CN114942947A

CN114942947A - Follow-up visit data processing method and system based on intelligent medical treatment

Info

Publication number: CN114942947A
Application number: CN202210512794.4A
Authority: CN
Inventors: 袁骏毅; 彭红; 侯晋; 俞惠丽; 宓林晖; 岑星星; 陈烨; 王毅豪
Original assignee: Shanghai Chest Hospital
Current assignee: Shanghai Chest Hospital
Priority date: 2022-07-04
Filing date: 2022-07-04
Publication date: 2022-08-26

Abstract

The embodiment of the invention discloses a follow-up visit data processing method and system based on intelligent medical treatment, wherein a medical tracking behavior sequence generated by behavior tracking of a follow-up visit patient to be subjected to follow-up visit by a plurality of intelligent medical service systems is obtained by combining a pre-configured follow-up visit strategy; generating medical interest point distribution of a target patient of a follow-up patient to be reviewed by combining the medical tracking behaviors covered by the tracked medical tracking behavior sequences; and determining one or more associated medical behaviors of the associated follow-up patients corresponding to the follow-up patients to be reviewed by combining the target patient medical interest point distribution, and analyzing whether a supplementary follow-up procedure needs to be performed on the follow-up patients to be reviewed by combining the one or more associated medical behaviors and the medical tracking behaviors covered by the plurality of medical tracking behavior sequences, wherein the one or more associated medical behaviors correspond to a plurality of different behavior dimensions of the associated follow-up patients to be reviewed respectively. Therefore, the initiation efficiency of the follow-up visit flow of the supplementary consultation can be improved.

Description

Follow-up visit data processing method and system based on intelligent medical treatment

Technical Field

The invention relates to the technical field of intelligent medical treatment, in particular to a follow-up visit data processing method and system based on intelligent medical treatment.

Background

The follow-up visit of medical treatment refers to an observation method for the patients who have been seen a doctor in the hospital to know the disease changes of the patients and guide the recovery of the patients regularly by communication or other means. On one hand, the follow-up visit can facilitate the long-term tracking observation of doctors aiming at postoperative patients, special disease groups, chronic diseases and the like, so that the doctors can know the disease changes of the patients in time and give treatment suggestions, and the patients with relapse and deterioration of the disease can be scheduled to be hospitalized again. On the other hand, effective execution of follow-up visits can improve the pre-medical and post-medical service level of a hospital, so that the hospital can master the first-hand patient data to carry out statistical analysis and experience accumulation, development of medical scientific research work and improvement of the service level of medical workers are facilitated, and the patient can be better served. For patients, follow-up may also help patients to recover better from the disease, so that patients may more clearly master their own treatment effect and stage outcome. Thus, medical follow-up is of paramount importance to both hospitals and patients.

However, in the related art, the follow-up visit suggestion is usually a preliminary suggestion given by a doctor based on own personal experience and the current state condition of a patient, in an actual scene, the patient may be outside the doctor and various behavior operations for other online intelligent medical service systems may exist, and further analysis for the medical tracking behaviors is lacking in the related art, so that a supplementary follow-up visit flow beyond the preliminary suggestion given by the doctor based on own personal experience and the current state condition of the patient is lacking, which causes an increase in communication cost between the doctor and the patient in a subsequent follow-up visit process, that is, in the related art, the efficiency of improving the initiation of the supplementary follow-up visit flow is low.

Disclosure of Invention

In view of this, the embodiment of the present invention provides a follow-up visit data processing method and system based on intelligent medical treatment.

In a first aspect, an embodiment of the present invention provides a follow-up visit data processing method based on smart medical treatment, applied to a follow-up visit data processing system based on smart medical treatment, including:

acquiring a medical tracking behavior sequence generated by behavior tracking of a plurality of intelligent medical service systems to-be-reviewed follow-up patients by combining a pre-configured follow-up strategy, wherein the plurality of intelligent medical service systems are respectively configured in page partitions with different behavior dimensions of a medical tracking process, the medical tracking process is used for performing medical behavior tracking on the to-be-reviewed follow-up patients with the end of a preliminary visit behavior, and the medical tracking behavior sequence comprises two or more medical tracking behaviors;

generating medical interest point distribution of the target patient of the follow-up patient to be reviewed by combining medical tracking behaviors covered by a plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems;

determining one or more associated medical behaviors of an associated follow-up patient corresponding to the follow-up patient to be re-diagnosed in combination with the target patient medical interest point distribution, and analyzing whether a supplementary follow-up procedure for the follow-up patient to be re-diagnosed is needed or not in combination with the one or more associated medical behaviors and the medical tracking behaviors covered by the plurality of medical tracking behavior sequences, wherein the one or more associated medical behaviors respectively correspond to a plurality of different behavior dimensions of the associated follow-up patient to be re-diagnosed;

and when the fact that a supplementary consultation follow-up procedure needs to be carried out on the follow-up patient to be subjected to the follow-up consultation is judged, generating reference prompt information based on the consultation follow-up plan corresponding to the past of the associated follow-up patient to be subjected to the follow-up consultation, and loading the reference prompt information into a medical file database of the follow-up patient to be subjected to the follow-up consultation.

In a possible implementation manner of the first aspect, the acquiring, by combining with a pre-configured follow-up visit policy, a medical tracking behavior sequence generated by the plurality of intelligent medical service systems performing behavior tracking on a to-be-follow-up visit patient specifically includes the following steps:

analyzing whether medical behavior tracking needs to be carried out on the follow-up patient to be re-diagnosed, and generating a corresponding medical behavior tracking instruction when the follow-up patient to be re-diagnosed needs to be subjected to medical behavior tracking;

the medical behavior tracking instruction is issued to each of the plurality of intelligent medical service systems, and after receiving the medical behavior tracking instruction, each of the plurality of intelligent medical service systems performs medical behavior tracking on the follow-up patient to be subjected to the follow-up examination, determines a corresponding medical tracking behavior sequence, and uploads the medical tracking behavior sequence to the follow-up examination follow-up data processing system based on the intelligent medical treatment;

and acquiring the medical tracking behavior sequence uploaded by each of the intelligent medical service systems in combination with the medical behavior tracking instruction.

In a possible implementation manner of the first aspect, the analyzing whether medical behavior tracking needs to be performed on the follow-up patient to be reviewed, and generating a corresponding medical behavior tracking instruction when it is determined that medical behavior tracking needs to be performed on the follow-up patient to be reviewed specifically includes the following steps:

analyzing whether first tracking trigger information sent by a related target medical behavior tracking subprogram is received, wherein the target medical behavior tracking subprogram is configured in the medical tracking process and used for generating the first tracking trigger information when the condition that the follow-up patient to be subjected to the follow-up diagnosis meets a preset behavior trigger condition is tracked;

if the first tracking trigger information sent by the target medical behavior tracking subprogram is received, judging that medical behavior tracking needs to be carried out on the follow-up patient to be reviewed, and if the first tracking trigger information sent by the target medical behavior tracking subprogram is not received, judging that medical behavior tracking does not need to be carried out on the follow-up patient to be reviewed;

and when the follow-up patient to be subjected to the follow-up examination needs to be subjected to medical behavior tracking, generating a medical behavior tracking instruction for performing medical behavior tracking on the follow-up patient to be subjected to the follow-up examination.

