CN108461130B - Intelligent scheduling method and system for treatment tasks - Google Patents
Intelligent scheduling method and system for treatment tasks Download PDFInfo
- Publication number
- CN108461130B CN108461130B CN201810050316.XA CN201810050316A CN108461130B CN 108461130 B CN108461130 B CN 108461130B CN 201810050316 A CN201810050316 A CN 201810050316A CN 108461130 B CN108461130 B CN 108461130B
- Authority
- CN
- China
- Prior art keywords
- treatment
- diagnosis
- similarity
- medical
- historical data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The invention provides a method and a system for intelligently scheduling a treatment task, and relates to the field of medical treatment. According to the intelligent scheduling method for the hospitalizing tasks, the recommended hospitalizing subjects are obtained by analyzing the historical data provided by the user and the historical data recorded in the system, the appointed hospitals are determined as candidates by combining the requested hospitalizing subjects provided by the patient, then the more reasonable hospitalizing consulting rooms are determined for the user through image recognition and data analysis, and the referred hospitalizing requests, the first electronic version historical data and the second electronic version historical data are sent to the consulting rooms, so that the reasonable degree of allocating the consulting rooms to the user is improved to a certain degree.
Description
Technical Field
The invention relates to the field of medical treatment, in particular to a method and a system for intelligently scheduling a treatment task.
Background
With the continuous progress of science and technology, medical technology has also been developed, various diagnosis and treatment equipment of hospitals are continuously updated, and the diagnosis level of doctors is increasingly improved. Meanwhile, with the continuous improvement of the living standard of people, the requirements of people on medical services are higher and higher. In order to better meet the needs of patients and provide the most accurate diagnosis and treatment services for patients, when doctors are determined for patients, the diagnosis and treatment history of the patients needs to be known for reference, so that the patients are allocated with the optimal doctors.
The inventors have found that the current way of assigning visits to patients is not ideal.
Disclosure of Invention
The invention aims to provide an intelligent scheduling method for a medical treatment task.
In a first aspect, an embodiment of the present invention provides an intelligent scheduling method for a medical treatment task, including:
acquiring a medical seeking request of a target user, wherein the medical seeking request carries an identity of the target user, biological authentication data, instant disease description and a medical seeking subject;
searching biological standard data corresponding to the identity in a database;
judging whether the similarity of the biological standard data and the biological authentication data exceeds a preset first threshold value or not;
if so, searching first electronic version historical data corresponding to the identity in a database, wherein the first electronic version historical data comprises diagnosis and treatment time, diagnosis and treatment places, diagnosis and treatment items, diagnosis and treatment processes and diagnosis and treatment results;
acquiring paper edition historical data provided by a target user;
photographing historical data of the paper edition to obtain a historical data photo;
performing character recognition on the historical data photo to obtain second electronic version historical data, wherein the second electronic version historical data comprises diagnosis and treatment time, diagnosis and treatment places, diagnosis and treatment items, diagnosis and treatment processes and diagnosis and treatment results;
determining a first pathological analysis result of a target user according to the first electronic edition historical data;
determining a second pathological analysis result of the target user according to the second electronic edition historical data;
determining a first weight corresponding to a first pathological analysis result according to the continuous degree of the diagnosis and treatment time of the first electronic version historical data and the time period of the diagnosis and treatment time;
determining a second weight corresponding to a second pathological analysis result according to the continuous degree of the diagnosis and treatment time of the second electronic version historical data and the time period of the diagnosis and treatment time;
determining a reference pathological analysis result according to the first pathological analysis result, the second pathological analysis result, the first weight and the second weight in a weighted averaging mode, wherein the reference pathological analysis result comprises the disease attack time, the treatment result, the treatment process, the treatment period and the medical treatment subject each time;
searching a diagnosis and treatment file with the matching degree with the reference pathological analysis result meeting the preset requirement in a database as a candidate diagnosis and treatment file by adopting a big data analysis mode;
judging whether the number of the candidate diagnosis and treatment files exceeds a preset second threshold value, if not, reducing the first threshold value, and re-executing the step of searching the candidate diagnosis and treatment files of which the matching degree with the reference pathological analysis result exceeds the first threshold value in the database; if yes, calculating the diagnosis and treatment success probability of each candidate diagnosis and treatment file according to the treatment period and the treatment result in each candidate diagnosis and treatment file;
determining recommended medical subjects according to the diagnosis and treatment success probability;
determining a first medical emergency degree according to the first pathological analysis result;
determining a second medical emergency degree according to the second pathological analysis result and the instant disease description;
if the recommended medical subject is the same as the requested medical subject, searching a first target hospital in the database according to the queuing condition and the diagnosis and treatment success rate of the target diagnosis room in the second hospital corresponding to the requested medical subject according to a first search rule; the target consulting room corresponds to a recommended medical subject;
if the recommended medical treatment subject is different from the requested medical treatment subject, judging whether at least one of the first medical treatment emergency degree or the second medical treatment emergency degree exceeds a preset threshold value; if yes, sending prompt information to the target user; if not, searching the first target hospital in the database according to the queuing condition, the diagnosis and treatment success rate and the first medical treatment emergency degree of the target diagnosis room in the first hospital corresponding to the medical treatment requesting subject according to the second searching rule;
calling video data of a plurality of target consulting rooms in a first target hospital;
extracting multiple frames of images from the video data according to the time sequence, wherein the interval between two adjacent images is preset time length;
according to the similarity of the multi-frame images, the actual queuing degree of each target consulting room;
determining the theoretical queuing degree of each target consulting room according to the acquired reservation condition of the target consulting room;
determining a target medical consulting room of a target user according to the actual queuing degree and the theoretical queuing degree;
and sending the medical request, the first electronic version historical data and the second electronic version historical data to a consulting room end in the target medical consulting room.
