CN113990514A - Abnormality detection device for doctor diagnosis and treatment behavior, computer device and storage medium - Google Patents
Abnormality detection device for doctor diagnosis and treatment behavior, computer device and storage medium Download PDFInfo
- Publication number
- CN113990514A CN113990514A CN202111262634.0A CN202111262634A CN113990514A CN 113990514 A CN113990514 A CN 113990514A CN 202111262634 A CN202111262634 A CN 202111262634A CN 113990514 A CN113990514 A CN 113990514A
- Authority
- CN
- China
- Prior art keywords
- medical record
- sample
- diagnosis
- treatment
- samples
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
Abstract
The embodiment of the application belongs to the field of artificial intelligence and digital medical treatment, and relates to an abnormity detection device, computer equipment and a storage medium for diagnosis and treatment behaviors of doctors, wherein the device comprises: the device comprises a sample set acquisition module, an information extraction module, a sample clustering module, a sample acquisition module, a category determination module, a specificity calculation module and an information generation module, wherein the device is used for extracting information of a plurality of medical record samples to obtain optimized medical record samples, and clustering the medical record samples to obtain optimized medical record samples of various clustering categories; acquiring a medical record sample to be detected related to a target doctor; determining the cluster type of the medical record sample to be detected; calculating a diagnosis and treatment specificity evaluation value of the medical record sample to be detected according to the clustering category; and generating diagnosis and treatment abnormity detection information of the target doctor according to the diagnosis and treatment specificity evaluation value. In addition, the present application also relates to a blockchain technique, and the medical record sample sets can be stored in the blockchain. The method and the device automatically realize the abnormity detection of the diagnosis and treatment behaviors of doctors.
Description
Technical Field
The present application relates to the field of digital medical technology, and in particular, to an abnormality detection apparatus, a computer device, and a storage medium for a doctor diagnosis and treatment behavior.
Background
With the development of computer technology, medical institutions such as hospitals increasingly use computers for fine management. Doctors are important supports of hospitals, and need to be examined and detected on a plurality of business levels such as medical insurance, medical health commission, internal management of hospitals and the like, wherein the diagnosis and treatment behaviors and the diagnosis and treatment process of the doctors are more important in examination and detection. However, the examination and the detection of the diagnosis and treatment behaviors of the doctor are mainly realized manually at present, and the examination and the detection depend on the experience of examination personnel, so that the detection effect is different, and the efficiency is lower.
Disclosure of Invention
An object of the embodiments of the present application is to provide an abnormality detection apparatus for a physician diagnosis and treatment behavior, a computer device, and a storage medium, so as to implement abnormality detection for the physician diagnosis and treatment behavior.
In order to solve the above technical problem, an embodiment of the present application further provides an abnormality detection apparatus for a doctor diagnosis and treatment behavior, which adopts the following technical scheme:
the system comprises a sample set acquisition module, a sample set acquisition module and a sample set analysis module, wherein the sample set acquisition module is used for acquiring a medical record sample set comprising a plurality of medical record samples;
the information extraction module is used for extracting information of each medical record sample in the medical record sample set according to a preset information dimension to obtain each optimized medical record sample;
the sample clustering module is used for clustering the optimized medical record samples to obtain optimized medical record samples of various clustering types;
the system comprises a sample acquisition module, a sample analysis module and a sample analysis module, wherein the sample acquisition module is used for acquiring a medical record sample to be detected related to a target doctor;
the category determination module is used for calculating the attribution degree of the medical record samples to be detected to each clustering category optimization medical record sample so as to determine the clustering category to which the medical record samples to be detected belong according to the attribution degree;
the specificity calculation module is used for calculating the diagnosis and treatment specificity evaluation value of the medical record sample to be detected according to the medical record sample to be detected and the determined clustering category;
and the information generating module is used for generating diagnosis and treatment abnormity detection information of the target doctor according to the diagnosis and treatment specificity evaluation value.
In order to solve the technical problem, an embodiment of the present application further provides a computer device, which includes a memory and a processor, where the memory stores computer-readable instructions, and the processor implements functions of each module and/or unit in the abnormality detection apparatus for a physician diagnosis and treatment behavior when executing the computer-readable instructions.
In order to solve the technical problem, an embodiment of the present invention further provides a computer-readable storage medium, where computer-readable instructions are stored on the computer-readable storage medium, and when the computer-readable instructions are executed by a processor, the computer-readable instructions implement the functions of the modules and/or units in the abnormality detection apparatus for physician diagnosis and treatment behavior.
Compared with the prior art, the embodiment of the application mainly has the following beneficial effects: the abnormity detection device for the diagnosis and treatment behaviors of physicians comprises a sample set acquisition module, an information extraction module, a sample clustering module, a sample acquisition module, a category determination module, a specificity calculation module and an information generation module, wherein after the device acquires a medical record sample set, information required for clustering is extracted from each medical record sample in the medical record sample set according to preset information dimensions to obtain a plurality of optimized medical record samples; clustering is carried out according to the commonalities among the optimized medical record samples by adopting a clustering algorithm to obtain optimized medical record samples of various clustering types; acquiring a medical record sample to be detected associated with a target doctor, optimizing the attribution degree of the medical record sample according to each cluster type, determining the cluster type to which the medical record sample belongs, and then calculating the difference between the medical record sample to be detected and the central point of the cluster type to obtain a diagnosis and treatment specificity evaluation value; the diagnosis and treatment specific evaluation value measures the fluctuation of diagnosis and treatment behaviors of doctors in a numerical form, and diagnosis and treatment abnormity monitoring information can be generated according to the diagnosis and treatment specific evaluation value, so that abnormity detection of the diagnosis and treatment behaviors of the doctors is automatically realized.