In a possible implementation manner of the first aspect, when it is determined that medical behavior tracking needs to be performed on the follow-up patient to be reviewed, the generating of the medical behavior tracking instruction for performing medical behavior tracking on the follow-up patient to be reviewed specifically includes the following steps:

for each intelligent medical service system in the plurality of intelligent medical service systems, determining page partition service node information of a page partition corresponding to the intelligent medical service system, and determining target behavior tracking quantity corresponding to the intelligent medical service system by combining the page partition service node information, wherein mapping relation exists between the page partition service node information and the target behavior tracking quantity;

for each of the plurality of intelligent medical service systems, combining the target behavior tracking number corresponding to the intelligent medical service system, generating a medical behavior tracking instruction corresponding to the intelligent medical service system and used for tracking the medical behavior of the follow-up patient to be reviewed, wherein the medical behavior tracking instruction carries the corresponding target behavior tracking number, the target behavior tracking number represents that the corresponding intelligent medical service system tracks the medical behavior of the follow-up patient to be reviewed, and a corresponding medical tracking behavior sequence is determined, and the behavior tracking number of the medical tracking behavior covered by the medical tracking behavior sequence is the corresponding target behavior tracking number.

In a possible implementation manner of the first aspect, the generating a target patient medical interest point distribution of the follow-up patient to be reviewed in conjunction with the medical tracking behavior covered by the plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems specifically includes the following steps:

for each medical tracking behavior sequence in a plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems, frequently screening the medical tracking behaviors covered by the medical tracking behavior sequence, and determining a medical tracking frequent behavior space corresponding to the medical tracking behavior sequence;

for each medical tracking behavior sequence in the multiple medical tracking behavior sequences tracked by the multiple intelligent medical service systems, performing patient interest point mining on a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence, and outputting target patient medical interest point information of the follow-up patient to be re-diagnosed on a behavior dimension corresponding to the medical tracking behavior sequence;

and generating the medical interest point distribution of the target patient of the follow-up patient to be reviewed by combining the medical interest point information of the target patient corresponding to each medical tracking behavior sequence in the plurality of medical tracking behavior sequences.

In a possible implementation manner of the first aspect, for each medical tracking behavior sequence in the plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems, frequent item screening is performed on the medical tracking behaviors covered by the medical tracking behavior sequence, and a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence is determined, which specifically includes the following steps:

for each medical tracking behavior sequence in the plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems, respectively performing frequent item support calculation on each medical tracking behavior covered by the medical tracking behavior sequence, and determining a behavior frequent item support corresponding to each medical tracking behavior;

for each medical tracking behavior sequence in the plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems, in the behavior frequent item support degrees corresponding to each medical tracking behavior covered by the medical tracking behavior sequence, generating a behavior frequent item support degree larger than a preset support degree as a target behavior frequent item support degree corresponding to the medical tracking behavior sequence, and outputting the medical tracking behavior corresponding to the target behavior frequent item support degree as a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence.

In a possible implementation manner of the first aspect, the determining, in combination with the target patient medical interest point distribution, one or more associated medical behaviors of an associated follow-up patient corresponding to the follow-up patient to be reviewed, and analyzing, in combination with the one or more associated medical behaviors and the medical tracking behaviors covered by the plurality of medical tracking behavior sequences, whether a supplementary follow-up procedure for the follow-up patient to be reviewed is required, specifically includes the following steps:

for each historical patient medical interest point distribution in a plurality of historical patient medical interest point distributions recorded in advance, calculating the coincidence rate between the historical patient medical interest point distribution and the target patient medical interest point distribution, and determining the coincidence rate of interest points corresponding to the historical patient medical interest point distribution;

generating an interest point coincidence rate larger than a preset coincidence rate as a target interest point coincidence rate in an interest point coincidence rate corresponding to each historical patient medical interest point distribution in the plurality of historical patient medical interest point distributions, determining the historical patient medical interest point distribution corresponding to the target interest point coincidence rate as an associated patient medical interest point distribution, and acquiring one or more associated medical behaviors corresponding to associated follow-up patients corresponding to the associated patient medical interest point distribution, wherein each associated medical behavior has corresponding behavior tracking and labeling information which expresses a behavior tracking dimension of the associated medical behavior;

for each medical tracking behavior sequence in the plurality of medical tracking behavior sequences, performing frequent item screening on the medical tracking behaviors covered by the medical tracking behavior sequence in the medical tracking behavior sequence, determining a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence, and determining a corresponding medical behavior tracking labeling information and a related medical behavior corresponding to the page service dimension in the one or more related medical behaviors by combining the page service dimension of a page partition corresponding to the medical tracking behavior sequence, wherein the corresponding medical tracking labeling information and the related medical behavior are used as a target related medical behavior space corresponding to the medical tracking behavior sequence;

for each medical tracking behavior sequence, calculating the coincidence rate between the medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence and the target associated medical behavior space corresponding to the medical tracking behavior sequence, and determining the behavior tracking coincidence rate corresponding to the medical tracking behavior sequence;

and analyzing whether a supplementary follow-up visit flow needs to be carried out on the follow-up patient to be subjected to the follow-up visit or not by combining the behavior tracking coincidence rate corresponding to each medical tracking behavior sequence.

In a possible implementation manner of the first aspect, for each medical tracking behavior sequence, calculating a coincidence rate between a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence and a target associated medical behavior space corresponding to the medical tracking behavior sequence, and determining a behavior tracking coincidence rate corresponding to the medical tracking behavior sequence specifically includes the following steps:

respectively determining the behavior tag attribute of each medical behavior member in the target associated medical behavior space, clustering the target associated medical behavior space by combining the behavior tag attribute of each medical behavior member, determining at least one corresponding associated medical behavior cluster, matching the behavior tag attributes of each medical behavior member corresponding to the same associated medical behavior cluster, wherein at least one medical behavior communication link is arranged between each medical behavior member of the same associated medical behavior cluster, and each medical behavior member covered by the medical behavior communication link is the same as the associated medical behavior cluster to which the corresponding medical behavior member belongs;

for each associated medical behavior cluster in the at least one associated medical behavior cluster, determining a cluster core behavior member of the associated medical behavior cluster, respectively calculating a first intra-cluster difference between each medical behavior member in the associated medical behavior cluster and the cluster core behavior member, and respectively combining the first intra-cluster difference between each medical behavior member in the associated medical behavior cluster and the cluster core behavior member corresponding to the associated medical behavior cluster to generate a first intra-cluster importance evaluation parameter corresponding to each medical behavior member, wherein a mapping relation exists between the first intra-cluster importance evaluation parameter and the first intra-cluster difference;

for each associated medical behavior cluster in the at least one associated medical behavior cluster, performing fusion calculation on the medical behavior engagement degrees of each medical behavior member in the associated medical behavior cluster by combining with the first intra-cluster importance evaluation parameter corresponding to each medical behavior member in the associated medical behavior cluster, determining the corresponding first importance fusion engagement degree, performing order sorting on the first importance fusion engagement degrees corresponding to each associated medical behavior cluster by combining with the spatial domain node of each associated medical behavior cluster in the target associated medical behavior space, and determining the first medical behavior engagement degree order information corresponding to the target associated medical behavior space;

combining the spatial domain nodes of each associated medical action cluster in the target associated medical action space, clustering the medical tracking frequent item behavior space, determining at least one medical tracking frequent item behavior cluster corresponding to the medical tracking frequent item behavior space, and for each medical frequent item behavior cluster, determining a cluster core behavior member of the medical frequent item behavior cluster, and respectively calculating the discrimination degrees in a second cluster between each medical behavior member in the medical frequent item behavior cluster and the cluster core behavior member, and generating second intra-cluster importance evaluation parameters corresponding to the medical behavior members in the medical frequent item behavior cluster by respectively combining the second intra-cluster discriminative power between the medical behavior members in the medical frequent item behavior cluster and the corresponding cluster core behavior members, a mapping relation exists between the second intra-cluster importance evaluation parameter and the second intra-cluster discrimination;

for each medical frequent item behavior cluster in the at least one medical frequent item behavior cluster, performing fusion calculation on the medical behavior engagement of each medical behavior member in the medical frequent item behavior cluster in combination with a second intra-cluster importance evaluation parameter corresponding to each medical behavior member in the medical frequent item behavior cluster, determining a second importance fusion engagement corresponding to the medical frequent item behavior cluster, performing order sorting on the second importance fusion engagement corresponding to each medical frequent item behavior cluster in combination with a spatial domain node of each medical frequent item behavior cluster in the medical tracking frequent item behavior space, and determining second medical behavior engagement order information corresponding to the medical tracking frequent item behavior space;

and calculating the association degree between the first medical behavior engagement degree sequence information and the second medical behavior engagement degree sequence information, and determining the coincidence rate between the medical tracking frequent item behavior space and the target associated medical behavior space as the behavior tracking coincidence rate corresponding to the corresponding medical tracking behavior sequence.