With reference to the first aspect, an embodiment of the present invention provides a first possible implementation manner of the first aspect, where the step of searching, in the database, a diagnosis and treatment archive whose matching degree with the reference pathological analysis result meets a preset requirement as the candidate diagnosis and treatment archive includes:
calculating a first similarity of each diagnosis and treatment file and a reference pathological analysis result according to a preset first algorithm, and calculating a treatment rational degree value of each diagnosis and treatment file according to a preset second algorithm;
grouping all diagnosis and treatment files according to diagnosis and treatment places of the diagnosis and treatment files, wherein the diagnosis and treatment places of the diagnosis and treatment files in each group of the diagnosis and treatment files are the same;
calculating the maximum likelihood estimation value of each file group by adopting a maximum likelihood estimation mode according to the first similarity and the treatment reasonable degree value;
and determining the diagnosis and treatment files in one or more groups as candidate diagnosis and treatment files according to the maximum likelihood estimation value of each group.
With reference to the first aspect, an embodiment of the present invention provides a second possible implementation manner of the first aspect, where the determining, according to the maximum likelihood estimation value of each group, a diagnosis and treatment archive in one or more designated groups as a candidate diagnosis and treatment archive includes:
if the distribution of the maximum likelihood estimation values of all candidate diagnosis and treatment files meets the preset condition, taking the diagnosis and treatment files in the diagnosis and treatment file group with the maximum likelihood estimation value exceeding the preset threshold value as the candidate diagnosis and treatment files;
if the distribution of the maximum likelihood estimation values of all candidate diagnosis and treatment files does not accord with the preset condition, calculating the first similarity of each diagnosis and treatment file and the reference pathological analysis result according to the preset first algorithm and calculating the treatment reasonable degree value of each diagnosis and treatment file according to the preset second algorithm according to the calculation conditions of the readjustment first algorithm and the second algorithm.
With reference to the first aspect, an embodiment of the present invention provides a third possible implementation manner of the first aspect, where performing character recognition on the historical data photo to obtain second electronic version historical data includes:
carrying out binarization processing on the historical data photo at least twice according to different binarization threshold values to obtain first identification graphs corresponding to the different binarization threshold values;
respectively carrying out character recognition on the character blocks on each first recognition graph according to the following mode to determine the character recognition result of each first recognition graph: respectively superposing the first template drawing with the character block on each first recognition drawing so as to determine the nearest reference distance between each first reference point on the character block and the character skeleton point in the template drawing, and determining the character block according to the nearest reference distance and the weight corresponding to the weight of each character skeleton point; the first reference point is a point at which the gradation value exceeds a predetermined threshold; the first template picture is obtained after big data analysis is carried out according to a doctor note database.
Calling a doctor semantic analysis model and a patient semantic analysis model from a doctor semantic analysis database and a patient semantic analysis database respectively;
and performing semantic analysis on each character recognition result by using a doctor semantic analysis model and a patient semantic analysis model respectively to determine second electronic version historical data.
With reference to the first aspect, an embodiment of the present invention provides a fourth possible implementation manner of the first aspect, where a character skeleton point includes a skeleton point and a peripheral point, a weight of the skeleton point is the highest of all reference points, and all skeleton points form a basic shape of a graph in the first template graph; the weight of the peripheral point is in negative correlation with the target distance, and the target distance is the distance between the peripheral point and the nearest skeleton point.
With reference to the first aspect, an embodiment of the present invention provides a fifth possible implementation manner of the first aspect, where the method further includes:
determining the weight value of each peripheral point according to the following modes:
carrying out binarization processing on the target candidate images for multiple times respectively according to different binarization threshold values to obtain second identification images corresponding to the different binarization threshold values;
assigning a weight to each peripheral point according to the following mode, wherein the weight of a first peripheral point is higher than that of a second peripheral point, and the first peripheral point is a reference peripheral point in a second identification image obtained by binarization processing by using a lower binarization threshold value; the second peripheral point is a peripheral point other than the reference peripheral point among peripheral points in the second recognition map obtained by the binarization processing using the higher binarization threshold value.
With reference to the first aspect, an embodiment of the present invention provides a sixth possible implementation manner of the first aspect, where the performing semantic analysis on each character recognition result by using a doctor semantic analysis model and a patient semantic analysis model respectively to determine the second electronic version history data includes:
analyzing each character recognition result by using a doctor semantic analysis model to obtain a plurality of first analysis results;
analyzing each character recognition result by using a patient semantic analysis model to obtain a plurality of second analysis results;
comparing the first analysis result with a doctor semantic standard template in a standard database to obtain a second similarity; comparing the first analysis result with a patient semantic standard template in a standard database to obtain a third similarity;
comparing the second analysis result with the doctor semantic standard template in the standard database to obtain a fourth similarity; comparing the second analysis result with a patient semantic standard template in a standard database to obtain a fifth similarity;
judging whether a first condition and a second condition are met simultaneously, wherein the first condition is that the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity exceeds a preset threshold value; the second condition is that the difference between the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity and the second large value is less than a predetermined threshold;
if yes, determining according to an analysis result corresponding to the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity; if not, readjusting the similarity calculation rule according to the year of the historical data of the paper edition.