Drawings
In order to more clearly illustrate the solution of the present application, the drawings needed for describing the embodiments of the present application will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and that other drawings can be obtained by those skilled in the art without inventive effort.
FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;
fig. 2 is a schematic structural view of an embodiment of an abnormality detection apparatus for medical actions of physicians according to the present application;
FIG. 3 is a schematic block diagram of one embodiment of a computer device according to the present application.
Detailed Description
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "including" and "having," and any variations thereof, in the description and claims of this application and the description of the above figures are intended to cover non-exclusive inclusions. The terms "first," "second," and the like in the description and claims of this application or in the above-described drawings are used for distinguishing between different objects and not for describing a particular order.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.
As shown in fig. 1, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 may have various communication client applications installed thereon, such as a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like.
The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture experts Group Audio Layer III, mpeg compression standard Audio Layer 3), MP4 players (Moving Picture experts Group Audio Layer IV, mpeg compression standard Audio Layer 4), laptop portable computers, desktop computers, and the like.
The server 105 may be a server providing various services, such as a background server providing support for pages displayed on the terminal devices 101, 102, 103.
The abnormality detection device for doctor diagnosis and treatment provided in the embodiment of the present application is generally provided in a server.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
With continued reference to fig. 2, a schematic structural diagram of an abnormality detection apparatus of physician's clinical behavior according to the present application is shown, which may include: a sample set obtaining module 201, an information extracting module 202, a sample clustering module 203, a sample obtaining module 204, a category determining module 205, a specificity calculating module 206 and an information generating module 207, wherein:
a sample set acquiring module 201, configured to acquire a medical record sample set including a plurality of medical record samples.
In this embodiment, an electronic device (for example, a server shown in fig. 1) on which the abnormality detection apparatus for doctor clinical behaviors operates may communicate with the terminal through a wired connection manner or a wireless connection manner. It should be noted that the wireless connection means may include, but is not limited to, a 3G/4G connection, a WiFi connection, a bluetooth connection, a WiMAX connection, a Zigbee connection, a uwb (ultra wideband) connection, and other wireless connection means now known or developed in the future.
Specifically, when abnormality detection of diagnosis and treatment behaviors of doctors is performed, a medical record sample set needs to be acquired first. The server can read the sample set of medical records from a specific storage space, such as a database.
It is emphasized that, to further ensure the privacy and security of the medical record sample sets, the medical record sample sets may also be stored in nodes of a block chain.
The block chain referred by the application is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
The information extraction module 202 is configured to perform information extraction on each medical record sample in the medical record sample set according to a preset information dimension, so as to obtain each optimized medical record sample.
Specifically, the medical record sample set comprises a plurality of medical record samples, the medical record samples comprise a plurality of information dimensions, and some information dimensions are irrelevant to clustering. Therefore, information extraction can be performed on each medical record sample according to the preset information dimension, and medical record information used for clustering is extracted from the medical record samples to obtain a plurality of optimized medical record samples.
And the sample clustering module 203 is used for clustering the optimized medical record samples to obtain optimized medical record samples of multiple clustering types.
Specifically, after the optimized medical record samples are obtained, the optimized medical record samples are clustered, and the optimized medical record samples are substantially clustered according to the characteristics of the specific information dimension, wherein the clustering can be unsupervised clustering. The present application does not directly classify the categories according to specific diseases, and it is considered that even the same disease may be different due to differences between patients in actual diagnosis and treatment. Therefore, compared with the classification of the disease class as the only element, the clustering scheme in the application has higher accuracy. After clustering is completed, optimized medical record samples of a plurality of clustering categories can be obtained, and each clustering category comprises a certain number of optimized medical record samples.
A sample acquiring module 204, configured to acquire a medical record sample to be detected associated with a target physician.
Specifically, after the clustering is completed, a medical record sample to be detected is obtained, where the medical record sample to be detected can be generated after a target physician treats a certain patient. And detecting whether the diagnosis and treatment behaviors of the target doctor are abnormal or not through the medical record sample to be detected.
The category determining module 205 is configured to calculate an attribution degree of the medical record sample to be detected to each cluster category optimized medical record sample, so as to determine the cluster category to which the medical record sample to be detected belongs according to the attribution degree.
Specifically, after the medical record sample to be detected is obtained, the cluster type to which the medical record sample to be detected belongs needs to be determined first. The cluster type to which the medical record sample to be detected belongs can also be determined through a clustering algorithm. In an embodiment, before clustering medical record samples to be detected, information extraction can also be performed on the medical record samples to be detected according to a preset information dimension; and then detecting whether the medical record sample to be detected after the information extraction has a missing value or an abnormal value, and performing clustering related processing on the medical record sample to be detected when the missing value or the abnormal value does not exist.