In a possible implementation manner of the first aspect, the step of mining the medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence to output the target patient medical interest point information of the follow-up patient to be reviewed on the behavior dimension corresponding to the medical tracking behavior sequence includes:

performing behavior tendency variable extraction on the medical tracking frequent item behavior space data corresponding to the medical tracking behavior sequence to determine a first behavior tendency variable map;

determining each positive behavior tendency point and each negative behavior tendency point in the medical tracking frequent item behavior space data by combining a first behavior tendency variable map;

determining each positive behavior tendency point and each first behavior tendency quantum map corresponding to each negative behavior tendency point in the first behavior tendency variable map by combining each positive behavior tendency point and each associated behavior knowledge point of each negative behavior tendency point in the medical tracking frequent item behavior space data;

respectively carrying out mining factor optimization on the first behavior tendency variable quantum map corresponding to each positive behavior tendency point and each negative behavior tendency point, and determining each second behavior tendency variable quantum map after the mining factor optimization of the first behavior tendency variable;

performing interest point mapping variable mining by combining the second behavior tendency variable quantum maps, and determining each target interest point mapping variable in the medical tracking frequent behavior space data;

determining each first behavior tendency variable sub-graph corresponding to each target interest point mapping variable in a first behavior tendency variable graph by combining the associated behavior knowledge points of each target interest point mapping variable in the medical tracking frequent behavior space data, and aggregating each first behavior tendency variable sub-graph corresponding to all target interest point mapping variables to generate a third behavior tendency variable graph;

extracting attention variables from the first action tendency variable map, and determining a fourth action tendency variable map; wherein, the map expression mode of the fourth line trend variable map is the same as that of the third line trend variable map;

aggregating the fourth behavior tendency variable map and the third behavior tendency variable map to generate a fifth behavior tendency variable map;

and (4) mining medical interest points by combining a fifth behavior tendency variable map, and outputting the medical interest point information of the target patient of the follow-up patient to be diagnosed on the behavior dimension corresponding to the medical tracking behavior sequence.

For example, in one possible implementation manner of the first aspect, the performing behavior trend variable extraction on the medical tracking frequent item behavior space data to determine a first behavior trend variable map includes:

transmitting the medical tracking frequent item behavior space data to a behavior tendency variable extraction branch of a target medical interest point mining model for performing behavior tendency variable extraction;

determining each positive behavior-trend point and each negative behavior-trend point in the medical tracking frequent item behavior-space data in combination with a first behavior-trend variable map, comprising:

transmitting the first behavior tendency variable map to a behavior tendency point extraction branch of the target medical interest point mining model to extract each positive behavior tendency point and each negative behavior tendency point in the medical tracking frequent item behavior space data;

and mining the mapping variables of the interest points by combining the second behavior tendency variable quantum maps, wherein the mining comprises the following steps:

transmitting each second behavior tendency variable quantum map to an interest point mapping variable extraction branch of the target medical interest point mining model to perform interest point mapping variable mining;

the medical interest point mining is carried out by combining the fifth behavior tendency variable map, and comprises the following steps:

and transmitting the fifth behavior tendency variable map to a medical interest point mining branch of the target medical interest point mining model for medical interest point mining.

For instance, in one possible implementation of the first aspect, the method further comprises:

and combining the calibrated medical tracking behavior space sample data sequence to carry out parameter layer optimization and selection on the calibrated medical interest point mining model and generate a converged target medical interest point mining model, wherein the steps of the method specifically comprise the following steps:

respectively extracting one piece of calibrated medical tracking behavior space sample data from the calibrated medical tracking behavior space sample data sequence, transmitting the calibrated medical tracking behavior space sample data to a behavior tendency variable extraction branch of the target medical interest point mining model, and extracting a behavior tendency variable to determine a first behavior tendency variable map of the transmitted calibrated medical tracking behavior space sample data;

transmitting the first behavior tendency variable map to a behavior tendency point extraction branch of the target medical interest point mining model, and determining each positive behavior tendency point and each negative behavior tendency point in transmitted calibrated medical tracking behavior space sample data;

determining each positive behavior tendency point and each first behavior tendency quantum map corresponding to each negative behavior tendency point in the first behavior tendency variable map by combining each positive behavior tendency point and each associated behavior knowledge point of each negative behavior tendency point in the transmitted sample data of the calibrated medical tracking behavior space;

respectively carrying out mining factor optimization on the first behavior tendency variable quantum maps corresponding to each positive behavior tendency point and each negative behavior tendency point, and determining each second behavior tendency variable quantum map after the mining factor optimization of the first behavior tendency variable;

transmitting each second behavior tendency variable quantum map to an interest point mapping variable extraction branch of the target medical interest point mining model, and determining each intermediate interest point mapping variable and intermediate interest point attribute of each intermediate interest point mapping variable in transmitted calibrated medical tracking behavior space sample data;

determining a mining loss value by combining each intermediate interest point mapping variable output by the interest point mapping variable extraction branch, each calibrated interest point mapping variable carried in the transmitted calibrated medical tracking behavior space sample data and the corresponding interest point attribute, and a pre-bound first medical interest point mining loss function, so as to determine a first medical interest point mining loss value;

determining each first behavior tendency variable sub-graph spectrum corresponding to each intermediate interest point mapping variable in the first behavior tendency variable graph spectrum by combining the associated behavior knowledge points of each intermediate interest point mapping variable in the transmitted calibrated medical tracking behavior space sample data, and aggregating each first behavior tendency variable sub-graph spectrum corresponding to all intermediate interest point mapping variables to generate a third behavior tendency variable graph spectrum;

transmitting a fifth behavior tendency variable map to a medical interest point mining branch of the target medical interest point mining model, and determining each intermediate interest point attribute contained in the transmitted calibrated medical tracking behavior space sample data;

determining a mining loss value by combining the attributes of each intermediate interest point contained in the transmitted calibrated medical tracking behavior space sample data generated by the medical interest point mining branch and the attributes of each calibrated interest point mapping variable carried in the transmitted calibrated medical tracking behavior space sample data and combining a pre-bound second medical interest point mining loss function, and determining a second medical interest point mining loss value;

weighting the first medical interest point mining loss value and the second medical interest point mining loss value, and carrying out model parameter layer optimization and selection on the target medical interest point mining model by combining the weighted interest point mining loss value;

and if the model parameter layer of the target medical interest point mining model is converged, outputting the corresponding target medical interest point mining model.

For example, in a possible implementation manner of the first aspect, after aggregating the first behavior tendency variable sub-graph spectrums corresponding to all the target interest point mapping variables to output the third behavior tendency variable graph, before the step of aggregating the fourth behavior tendency variable graph and the third behavior tendency variable graph, the method further includes:

performing behavior tendency variable expansion on the third behavior tendency variable map, determining an expanded behavior tendency variable of each behavior tendency variable in the third behavior tendency variable map, performing association fusion on each behavior tendency variable in the third behavior tendency variable map and the expanded behavior tendency variable thereof respectively, and determining an expanded behavior tendency variable map of the third behavior tendency variable map;

the aggregating the fourth behavior tendency variable map with the third behavior tendency variable map comprises:

and aggregating the fourth behavior tendency variable map with the expanded behavior tendency variable map of the third behavior tendency variable map.