With reference to the first aspect, an embodiment of the present invention provides a seventh possible implementation manner of the first aspect, where the medical request is encrypted.
With reference to the first aspect, an embodiment of the present invention provides an eighth possible implementation manner of the first aspect, wherein an update period of the semantic analysis model of the doctor is 1 day.
In a second aspect, an embodiment of the present invention further provides a system for scheduling a medical treatment task, including a system end and a consulting room end;
the system side is used for executing corresponding operations according to the method of the first aspect;
and the clinic end is used for receiving and displaying the medical request, the first electronic edition historical data and the second electronic edition historical data.
According to the intelligent scheduling method for the hospitalizing tasks, provided by the embodiment of the invention, the recommended hospitalizing subjects are obtained by analyzing the historical data provided by the user and the historical data recorded in the system, the appointed hospitals are determined as candidates by combining the requested hospitalizing subjects provided by the patient, then, more reasonable hospitalizing rooms are determined for the user through image recognition and data analysis, and the referred hospitalizing requests, the first electronic edition historical data and the second electronic edition historical data are sent to the consulting rooms, so that the reasonable degree of allocating the consulting rooms for the user is improved to a certain degree.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 shows an architecture diagram of a system to which the method provided herein is applicable;
FIG. 2 is a diagram illustrating specific styles of character templates used in the method provided by the present application;
FIG. 3 is a diagram illustrating a character template with a skeleton used in the method provided by the present application;
fig. 4 shows a schematic diagram of a character template used in the method provided by the present application, which contains skeleton points and peripheral points.
Detailed Description
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 only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
In the related art, a method of registering a user to a designated hospital is usually adopted to assign a designated department to the user for diagnosis and treatment, and a method of making a network appointment for diagnosis and treatment also appears in the subsequent process.
Aiming at the situation, the application provides an intelligent scheduling method for the treatment tasks, which comprises the following steps:
acquiring a medical seeking request of a target user, wherein the medical seeking request carries an identity of the target user, biological authentication data, instant disease description and a medical seeking subject;
searching biological standard data corresponding to the identity in a database;
judging whether the similarity of the biological standard data and the biological authentication data exceeds a preset first threshold value or not;
if so, searching first electronic version historical data corresponding to the identity in a database, wherein the first electronic version historical data comprises diagnosis and treatment time, diagnosis and treatment places, diagnosis and treatment items, diagnosis and treatment processes and diagnosis and treatment results;
acquiring paper edition historical data provided by a target user;
photographing historical data of the paper edition to obtain a historical data photo;
performing character recognition on the historical data photo to obtain second electronic version historical data, wherein the second electronic version historical data comprises diagnosis and treatment time, diagnosis and treatment places, diagnosis and treatment items, diagnosis and treatment processes and diagnosis and treatment results;
determining a first pathological analysis result of a target user according to the first electronic edition historical data;
determining a second pathological analysis result of the target user according to the second electronic edition historical data;
determining a first weight corresponding to a first pathological analysis result according to the continuous degree of the diagnosis and treatment time of the first electronic version historical data and the time period of the diagnosis and treatment time;
determining a second weight corresponding to a second pathological analysis result according to the continuous degree of the diagnosis and treatment time of the second electronic version historical data and the time period of the diagnosis and treatment time;
determining a reference pathological analysis result according to the first pathological analysis result, the second pathological analysis result, the first weight and the second weight in a weighted averaging mode, wherein the reference pathological analysis result comprises the disease attack time, the treatment result, the treatment process, the treatment period and the medical treatment subject each time;
searching the diagnosis and treatment files with the matching degree with the reference pathological analysis result exceeding a third threshold value in the database as candidate diagnosis and treatment files in a big data analysis mode;
judging whether the number of the candidate diagnosis and treatment files exceeds a preset second threshold value, if not, reducing a third threshold value, and re-executing the step of searching the diagnosis and treatment files with the matching degree with the reference pathological analysis result exceeding the third threshold value in the database as the candidate diagnosis and treatment files; if yes, calculating the diagnosis and treatment success probability of each candidate diagnosis and treatment file according to the treatment period and the treatment result in each candidate diagnosis and treatment file;
determining recommended medical subjects according to the diagnosis and treatment success probability;
determining a first medical emergency degree according to the first pathological analysis result;
determining a second medical emergency degree according to the second pathological analysis result and the instant disease description;
if the recommended medical subject is the same as the requested medical subject, searching a first target hospital in the database according to the queuing condition and the diagnosis and treatment success rate of the target diagnosis room in the second hospital corresponding to the requested medical subject according to a first search rule; the target consulting room corresponds to a recommended medical subject;
if the recommended medical treatment subject is different from the requested medical treatment subject, judging whether at least one of the first medical treatment emergency degree or the second medical treatment emergency degree exceeds a preset threshold value; if yes, sending prompt information to the target user; if not, searching the first target hospital in the database according to the queuing condition, the diagnosis and treatment success rate and the first medical treatment emergency degree of the target diagnosis room in the first hospital corresponding to the medical treatment requesting subject according to the second searching rule;
calling video data of a plurality of target consulting rooms in a first target hospital;
extracting multiple frames of images from the video data according to the time sequence, wherein the interval between two adjacent images is preset time length;
according to the similarity of the multi-frame images, the actual queuing degree of each target consulting room;
determining the theoretical queuing degree of each target consulting room according to the acquired reservation condition of the target consulting room;
determining a target medical consulting room of a target user according to the actual queuing degree and the theoretical queuing degree;
and sending the medical request, the first electronic version historical data and the second electronic version historical data to a consulting room end in the target medical consulting room.