And calculating the attribution degree of the medical record samples to be detected to each cluster type optimization medical record sample through a clustering algorithm, and selecting the cluster type corresponding to the maximum attribution degree as the cluster type to which the medical record samples to be detected belong.
And the specificity calculation module 206 is configured to calculate a diagnosis and treatment specificity evaluation value of the medical record sample to be detected according to the medical record sample to be detected and the determined cluster type.
The diagnosis and treatment specificity evaluation value can be a numerical value and is used for measuring the difference between the medical record sample to be detected and the cluster type to which the medical record sample belongs.
Specifically, after the cluster type to which the medical record sample to be detected belongs is determined, the difference between the medical record sample to be detected and the cluster type can be calculated through a clustering algorithm, so that the diagnosis and treatment specificity evaluation value of the medical record sample to be detected is obtained. For example, the distance between the medical record sample to be detected and the determined cluster category can be calculated by a K-means clustering algorithm (K-means clustering algorithm, which is a clustering analysis algorithm for iterative solution) as the diagnosis and treatment specificity evaluation value.
And an information generating module 207, configured to generate diagnosis and treatment abnormality detection information of the target physician according to the diagnosis and treatment specificity evaluation value. Specifically, the diagnosis and treatment specificity evaluation value is used for measuring the difference between the medical record sample to be detected and the cluster type to which the medical record sample belongs, and the larger the diagnosis and treatment specificity evaluation value is, the larger the difference between the medical record sample to be detected and the optimized medical record sample of the cluster type to which the medical record sample belongs is, the larger the diagnosis and treatment behavior fluctuation of a target doctor is, and abnormal diagnosis and treatment behaviors may exist. After the diagnosis and treatment specific evaluation value is calculated, the diagnosis and treatment specific evaluation value can be used as diagnosis and treatment abnormity detection information of a target doctor.
Or, comparing the diagnosis and treatment specific evaluation value with a preset standard evaluation value, wherein the standard evaluation value is at least one, the standard evaluation value can form a plurality of evaluation intervals, and the evaluation intervals divide diagnosis and treatment behaviors into a plurality of states. For example, the states corresponding to the evaluation intervals may include: standard, normal, fluctuating significantly, abnormal. Comparing the diagnosis and treatment specific evaluation value with each standard evaluation value, determining an evaluation interval where the diagnosis and treatment specific evaluation value is located, and generating diagnosis and treatment abnormal detection information according to the state corresponding to the evaluation interval and the diagnosis and treatment specific evaluation value; when the diagnosis and treatment specificity evaluation value is in a standard and normal interval, the diagnosis and treatment behavior of the target doctor is normal; when the diagnosis and treatment specific evaluation value is in an abnormal interval with large fluctuation, the diagnosis and treatment behavior of a doctor is abnormal.
In this embodiment, the abnormality detection device for doctor diagnosis and treatment behavior includes a sample set acquisition module, an information extraction module, a sample clustering module, a sample acquisition module, a category determination module, a specificity calculation module, and an information generation module, and after acquiring a medical record sample set, the device extracts information required for clustering from each medical record sample in the medical record sample set according to a preset information dimension to obtain a plurality of optimized medical record samples; clustering is carried out according to the commonalities among the optimized medical record samples by adopting a clustering algorithm to obtain optimized medical record samples of various clustering types; acquiring a medical record sample to be detected associated with a target doctor, optimizing the attribution degree of the medical record sample according to each cluster type, determining the cluster type to which the medical record sample belongs, and then calculating the difference between the medical record sample to be detected and the central point of the cluster type to obtain a diagnosis and treatment specificity evaluation value; the diagnosis and treatment specific evaluation value measures the fluctuation of diagnosis and treatment behaviors of doctors in a numerical form, and diagnosis and treatment abnormity monitoring information can be generated according to the diagnosis and treatment specific evaluation value, so that abnormity detection of the diagnosis and treatment behaviors of the doctors is automatically realized.
Further, the information extraction module 202 may include: the information extraction submodule and the medical record processing submodule, wherein:
and the information extraction submodule is used for extracting information of all medical record samples in the medical record sample set based on the information dimension of the patient, the diagnosis and treatment information dimension and the diagnosis and treatment effect dimension to obtain a plurality of candidate medical record samples.
And the medical record processing submodule is used for preprocessing the obtained candidate medical record samples to obtain optimized medical record samples.
Specifically, when information extraction is performed on each medical record sample, information extraction can be performed from three large dimensions, namely a patient information dimension, a diagnosis and treatment information dimension and a diagnosis and treatment effect dimension. A variety of specific features may be included in each dimension.
Wherein the characteristics of the patient information dimension are related to basic information of the patient, such as characteristics of age, blood type, presence or absence of allergy, and the like.
The features involved in the clinical information dimension may include: the disease diagnosis result (may be a specific disease type), the disease category (a general category of a specific disease, for example, diabetes belongs to endocrine diseases), disease outcome information, a diagnosis means (for example, blood sugar determination), a diagnosis means (for example, which treatment a patient has performed, which medicine has been taken), and the like.