For example, in one possible implementation of the first aspect, the method further comprises:

adding interest point influence variables of each medical interest point target in sample data of each calibrated medical tracking behavior space;

after determining each positive behavior tendency point and each first behavior tendency variation sub-map corresponding to each negative behavior tendency point in the first behavior tendency variable map and before weighting the first medical interest point mining loss value and the second medical interest point mining loss value, the method further includes:

respectively carrying out mining factor optimization on the second behavior tendency variables on each positive behavior tendency point and each first behavior tendency variable quantum map corresponding to each negative behavior tendency point, and determining each sixth behavior tendency variable quantum map after the mining factor optimization of the second behavior tendency variables;

transmitting each sixth behavior tendency variable quantum map to an interest point influence variable extraction branch of the target medical interest point mining model, and determining interest point influence variables and intermediate interest point attributes of each medical interest point target in transmitted calibrated medical tracking behavior space sample data;

determining mining loss values by combining the interest point influence variables and the intermediate interest point attributes of the medical interest point targets in the transmitted calibrated medical tracking behavior space sample data obtained by the interest point influence variable extraction branch and the interest point influence variables and the interest point attributes of the medical interest point targets carried in the transmitted calibrated medical tracking behavior space sample data, and determining the mining loss values of the third medical interest points by combining a pre-bound third medical interest point mining loss function;

weighting the first medical point of interest mining loss value and the second medical point of interest mining loss value comprises:

weighting the first medical point of interest mining loss value, the second medical point of interest mining loss value, and the third medical point of interest mining loss value.

In a second aspect, an embodiment of the present invention provides a follow-up visit data processing system based on smart medical treatment, including:

a processor;

a memory in which a computer program is stored, wherein the computer program realizes the return visit data processing method based on intelligent medical treatment according to the first aspect when executed.

As described above, in the embodiment of the present invention, a pre-configured follow-up visit policy is combined to obtain a medical tracking behavior sequence generated by behavior tracking of a plurality of intelligent medical service systems for a follow-up visit patient to be treated; generating medical interest point distribution of a target patient of a follow-up patient to be reviewed by combining the medical tracking behaviors covered by the tracked medical tracking behavior sequences; and determining one or more associated medical behaviors of the associated follow-up patients corresponding to the follow-up patients to be reviewed by combining the target patient medical interest point distribution, and analyzing whether a supplementary follow-up procedure needs to be performed on the follow-up patients to be reviewed by combining the one or more associated medical behaviors and the medical tracking behaviors covered by the plurality of medical tracking behavior sequences, wherein the one or more associated medical behaviors correspond to a plurality of different behavior dimensions of the associated follow-up patients to be reviewed respectively. Therefore, the initiation efficiency of the supplementary consultation follow-up procedure can be improved.

Drawings

Fig. 1 is a schematic flowchart illustrating steps of a follow-up visit data processing method based on intelligent medical treatment according to an embodiment of the present invention;

fig. 2 is a block diagram illustrating a structure of a follow-up visit data processing system based on smart medical treatment for performing the follow-up visit data processing method based on smart medical treatment in fig. 1 according to an embodiment of the present invention.

Detailed description of the steps

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments obtained by a person of ordinary skill in the art by means of the embodiments of the present invention without making any creative effort fall within the protection scope of the present invention.

The STEP110, in combination with a pre-configured follow-up visit policy, obtains a medical tracking behavior sequence generated by the behavior tracking of the multiple intelligent medical service systems on the to-be-follow-up visit patients.

For some exemplary design ideas, the repeated-consultation follow-up data processing system based on smart medical treatment may obtain, in combination with a pre-configured repeated-consultation follow-up strategy, a medical tracking behavior sequence generated by the multiple smart medical service systems performing behavior tracking on the to-be-repeated-consultation follow-up patients. The intelligent medical service systems are respectively configured in page partitions with different behavior dimensions of a medical tracking process, the medical tracking process is used for tracking medical behaviors of the follow-up patients to be reviewed after the initial visit behavior is finished, and the medical tracking behavior sequence comprises two or more medical tracking behaviors.

And the STEP120 generates the target patient medical interest point distribution of the follow-up patient to be reviewed in combination with the medical tracking behaviors covered by the plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems.

For some exemplary design ideas, the repeated visit data processing system based on smart medical treatment may combine medical tracking behaviors covered by a plurality of medical tracking behavior sequences tracked by the plurality of smart medical service systems to generate a target patient medical interest point distribution of the follow-up patient to be repeated.

STEP130, determining one or more relevant medical behaviors of the relevant follow-up patient to be re-diagnosed corresponding to the follow-up patient to be re-diagnosed by combining the target patient medical interest point distribution, and analyzing whether a supplementary follow-up procedure needs to be performed on the follow-up patient to be re-diagnosed by combining the one or more relevant medical behaviors and the medical tracking behaviors covered by the plurality of medical tracking behavior sequences.

For some exemplary design ideas, the system for processing follow-up visit data based on smart medical treatment may determine, in combination with the target patient medical interest point distribution, one or more associated medical behaviors of an associated follow-up visit patient corresponding to the follow-up visit patient to be reviewed, and analyze, in combination with the one or more associated medical behaviors and the medical tracking behaviors covered by the plurality of medical tracking behavior sequences, whether a supplementary follow-up visit procedure needs to be performed on the follow-up visit patient to be reviewed. Wherein the one or more associated medical behaviors correspond to a plurality of different behavioral dimensions of the associated follow-up patient, respectively.

STEP140, when it is determined that a supplementary follow-up procedure needs to be performed on the to-be-reviewed follow-up patient, generating reference prompt information based on the follow-up plan corresponding to the past of the associated follow-up patient, and loading the reference prompt information into a medical archive database of the to-be-reviewed follow-up patient.

Based on the above steps, in this embodiment, a medical tracking behavior sequence generated by behavior tracking of a plurality of intelligent medical service systems for a follow-up patient to be reviewed may be obtained by combining a pre-configured follow-up visit policy, and then, a target patient medical interest point distribution of the follow-up patient to be reviewed may be generated by combining medical tracking behaviors covered by a plurality of tracked medical tracking behavior sequences, so that one or more associated medical behaviors of the associated follow-up patient corresponding to the follow-up patient to be reviewed may be determined by combining the target patient medical interest point distribution, and whether a supplementary follow-up visit process needs to be performed on the follow-up patient to be reviewed by combining the one or more associated medical behaviors and the medical tracking behaviors covered by the plurality of medical tracking behavior sequences is analyzed. By adopting the technical scheme, the initiation efficiency of the follow-up visit flow of the supplementary consultation can be improved.

For some exemplary design considerations, STEP110 may include the following:

firstly, analyzing whether medical behavior tracking needs to be carried out on a follow-up patient to be reviewed, and generating a corresponding medical behavior tracking instruction when judging that the medical behavior tracking needs to be carried out on the follow-up patient to be reviewed;

secondly, issuing the medical action tracking instruction to each of the plurality of intelligent medical service systems, and after receiving the medical action tracking instruction, each of the plurality of intelligent medical service systems performs medical action tracking on the follow-up patient to be subjected to follow-up visit, determines a corresponding medical tracking action sequence, and uploads the medical action tracking sequence to the follow-up visit data processing system based on intelligent medical treatment;

and then acquiring the medical tracking behavior sequence uploaded by each intelligent medical service system in the plurality of intelligent medical service systems in combination with the medical behavior tracking instruction.