The target medical seeking request is usually sent by a user through a device controlled by the user, wherein the device is an intelligent electronic device such as a PC (personal computer), a mobile terminal and the like.
The identity, the biometric authentication data, the instant medical description and the requested medical subject carried in the medical request are provided by the user, and specifically, the system can automatically display forms and guide information on a device operated by the user so as to facilitate the targeted filling of the user.
The identification may be identification card information, identification code, etc. of the user, which can distinguish the user from other users. The biometric authentication data such as fingerprint data or iris data can be collected by the user through the terminal and uploaded to the system. The instant medical statement is usually provided by the user, but it should be understood that the instant medical statement is only used as a reference because the opinion provided by the user is not sufficiently accurate. Similarly, the requested medical subjects are also provided by the user with the same lesser accuracy.
The biometric standard data corresponds to the biometric authentication data, and the type of the two should be the same.
It is determined whether the similarity of the biometric standard data and the biometric authentication data exceeds a preset first threshold, i.e. to see whether the two data are sufficiently similar, and if so, the subsequent steps may be performed.
The first electronic version history data refers to the recorded condition when the user visits a doctor, and compared with the second electronic version history data, the first electronic version history data is usually directly recorded into the system by a hospital in an electronic information mode, and the second electronic version history data is obtained by recording on a medical record held by the user when the user visits a doctor.
The pathological analysis result refers to the process and result of a specific diagnosis.
And searching diagnosis and treatment files with matching degree meeting preset requirements with the reference pathological analysis result in the database as candidate diagnosis and treatment files by adopting a big data analysis mode. This step is actually to find out in the database which users have similar diagnosis processes to the current user, and if they are similar, it can be used as a reference. For example, a similar diagnosis and treatment process of a patient with successful treatment can be selected as a reference, so as to improve the success rate.
The recommended medical subjects are the results obtained after theoretical analysis is carried out according to big data and models. The user's own choice should also be considered in place, i.e. the requesting medical subject should also be considered. Therefore, the final calculation is performed according to a specific algorithm, and the weight values may be adjusted accordingly.
Furthermore, it is not accurate enough to determine the number of hospitalizations in the office simply by means of the reservation system of the hospital. Therefore, in the scheme, a mode of video capture by a camera is adopted, the actual number of people in the consulting room is determined through image comparison, and the queuing time for seeking medical treatment in the consulting room is more accurately judged by combining the number of reserved people. And finally, the required medical treatment request, the first electronic edition historical data and the second electronic edition historical data are sent to a medical treatment room end in the target medical treatment room. The clinic end is intelligent equipment arranged in the target medical consulting room.
When the similarity of multiple frames of images is identified, the moving foreground images in the images can be determined in a specific pixel comparison mode, and the number of pedestrians in the foreground images can be determined by performing feature analysis on the foreground images.
Preferably, searching the diagnosis and treatment archive with the matching degree with the reference pathological analysis result meeting the preset requirement in the database as the candidate diagnosis and treatment archive comprises:
calculating a first similarity of each diagnosis and treatment file and a reference pathological analysis result according to a preset first algorithm, and calculating a treatment rational degree value of each diagnosis and treatment file according to a preset second algorithm;
grouping all diagnosis and treatment files according to diagnosis and treatment places of the diagnosis and treatment files, wherein the diagnosis and treatment places of the diagnosis and treatment files in each group of the diagnosis and treatment files are the same;
calculating the maximum likelihood estimation value of each file group by adopting a maximum likelihood estimation mode according to the first similarity and the treatment reasonable degree value;
and determining the diagnosis and treatment files in one or more groups as candidate diagnosis and treatment files according to the maximum likelihood estimation value of each group.
The purpose of using the first algorithm is to determine the diagnosis and treatment file which is most similar to the reference pathological analysis result, the reasonable degree value refers to the value of the diagnosis and treatment file, the factors influencing the treatment reasonable degree value mainly comprise the treatment result and the disease condition, the better the treatment result is, the higher the treatment reasonable degree value is, and the more similar the disease condition is, the higher the treatment reasonable degree value is.
The main purpose of grouping the diagnosis and treatment files is to divide areas/hospitals, and if there are files in different areas in the file group, it is meaningless to estimate the probability, and it is still impossible to determine where to visit the doctor. Therefore, the maximum likelihood estimation of the profile sets is determined in the subsequent process by means of maximum likelihood estimation, i.e. which profile set has the better result (the value can represent the treatment success rate). And finally, determining the most ideal file as a candidate diagnosis and treatment file.
