data collection and conversion; the data standard association model integrates data and realizes data processing and use by type, source, time, patient and other dimensions.

[0052] Among them, the patient side supports WAP, WeChat applet, and APP. The main functions include patient registration, hospital department switching, medical education query, message reminder, diary card, online message, audio and video follow-up, scale follow-up, report management, intelligent consultation, etc., and realizes the collection, aggregation and screening of diagnosis and treatment data through multiple channels. The doctor side supports PC web pages, WAP, WeChat applet, and APP. The main functions include hospital department management, patient management, patient message management, audio and video follow-up, dynamic scale configuration, scale follow-up plan, intelligent voice model configuration, and intelligent voice follow-up plan. Through the configuration of various models and plans, the collection and association of multi-channel follow-up data of patients can be realized. Function access standard module is used for the access and use of basic component capabilities. The default component library realizes the basic capabilities of data collection, conversion and transmission, including dynamic configuration form, dynamic configuration of collected data; SMS sending and receiving capabilities; intelligent voice outbound calling capabilities; real-time audio and video communication capabilities; image text recognition capabilities; and voice-to-text capabilities. The data standard association model integrates data and processes all component library data according to type, source, time, patient and other dimensions to achieve data integration and use. Through the association model, the collected data is summarized to achieve data initialization of the algorithm model, ultimately supporting early disease screening.

[0053] An exemplary medical follow-up system, in terms of business functions, may include: medical institution management, patient file management, AI voice outbound calling, follow-up plan definition, video follow-up, transcription of follow-up results, digital doctor intelligent Q&A, intelligent identification of examination results, education plan definition, statistical analysis, OCR recognition, microservice application, in-hospital system intercommunication, intelligent early warning and other functions.

[0054] In terms of medical institution management, it includes: department management and staff management; department management can use departments as independent management units to enter and manage basic information of each department; staff management can manage staff within medical institutions, and the department administrator is responsible for the staff configuration of the department. In terms of patient file management, it includes: patient file import, patient file collection, patient file management and patient grouping; patient file import can obtain basic patient data by importing excel spreadsheets; patient file collection can be completed by patients manually entering patient information, combined with OCR recognition capabilities, taking photos of patients' ID cards, report sheets, etc.; patient file management can manage patient files under the corresponding department, including patient diseases, diagnosis results, previous medical history and other data; patient grouping can support group management of patients according to different patient conditions, and match corresponding follow-up plans and education plans according to grouping conditions. In terms of AI voice outbound calls, it supports AI voice outbound call follow-up and SMS reminders, and long-term follow-up of patients, which can support private deployment. In terms of follow-up plan definition, it supports custom configuration of follow-up processes, including the configuration of follow-up stages, time frequency, follow-up methods, follow-up scales, etc. For patients in different groups, different follow-up processes and plans can be used to manage patients; it supports custom configuration of standard scales, and configures the name of the scale, title and other basic information; in each scale, it supports configuration of questions, such as single selection, multiple selection, uploading pictures, filling in text, etc. In terms of video follow-up, it supports video calls, and supports multi-terminal two-way video calls for consultation, follow-up, and consultation on PC, PAD, mobile phones, etc.; it supports video recording and transfer; through video calls, video consultation, follow-up and consultation between doctors and patients can be realized. In terms of follow-up result transcription, it supports video and voice follow-up text transcription, which can be intelligently filled in the follow-up table, and the follow-up results can be reviewed and listened to again, and the transcription results can be modified. In terms of intelligent question and answer of digital doctors, according to the needs of various departments of the hospital, virtual doctor digital people can be combined with AI voice, voice recognition, and speech library to conduct multiple rounds of dialogue and answer with patients, actively guide follow-up and consult pre-hospital and post-hospital service processes and problems. In terms of intelligent recognition of examination results, it supports the recognition of basic information of routine reports such as blood routine and color ultrasound, supports report data verification, and can verify whether the uploaded report is correct through the report name text on the report. In terms of the definition of education programs, it supports centralized management of education knowledge in pictures, texts, and videos, and can classify education knowledge accordingly, configure the name and knowledge content of the program, support custom configuration of knowledge push rules, support setting different push rules, such as setting the patient grouping, push time, push knowledge content, frequency, etc., support matching the corresponding education program with the medical record information of the questioner, and intelligently push relevant education knowledge. In terms of statistical analysis, statistical analysis of information can be completed based on the collected scales and patient data, such as the patient's scale filling data.In terms of OCR recognition, OCR recognition of standard documents such as ID cards is supported. In terms of microservice applications, unified call services can be provided to provide support for the addition of subsequent functions and modules and continuous upgrades and optimizations. In terms of intercommunication between hospital systems, data intercommunication capabilities with hospital systems such as HIS, EMR, etc. can be reserved. In terms of intelligent early warning, patient follow-up data can be collected through scales, voice, audio and video, OCR, etc., and the problem data can be divided into 4 categories through cleaning and conversion: core problems, family genetic problems, intrinsic physical problems and living environment problems. The early screening early warning coefficient is calculated according to the formula, and the early screening early warning coefficient is compared with the selected threshold to obtain the final early screening analysis results and generate early warning information; the generated early warning information can be provided to patients, patients' families, doctors, etc. through SMS, mini-program messages, PC-side messages, etc. Real-time alarm push; the early warning mode can be configured through SMS, system messages, etc.; it can support the configuration of early warning message templates; it supports doctors, patients, and family members to receive corresponding early warning prompts and view early warning details according to the configured early warning mode.