For some exemplary design ideas, the step of analyzing whether medical behavior tracking needs to be performed on the follow-up patient to be reviewed, and generating a corresponding medical behavior tracking instruction when it is determined that medical behavior tracking needs to be performed on the follow-up patient to be reviewed may include the following steps:

firstly, whether first tracking trigger information sent by a related target medical behavior tracking subprogram is received or not is analyzed, wherein the target medical behavior tracking subprogram is configured in the medical tracking process and used for generating the first tracking trigger information when the condition that the follow-up patients to be reviewed meet preset behavior trigger conditions is tracked;

secondly, if the first tracking trigger information sent by the target medical behavior tracking subprogram is received, judging that medical behavior tracking needs to be carried out on the follow-up patient to be reviewed, and if the first tracking trigger information sent by the target medical behavior tracking subprogram is not received, judging that medical behavior tracking does not need to be carried out on the follow-up patient to be reviewed;

and then, when the follow-up patient to be reviewed needs to be tracked for medical behavior, generating a medical behavior tracking instruction for tracking the medical behavior of the follow-up patient to be reviewed.

For some exemplary design ideas, the step of generating a medical behavior tracking instruction for performing medical behavior tracking on the follow-up patient to be reviewed when it is determined that the medical behavior tracking on the follow-up patient to be reviewed is required may include the following steps:

firstly, for each intelligent medical service system in the plurality of intelligent medical service systems, determining page partition service node information of a page partition corresponding to the intelligent medical service system, and determining target behavior tracking quantity corresponding to the intelligent medical service system by combining the page partition service node information, wherein mapping relation exists between the page partition service node information and the target behavior tracking quantity;

secondly, for each of the plurality of intelligent medical service systems, combining the target behavior tracking number corresponding to the intelligent medical service system, generating a medical behavior tracking instruction corresponding to the intelligent medical service system and used for tracking the medical behavior of the follow-up patient to be reviewed, wherein the medical behavior tracking instruction carries the corresponding target behavior tracking number, the target behavior tracking number represents that the corresponding intelligent medical service system tracks the medical behavior of the follow-up patient to be reviewed, and a corresponding medical tracking behavior sequence is determined, and the behavior tracking number of the medical tracking behavior covered by the medical tracking behavior sequence is the corresponding target behavior tracking number.

For some exemplary design considerations, STEP120 may include the following:

firstly, for each medical tracking behavior sequence in a plurality of medical tracking behavior sequences tracked by a plurality of intelligent medical service systems, frequently screening the medical tracking behaviors covered by the medical tracking behavior sequence, and determining a medical tracking frequent behavior space corresponding to the medical tracking behavior sequence;

secondly, for each medical tracking behavior sequence in a plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems, performing patient interest point mining on a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence, and outputting target patient medical interest point information of the follow-up patients to be reviewed on a behavior dimension corresponding to the medical tracking behavior sequence;

then, combining the target patient medical interest point information corresponding to each of the plurality of medical tracking behavior sequences, generating a target patient medical interest point distribution of the follow-up patient to be reviewed (for example, the target patient medical interest point information may be spliced to determine a three-dimensional target patient medical interest point distribution).

For some exemplary design ideas, the step of performing frequent item screening on medical tracking behaviors included in the medical tracking behavior sequences for each of the plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems, and determining a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequences may include the following steps:

firstly, respectively carrying out frequent item support calculation on each medical tracking behavior covered by a plurality of medical tracking behavior sequences tracked by a plurality of intelligent medical service systems, and determining the behavior frequent item support corresponding to each medical tracking behavior;

secondly, for each medical tracking behavior sequence in the plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems, in the behavior frequent item support degrees corresponding to each medical tracking behavior covered by the medical tracking behavior sequence, generating a behavior frequent item support degree larger than a preset support degree as a target behavior frequent item support degree corresponding to the medical tracking behavior sequence, and outputting the medical tracking behavior corresponding to the target behavior frequent item support degree as a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence.

For some exemplary design considerations, STEP130 may include the following:

firstly, for each historical patient medical interest point distribution in a plurality of historical patient medical interest point distributions recorded in advance, calculating the coincidence rate between the historical patient medical interest point distribution and the target patient medical interest point distribution, and determining the coincidence rate of interest points corresponding to the historical patient medical interest point distribution;

secondly, generating an interest point coincidence rate which is larger than a preset coincidence rate in an interest point coincidence rate corresponding to each historical patient medical interest point distribution in the plurality of historical patient medical interest point distributions as a target interest point coincidence rate, determining the historical patient medical interest point distribution corresponding to the target interest point coincidence rate as an associated patient medical interest point distribution, and acquiring one or more associated medical behaviors corresponding to associated follow-up patients corresponding to the associated patient medical interest point distribution, wherein each associated medical behavior has corresponding behavior tracking and labeling information which expresses a behavior tracking dimension of the associated medical behavior;

then, for each medical tracking behavior sequence in the plurality of medical tracking behavior sequences, performing frequent item screening on medical tracking behaviors covered by the medical tracking behavior sequence in the medical tracking behavior sequence, determining a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence, and determining a corresponding behavior tracking labeling information and a corresponding medical behavior corresponding to the page service dimension in the one or more associated medical behaviors by combining the page service dimension of a page partition corresponding to the medical tracking behavior sequence, wherein the corresponding behavior tracking labeling information and the corresponding associated medical behavior are used as a target associated medical behavior space corresponding to the medical tracking behavior sequence;

on the basis, for each medical tracking behavior sequence, calculating the coincidence rate between the medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence and the target associated medical behavior space corresponding to the medical tracking behavior sequence, and determining the behavior tracking coincidence rate corresponding to the medical tracking behavior sequence;

and finally, analyzing whether a supplementary follow-up visit process needs to be carried out on the follow-up patient to be re-diagnosed or not by combining the behavior tracking coincidence rate corresponding to each medical tracking behavior sequence (if the behavior tracking coincidence rate corresponding to each medical tracking behavior sequence is larger than the preset coincidence rate, the supplementary follow-up visit process does not need to be carried out on the follow-up patient to be re-diagnosed).

For some exemplary design ideas, for each medical tracking behavior sequence, the step of calculating a coincidence rate between a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence and a target associated medical behavior space corresponding to the medical tracking behavior sequence, and determining a behavior tracking coincidence rate corresponding to the medical tracking behavior sequence may include the following steps:

firstly, respectively determining the behavior tag attribute of each medical behavior member in the target associated medical behavior space, clustering the target associated medical behavior space by combining the behavior tag attribute of each medical behavior member, determining at least one corresponding associated medical behavior cluster, matching the behavior tag attributes of each medical behavior member corresponding to the same associated medical behavior cluster, wherein at least one medical behavior communication link is arranged between each medical behavior member of the same associated medical behavior cluster, and each medical behavior member covered by the medical behavior communication link is the same as the associated medical behavior cluster to which the corresponding medical behavior member belongs;

secondly, for each associated medical behavior cluster in the at least one associated medical behavior cluster, determining a cluster core behavior member of the associated medical behavior cluster, respectively calculating a first intra-cluster difference between each medical behavior member in the associated medical behavior cluster and the cluster core behavior member, and respectively combining the first intra-cluster difference between each medical behavior member in the associated medical behavior cluster and the cluster core behavior member corresponding to the associated medical behavior cluster to generate a first intra-cluster importance evaluation parameter corresponding to each medical behavior member, wherein a mapping relation exists between the first intra-cluster importance evaluation parameter and the first intra-cluster difference;

then, for each associated medical behavior cluster in the at least one associated medical behavior cluster, performing fusion calculation on the medical behavior engagement degrees of each medical behavior member in the associated medical behavior cluster by combining with a first intra-cluster importance evaluation parameter corresponding to each medical behavior member in the associated medical behavior cluster, determining a corresponding first importance fusion engagement degree, performing order arrangement on the first importance fusion engagement degrees corresponding to each associated medical behavior cluster by combining with a space domain node of each associated medical behavior cluster in the target associated medical behavior space, and determining first medical behavior engagement degree order information corresponding to the target associated medical behavior space;