Preferably, the step of determining the diagnosis and treatment documents in one or more designated groups as candidate diagnosis and treatment documents according to the maximum likelihood estimation value of each group includes:
if the distribution of the maximum likelihood estimation values of all candidate diagnosis and treatment files meets the preset condition, taking the diagnosis and treatment files in the diagnosis and treatment file group with the maximum likelihood estimation value exceeding the preset threshold value as the candidate diagnosis and treatment files;
if the distribution of the maximum likelihood estimation values of all candidate diagnosis and treatment files does not accord with the preset condition, calculating the first similarity of each diagnosis and treatment file and the reference pathological analysis result according to the preset first algorithm and calculating the treatment reasonable degree value of each diagnosis and treatment file according to the preset second algorithm according to the calculation conditions of the readjustment first algorithm and the second algorithm.
Whether the distribution of the maximum likelihood estimated values of all candidate diagnosis and treatment files meets the preset condition is judged, whether the maximum likelihood estimated values are similar to the historical records is mainly judged, and because the success rate of a certain area or hospital or the ratio of success probabilities among different hospitals is not excessively changed generally, whether previous calculation is wrong or not can be distinguished by means of the historical records.
It can also be judged according to the relative size of the numerical value, mainly because the medical conditions between different hospitals of the same type (like a region/the same level) are basically the same, and therefore, the calculated result should be basically the same. Therefore, whether the calculation result is accurate or not can be verified in this way.
Preferably, the character recognition of the historical data photo to obtain the second electronic edition historical data comprises:
carrying out binarization processing on the historical data photo at least twice according to different binarization threshold values to obtain first identification graphs corresponding to the different binarization threshold values;
respectively carrying out character recognition on the character blocks on each first recognition graph according to the following mode to determine the character recognition result of each first recognition graph: respectively superposing the first template drawing with the character block on each first recognition drawing so as to determine the nearest reference distance between each first reference point on the character block and the character skeleton point in the template drawing, and determining the character block according to the nearest reference distance and the weight corresponding to each character skeleton point; the first reference point is a point at which the gradation value exceeds a predetermined threshold; the first template picture is obtained after big data analysis is carried out according to a doctor note database.
Calling a doctor semantic analysis model and a patient semantic analysis model from a doctor semantic analysis database and a patient semantic analysis database respectively;
and performing semantic analysis on each character recognition result by using a doctor semantic analysis model and a patient semantic analysis model respectively to determine second electronic version historical data.
Preferably, the character skeleton points include skeleton points and peripheral points, the weight of the skeleton points is the highest of all the reference points, and all the skeleton points form the basic shape of the graph in the first template graph; the weight of the peripheral point is in negative correlation with the target distance, and the target distance is the distance between the peripheral point and the nearest skeleton point.
Preferably, the method provided by the present application further comprises:
determining the weight value of each peripheral point according to the following modes:
carrying out binarization processing on the target candidate images for multiple times respectively according to different binarization threshold values to obtain second identification images corresponding to the different binarization threshold values;
assigning a weight to each peripheral point according to the following mode, wherein the weight of a first peripheral point is higher than that of a second peripheral point, and the first peripheral point is a reference peripheral point in a second identification image obtained by binarization processing by using a lower binarization threshold value; the second peripheral point is a peripheral point other than the reference peripheral point among peripheral points in the second recognition map obtained by the binarization processing using the higher binarization threshold value.
As shown in fig. 2, a specific style of the template is shown, the template may be any number, but considering the possible characteristics of the handwriting of the doctor, a standard character library may be established in advance for the doctor, and the characters in the standard character library may be used as the template.
As shown in fig. 3, the skeleton region refers to the innermost small region in the graph, and the non-skeleton region is the foreground region except for the skeleton region. Meanwhile, the skeleton region should be able to correctly reflect the basic shape of the figure, as shown in fig. 3, the white part inside the figure 6 is its skeleton, the figure presented by the skeleton is still 6, and the skeleton is contained inside the non-skeleton region. The recognition result map may be divided into two parts, one part is a skeleton region, and the other part is a peripheral region, where the peripheral region includes a non-skeleton region and a background region.
As shown in fig. 4, which shows a graph in which the template map (left map) is converted into skeleton points and a peripheral point map (right map), in fig. 4, points closer to the skeleton area in fig. 3 have higher weight values. The areas of the image obtained by using different thresholds for binarization are different, and further, the weight value can be given to the peripheral point in the newly appeared area through the step-type binarization mode.
Preferably, the step of performing semantic analysis on each character recognition result by using a doctor semantic analysis model and a patient semantic analysis model respectively to determine the second electronic version history data comprises:
analyzing each character recognition result by using a doctor semantic analysis model to obtain a plurality of first analysis results;
analyzing each character recognition result by using a patient semantic analysis model to obtain a plurality of second analysis results;
comparing the first analysis result with a doctor semantic standard template in a standard database to obtain a second similarity; comparing the first analysis result with a patient semantic standard template in a standard database to obtain a third similarity;
comparing the second analysis result with the doctor semantic standard template in the standard database to obtain a fourth similarity; comparing the second analysis result with a patient semantic standard template in a standard database to obtain a fifth similarity;
judging whether a first condition and a second condition are met simultaneously, wherein the first condition is that the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity exceeds a preset threshold value; the second condition is that the difference between the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity and the second large value is less than a predetermined threshold;
if yes, determining according to an analysis result corresponding to the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity; if not, readjusting the similarity calculation rule according to the year of the historical data of the paper edition.