[0055] Regarding follow-up data collection, Figure 6 As shown, the system can realize doctor-patient interaction through the scale module (i.e., questionnaire filling form), and issue warnings for excessive data through the early warning module, and notify doctors and patients in time. It includes doctor module, scale module, follow-up module, project module, message notification module, patient module, statistical analysis module, and early warning module.

[0056] 1. After the doctor enters the system, he can configure dynamic scales of different types and styles through the scale module according to his own needs. The scale can be divided into two types: active push and non-active push. The active push requires the doctor to push it to the patient before the patient can fill in the scale. Non-active push means that when the doctor binds this patient, the patient can fill in the scale. The specific binding method is based on actual needs. The scale has basic functions such as single selection, multiple selection, date, single-line input, multi-line input, counter, etc. Doctors can combine these basic functions according to actual needs to form their own dynamic scale. It can save the complete page JSON string through the database to realize the scale function.

[0057] 2. After the doctor has built the scale, it will be pushed to the patient in an active push / inactive push manner according to different types. The scale of this application can be classified into questionnaire scales and project scales. For project scales, non-active push can be used. When the doctor adds a patient through the patient module and binds the patient, project and project scale through the project module, the patient can enter the system and fill in the scale in the patient center. For questionnaire scales, active push can be used. Different push rules can be configured through the push module to achieve regular push, which is convenient for long-term observation of the patient's condition. After the push, a message is sent to the patient through the message module to remind the patient to fill in the scale.

[0058] 3. In addition to doctors creating scales and pushing them to patients to fill in, patients can also fill in different types and styles of scales through the scale module according to their actual situation to record their own conditions, which will help doctors conduct data analysis. Doctors can also fill in some scales themselves through the scale module to record the basic conditions of patients.

[0059] 4. When submitting the scale, the warning module configures the warning rules and detects abnormalities, and sends warning messages to patients and doctors through the message notification module.

[0060] 5. The data collected by the scale module can be processed by the statistical analysis module, and the doctor can see the data intuitively.

[0061] Regarding the statistical analysis of follow-up data, Figure 7 As shown, doctor-patient interaction is achieved by combining voice function, video function and OCR function, and early warning rules are used to issue early warnings for some data exceeding the standard, so as to notify doctors and patients in time.

[0062] 1. The doctor establishes an outbound call model through the intelligent voice outbound call module and synchronizes it to the third-party voice function. The doctor determines the time and template for speaking to the patient through the voice outbound call module. The doctor then conducts voice follow-up with the patient through the voice function and records the follow-up data.

[0063] 2. The doctor uses the audio and video module to schedule a video consultation with the patient. When the consultation time comes, the doctor conducts a video consultation through the video function to understand and deal with the patient's problems in a timely manner and record follow-up data.

[0064] 3. Patients can upload examination reports through the picture module, and doctors can view them through the picture module and parse the content through the OCR function to record follow-up data.

[0065] 4. The follow-up data collected above are analyzed by the medical statistical analysis algorithm of the statistical analysis module. Medical statistical analysis algorithms play an important role in medical research and clinical practice, and can help doctors and researchers understand the distribution of diseases, risk factors, treatment effects, etc. The following methods can be combined for statistical analysis.

[0066] a. Descriptive statistical analysis: Descriptive statistical analysis is used to summarize and describe the basic characteristics of data, including mean, median, standard deviation, minimum, maximum, etc. That is, the average age, median age, standard deviation, minimum age, and maximum age of the patients can be obtained by their age. The corresponding values ​​can be obtained by the following algorithm:

[0067] Mean:

[0068]

[0069] Where n is the number of patients, x i is a sample of the ith value.

[0070] Median:

[0071] Sort the data from small to large and find the value in the middle. If the sample size n is an odd number, the median is the values; if n is even, the median is the The value and The average of the values

[0072] Standard Deviation:

[0073]

[0074] Minimum:

[0075]

[0076] Maximum:

[0077]

[0078] b. Hypothesis testing: Hypothesis testing is used to determine whether the difference between two or more samples is significant.

[0079] t-test: used to compare whether the means of two independent samples are significantly different.

[0080]

[0081] in, and is the mean of the two samples, s 1 and 2 is the standard deviation of the two samples, n 1 and n 1 Is the number of two samples. Used to compare the blood pressure difference between two groups of patients (treatment group and control group).

[0082] Chi-Square Test:

[0083] Used to test the association between two categorical variables

[0084]

[0085] Among them, O ij is the observed frequency, E ijis the expected frequency, r is the number of rows, and c is the number of columns. It is used to test the association between smoking and lung cancer, etc.

[0086] c. Regression analysis: Regression analysis is used to study the relationship between a dependent variable and one or more independent variables.

[0087] Linear Regression:

[0088] y=β 0 +β 1 x 1 +β 2 x 2 +…+β p x p +∈

[0089] Among them, y is the dependent variable, x 1 ,x 2 ,…,x p is the independent variable, β 0 ,β 1 ,…,β p is the regression coefficient, and ∈ is the error term. It is used to study the relationship between the patient's age, weight, and blood pressure.

[0090] Logistic Regression:

[0091]

[0092] Used for binary classification problems. Used to predict whether a patient has heart disease.

[0093] d. Survival analysis: Survival analysis is used to study the time to occurrence of events, such as the survival time of a patient.

[0094] Cox Proportional Hazards Model:

[0095]

[0096] The relationship between the patient's survival time and factors such as age, gender, and treatment regimen can be studied.

[0097] 5. Use the early warning rules of the early warning module to issue early warning notices for sensitive data. Early warning rules are very important in the medical field and can help doctors and medical institutions to promptly identify potential health problems of patients and take preventive measures. This purpose is achieved through the following common medical early warning rule algorithms.

[0098] a. Threshold-based warning rule: When a physiological indicator x exceeds the preset threshold T, a warning is triggered. For example, if the patient's blood pressure output is 140 / 90 mmHg, and the normal blood pressure value of the human body is less than 120 / 80 mmHg, the blood pressure exceeds the threshold and triggers a warning;

[0099]

[0100] b. Early warning based on statistics: Use statistical methods (such as standard deviation, mean, median, etc.) to detect abnormal values. For example, by calculating the mean and standard deviation of the patient's blood pressure and setting upper and lower thresholds, abnormal values ​​can be effectively detected and early warnings can be triggered. Assuming the mean is μ and the standard deviation is σ, set the upper threshold to μ+kσ and the lower threshold to μ-kσ, and the input value is x;

[0101]

[0102] c. Warning based on time series analysis: predict future values ​​through time series models (such as ARIMA), and trigger warnings when there is a large deviation between the predicted value and the actual value. is the predicted value, y t is the actual value, ∈ is the allowable error range;

[0103]

[0104] d. Early warning based on machine learning: Use machine learning models (such as logistic regression, random forest, neural network, etc.) to predict the probability of an event. Assume that P(y=1|x) is the probability of an event, and T is the threshold. x is the input feature vector, which contains multiple features;

[0105]

[0106] The above commonly used statistical analysis methods help doctors and medical institutions to discover potential health problems of patients and take preventive measures. In addition to the above commonly used statistical analysis methods, this application proposes early screening and warning of follow-up data.

[0107] Regarding follow-up data early screening warning, such as Figure 8 As shown, Figure 8 This is a flowchart for early disease screening. The overall process is to first collect patient follow-up data through scales, voice, audio and video, OCR, etc. Through data cleaning and conversion, the results of each question are converted into corresponding score values, and divided into 4 types of data (core problem data, environmental factor problem data, intrinsic factor problem data, family factor problem data), and then calculate the early screening warning coefficient C. Finally, the early screening warning coefficient C and the selected threshold C 阈值 By comparison, early screening analysis results can be obtained to achieve early screening of diseases during follow-up.

[0108] It is understandable that this exemplary medical follow-up system, 1. proposes a set of funnel screening models that go down layer by layer, from the collection of scales (text), to voice collection (voice-to-text analysis), to report collection (report recognition analysis), to video follow-up (equipment indicator access), and builds a complete data screening model to assist doctors in completing machine screening for early disease screening. Doctors can quickly complete data processing and analysis based on different indicators to obtain abnormal patient cases. 2. It integrates multiple technical capabilities such as OCR recognition, voice-to-text, and AI voice outbound calls to process unstructured data that occurs during follow-up, complete the collection process of scales, voices, and medical images, and form a set of standardized, structured, and normalized capabilities for follow-up data collection.

[0109] The above specific implementation methods further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A medical follow-up method, characterized in that: Execution based on capability platform, including: Receiving input data from the patient, the input data including: form data, picture data, and audio and video data, and converting and storing the input data into follow-up data; receiving a follow-up request from a doctor, establishing an audio and video communication with a patient according to the follow-up request, and converting and storing data in the audio and video communication into follow-up data; Receive a follow-up plan from the doctor, the follow-up plan including: a follow-up scale, a push time, and a push group, push the follow-up scale to the patient end of the push group according to the push time and the push group, and convert and store the follow-up scale sent back by the patient end into follow-up data; Conduct early screening analysis based on the follow-up data, taking into account core factors, family genetic factors, intrinsic physical factors and living environment factors; Generate early warning signals based on early screening analysis results, and issue warnings to doctors and patients based on built-in warning modes.

2. A medical follow-up method according to claim 1, characterized in that: Early screening analysis is performed based on the follow-up data, taking into account core factors, family genetic factors, intrinsic physical factors and living environment factors; including: Extracting answers to core questions, family genetic questions, intrinsic physical questions and living environment questions from the follow-up data and quantifying the answers into numerical values ​​to obtain answer quantification values; Calculate the weights of core issues, family genetic issues, inherent physical issues, and living environment issues; The early screening warning coefficient is obtained by weighted summing the quantitative values ​​of the answers to the core questions, family genetic questions, internal physical problems, and living environment questions and the corresponding weight values; The early screening warning coefficient is compared with the selected threshold value to obtain the early screening analysis result.

3. A medical follow-up method according to claim 2, characterized in that: Calculate the weights of core issues, family genetic issues, inherent physical issues, and living environment issues; including: Obtain follow-up data of patients and non-patients, and extract quantitative values ​​of answers of patients and non-patients for various questions from the follow-up data; For each core question, the weight value of the core question is obtained by dividing the difference between the mean of the quantitative values ​​of the answers of all patients to the core question and the mean of the quantitative values ​​of the answers of all non-patients to the core question by the difference between the upper and lower limits of the quantitative values ​​of the answers to the core question; For any of the family genetic problems, internal physical problems and living environment problems, the quantitative mean of each patient's answer to the problem is calculated based on all the quantitative values ​​of each patient's answer to the problem, and the quantitative mean of all patients' answers to the problem is calculated based on the quantitative mean of each patient's answer to the problem. The weight value of the problem is obtained by dividing the difference between the quantitative mean of all patients' answers to the problem and the quantitative mean of all non-sick people's answers to the problem by the difference between the upper and lower limits of the quantitative value of the answer to the problem. The dynamic adjustment factor is calculated based on the sum of the weight values ​​of all core issues, and the weight values ​​of the family genetic problems, inherent physical problems and living environment problems are adjusted based on the dynamic adjustment factor.

4. A medical follow-up method according to claim 3, characterized in that: Comparing the early screening warning coefficient with the selected threshold value to obtain the early screening analysis result; including: Calculate the early screening warning coefficients of all patients based on the follow-up data of all patients, and sort the early screening warning coefficients of all patients by coefficient size; Select the early screening warning coefficient in order from small to large, calculate the difference between the selected early screening warning coefficient and the average value of the early screening warning coefficients of all patients, and the difference between the selected early screening warning coefficient and the adjacent larger early screening warning coefficient, until both differences are less than or equal to the set threshold, at which time the selected early screening warning coefficient is the selected threshold; The early screening warning coefficient calculated from the follow-up data of patients with undetermined illness is compared with the selected threshold. When the early screening warning coefficient is greater than or equal to the selected threshold, it indicates the presence of abnormal risk; when the early screening warning coefficient is less than the selected threshold, it indicates the absence of abnormal risk.

5. A medical follow-up method according to claim 1, characterized in that: Also includes: Receive the doctor's education plan, the education plan includes: education content, push time and push group, and push the education content to the patient end of the corresponding push group according to the push time and push group.

6. A medical follow-up system, characterized in that: include: The patient end is used for the patient to enter data, and the entered data includes: form data, picture data, and audio and video data; The doctor side is used to group patients, initiate follow-up requests and create follow-up plans. The follow-up requests include video requests and audio requests. The follow-up plans include follow-up scales, push time and push grouping. The capability platform side is used to receive the input data from the patient side and convert and store it into follow-up data, create audio and video communication with the patient side according to the follow-up request from the doctor side and convert and store the data in the audio and video communication into follow-up data, push the follow-up scale to the patient side in the push group according to the follow-up plan of the doctor side and convert and store the returned follow-up scale into follow-up data, conduct early screening analysis based on the follow-up data considering core factors, family genetic factors, intrinsic physical factors and living environment factors, generate early warning signals based on the early screening analysis results, and issue warnings to the doctor side and the patient side according to the built-in warning mode; The management side is used to manage patient and doctor data by department.

7. A medical follow-up system according to claim 6, characterized in that: The capability platform includes: The early warning answer quantification module is used to extract answers to core questions, family genetic questions, intrinsic physical questions and living environment questions from the follow-up data, quantify the answers into numerical values, and obtain answer quantification values; The early warning weight calculation module is used to calculate the weight values ​​of core problems, family genetic problems, inherent physical problems and living environment problems; The early warning coefficient calculation module is used to obtain the early screening early warning coefficient by weighted summing up the quantitative values ​​of the answers to the core questions, family genetic questions, intrinsic physical problems, and living environment questions and the corresponding weight values; The early screening warning analysis module is used to compare the early screening warning coefficient with the selected threshold value to obtain the early screening analysis result.

8. A medical follow-up system according to claim 7, characterized in that: The warning weight calculation module on the capability platform is specifically used for: Obtain follow-up data of patients and non-patients, and extract quantitative values ​​of answers of patients and non-patients for various questions from the follow-up data; For each core question, the weight value of the core question is obtained by dividing the difference between the mean of the quantitative values ​​of the answers of all patients to the core question and the mean of the quantitative values ​​of the answers of all non-patients to the core question by the difference between the upper and lower limits of the quantitative values ​​of the answers to the core question; For any of the family genetic problems, internal physical problems and living environment problems, the quantitative mean of each patient's answer to the problem is calculated based on all the quantitative values ​​of each patient's answer to the problem, and the quantitative mean of all patients' answers to the problem is calculated based on the quantitative mean of each patient's answer to the problem. The weight value of the problem is obtained by dividing the difference between the quantitative mean of all patients' answers to the problem and the quantitative mean of all non-sick people's answers to the problem by the difference between the upper and lower limits of the quantitative value of the answer to the problem. The dynamic adjustment factor is calculated based on the sum of the weight values ​​of all core issues, and the weight values ​​of the family genetic problems, inherent physical problems and living environment problems are adjusted based on the dynamic adjustment factor.

9. A medical follow-up system according to claim 8, characterized in that: The early screening warning analysis module is specifically used for: Calculate the early screening warning coefficients of all patients based on the follow-up data of all patients, and sort the early screening warning coefficients of all patients by coefficient size; Select the early screening warning coefficient in order from small to large, calculate the difference between the selected early screening warning coefficient and the average value of the early screening warning coefficients of all patients, and the difference between the selected early screening warning coefficient and the adjacent larger early screening warning coefficient, until both differences are less than or equal to the set threshold, at which time the selected early screening warning coefficient is the selected threshold; The early screening warning coefficient calculated from the follow-up data of patients with undetermined illness is compared with the selected threshold. When the early screening warning coefficient is greater than or equal to the selected threshold, it indicates the presence of abnormal risk; when the early screening warning coefficient is less than the selected threshold, it indicates the absence of abnormal risk.

10. A medical follow-up system according to claim 6, characterized in that: In the system: The doctor end is also used to create an education plan, which includes: education content, push time and push group; The capability platform end is also used to push the education content to the patient end of the corresponding push group according to the education push time and education object.

Images