on the basis, combining the spatial domain nodes of each associated medical action cluster in the target associated medical action space, clustering the medical tracking frequent item behavior space, determining at least one medical tracking frequent item behavior cluster corresponding to the medical tracking frequent item behavior space, and for each medical frequent item behavior cluster, determining a cluster core behavior member of the medical frequent item behavior cluster, and respectively calculating the discrimination degrees in a second cluster between each medical behavior member in the medical frequent item behavior cluster and the cluster core behavior member, and generating second intra-cluster importance evaluation parameters corresponding to the medical behavior members in the medical frequent item behavior cluster by respectively combining the second intra-cluster discriminative power between the medical behavior members in the medical frequent item behavior cluster and the corresponding cluster core behavior members, a mapping relation exists between the second intra-cluster importance assessment parameter and the second intra-cluster discriminative power;

further, for each medical frequent item behavior cluster in the at least one medical frequent item behavior cluster, performing fusion calculation on the medical behavior engagement of each medical frequent item behavior member in the medical frequent item behavior cluster in combination with a second intra-cluster importance evaluation parameter corresponding to each medical behavior member in the medical frequent item behavior cluster, determining a second importance fusion engagement corresponding to the medical frequent item behavior cluster, performing order sorting on the second importance fusion engagement corresponding to each medical frequent item behavior cluster in combination with a spatial domain node of each medical frequent item behavior cluster in the medical tracking frequent item behavior space, and determining second medical behavior engagement order information corresponding to the medical tracking frequent item behavior space;

and finally, calculating the association degree between the first medical behavior engagement degree sequence information and the second medical behavior engagement degree sequence information, and determining the coincidence rate between the medical tracking frequent item behavior space and the target associated medical behavior space as the behavior tracking coincidence rate corresponding to the corresponding medical tracking behavior sequence.

For another exemplary design idea (which may be implemented in an alternative manner to the design idea described above), for each medical tracking behavior sequence, the step of calculating a coincidence rate between a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence and a target associated medical behavior space corresponding to the medical tracking behavior sequence, and determining a behavior tracking coincidence rate corresponding to the medical tracking behavior sequence may also include the following steps:

firstly, respectively determining the behavior tag attribute of each medical behavior member in the target associated medical behavior space, clustering the target associated medical behavior space by combining the behavior tag attribute of each medical behavior member, determining at least one associated medical behavior cluster corresponding to the target associated medical behavior space, matching the behavior tag attributes of each medical behavior member corresponding to the same associated medical behavior cluster, and at least one medical behavior communication link is arranged between each medical behavior member of the same associated medical behavior cluster, wherein each medical behavior member covered by the medical behavior communication link is the same as the associated medical behavior cluster to which the corresponding medical behavior member belongs;

secondly, for each associated medical behavior cluster in the at least one associated medical behavior cluster, generating each clustered edge medical behavior member and a clustered core behavior member in the associated medical behavior cluster, and determining whether to take the clustered edge medical behavior member as a first target clustered edge medical behavior member corresponding to the associated medical behavior cluster by combining the intra-cluster differentiation between the clustered edge medical behavior member and the clustered core behavior member;

then, for each first target cluster edge medical behavior member corresponding to each associated medical behavior cluster in the at least one associated medical behavior cluster, taking the first target cluster edge medical behavior member as a reference member, extracting affected medical behavior members in the target associated medical behavior space respectively according to a plurality of target behavior affected areas with different behavior dimensions, determining a plurality of first medical behavior member sequences corresponding to the first target cluster edge medical behavior member, respectively calculating the mean participation degree and the discrete participation degree of the medical behavior participation degrees of the medical behavior members covered by each first medical behavior member sequence, determining the medical behavior mean participation degree and the medical behavior discrete participation degree corresponding to each first medical behavior member sequence, and combining the plurality of medical behavior mean participation degrees corresponding to the plurality of first medical behavior member sequences corresponding to the first target cluster edge medical behavior member, ranking based on the corresponding degree of participation, determining first medical behavior average degree of participation ranking information corresponding to the first target cluster edge medical behavior member, wherein the medical behavior discrete degree of participation corresponding to each first medical behavior member sequence is used as ranking marking information of the corresponding medical behavior average degree of participation in the first medical behavior average degree of participation ranking information;

on the basis, by combining with behavior space domain nodes of each first target cluster edge medical behavior member in the target associated medical behavior space corresponding to each associated medical behavior cluster in the at least one associated medical behavior cluster, determining each corresponding second target cluster edge medical behavior member in the medical tracking frequent item behavior space, regarding each second target cluster edge medical behavior member, taking the second target cluster edge medical behavior member as a reference member, extracting affected medical behavior members in the medical tracking frequent item behavior space by respectively taking target behavior influence areas with different behavior dimensions, determining a plurality of second medical behavior member sequences corresponding to the second target cluster edge medical behavior members, and respectively calculating the mean participation degree and the dispersion participation degree of the medical behavior participation degrees of the medical behavior members covered by each second medical behavior member sequence, determining average medical behavior participation and discrete medical behavior participation corresponding to each second medical behavior member sequence, and determining second medical behavior average participation ranking information corresponding to the second target clustering edge medical behavior member by combining a plurality of average medical behavior participation corresponding to a plurality of second medical behavior member sequences corresponding to the second target clustering edge medical behavior member and ranking based on the corresponding participation magnitude relation, wherein the discrete medical behavior participation corresponding to each second medical behavior member sequence is used as ranking marking information of the corresponding average medical behavior participation in the second medical behavior ranking information;

finally, for each group of the first target cluster edge medical treatment behavior member and the second target cluster edge medical treatment behavior member having a corresponding relationship, calculating a degree of association between the average degree of participation ranking of the first medical treatment behavior corresponding to the first target cluster edge medical treatment behavior member and the average degree of participation ranking of the second medical treatment behavior corresponding to the second target cluster edge medical treatment behavior member (for example, it may be determined whether the average degree of participation of the medical treatment behavior and the ranking marking information of the corresponding ranking nodes are the same, then, counting the number ratio of the same ranking nodes, and determining the degree of association of positive correlation by combining the number ratio), and combining the degrees of association (such as the average degree of participation of the degree of association and the like) corresponding to the first target cluster edge medical treatment behavior member and the second target cluster edge medical treatment behavior member having a corresponding relationship, and generating a coincidence rate between the medical tracking frequent item behavior space and the target associated medical behavior space as a behavior tracking coincidence rate corresponding to the corresponding medical tracking behavior sequence.

In a possible implementation manner, an exemplary implementation manner of mining the medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence to output the target patient medical interest point information of the follow-up patient to be diagnosed in the behavior dimension corresponding to the medical tracking behavior sequence may be referred to in the following examples.

(1) Performing behavior tendency variable extraction on the medical tracking frequent item behavior space data corresponding to the medical tracking behavior sequence to determine a first behavior tendency variable map;

(2) determining each positive behavior tendency point and each negative behavior tendency point in the medical tracking frequent item behavior space data by combining a first behavior tendency variable map;

(3) determining each positive behavior tendency point and each first behavior tendency variation sub-map corresponding to each negative behavior tendency point in a first behavior tendency variable map by combining each positive behavior tendency point and each negative behavior tendency point with associated behavior knowledge points in the medical tracking frequent behavior space data;

(4) respectively carrying out mining factor optimization on the first behavior tendency variable quantum map corresponding to each positive behavior tendency point and each negative behavior tendency point, and determining each second behavior tendency variable quantum map after the mining factor optimization of the first behavior tendency variable;

(5) carrying out interest point mapping variable mining by combining the second behavior tendency variable quantum maps, and determining each target interest point mapping variable in the medical tracking frequent behavior space data;

(6) determining each first behavior tendency variable sub-map corresponding to each target interest point mapping variable in the first behavior tendency variable map by combining the associated behavior knowledge points of each target interest point mapping variable in the medical tracking frequent behavior space data, and aggregating each first behavior tendency variable sub-map corresponding to all target interest point mapping variables to generate a third behavior tendency variable map;

(7) extracting attention variables from the first action tendency variable map, and determining a fourth action tendency variable map; wherein, the map expression mode of the trend variable map of the fourth line is the same as that of the trend variable map of the third line;

(8) aggregating the fourth behavior tendency variable map and the third behavior tendency variable map to generate a fifth behavior tendency variable map;

(9) and (4) mining medical interest points by combining a fifth behavior tendency variable map, and outputting the medical interest point information of the target patient of the follow-up patient to be diagnosed on the behavior dimension corresponding to the medical tracking behavior sequence.

For example, in one possible implementation, performing behavior trend variable extraction on the medical tracking frequent item behavior space data to determine a first behavior trend variable map includes: and transmitting the medical tracking frequent item behavior space data to a behavior tendency variable extraction branch of the target medical interest point mining model for performing behavior tendency variable extraction. Determining each positive behavior-trend point and each negative behavior-trend point in the medical tracking frequent item behavior space data in combination with a first behavior-trend variable map, comprising: and transmitting the first behavior tendency variable map to a behavior tendency point extraction branch of the target medical interest point mining model to extract each positive behavior tendency point and each negative behavior tendency point in the medical tracking frequent item behavior space data.

And carrying out interest point mapping variable mining by combining the second behavior tendency variable quantum maps, wherein the interest point mapping variable mining comprises the following steps: and transmitting each second behavior tendency variable quantum graph to an interest point mapping variable extraction branch of the target medical interest point mining model for interest point mapping variable mining. And (3) combining a fifth behavior tendency variable map to perform medical interest point mining, which comprises the following steps: and transmitting a fifth behavior tendency variable map to a medical interest point mining branch of the target medical interest point mining model for medical interest point mining.

For example, in one possible implementation, the training step of the target medical interest point mining model can be seen in the following examples.

(1) The method comprises the steps of optimizing and selecting a parameter layer of a calibrated medical interest point mining model by combining a calibrated medical tracking behavior space sample data sequence, and generating a converged target medical interest point mining model, and specifically comprises the following steps:

(2) respectively extracting one piece of calibrated medical tracking behavior space sample data from the calibrated medical tracking behavior space sample data sequence, transmitting the calibrated medical tracking behavior space sample data to a behavior tendency variable extraction branch of the target medical interest point mining model, and extracting a behavior tendency variable to determine a first behavior tendency variable map of the transmitted calibrated medical tracking behavior space sample data;

(3) transmitting the first behavior tendency variable map to a behavior tendency point extraction branch of the target medical interest point mining model, and determining each positive behavior tendency point and each negative behavior tendency point in transmitted calibrated medical tracking behavior space sample data;

(4) determining each positive behavior tendency point and each first behavior tendency quantum map corresponding to each negative behavior tendency point in the first behavior tendency variable map by combining each positive behavior tendency point and each associated behavior knowledge point of each negative behavior tendency point in the transmitted sample data of the calibrated medical tracking behavior space;

(5) respectively carrying out mining factor optimization on the first behavior tendency variable quantum map corresponding to each positive behavior tendency point and each negative behavior tendency point, and determining each second behavior tendency variable quantum map after the mining factor optimization of the first behavior tendency variable;

(6) transmitting each second behavior tendency variable quantum map to an interest point mapping variable extraction branch of the target medical interest point mining model, and determining each intermediate interest point mapping variable in transmitted calibrated medical tracking behavior space sample data and intermediate interest point attributes of each intermediate interest point mapping variable;

(7) determining a mining loss value by combining each intermediate interest point mapping variable output by the interest point mapping variable extraction branch, each calibrated interest point mapping variable carried in the transmitted calibrated medical tracking behavior space sample data and the corresponding interest point attribute, and a pre-bound first medical interest point mining loss function, so as to determine a first medical interest point mining loss value;

(8) determining each first behavior tendency variable quantum graph corresponding to each intermediate interest point mapping variable in the first behavior tendency variable graph by combining the associated behavior knowledge points of each intermediate interest point mapping variable in the transmitted sample data of the calibrated medical tracking behavior space, and aggregating each first behavior tendency variable sub-graph spectrum corresponding to all intermediate interest point mapping variables to generate a third behavior tendency variable graph;

(9) extracting attention variables from the first action tendency variable map, and determining a fourth action tendency variable map; wherein, the map expression mode of the fourth line trend variable map is the same as that of the third line trend variable map;

(10) aggregating the fourth behavior tendency variable map and the third behavior tendency variable map to generate a fifth behavior tendency variable map;

(11) transmitting a fifth behavior tendency variable map to a medical interest point mining branch of the target medical interest point mining model, and determining each intermediate interest point attribute contained in the transmitted calibrated medical tracking behavior space sample data;

(12) determining a digging loss value by combining the attributes of each intermediate interest point contained in the transmitted calibrated medical tracking behavior space sample data generated by the medical interest point digging branch and the attributes of each calibrated interest point mapping variable carried in the transmitted calibrated medical tracking behavior space sample data and a pre-bound second medical interest point digging loss function, and determining a second medical interest point digging loss value;

(13) weighting the first medical interest point mining loss value and the second medical interest point mining loss value, and carrying out model parameter layer optimization and selection on the target medical interest point mining model by combining the weighted interest point mining loss value;

(14) and if the model parameter layer of the target medical interest point mining model is converged, outputting the corresponding target medical interest point mining model.

For example, in one possible implementation, after the third behavioral tendency variable map is output by aggregating the first behavioral tendency variable sub-map spectra corresponding to all target interest point mapping variables, and before the step of aggregating the fourth behavioral tendency variable map and the third behavioral tendency variable map, the third behavioral tendency variable map may be subjected to behavioral tendency variable expansion, the expanded behavioral tendency variables of each behavioral tendency variable in the third behavioral tendency variable map are determined, and each behavioral tendency variable in the third behavioral tendency variable map is associated and fused with the expanded behavioral tendency variable thereof, so as to determine the expanded behavioral tendency variable map of the third behavioral tendency variable map. Thus, aggregating the fourth behavior tendency variable map with the third behavior tendency variable map comprises: and aggregating the fourth behavior tendency variable map with the expanded behavior tendency variable map of the third behavior tendency variable map.

For example, in a possible implementation manner, the present embodiment may further add the point of interest influence variable of each medical point of interest target to the sample data of each calibrated medical tracking behavior space.

After determining each positive behavior tendency point and each first behavior tendency quantum graph corresponding to each negative behavior tendency point in the first behavior tendency variable graph and before weighting the first medical interest point mining loss value and the second medical interest point mining loss value, the embodiment may further perform mining factor optimization of the second behavior tendency variable on each first behavior tendency quantum graph corresponding to each positive behavior tendency point and each negative behavior tendency point, and determine each sixth behavior tendency quantum graph after the mining factor optimization of the second behavior tendency variable; transmitting each sixth behavior tendency variable quantum map to an interest point influence variable extraction branch of the target medical interest point mining model, and determining interest point influence variables and intermediate interest point attributes of each medical interest point target in transmitted calibrated medical tracking behavior space sample data; and determining a mining loss value by combining the interest point influence variable and the intermediate interest point attribute of each medical interest point target in the transmitted calibrated medical tracking behavior space sample data obtained by the interest point influence variable extraction branch and the interest point influence variable and the interest point attribute of each medical interest point target carried in the transmitted calibrated medical tracking behavior space sample data, and combining a pre-bound third medical interest point mining loss function to determine the mining loss value of the third medical interest point.

On the basis, weighting the first medical interest point mining loss value and the second medical interest point mining loss value comprises the following steps: weighting the first medical point of interest mining loss value, the second medical point of interest mining loss value, and the third medical point of interest mining loss value.

Based on the above description, in another embodiment, the present invention further provides a follow-up visit data Processing system based on smart medical treatment, referring to fig. 2, fig. 2 is a structural diagram of the follow-up visit data Processing system 100 based on smart medical treatment provided in the embodiment of the present invention, the follow-up visit data Processing system 100 based on smart medical treatment may generate a relatively large difference due to different configurations or performances, and may include one or more Central Processing Units (CPUs) 112 (e.g., one or more processors) and a memory 111. Wherein the memory 111 may be a transient storage or a persistent storage. The program stored in memory 111 may include one or more modules, each of which may include a series of instructions operating on the smart medical-based follow-up data processing system 100. Further, the central processor 112 may be configured to communicate with the memory 111, and execute a series of instruction operations in the memory 111 on the data processing system 100 for follow-up visit based on smart medical treatment.

The smart medical-based follow-up data processing system 100 may also include one or more power supplies, one or more communication units 113, one or more delivery-to-output interfaces, and/or one or more operating systems, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, and the like.

The steps executed by the revisit data processing system based on smart medical treatment in the above embodiment can be combined with the structure of the revisit data processing system based on smart medical treatment shown in fig. 2.

In addition, the embodiment of the present invention further provides a storage medium, where the storage medium is used to store a computer program, and the computer program is used to execute the method provided by the above embodiment.

Embodiments of the present invention also provide a computer program product including instructions, which when run on a computer, cause the computer to perform the method provided by the above embodiments.

Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium may be at least one of the following media: various media that can store program codes, such as Read-only Memory (ROM), RAM, magnetic disk, or optical disk.

It should be noted that, in the present specification, each embodiment is described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on differences from other embodiments. In particular, for the apparatus and system embodiments, since they are substantially similar to the method embodiments, they are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described embodiments of the apparatus and system are merely illustrative, and units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the modules may be selected based on actual needs to achieve the purpose of the solution of the embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

The above description is only one specific implementation procedure of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A consultation follow-up data processing method based on intelligent medical treatment is applied to a consultation follow-up data processing system based on intelligent medical treatment, and is characterized by comprising the following steps:

acquiring a medical tracking behavior sequence generated by behavior tracking of a plurality of intelligent medical service systems to-be-retested follow-up patients by combining a pre-configured follow-up strategy, wherein the plurality of intelligent medical service systems are respectively configured on different behavior dimensionalities of a medical tracking process, the medical tracking process is used for performing medical behavior tracking on the to-be-retested follow-up patients with the finished preliminary visit behavior, and the medical tracking behavior sequence comprises two or more medical tracking behaviors;

and when the supplementary follow-up visit flow needs to be carried out on the follow-up patient to be subjected to the follow-up visit, generating reference prompt information based on the corresponding follow-up visit plan of the related follow-up visit patient and loading the reference prompt information into a medical file database of the follow-up visit patient to be subjected to the follow-up visit.

2. The revisiting follow-up data processing method based on intelligent medical treatment as claimed in claim 1, wherein the step of obtaining a medical tracking behavior sequence generated by the behavior tracking of the multiple intelligent medical service systems on the to-be-revising follow-up patients in combination with a pre-configured revisiting follow-up strategy specifically comprises the following steps:

and acquiring the medical tracking behavior sequence uploaded by each intelligent medical service system in the plurality of intelligent medical service systems in combination with the medical behavior tracking instruction.

3. The follow-up visit data processing method based on intelligent medical treatment as claimed in claim 2, wherein the analysis of whether medical behavior tracking needs to be performed on the follow-up patient to be reviewed, and when it is determined that medical behavior tracking needs to be performed on the follow-up patient to be reviewed, a corresponding medical behavior tracking instruction is generated, specifically comprising the following steps:

if the first tracking trigger information sent by the target medical behavior tracking subprogram is received, judging that medical behavior tracking needs to be carried out on the follow-up patient to be re-diagnosed, and if the first tracking trigger information sent by the target medical behavior tracking subprogram is not received, judging that medical behavior tracking does not need to be carried out on the follow-up patient to be re-diagnosed;

4. The follow-up visit data processing method based on intelligent medical treatment as claimed in claim 3, wherein when it is determined that the follow-up visit patient to be reviewed needs to be followed up, a medical behavior following instruction for following the medical behavior of the follow-up visit patient to be reviewed is generated, and the method specifically comprises the following steps:

5. The method for processing follow-up visit data based on intelligent medical treatment according to claim 1, wherein the step of generating the medical interest point distribution of the target patient of the follow-up visit patient in combination with the medical tracking behaviors covered by the medical tracking behavior sequences tracked by the intelligent medical service systems comprises the following steps:

for each medical tracking behavior sequence in the multiple medical tracking behavior sequences tracked by the multiple intelligent medical service systems, performing patient interest point mining on a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence, and outputting target patient medical interest point information of the follow-up patient to be reviewed on a behavior dimension corresponding to the medical tracking behavior sequence;

6. The revisit follow-up visit data processing method based on intelligent medical treatment according to claim 5, wherein for each medical tracking behavior sequence in the plurality of medical tracking behavior sequences tracked by the plurality of intelligent medical service systems, frequent item screening is performed on the medical tracking behavior covered by the medical tracking behavior sequence, and a medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence is determined, specifically comprising the following steps:

7. The follow-up visit data processing method based on intelligent medical treatment according to any one of claims 1 to 6, wherein the method for determining one or more associated medical behaviors of an associated follow-up visit patient corresponding to the follow-up visit patient to be reviewed in combination with the target patient medical interest point distribution, and analyzing whether a supplementary follow-up visit procedure for the follow-up visit patient to be reviewed is required in combination with the one or more associated medical behaviors and the medical tracking behaviors covered by the plurality of medical tracking behavior sequences, specifically comprises the following steps:

8. The revisit follow-up visit data processing method based on intelligent medical treatment as claimed in claim 7, wherein for each medical tracking behavior sequence, the coincidence rate between the medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence and the target associated medical behavior space corresponding to the medical tracking behavior sequence is calculated, and the behavior tracking coincidence rate corresponding to the medical tracking behavior sequence is determined, specifically comprising the following steps:

9. The follow-up visit data processing method based on intelligent medical treatment as claimed in claim 5, wherein the step of mining the medical tracking frequent item behavior space corresponding to the medical tracking behavior sequence for the patient interest points and outputting the target patient medical interest point information of the follow-up visit patient to be reviewed in the behavior dimension corresponding to the medical tracking behavior sequence comprises:

determining each positive behavior tendency point and each first behavior tendency variation sub-map corresponding to each negative behavior tendency point in a first behavior tendency variable map by combining each positive behavior tendency point and each negative behavior tendency point with associated behavior knowledge points in the medical tracking frequent behavior space data;

carrying out interest point mapping variable mining by combining the second behavior tendency variable quantum maps, and determining each target interest point mapping variable in the medical tracking frequent behavior space data;

determining each first behavior tendency variable sub-map corresponding to each target interest point mapping variable in the first behavior tendency variable map by combining the associated behavior knowledge points of each target interest point mapping variable in the medical tracking frequent behavior space data, and aggregating each first behavior tendency variable sub-map corresponding to all target interest point mapping variables to generate a third behavior tendency variable map;

10. A follow-up visit data processing system based on intelligent medical treatment is characterized by comprising:

a processor;

a memory having stored therein a computer program which, when executed, implements the intelligent medical treatment-based follow-up data processing method according to any one of claims 1 to 9.