When semantic analysis is performed, the purpose of setting the two semantic models is to more accurately analyze words and sentences. This is mainly because the professional words used by the doctor are many, if the semantic model is not established separately, the result obtained by the analysis may be paradoxical, and because the patient may come from different regions and the dialects of different regions have certain differences, the semantic model may be adjusted appropriately according to the differences of the regions, or the result of the semantic analysis of the doctor is mainly used in the final judgment process.
The reason why the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity exceeds the predetermined threshold is mainly to investigate whether the highest similarity is reliable enough, and if the maximum value of the similarity is still low, the similarity is not reliable and the result cannot be adopted.
Similarly, because there are the similarity of the semantic analysis results of two doctors and the similarity of the semantic analysis results of two patients, the similarity of the semantic analysis results of the same type should be consistent, and the condition that one similarity is very high and the rest are very low is not caused, so that the characteristics can be used for judging the semantic analysis results.
Preferably, the medical request is encrypted.
Preferably, the update period of the semantic analysis model of the doctor is 1 day.
Based on the method, the application also provides a diagnosis task scheduling system, which comprises a system end and a consulting room end;
the system end is used for executing corresponding operation according to the method;
and the clinic end is used for receiving and displaying the medical request, the first electronic edition historical data and the second electronic edition historical data.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions 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 (10)
1. An intelligent scheduling method for a doctor seeing task is characterized by comprising the following steps:
acquiring a medical seeking request of a target user, wherein the medical seeking request carries an identity of the target user, biological authentication data, instant disease description and a medical seeking subject;
searching biological standard data corresponding to the identity in a database;
judging whether the similarity of the biological standard data and the biological authentication data exceeds a preset first threshold value or not;
if so, searching first electronic version historical data corresponding to the identity in a database, wherein the first electronic version historical data comprises diagnosis and treatment time, diagnosis and treatment places, diagnosis and treatment items, diagnosis and treatment processes and diagnosis and treatment results;
acquiring paper edition historical data provided by a target user;
photographing historical data of the paper edition to obtain a historical data photo;
performing character recognition on the historical data photo to obtain second electronic version historical data, wherein the second electronic version historical data comprises diagnosis and treatment time, diagnosis and treatment places, diagnosis and treatment items, diagnosis and treatment processes and diagnosis and treatment results;
determining a first pathological analysis result of a target user according to the first electronic edition historical data;
determining a second pathological analysis result of the target user according to the second electronic edition historical data;
determining a first weight corresponding to a first pathological analysis result according to the continuous degree of the diagnosis and treatment time of the first electronic version historical data and the time period of the diagnosis and treatment time;
determining a second weight corresponding to a second pathological analysis result according to the continuous degree of the diagnosis and treatment time of the second electronic version historical data and the time period of the diagnosis and treatment time;
determining a reference pathological analysis result according to the first pathological analysis result, the second pathological analysis result, the first weight and the second weight in a weighted averaging mode, wherein the reference pathological analysis result comprises the disease attack time, the treatment result, the treatment process, the treatment period and the medical treatment subject each time;
searching the diagnosis and treatment files with the matching degree with the reference pathological analysis result exceeding a third threshold value in the database as candidate diagnosis and treatment files in a big data analysis mode;
judging whether the number of the candidate diagnosis and treatment files exceeds a preset second threshold value, if not, reducing a third threshold value, and re-executing the step of searching the diagnosis and treatment files with the matching degree with the reference pathological analysis result exceeding the third threshold value in the database as the candidate diagnosis and treatment files; if yes, calculating the diagnosis and treatment success probability of each candidate diagnosis and treatment file according to the treatment period and the treatment result in each candidate diagnosis and treatment file;
determining recommended medical subjects according to the diagnosis and treatment success probability;
determining a first medical emergency degree according to the first pathological analysis result;
determining a second medical emergency degree according to the second pathological analysis result and the instant disease description;
if the recommended medical subject is the same as the requested medical subject, searching a first target hospital in the database according to the queuing condition and the diagnosis and treatment success rate of the target diagnosis room in the second hospital corresponding to the requested medical subject according to a first search rule; the target consulting room corresponds to a recommended medical subject;
if the recommended medical treatment subject is different from the requested medical treatment subject, judging whether at least one of the first medical treatment emergency degree or the second medical treatment emergency degree exceeds a preset threshold value; if yes, sending prompt information to the target user; if not, searching the first target hospital in the database according to the queuing condition, the diagnosis and treatment success rate and the first medical treatment emergency degree of the target diagnosis room in the first hospital corresponding to the medical treatment requesting subject according to the second searching rule;
calling video data of a plurality of target consulting rooms in a first target hospital;
extracting multiple frames of images from the video data according to the time sequence, wherein the interval between two adjacent images is preset time length;
determining the actual queuing degree of each target consulting room according to the similarity of the multi-frame images;
determining the theoretical queuing degree of each target consulting room according to the acquired reservation condition of the target consulting room;
determining a target medical consulting room of a target user according to the actual queuing degree and the theoretical queuing degree;
and sending the medical request, the first electronic version historical data and the second electronic version historical data to a consulting room end in the target medical consulting room.