The features involved in the diagnostic effect dimension may include: the hospitalization cost, the hospitalization time, the adverse outcome of the disease, the presence or absence of a return visit, etc. of the patient.
After information extraction, firstly obtaining a candidate medical record sample, and then preprocessing the candidate medical record sample. The preprocessing can be data cleaning, for example, detecting missing values and abnormal values in the candidate medical record samples, and deleting the candidate medical record samples with the missing values and the abnormal values to obtain optimized medical record samples for clustering.
In this embodiment, information required for clustering is extracted from the patient information dimension, the diagnosis and treatment information dimension, and the diagnosis and treatment effect dimension to obtain candidate medical record samples, and then preprocessing is performed to screen the candidate medical record samples to obtain optimized medical record samples, so that data preparation is performed for clustering.
Further, the sample clustering module 203 may include: a weight addition submodule and a sample clustering submodule, wherein:
and the weight adding submodule is used for adding characteristic weight to each characteristic in each optimized medical record sample.
And the sample clustering submodule is used for clustering the optimized medical record samples with the characteristic weight through a preset clustering algorithm to obtain optimized medical record samples of various clustering categories.
Specifically, feature weights can be added to various features in the optimized medical record sample so as to differentiate different features and strengthen the action weight of a part of features in clustering. The clustering can be realized by a preset clustering algorithm, and in one embodiment, the clustering algorithm adopted when clustering the optimized medical record samples can be a K-means clustering algorithm. When the distance is calculated by adopting a K-means clustering algorithm, the characteristic weight is used as the coefficient of each distance to be added in the calculation, thereby exerting the utility.
The method and the device can adopt a K-means clustering algorithm to perform unsupervised clustering on the optimized medical record samples. The K-means clustering algorithm divides the optimized medical record samples into K clustering categories, namely K clusters, and can calculate the error sum of squares SSE (for each cluster, the sum of squares of the distances from the samples in the cluster to the cluster center is calculated, and then the sum of squares of each cluster is added to obtain the error sum of squares of the clustering results) as a target function. Generally, there are a limited number of types of diseases that are involved by a physician in a department, and thus there is a limited number of types of diseases that are covered by optimized medical record samples. The number of the clustering classes K is usually larger than the number of disease classes covered by the optimized medical record samples, the K can be gradually increased on the basis of the number of the disease classes, SSE corresponding to each K is calculated, when the change of the SSE tends to be flat or the change of the SSE is smaller than a preset fluctuation threshold value, iteration is stopped, and the K at the moment is selected as the number of the final clustering classes.
In this embodiment, a feature weight can be added to each feature in the optimized medical record samples, and the optimized medical record samples with the feature weight are clustered, so that each feature can be differentiated, and the function of part of features in clustering is enhanced.
Further, the weight adding sub-module may include: an identification detection unit, a weight addition unit and a weight determination unit, wherein:
and the identification detection unit is used for detecting whether the scene identification exists or not.
And the weight adding unit is used for adding characteristic weight to each characteristic in each optimized medical record sample according to the scene identifier when the scene identifier exists.
And the weight determining unit is used for determining the feature weight of each feature in each optimized medical record sample based on a preset feature weight algorithm when the scene identifier does not exist.
Specifically, there may be a variety of application scenarios with different trends for abnormal detection of physician diagnosis and treatment behaviors, such as medical insurance, health care committee, and hospital internal management. The features that are biased also differ for different application scenarios. For example, in a medical insurance scenario, one may focus more on features that are dimensionally related to the clinical outcome.
Whether a scene identifier exists can be detected first, and the scene identifier is used for identifying an application scene. In an embodiment, the abnormality detection apparatus for a physician diagnosis and treatment behavior obtains a medical record data set according to the received abnormality detection request, and at this time, it may be detected whether a scene identifier exists in the abnormality detection request. Under different application scenes, the feature weights of various features are preset. And if the scene identification exists, adding preset feature weight to each feature in the optimized medical record sample according to the scene identification.
If the scene identification does not exist, the feature weight of each feature can be calculated through a preset feature weight algorithm. In one embodiment, the feature weight algorithm may be a Relief algorithm that may assign different weights to various features by correlation between certain features and the same class of features. When the Relief algorithm is used, the same initial weight is given to each feature, and then three feature combinations are divided based on the patient information dimension, the diagnosis and treatment information dimension and the diagnosis and treatment effect dimension. Based on the feature combinations, the feature weight of each feature is calculated.
By means of a Relief algorithm, a sample R is randomly selected from any one feature combination D, the nearest sample H of the sample R is searched from D, and the nearest sample M of R is searched from other feature combinations. If R and H are less distant on a feature than R and M, indicating that the feature is useful for distinguishing between the closest neighbors of the same class and a different class, the weight of the feature may be increased, and conversely, the weight of the feature may be decreased. Wherein the distance may be measured in similarity between feature data. And carrying out a plurality of iterations on the processes to finally obtain the weight of each feature, wherein the larger the weight is, the stronger the capability of representing the corresponding feature in classification is.
It is to be understood that, in clustering, a feature weight may not be added to each feature.
In this embodiment, whether a scene identifier exists or not can be detected, a feature weight is added to each feature according to the scene identifier in combination with objective requirements, or a feature weight is added to each feature through a preset feature weight algorithm, so that the adding mode of the feature weight is enriched.
Further, the category determining module 205 may include: the sub-module is selected to extraction submodule, affiliation degree operator module and category, wherein:
and the extraction submodule is used for extracting the patient information and the diagnosis and treatment information from the medical record sample to be detected.
And the attribution degree calculation operator module is used for calculating the attribution degree of the medical record samples to be detected to the optimized medical record samples of each cluster type based on the extracted patient information and diagnosis and treatment information.
And the category selection submodule is used for selecting the cluster category corresponding to the maximum attribution degree as the cluster category to which the medical record sample to be detected belongs.
The attribution degree measures the similarity between the medical record samples to be detected and the optimized medical record samples of each cluster type.
Specifically, the features of the patient information dimension and the diagnosis and treatment information dimension can be extracted from the medical record samples to be detected, and the features of the patient information dimension and the diagnosis and treatment information dimension can be extracted from the optimized medical record samples of each cluster category. And calculating the attribution degree of the medical record samples to be detected to the optimized medical record samples of each cluster type based on the characteristics of the information dimension and the diagnosis and treatment information dimension of the patient.
The attribution degree can be a numerical value, and the larger the numerical value is, the more similar the medical record sample to be detected and the optimized medical record sample in the cluster type represented by the attribution degree are. The cluster category corresponding to the maximum attribution degree can be selected as the cluster category to which the medical record sample to be detected belongs.
Further, the attribution degree operator module may include: a distance calculation unit or a number calculation unit, wherein:
and the distance calculation unit is used for calculating the distance from the medical record sample to be detected to each cluster type optimized medical record sample based on the extracted patient information and the diagnosis and treatment information, and generating the attribution degree of the medical record sample to be detected to each cluster type optimized medical record sample based on the distance.
And the quantity calculating unit is used for calculating the quantity of the samples of the nearest samples of the medical record samples to be detected in the optimized medical record samples of each cluster type based on the extracted patient information and the diagnosis and treatment information, and taking the quantity of the samples as the attribution degree of the medical record samples to be detected to the optimized medical record samples of each cluster type.
Specifically, when the attribution degree is calculated, the central point (i.e. the clustering center) of the optimized medical record sample of each clustering category can be calculated through a K-means clustering algorithm, then the distance from the medical record sample to be detected to each central point is calculated, and the opposite number of the distances is used as the attribution degree. The smaller the distance from the medical record sample to be detected to the center point of which cluster type is, the larger the attribution degree is, and the cluster type corresponding to the maximum attribution degree is selected as the cluster type to which the medical record sample to be detected belongs.
When the attribution degree is calculated, a KNN algorithm (K-nearest neighbor classification algorithm) can be adopted to calculate the distance from the medical record sample to be detected to each optimized medical record sample, the distances are sorted from small to large, the number of optimized medical record samples of each clustering class in the first K distances is counted, and the number is used as the attribution degree of the medical record sample to be detected to each clustering class optimized medical record sample.
No matter the K-means clustering algorithm or the KNN algorithm, the characteristics of the information dimension of the patient and the diagnosis and treatment information dimension are converted into vectors for calculation.
In the embodiment, the distance from the medical record sample to be detected to the central point of each cluster type optimized medical record sample is calculated through a K-means algorithm, or the number of each cluster type optimized medical record sample in the nearest sample of the medical record sample to be detected is calculated through a KNN algorithm, so that the calculation of the attribution degree of the medical record sample to be detected to each cluster type optimized medical record sample is realized.
In this embodiment, the attribution degree of the medical record samples to be detected to each cluster type is calculated based on the patient information and the diagnosis and treatment information, and the cluster type to which the medical record samples to be detected belong can be accurately determined according to the attribution degree.
Further, the specificity calculation module 206 may include: a sample mapping submodule, a vector determination submodule and a distance calculation submodule, wherein:
and the sample mapping submodule is used for mapping the medical record sample to be detected into a first characterization vector.
And the vector determination submodule is used for determining a second characterization vector corresponding to the central point of the clustering category based on each optimized medical record sample in the clustering category.
And the distance calculation submodule is used for calculating the distance between the first characterization vector and the second characterization vector and taking the distance as the specificity evaluation value of the medical record sample to be detected.
Specifically, the features of each dimension in the medical record sample to be detected are mapped into a vector according to a preset quantization rule, so that a first characterization vector is obtained. Similarly, mapping the feature of each dimension in each optimized medical record sample in the cluster category to be a vector, and averaging the vectors generated by each optimized medical record sample to obtain a second characterization vector corresponding to the central point (i.e. the cluster center) of the optimized medical record sample in the cluster category.
And calculating the distance between the first characterization vector and the second characterization vector, specifically the Euclidean distance, and taking the calculated distance as the specificity evaluation value of the medical record sample to be detected. It can be understood that the smaller the distance is, the smaller the distance between the medical record sample to be detected and the central point is, and the smaller the volatility and the difference of the medical record sample to be detected are compared with the cluster type.
In this embodiment, the characterization vectors of the medical record samples to be detected and the cluster category center points are calculated, and the distance between the vectors is calculated, which reflects the distance between the medical record samples to be detected and the center points, and can be used as a diagnosis and treatment specificity evaluation value for measuring the diagnosis and treatment behaviors of physicians.
Further, the apparatus 200 for detecting abnormality of medical treatment behavior of a doctor may further include: early warning generation module and information transmission module, wherein:
and the early warning generation module is used for generating early warning information according to the medical record sample to be detected, the cluster type of the medical record sample to be detected and the diagnosis and treatment specificity evaluation value when the target doctor is determined to be in the diagnosis and treatment abnormal state according to the diagnosis and treatment abnormal detection information.
And the information sending module is used for sending the early warning information to the terminal logged in by the target account.
Specifically, when the target doctor is determined to be in the abnormal diagnosis and treatment state according to the abnormal diagnosis and treatment information, the early warning information is generated according to the medical record sample to be detected, the cluster type to which the medical record sample to be detected belongs and the diagnosis and treatment specificity evaluation value. The abnormality may mean that the diagnosis-specific evaluation value is greater than a predetermined threshold value of evaluation values. In one embodiment, the early warning information may further include optimized medical record samples included in the cluster category.
The early warning information is sent to the terminal logged in by the target account, the terminal triggers an alarm state and displays the received early warning information, so that related personnel can conveniently perform abnormal investigation and timely confirm the reason of abnormal diagnosis and treatment behaviors.
In this embodiment, the generated early warning information is sent to a terminal where a target account logs in, so that exception troubleshooting can be performed in time, and the occurrence of diagnosis and treatment behavior exception can be reduced.
The embodiment of the application can acquire and process related data based on an artificial intelligence technology. For example, the required data can be identified and extracted from the medical record data for calculation by natural language processing technology in artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
Those skilled in the art will appreciate that the functions of the modules and/or units in the above embodiments may be implemented by hardware associated with computer readable instructions, which may be stored in a computer readable storage medium, and when executed, the functions of the modules and/or units in the above embodiments may be included. The storage medium may be a non-volatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a Random Access Memory (RAM).
In order to solve the technical problem, an embodiment of the present application further provides a computer device. Referring to fig. 3, fig. 3 is a block diagram of a basic structure of a computer device according to the present embodiment.
The computer device 3 comprises a memory 31, a processor 32, a network interface 33 communicatively connected to each other via a system bus. It is noted that only the computer device 3 having the components 31-33 is shown in the figure, but it is to be understood that not all of the shown components are required to be implemented, and that more or less components may be implemented instead. As will be understood by those skilled in the art, the computer device is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction, and the hardware includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), an embedded device, and the like.
The computer device can be a desktop computer, a notebook, a palm computer, a cloud server and other computing devices. The computer equipment can carry out man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch panel or voice control equipment and the like.
The memory 31 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 31 may be an internal storage unit of the computer device 3, such as a hard disk or a memory of the computer device 3. In other embodiments, the memory 31 may also be an external storage device of the computer device 3, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the computer device 3. Of course, the memory 31 may also comprise both an internal storage unit of the computer device 3 and an external storage device thereof. In this embodiment, the memory 31 is generally used to store an operating system and various types of application software installed in the computer device 3, so as to implement the functions of each module and/or unit in the above-mentioned abnormality detection of doctor diagnosis and treatment behavior. Further, the memory 31 may also be used to temporarily store various types of data that have been output or are to be output.
The processor 32 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip in some embodiments. The processor 32 is typically used to control the overall operation of the computer device 3. In this embodiment, the processor 32 is configured to execute the computer readable instructions or the processing data stored in the memory 31 to implement the functions of each module and/or unit in the abnormality detection of the doctor diagnosis and treatment behavior.
The network interface 33 may comprise a wireless network interface or a wired network interface, and the network interface 33 is generally used for establishing communication connection between the computer device 3 and other electronic devices.
In this embodiment, when the processor executes the computer readable instructions stored in the memory, the functions of the modules and/or units in the abnormal detection of the doctor diagnosis and treatment behavior according to the above embodiment are realized, and after a medical record sample set is obtained, information required for clustering is extracted from medical record samples in the medical record sample set according to preset information dimensions, so as to obtain a plurality of optimized medical record samples; clustering is carried out according to the commonalities among the optimized medical record samples by adopting a clustering algorithm to obtain optimized medical record samples of various clustering types; acquiring a medical record sample to be detected associated with a target doctor, optimizing the attribution degree of the medical record sample according to each cluster type, determining the cluster type to which the medical record sample belongs, and then calculating the difference between the medical record sample to be detected and the central point of the cluster type to obtain a diagnosis and treatment specificity evaluation value; the diagnosis and treatment specific evaluation value measures the fluctuation of diagnosis and treatment behaviors of doctors in a numerical form, and diagnosis and treatment abnormity monitoring information can be generated according to the diagnosis and treatment specific evaluation value, so that abnormity detection of the diagnosis and treatment behaviors of the doctors is automatically realized.
The present application further provides another embodiment, which is to provide a computer-readable storage medium, where the computer-readable storage medium stores computer-readable instructions, where the computer-readable instructions are executable by at least one processor, so that the at least one processor performs the functions of the modules and/or units in the above-mentioned abnormality detection of physician diagnosis and treatment behavior, and after acquiring a medical record sample set, extracts information required for clustering from medical record samples in the medical record sample set according to preset information dimensions, so as to obtain a plurality of optimized medical record samples; clustering is carried out according to the commonalities among the optimized medical record samples by adopting a clustering algorithm to obtain optimized medical record samples of various clustering types; acquiring a medical record sample to be detected associated with a target doctor, optimizing the attribution degree of the medical record sample according to each cluster type, determining the cluster type to which the medical record sample belongs, and then calculating the difference between the medical record sample to be detected and the central point of the cluster type to obtain a diagnosis and treatment specificity evaluation value; the diagnosis and treatment specific evaluation value measures the fluctuation of diagnosis and treatment behaviors of doctors in a numerical form, and diagnosis and treatment abnormity monitoring information can be generated according to the diagnosis and treatment specific evaluation value, so that abnormity detection of the diagnosis and treatment behaviors of the doctors is automatically realized.
Through the above description of the embodiments, those skilled in the art will clearly understand that the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present application.
It is to be understood that the above-described embodiments are merely illustrative of some, but not restrictive, of the broad invention, and that the appended drawings illustrate preferred embodiments of the invention and do not limit the scope of the invention. This application is capable of embodiments in many different forms and is provided for the purpose of enabling a thorough understanding of the disclosure of the application. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to one skilled in the art that the present application may be practiced without modification or with equivalents of some of the features described in the foregoing embodiments. All equivalent structures made by using the contents of the specification and the drawings of the present application are directly or indirectly applied to other related technical fields and are within the protection scope of the present application.
Claims (10)
1. An abnormality detection device for doctor diagnosis and treatment behavior, comprising:
the system comprises a sample set acquisition module, a sample set acquisition module and a sample set analysis module, wherein the sample set acquisition module is used for acquiring a medical record sample set comprising a plurality of medical record samples;
the information extraction module is used for extracting information of each medical record sample in the medical record sample set according to a preset information dimension to obtain each optimized medical record sample;
the sample clustering module is used for clustering the optimized medical record samples to obtain optimized medical record samples of various clustering types;
the system comprises a sample acquisition module, a sample analysis module and a sample analysis module, wherein the sample acquisition module is used for acquiring a medical record sample to be detected related to a target doctor;
the category determination module is used for calculating the attribution degree of the medical record samples to be detected to each clustering category optimization medical record sample so as to determine the clustering category to which the medical record samples to be detected belong according to the attribution degree;
the specificity calculation module is used for calculating the diagnosis and treatment specificity evaluation value of the medical record sample to be detected according to the medical record sample to be detected and the determined clustering category;
and the information generating module is used for generating diagnosis and treatment abnormity detection information of the target doctor according to the diagnosis and treatment specificity evaluation value.
2. The apparatus for detecting abnormality in medical diagnosis and treatment behavior according to claim 1, wherein the information extraction module includes:
the information extraction submodule is used for extracting information of all medical record samples in the medical record sample set based on the information dimension of a patient, the diagnosis and treatment information dimension and the diagnosis and treatment effect dimension to obtain a plurality of candidate medical record samples;
and the medical record processing submodule is used for preprocessing the obtained candidate medical record samples to obtain optimized medical record samples.
3. The apparatus for detecting abnormalities in diagnostic and therapeutic actions of physicians according to claim 1, wherein said sample clustering means comprises:
the weight adding submodule is used for adding feature weights to the features in the optimized medical record samples;
and the sample clustering submodule is used for clustering the optimized medical record samples with the characteristic weight through a preset clustering algorithm to obtain optimized medical record samples of various clustering categories.
4. The apparatus for detecting abnormality in medical diagnosis and treatment behavior according to claim 3, wherein the weight adding sub-module includes:
the identification detection unit is used for detecting whether the scene identification exists or not;
the weight adding unit is used for adding feature weights to the features in the optimized medical record samples according to the scene identifiers when the scene identifiers exist;
and the weight determining unit is used for determining the characteristic weight of each characteristic in each optimized medical record sample based on a preset characteristic weight algorithm when the scene identifier does not exist.
5. The apparatus for detecting abnormality in medical diagnosis behavior according to claim 1, wherein the category determination module includes:
the extraction submodule is used for extracting patient information and diagnosis and treatment information from the medical record sample to be detected;
the attribution degree calculation operator module is used for calculating the attribution degree of the medical record samples to be detected to the optimized medical record samples of each cluster type based on the extracted patient information and the diagnosis and treatment information;
and the category selection submodule is used for selecting the cluster category corresponding to the maximum attribution degree as the cluster category to which the medical record sample to be detected belongs.
6. The apparatus for detecting abnormality in medical diagnosis and treatment behavior according to claim 5, wherein the attribution degree calculating sub-module includes:
the distance calculation unit is used for calculating the distance from the medical record sample to be detected to each cluster type optimized medical record sample based on the extracted patient information and the diagnosis and treatment information, and generating the attribution degree of the medical record sample to be detected to each cluster type optimized medical record sample based on the distance;
alternatively, the first and second electrodes may be,
and the quantity calculating unit is used for calculating the quantity of the samples of the nearest samples of the medical record samples to be detected in the optimized medical record samples of each cluster type based on the extracted patient information and the diagnosis and treatment information, and taking the quantity of the samples as the attribution degree of the medical record samples to be detected to the optimized medical record samples of each cluster type.
7. The apparatus for detecting abnormality in medical diagnosis behavior according to claim 1, wherein the specificity calculation module includes:
the sample mapping submodule is used for mapping the medical record sample to be detected into a first characterization vector;
the vector determination submodule is used for determining a second characterization vector corresponding to the central point of the clustering category based on each optimized medical record sample in the clustering category;
and the distance calculation submodule is used for calculating the distance between the first characterization vector and the second characterization vector and taking the distance as the specificity evaluation value of the medical record sample to be detected.
8. The apparatus for detecting an abnormality in a physician's clinical behavior according to claim 1, further comprising:
the early warning generation module is used for generating early warning information according to the medical record sample to be detected, the cluster type of the medical record sample to be detected and the diagnosis and treatment specificity evaluation value when the target doctor is determined to be in the diagnosis and treatment abnormal state according to the diagnosis and treatment abnormal detection information;
and the information sending module is used for sending the early warning information to a terminal logged in by the target account.
9. A computer device comprising a memory and a processor, wherein the memory stores computer readable instructions, and the processor when executing the computer readable instructions realizes the functions of each module and/or unit in the abnormality detection apparatus of physician diagnosis and treatment behavior according to any one of claims 1 to 8.
10. A computer-readable storage medium, wherein the computer-readable storage medium stores thereon computer-readable instructions, which when executed by a processor, implement the functions of the modules and/or units of the abnormality detection apparatus of physician diagnosis and treatment behavior according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111262634.0A CN113990514A (en) | 2021-10-28 | 2021-10-28 | Abnormality detection device for doctor diagnosis and treatment behavior, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111262634.0A CN113990514A (en) | 2021-10-28 | 2021-10-28 | Abnormality detection device for doctor diagnosis and treatment behavior, computer device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113990514A true CN113990514A (en) | 2022-01-28 |
Family
ID=79743318
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111262634.0A Pending CN113990514A (en) | 2021-10-28 | 2021-10-28 | Abnormality detection device for doctor diagnosis and treatment behavior, computer device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113990514A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117056740A (en) * | 2023-08-07 | 2023-11-14 | 北京东方金信科技股份有限公司 | Method, system and readable medium for calculating table similarity in data asset management |
-
2021
- 2021-10-28 CN CN202111262634.0A patent/CN113990514A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117056740A (en) * | 2023-08-07 | 2023-11-14 | 北京东方金信科技股份有限公司 | Method, system and readable medium for calculating table similarity in data asset management |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fang et al. | Computational health informatics in the big data age: a survey | |
| CN111145910A (en) | Abnormal case identification method and device based on artificial intelligence and computer equipment | |
| CN108899064A (en) | Electronic health record generation method, device, computer equipment and storage medium | |
| US10191956B2 (en) | Event detection and characterization in big data streams | |
| WO2021135449A1 (en) | Deep reinforcement learning-based data classification method, apparatus, device, and medium | |
| US20190237200A1 (en) | Recording medium recording similar case retrieval program, information processing apparatus, and similar case retrieval method | |
| CN110752027B (en) | Electronic medical record data pushing method, device, computer equipment and storage medium | |
| Gezimati et al. | Internet of things enabled framework for terahertz and infrared cancer imaging | |
| D’Amario et al. | GENERATOR HEART FAILURE DataMart: An integrated framework for heart failure research | |
| CN113990514A (en) | Abnormality detection device for doctor diagnosis and treatment behavior, computer device and storage medium | |
| US20120209620A1 (en) | Detecting unexpected healthcare utilization by constructing clinical models of dominant utilization groups | |
| CN114664458A (en) | Patient classification device, computer device and storage medium | |
| Kaur et al. | Critical Review on Data Mining in Healthcare Sector | |
| CN113821641B (en) | Method, device, equipment and storage medium for classifying medicines based on weight distribution | |
| CN114141358A (en) | Disease diagnosis apparatus based on knowledge map, computer device, and storage medium | |
| CN114822741A (en) | Processing device, computer equipment and storage medium of patient classification model | |
| CN113780855A (en) | Medical institution supervision method and device, computer equipment and storage medium | |
| Weiss et al. | Managing healthcare costs by peer-group modeling | |
| CN113780457A (en) | Method, device, equipment and medium for detecting abnormity of traditional Chinese medicine resource consumption | |
| CN113707304A (en) | Triage data processing method, device, equipment and storage medium | |
| Jiang et al. | Timeline Registration for Electronic Health Records | |
| Pivithuru et al. | E-patient Card: An Integrated Electronic Health Recording System for Patient | |
| CN113689143B (en) | Medical expense anomaly detection method, device, computer equipment and storage medium | |
| CN116364223B (en) | Feature processing method, device, computer equipment and storage medium | |
| CN113327655B (en) | Outlier detection method, device, equipment and medium for multidimensional data |
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 |