2. The method according to claim 1, wherein the step of searching the diagnosis and treatment archive with matching degree meeting the preset requirement with the reference pathological analysis result in the database as the candidate diagnosis and treatment archive comprises:
calculating a first similarity of each diagnosis and treatment file and a reference pathological analysis result according to a preset first algorithm, and calculating a treatment rational degree value of each diagnosis and treatment file according to a preset second algorithm;
grouping all diagnosis and treatment files according to diagnosis and treatment places of the diagnosis and treatment files, wherein the diagnosis and treatment places of the diagnosis and treatment files in each group of the diagnosis and treatment files are the same;
calculating the maximum likelihood estimation value of each file group by adopting a maximum likelihood estimation mode according to the first similarity and the treatment reasonable degree value;
and determining the diagnosis and treatment files in one or more groups as candidate diagnosis and treatment files according to the maximum likelihood estimation value of each group.
3. The method of claim 2, wherein the step of determining the one or more assigned groups of medical profiles as candidate medical profiles based on the maximum likelihood estimates for each group comprises:
if the distribution of the maximum likelihood estimation values of all candidate diagnosis and treatment files meets the preset condition, taking the diagnosis and treatment files in the diagnosis and treatment file group with the maximum likelihood estimation value exceeding the preset threshold value as the candidate diagnosis and treatment files;
if the distribution of the maximum likelihood estimation values of all candidate diagnosis and treatment files does not accord with the preset condition, calculating the first similarity of each diagnosis and treatment file and the reference pathological analysis result according to the preset first algorithm and calculating the treatment reasonable degree value of each diagnosis and treatment file according to the preset second algorithm according to the calculation conditions of the readjustment first algorithm and the second algorithm.
4. The method of claim 1, wherein performing text recognition on the historical data photo to obtain a second electronic version of the historical data comprises:
carrying out binarization processing on the historical data photo at least twice according to different binarization threshold values to obtain first identification graphs corresponding to the different binarization threshold values;
respectively carrying out character recognition on the character blocks on each first recognition graph according to the following mode to determine the character recognition result of each first recognition graph: respectively superposing the first template drawing with the character block on each first recognition drawing so as to determine the nearest reference distance between each first reference point on the character block and the character skeleton point in the template drawing, and determining the character block according to the nearest reference distance and the weight corresponding to each character skeleton point; the first reference point is a point at which the gradation value exceeds a predetermined threshold; the first template picture is obtained after big data analysis is carried out according to a doctor note database;
calling a doctor semantic analysis model and a patient semantic analysis model from a doctor semantic analysis database and a patient semantic analysis database respectively;
and performing semantic analysis on each character recognition result by using a doctor semantic analysis model and a patient semantic analysis model respectively to determine second electronic version historical data.
5. The method according to claim 4, wherein the character skeleton points comprise skeleton points and peripheral points, the weight of the skeleton points is the highest among all reference points, and all the skeleton points form the basic shape of the graph in the first template graph; the weight of the peripheral point is in negative correlation with the target distance, and the target distance is the distance between the peripheral point and the nearest skeleton point.
6. The method of claim 5, further comprising:
determining the weight value of each peripheral point according to the following modes:
carrying out binarization processing on the target candidate images for multiple times respectively according to different binarization threshold values to obtain second identification images corresponding to the different binarization threshold values;
assigning a weight to each peripheral point according to the following mode, wherein the weight of a first peripheral point is higher than that of a second peripheral point, and the first peripheral point is a reference peripheral point in a second identification image obtained by binarization processing by using a lower binarization threshold value; the second peripheral point is a peripheral point other than the reference peripheral point among peripheral points in the second recognition map obtained by the binarization processing using the higher binarization threshold value.
7. The method of claim 4, wherein the step of performing semantic analysis using a physician semantic analysis model and a patient semantic analysis model for each of the text recognition results to determine the second electronic version of the historical data comprises:
analyzing each character recognition result by using a doctor semantic analysis model to obtain a plurality of first analysis results;
analyzing each character recognition result by using a patient semantic analysis model to obtain a plurality of second analysis results;
comparing the first analysis result with a doctor semantic standard template in a standard database to obtain a second similarity; comparing the first analysis result with a patient semantic standard template in a standard database to obtain a third similarity;
comparing the second analysis result with the doctor semantic standard template in the standard database to obtain a fourth similarity; comparing the second analysis result with a patient semantic standard template in a standard database to obtain a fifth similarity;
judging whether a first condition and a second condition are met simultaneously, wherein the first condition is that the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity exceeds a preset threshold value; the second condition is that the difference between the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity and the second large value is less than a predetermined threshold;
if yes, determining according to an analysis result corresponding to the maximum value of the second similarity, the third similarity, the fourth similarity and the fifth similarity; if not, readjusting the similarity calculation rule according to the year of the historical data of the paper edition.
8. The method of claim 1,
the medical request is encrypted.
9. The method of claim 4,
the updating period of the semantic analysis model of the doctor is 1 day.
10. A system for scheduling a treatment task comprises a system end and a treatment room end;
the system end is used for executing corresponding operations according to the method of any one of claims 1 to 9;
and the clinic end is used for receiving and displaying the medical request, the first electronic edition historical data and the second electronic edition historical data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810050316.XA CN108461130B (en) | 2018-01-18 | 2018-01-18 | Intelligent scheduling method and system for treatment tasks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810050316.XA CN108461130B (en) | 2018-01-18 | 2018-01-18 | Intelligent scheduling method and system for treatment tasks |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108461130A CN108461130A (en) | 2018-08-28 |
| CN108461130B true CN108461130B (en) | 2020-09-18 |
Family
ID=63220805
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810050316.XA Active CN108461130B (en) | 2018-01-18 | 2018-01-18 | Intelligent scheduling method and system for treatment tasks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108461130B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110164523A (en) * | 2019-05-21 | 2019-08-23 | 山东大学 | A kind of intelligent health analysis method and system with intelligence function |
| CN110516161B (en) * | 2019-08-30 | 2021-06-01 | 深圳前海微众银行股份有限公司 | Recommendation method and device |
| CN112530565A (en) * | 2020-04-18 | 2021-03-19 | 赵芳 | Medical resource allocation method applied to user behavior analysis and cloud computing server |
| CN112613396B (en) * | 2020-12-19 | 2022-10-25 | 河北志晟信息技术股份有限公司 | Task emergency degree processing method and system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20010109031A (en) * | 2000-06-01 | 2001-12-08 | 이창헌 | The medical service providing system through the internet |
| CN102004862A (en) * | 2010-12-14 | 2011-04-06 | 王兴强 | Clinical method and system for patient to acquire right to know medical treatment |
| JP2012009032A (en) * | 2011-07-12 | 2012-01-12 | Takuo Tanaka | Medical diagnosis support system |
| CN103559637A (en) * | 2013-11-13 | 2014-02-05 | 王竞 | Method and system for recommending doctor for patient |
| CN104680458A (en) * | 2015-02-09 | 2015-06-03 | 李宏强 | Method for recommending medical-seeking departments, hospitals or doctors to public based on massive prescriptions |
| CN106326887A (en) * | 2016-08-29 | 2017-01-11 | 东方网力科技股份有限公司 | Method and device for checking optical character recognition result |
| CN107103201A (en) * | 2017-05-10 | 2017-08-29 | 北京大数医达科技有限公司 | Generation method, device and the medical path air navigation aid of medical guidance path |
| CN107463779A (en) * | 2017-07-31 | 2017-12-12 | 合肥桥旭科技有限公司 | A kind of medical information integration system based on Internet of Things |
-
2018
- 2018-01-18 CN CN201810050316.XA patent/CN108461130B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20010109031A (en) * | 2000-06-01 | 2001-12-08 | 이창헌 | The medical service providing system through the internet |
| CN102004862A (en) * | 2010-12-14 | 2011-04-06 | 王兴强 | Clinical method and system for patient to acquire right to know medical treatment |
| JP2012009032A (en) * | 2011-07-12 | 2012-01-12 | Takuo Tanaka | Medical diagnosis support system |
| CN103559637A (en) * | 2013-11-13 | 2014-02-05 | 王竞 | Method and system for recommending doctor for patient |
| CN104680458A (en) * | 2015-02-09 | 2015-06-03 | 李宏强 | Method for recommending medical-seeking departments, hospitals or doctors to public based on massive prescriptions |
| CN106326887A (en) * | 2016-08-29 | 2017-01-11 | 东方网力科技股份有限公司 | Method and device for checking optical character recognition result |
| CN107103201A (en) * | 2017-05-10 | 2017-08-29 | 北京大数医达科技有限公司 | Generation method, device and the medical path air navigation aid of medical guidance path |
| CN107463779A (en) * | 2017-07-31 | 2017-12-12 | 合肥桥旭科技有限公司 | A kind of medical information integration system based on Internet of Things |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108461130A (en) | 2018-08-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108461130B (en) | Intelligent scheduling method and system for treatment tasks | |
| CN112071425B (en) | Data processing method and device, computer equipment and storage medium | |
| EP3076320B1 (en) | Individual identification device, and identification threshold setting method | |
| CN104657705B (en) | Pattern recognition device and data entry method towards pattern recognition device | |
| US8438182B2 (en) | Patient identification | |
| CN111241265A (en) | Information recommendation method, equipment, storage medium and device | |
| Kuru et al. | Biomedical visual data analysis to build an intelligent diagnostic decision support system in medical genetics | |
| CN108461110B (en) | Medical information processing method, device and equipment | |
| CN108197592B (en) | Information acquisition method and device | |
| CN111344715A (en) | Object recognition system and method | |
| CN112634889B (en) | Electronic case input method, device, terminal and medium based on artificial intelligence | |
| CN111104825A (en) | Face registry updating method, device, equipment and medium | |
| CN109829072A (en) | Construct atlas calculation and relevant apparatus | |
| CN112614578A (en) | Doctor intelligent recommendation method and device, electronic equipment and storage medium | |
| KR101653505B1 (en) | Computing apparatus and method for diagnosing status of health based on image | |
| KR101126704B1 (en) | Online client diagnosis system and method thereof | |
| CN115588492A (en) | Diagnosis guiding method and system based on artificial intelligence | |
| CN115662562A (en) | Medical record diagnosis and treatment data management method, device, equipment and storage medium | |
| KR20220000851A (en) | Dermatologic treatment recommendation system using deep learning model and method thereof | |
| CN113707304A (en) | Triage data processing method, device, equipment and storage medium | |
| CN112801994A (en) | Bone age assessment method and system | |
| KR20080109425A (en) | Methods and system for extracting facial features and verifying sasang constitution through image recognition | |
| CN116072303B (en) | Medical information card data identification system and method for hospital information department | |
| CN114722977B (en) | Medical object classification method and device, electronic equipment and storage medium | |
| US20240112146A1 (en) | Work management device, work management method, and recording medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |