CN111599427A - A recommended method, device, electronic device and storage medium for unified diagnosis - Google Patents

Abstract

The invention relates to the technical field of auxiliary diagnosis of diseases in health management, in particular to a recommended method, device, electronic device and storage medium for unified diagnosis. According to the diagnostic codes of the subjects, the similarity measurement result between the diagnostic codes and the primary/secondary typical disease diagnostic code sets in the cluster C _k is determined; the cluster C _k is based on the difference between the diagnostic information of N subjects. K clusters divided by the similarity of the The diagnostic codes are visually analyzed in the ontology structure of the ICD. According to the semantic relationship between the diagnostic codes under the same branch, the nearest parent node is identified, and the disease type is obtained and reordered to obtain the unified diagnosis of the subject, which solves the problem of the same subject. However, there is a technical problem that the unified diagnosis results given by different chief inspectors are quite different.

Description

A recommended method, device, electronic device and storage medium for unified diagnosis

technical field

The invention relates to the technical field of auxiliary diagnosis of diseases in health management, in particular to a recommended method, device, electronic device and storage medium for unified diagnosis.

Background technique

A health examination report is a medical document issued by a health management (physical examination) institution to the subject, including the first page of the health examination report, the main examination report, the physical examination records, laboratory and medical imaging examination reports, etc. As the core component of the physical examination report, the main inspection report has no specific implementation rules for reference. Unified diagnosis is the main content of the main inspection report. The chief inspector should follow the basic principles of clinical diagnosis thinking, especially the principle of monism, reasonably classify diseases according to the system, and try to use one disease to summarize or explain multiple clinical manifestations of the disease. , Reasonable classification can make the main inspection report clear. At present, the main inspection reports of various medical institutions are mostly completed by borrowing relevant software, and the diagnosis results of diseases are mostly a list of various positive inspection results.

In practical work, the chief examiner usually needs to strictly check the basic information of the examinee, various inspection results, summarize the primary and secondary diagnoses and positive findings, and give the corresponding diagnosis results based on their own medical knowledge.

In practice, the inventor found that the above-mentioned prior art has the following defects:

Due to the differences in the professional level of the chief inspector, the different inspection items of the subjects, the variation of primary/secondary diagnosis and positive signs, the existing unified diagnosis methods that only rely on the professional knowledge of the chief inspector are time-consuming, labor-intensive, and inefficient. , time-consuming, and at the same time, for the same subject's physical examination conclusions, there are also great differences in the unified diagnosis results given by different chief examiners.

SUMMARY OF THE INVENTION

In order to solve the above-mentioned technical problems, the purpose of the present invention is to provide a recommended method, device, equipment and storage medium for unified diagnosis. The technical solutions adopted are as follows:

In a first aspect, an embodiment of the present invention provides a recommended method for unified diagnosis, which includes the following steps:

According to the diagnostic codes of the subjects, the similarity measurement result between the diagnostic codes and the primary/secondary typical disease diagnostic code sets in the cluster C _k is determined; the cluster C _k is based on the difference between the diagnostic information of N subjects. K clusters divided by similarity;

Select the cluster to which the similarity value greater than the preset threshold belongs, and obtain the disease co-occurrence pattern corresponding to each cluster from the unified diagnosis database;

Visually analyze the obtained disease co-occurrence pattern in combination with the diagnosis code in the ontology structure of the ICD, identify the nearest parent node according to the semantic relationship between the diagnosis codes under the same branch, obtain the disease type and reorder to obtain the subject unified diagnosis.

In a second aspect, an embodiment of the present invention provides a recommendation device for unified diagnosis, the device comprising:

The similarity measurement module is used to determine the similarity measurement result of the diagnosis code and the primary/secondary typical disease diagnosis code set in the cluster C _k according to the diagnosis code of the examinee; the cluster C _k is based on N K clusters divided by the similarity between subjects' diagnostic information;

a matching detection module for selecting clusters to which a similarity value greater than a preset threshold belongs, and obtaining a disease co-occurrence pattern corresponding to each cluster from the unified diagnosis database; and

The recommendation module is used to visually analyze the obtained disease co-occurrence pattern in combination with the diagnostic codes in the ICD ontology structure, identify the nearest parent node according to the semantic relationship between the diagnostic codes under the same branch, obtain the disease types and reorder them A unified diagnosis for the subject is obtained.

In a third aspect, an embodiment of the present invention provides an electronic device, including:

processing unit;

a storage unit for storing instructions executable by the processing unit;

Wherein, the processing unit is configured to: execute any of the methods described above.

In a fourth aspect, an embodiment of the present invention provides a storage medium, where computer-readable program instructions are stored in the storage medium, and when the program instructions are executed by a processing unit, any one of the methods described above is implemented.

The present invention has the following beneficial effects:

In the embodiment of the present invention, the disease co-occurrence pattern corresponding to each cluster is obtained from the unified diagnosis database by measuring the similarity between the diagnostic code of the subject and the primary/secondary typical disease diagnosis code set, and the obtained disease The co-occurrence patterns combined with the diagnostic codes are used for visual analysis in the ontology structure of the ICD, and then the disease types are obtained and rearranged to obtain the unified diagnosis of the subject. Aiming at the diagnosis of various diseases and positive signs of the examinee, it provides auxiliary reference for the primary examiner to quickly and automatically determine the unified diagnosis of major diseases, and according to the unified diagnosis and its ranking results, the examinee can clearly understand the main problems of their own health. with secondary issues. It solves the technical problem that for the same subject, the unified diagnosis results given by different chief inspectors are quite different.

Description of drawings

In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or in the prior art, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description The drawings are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without any creative effort.

FIG. 1 is a flowchart of a recommended method for unified diagnosis provided by an embodiment of the present invention;

2 is a flowchart of a method for constructing a unified diagnostic library provided by an embodiment of the present invention;

3 is a flowchart of a method for obtaining a primary/secondary typical disease diagnosis code set provided by an embodiment of the present invention;

4 is an overall flowchart of a method for constructing a unified diagnosis library and a recommended method for unified diagnosis provided by an embodiment of the present invention;

FIG. 5 is a partial structure of the ICD-10 disease classification system provided by an embodiment of the present invention;

6 is a schematic structural diagram of a typical disease diagnosis code set for three clusters provided by another embodiment of the present invention;

7 is a schematic structural diagram of a typical disease co-occurrence pattern of cluster 3 provided by another embodiment of the present invention;

8 is a schematic structural diagram of a unified diagnosis result about the patient provided by another embodiment of the present invention;

9 is a structural block diagram of a recommendation device for unified diagnosis provided by an embodiment of the present invention;

FIG. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed ways

In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following describes a recommended method, device, electronic device and storage medium for a unified diagnosis according to the present invention with reference to the accompanying drawings and preferred embodiments. , its specific implementation, structure, features and effects are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" are not necessarily referring to the same embodiment. Furthermore, the particular features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

Unless otherwise defined, 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 invention belongs.

The following specifically describes a specific scheme of a recommended method, apparatus, electronic device, and storage medium for unified diagnosis provided by the embodiments of the present invention with reference to the accompanying drawings.

Please refer to FIG. 1 , which shows a flowchart of a recommended method for unified diagnosis provided by an embodiment of the present invention. The embodiment recommends a reasonable unified diagnosis result through the similarity measurement of the disease diagnosis code, the visual analysis and the evaluation of the chief examiner, and the method for recommending the unified diagnosis includes the following steps:

Step S001, according to the diagnostic code of the examinee, determine the similarity measurement result of the diagnostic code and the primary/secondary typical disease diagnosis code set in the cluster C _k ; the cluster C _k is based on the diagnosis of N subjects. The similarity between the information is divided into K clusters.

Step S002, select the cluster to which the similarity value greater than the preset threshold belongs, and obtain the disease co-occurrence pattern corresponding to each cluster from the unified diagnosis database.

Step S003, the acquired disease co-occurrence pattern is combined with the diagnosis code to perform a visual analysis in the ICD ontology structure, according to the semantic relationship between the diagnosis codes under the same branch, identify the nearest parent node, obtain the disease type and reorder to obtain the disease type. Disease co-occurrence patterns in subjects.

Specifically in this embodiment, after the new subject has undergone a health checkup, the chief examiner determines a series of primary/secondary diagnoses and diagnostic codes for positive findings according to the subject's current examination results, firstly according to the similarity measure Methods Calculate the similarity between the subject and the primary/secondary typical disease diagnosis code set of each cluster, and obtain a set of similarity measurement results. Sort the similarity measurement results, identify the highest similarity value or the top five similarity values, and assign the result of the cluster to which it belongs to the subject, that is, recommend the typical disease co-occurrence pattern in the unified diagnosis database to the main examiner. physician. Visually analyze the subject's disease diagnosis code and the recommended typical disease co-occurrence pattern in the ICD ontology structure, according to the conceptual semantic relationship between disease diagnosis codes under the same branch, automatically identify the nearest parent node, and define a new disease Type and its order, determine the unified diagnosis for this subject. The chief examiner can correct, classify, evaluate and define the obtained disease co-occurrence patterns according to their own domain knowledge, and then obtain a unified diagnosis result for the subject's physical examination.

To sum up, the embodiment of the present invention obtains the disease co-occurrence pattern corresponding to each cluster from the unified diagnostic database by measuring the similarity between the diagnostic code of the subject and the primary/secondary typical disease diagnostic code set, The obtained disease co-occurrence patterns combined with diagnostic codes were visualized in the ontology structure of the ICD, and then the disease types were obtained and reordered to obtain the disease co-occurrence patterns of the subject. Aiming at the diagnosis of various diseases and positive signs of the examinee, it provides auxiliary reference for the primary examiner to quickly and automatically determine the unified diagnosis of major diseases, and according to the unified diagnosis and its ranking results, the examinee can clearly understand the main problems of their own health. with secondary issues. It solves the technical problem that for the same subject, the unified diagnosis results given by different chief inspectors are quite different.

Please refer to FIGS. 2 to 4 , as a preferred embodiment of the present invention, taking the ICD-10 disease classification system as an example, the method for constructing a unified diagnosis library in the above step S002 includes the following steps:

In step S201, the main inspection report of the subject is mapped to the ICD disease system, which is represented by the disease diagnosis code.

Specifically, the current primary/secondary diagnosis and positive signs of the subject are expressed in the form of disease diagnosis codes (such as ICD-10) in the standard primary examination report. Usually, a subject is marked as a set of a series of disease codes, and the codes have a certain sequence, and the higher the diagnostic code is, the main disease type or positive characteristic of the subject is represented. The subject's diagnostic information is defined as:

Di={(d ₁ ,Seq(d ₁ )),(d ₂ ,Seq(d ₂ )),…}

Wherein, d ₁ represents the disease code of the subject, and Seq(d ₁ ) represents the order of the disease code of the subject.

Step S202, construct a disease coding ontology structure on the basis of the ICD disease classification system according to the disease diagnosis code.

In order to measure the similarity of the diagnostic information of the subjects, it is necessary to construct a disease coding ontology structure (tree structure) on the basis of the ICD disease classification system. Taking ICD-10 as an example, please refer to Figure 5, which shows the local structure of the ICD-10 disease classification system, including 22 chapters (the first level-1), 262 sections (the second level-2), 2051 There are 3-bit categories (the third level-3), 9505 4-bit subcategories (the fourth level-4), and 22908 6-bit extended codes (the fifth level-5).

Step S203, the similarity measure of the semantic relationship of the disease diagnosis code in the disease code ontology structure.

The ICD-10 disease classification system is a hierarchical tree with a classification system. The diagnostic codes under the same branch have a certain similarity. The embodiment of the present invention adopts a similarity measurement method based on semantic relationship, and the similarity measurement method includes disease diagnosis. The measure of the information content of the codes, the measure of the similarity between the codes of the disease diagnosis, and the measure of the similarity between the sets of the codes of the disease diagnosis.

For the measurement method of disease diagnosis coding information, specifically, in the ICD-10 disease classification system, each code represents a concept, and there is semantic similarity between classification concepts, and concepts in the same branch are more semantic than concepts in different branches. resemblance. The embodiment of the present invention adopts a hierarchical depth measurement method, that is, each layer is assigned a specific numerical value, and the deeper the conceptual level, the greater the numerical value. For an ICD- ₁₀ code the d1 information volume is defined as:

IC(d ₁ )=path(d ₁ →r)

Among them, r represents the root node of the ICD-10 disease classification system, and path(·) is defined as the path length from the ICD-10 code d ₁ to the root node r. Therefore, the information volume of the root node is 0, the information volume of the first-level chapter is 1, the information volume of the second-level section is 2, the information volume of the third-level category is 3, and the information volume of the fourth-level sub-object is 4. , and the information amount of the fifth layer extended coding is 5.

For the method for measuring the similarity between the disease diagnosis codes, specifically, the similarity between the two codes is measured according to the information amount of the disease diagnosis codes. The embodiment of the present invention adopts the method of the nearest parent node of two semantic concepts to calculate the similarity, and the similarity between two codes is defined as:

Among them, d ₁ and d ₂ are the two disease diagnosis codes in the ICD-10 disease classification system, and LCA(d ₁ , d ₂ ) is defined as the nearest parent node of the disease diagnosis code d ₁ and the disease diagnosis code d ₂ , if d ₁ =d ₂ , then LCA(d ₁ ,d ₂ )=d ₁ =d ₂ , IC(LCA(d ₁ ,d ₂ ))=IC(d ₁ )=IC(d ₂ ); if d ₁ ≠d ₂ , and LCA(d ₁ , d ₂ )=root node, then IC(LCA(d ₁ ,d ₂ ))=0.

For the similarity measurement method between disease diagnosis code sets, specifically, in the health examination report, the subject is usually diagnosed as a set of disease diagnosis code sets, and by measuring the similarity of the two disease diagnosis code sets to represent the subject. Similarity of disease diagnosis information among patients. In the embodiment of the present invention, a pairwise average value of the most similar codes is used to measure the similarity between code sets, and it is assumed that the diagnostic information of subject i and subject j is defined as Di′ _i ={d _i1 ,d _i2 ,...,d _ig ,...} and Di' _j ={d _j1 ,d _j2 ,...,d _jh ,...} (the order of disease diagnosis codes is not considered here), then the diagnosis of subject i and subject j Information similarity is defined as:

Among them, Di' _i | is the number of disease diagnosis codes in the disease diagnosis information of subject i, |Di' _j | is the number of disease diagnosis codes in the disease diagnosis information of subject j, and _dig is the number of disease diagnosis codes of subject i The gth disease diagnosis code of d _jh is the hth disease diagnosis code of subject j.

By measuring the pairwise similarity of the diagnostic information of all the physical examination populations, the similarity matrix S of the physical examination population is obtained.

Step S204, according to the similarity measurement result, cluster the subjects to obtain the cluster C _K .

On the basis of the similarity matrix of diagnostic information of the physical examination population, a clustering algorithm is used to cluster similar subjects into a group to ensure that subjects within a cluster are similar and subjects outside a cluster are not similar.

Clustering algorithm is an effective unsupervised learning algorithm, which is mainly divided into division method, hierarchical method, density-based method, etc. The most common ones are K-means, hierarchical clustering, and density peak-based clustering. The embodiment of the present invention takes AP clustering (affine propagation clustering) as an example, and divides the medical examination population into different clusters.

In AP clustering, the selection of the number of clusters is controlled by the value of the preference parameter p. In the embodiment of the present invention, the preference coefficient p _c is used to control the number of clusters:

p=mean(S)-p _c ·N

Among them, S is the similarity matrix of the diagnostic information of all physical examination populations, and N is the number of subjects.

According to the disease diagnosis information of the subjects, the N subjects are divided into K clusters (C ₁ , C ₂ , ..., C _K ), and the number of subjects in each cluster is defined as:

Here, C _k (Di' _j ) represents the subject j classified into the cluster C _K , and E(C _K ) represents the population representative of the cluster C _k . λ(·) is an indicative function, when subject j is divided into clusters C _k , C _k (Di′ _j )=E(C _k ), λ(C _k (Di′ _j ), E(C _k ) ))=1, otherwise λ(C _k (Di′ _j ), E(C _k ))=0. In other words, when subject j is divided into cluster C _k , it is assigned the class label E(C _K ) of this cluster, that is, the representative population of this cluster indicates that this subject is divided into this cluster.

Step S205, obtaining a primary/secondary typical disease diagnosis code set.

Referring to Figure 3, the method for obtaining the primary/secondary typical disease diagnosis code set includes the following steps:

Step S301 , in the cluster C _k , the representative of the population whose similarity with the subject is greater than the similarity threshold is selected as the core physical examination population.

Due to the complex semantic relationship in the ICD-10 disease classification system, there are obvious differences between the codes at different levels, so that the ICD-10 disease diagnosis codes that simply use the population representative cannot fully describe the common characteristics of the clusters to which they belong. The embodiment of the present invention adopts a method of defining a core region from each cluster, selecting subjects with a high similarity to the representative of the cluster, analyzing their diagnostic information, and extracting high-frequent disease diagnosis codes to represent the common features of the cluster. .

Define the core area of cluster C _k , that is, the selection of the core physical examination population:

Core _k ={Di′ _j |S(Di′ _j ,E(C _k ))≥τ _k }

Among them, E(C _k ) represents the crowd representative of cluster C _k , and τ _k is the similarity threshold set in advance.

Step S302, selecting the diagnostic codes of the subjects whose frequency is greater than the frequency threshold in the core physical examination population to form a typical disease diagnostic code set.

The frequency of disease diagnosis code d _h in cluster C _k is defined as:

Among them, |Core _k | is the number of core physical examination populations in cluster C _k . λ(·) is an indicative function. When the disease diagnosis code d _h belongs to the diagnosis code of the jth subject, λ(d _h , Di′ _j )=1, otherwise λ(d _h , Di′ _j )= 0.

As an example, if there are 100 subjects in cluster C _k , and 50 subjects have the diagnosis code of hypertension ( _dh ), the return probability is 50/100=0.5.

By setting the frequency threshold δ of disease diagnosis codes, the typical disease diagnosis code set of cluster C _k is defined:

TICDS _k ={d _h |Frequency _k (d _h )>δ}

Step S303, calculate the average order of diagnostic codes of each typical disease in the cluster C _k . The smaller the average order of typical disease diagnosis codes, the greater the probability of being the main diagnosis.

Define the average order of diagnostic codes for each typical disease in cluster C _k :

为典型疾病诊断编码d_h在受检者 j的诊断信息Di_j中的次序。Among them, H' is the number of typical disease diagnosis code sets,

The order of the typical disease diagnosis code _dh in the diagnostic information Di _j of subject j.

Step S304, according to the average order, reorder the typical disease diagnosis code sets extracted from the cluster C _k to obtain the main disease diagnosis type and the secondary disease diagnosis type in the cluster C _k .

By reordering the typical disease diagnosis code sets extracted from the cluster C _k , the main disease diagnosis types and the secondary disease diagnosis types in the cluster can be identified.

Define the ranking function Rank( ) to obtain the primary/secondary typical disease diagnosis code set with sequence features:

The return value of the Average _k (d ₁ ) function is a positive number.

As an example, if there are three diseases, A, B, and C, and the corresponding averages are 5.3, 2.5, and 3.6, respectively, then after the Rank function is reordered, the order becomes B, C, and A, and the higher the order is. The more likely it is the primary diagnosis.

Step S206, display the primary/secondary typical disease diagnosis code set of cluster C _k in the ICD ontology structure, identify the nearest parent node according to the conceptual semantic relationship between the disease diagnosis codes under the same branch, obtain the new disease type, and assign the new disease type to the new disease type. Reordering yields typical disease co-occurrence patterns.

Specifically, combined with visual analysis technology, the primary/secondary typical disease diagnosis code set of cluster C _k is displayed in the ICD-10 disease classification system, and the nearest parent node is automatically identified according to the semantic relationship between the diagnosis codes of diseases under the same branch. Define a new disease type e _i , i.e.

e _i =LCA(d ₁ ,d ₂ ,...,d _m )

Among them, d ₁ , d ₂ ...... d _m are the disease diagnosis codes in m ICD-10 systems, and LCA(d ₁ , d ₂ , ..., d _m ) is the nearest code d ₁ , d ₂ ...... d _m parent node.

Therefore, the primary/secondary typical disease diagnosis code set of cluster C _k is transformed into a set of new disease types {e ₁ ,e ₂ ,…,e _m0 ,}, and by reordering, the typical disease co-occurrence pattern is obtained, That is, the preliminary results of a unified diagnosis.

Step S207: Receive an auxiliary operation of the chief examiner, adjust the typical disease co-occurrence pattern, and obtain a unified diagnosis database.

According to their own domain knowledge, the chief examiner can correct, classify, evaluate and define the identified typical disease co-occurrence patterns, and obtain the final unified diagnosis result.

By summarizing the typical disease co-occurrence patterns of all clusters, a typical disease co-occurrence pattern library for the selected physical examination population, that is, a unified diagnosis library is constructed.

The following takes the clinical diagnosis data of electronic medical records as an example to verify the feasibility of the experimental results.

The diagnostic information of 4418 patients with uremia in the publicly available MIMIC-III dataset was selected, and the diagnostic information was represented in the form of ICD-9 codes (pure numbers, no letters), through the diagnostic information similarity measure, AP clustering, main/ The secondary typical disease diagnosis code set is extracted, and finally three clusters are determined, of which the typical disease diagnosis code set is shown in Figure 6.

Taking the patient diagnosis information in cluster 3 as an example, the typical disease diagnosis code set and its sequence are visually analyzed in the ICD-9 ontology structure, as shown in Figure 7, in which the unified diagnosis is urogenital system disease, circulatory system disease, As well as acute respiratory failure, secondary diagnoses (complications) are sepsis, sepsis, and anemia.

For newly admitted patients, their diagnostic codes are 519.09, 518.81, 491.21, 38.9, 995.92, 785.52, 584.9, 482.1, 427.31, 519.19, and the similarity between the patient's diagnostic information and the typical disease diagnostic code set of 3 clusters is calculated. The measurement result shows that the similarity with cluster 3 is the largest, and the unified diagnosis of cluster 3 is recommended to this patient. Therefore, the unified diagnosis result of this patient is the fusion of the diagnosis result of cluster 3 and the diagnosis result of this examination, that is, the main diagnosis is respiratory system. Diseases, urinary system diseases, and circulatory system diseases, and secondary diagnoses (complications) are sepsis, sepsis, and anemia, as shown in FIG. 8 .

Referring to FIG. 9, based on the same inventive concept as the above method embodiments, another embodiment of the present invention further provides a recommendation apparatus for unified diagnosis, the apparatus includes a similarity measurement module 901, a matching detection module 902 and a recommendation module 903.

Specifically, the similarity measurement module 901 is configured to determine the similarity measurement result between the diagnosis code and the primary/secondary typical disease diagnosis code set in the cluster C _k according to the diagnostic code of the examinee; the cluster C _k is K clusters divided according to the similarity between N subjects' diagnostic information. The matching detection module 902 is configured to select the cluster to which the similarity value greater than the preset threshold belongs, and obtain the disease co-occurrence pattern corresponding to each cluster from the unified diagnosis database. The recommendation module 903 is used to perform a visual analysis on the obtained disease co-occurrence pattern in combination with the diagnosis codes in the ICD ontology structure, identify the nearest parent node according to the semantic relationship between the diagnosis codes under the same branch, obtain the disease types and reorder them Obtain the disease co-occurrence pattern of the subject.

Preferably, the device further includes a unified diagnosis library building module, and the unified diagnosis library building module includes a typical disease co-occurrence pattern acquisition module and a unified diagnosis library generation module. Specifically, the typical disease co-occurrence pattern acquisition module is used to display the primary/secondary typical disease diagnosis code set in the ICD ontology structure, identify the nearest parent node according to the semantic relationship between the disease diagnosis codes under the same branch, and obtain a new disease Types, reordering new disease types to obtain typical disease co-occurrence patterns. The unified diagnosis library generation module is used for receiving the auxiliary operation of the chief examiner, and adjusting the typical disease co-occurrence pattern to obtain the unified diagnosis library.

Preferably, the device further comprises a primary/secondary typical disease diagnosis code set acquisition module, the primary/secondary typical disease diagnosis code set acquisition module includes a core physical examination population determination module, a typical disease diagnosis code set generation module, an average order calculation module and Get the module. Specifically, the core physical examination population determination module is used to select the representative of the population whose similarity with the subject is greater than the similarity threshold in the cluster C _k as the core physical examination population. The typical disease diagnosis code set generation module is used to select the diagnosis codes of the subjects whose frequency of occurrence is greater than the frequency threshold in the core physical examination population to form a typical disease diagnosis code set. The mean order calculation module is used to calculate the mean order of diagnostic codes for each typical disease in cluster C _k . The obtaining module is configured to reorder the typical disease diagnosis code sets extracted from the cluster C _k according to the average order, so as to obtain the main disease diagnosis type and the secondary disease diagnosis type in the cluster C _k .

Preferably, the device further includes a similarity measurement module, the similarity measurement module includes an information amount measurement module for disease diagnosis codes, a similarity measurement module between disease diagnosis codes, and a similarity measurement module between disease diagnosis code sets.

Please refer to FIG. 10. FIG. 10 shows a possible schematic structural diagram of the electronic device involved in the above embodiment. The electronic device may include: a processing unit 1001 , a storage unit 1002 and a communication unit 1003 . The processing unit 1001 may be arranged to communicate with the storage unit 1002 . The storage unit 1002 is used for storing executable instructions and/or program codes of the processing unit 1001, wherein the processing unit is configured to execute a recommended method for unified diagnosis provided by any one of the above method embodiments. The communication unit 1003 is used to support the communication between the electronic device and other network entities to realize functions such as data exchange. For example, the communication module 1003 supports the communication between the electronic device and other intelligent terminals to realize the data exchange function.

The processing unit 1001 may be a processor or a controller. The communication module 1003 may be a transceiver, an RF circuit, a communication interface, or the like. The storage unit 1002 may be a memory.

FIG. 10 is only a possible implementation manner of the embodiment of the present application. In practical applications, the electronic device may further include more or less components, which is not limited here.

It should be noted that the electronic device may be a server or an intelligent terminal, and the intelligent terminal may be a computer, a tablet computer, a smart phone, or the like.

Embodiments of the present invention further provide a storage medium, where computer-readable program instructions are stored in the storage medium, and when the program instructions are executed by the processing unit, a unified diagnosis recommendation provided in any one of the foregoing embodiments is implemented. method. For example, the computer-readable storage medium may be Read-Only Memory (ROM), Random Access Memory (RAM), Compact Disc Read-Only Memory (CD-ROM), magnetic tape , floppy disks and optical data storage devices.

It should be noted that: the above-mentioned order of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that in the embodiments and still achieve desirable results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the apparatus and server embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and for related parts, please refer to the partial descriptions of the method embodiments.

Those of ordinary skill in the art can understand that all or part of the steps of implementing the above embodiments can be completed by hardware, or can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. The storage medium mentioned may be a read-only memory, a magnetic disk or an optical disk, etc.

The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.

Claims (10)

1. A method for recommending unified diagnosis, comprising the steps of:

determining a diagnostic code and cluster C based on the diagnostic code of the subject_kSimilarity measure results of the primary/secondary canonical disease diagnosis coding set in (1); the cluster C_kIs K clusters divided according to the similarity between the diagnostic information of N subjects;

selecting a cluster to which the similarity value greater than a preset threshold belongs, and acquiring a disease co-occurrence mode corresponding to each cluster from a unified diagnosis library;

and carrying out visual analysis on the ICD body structure by combining the acquired disease co-occurrence mode with the diagnosis codes, identifying the nearest father node according to the semantic relation among the diagnosis codes under the same branch, acquiring the disease type and reordering to acquire the unified diagnosis of the examinee.

2. The recommendation method for unified diagnosis according to claim 1, wherein the unified diagnosis library is constructed by the following steps:

displaying the main/secondary typical disease diagnosis code sets in the ICD body structure, identifying a nearest father node according to the semantic relation between the disease diagnosis codes under the same branch to obtain a new disease type, and reordering the new disease types to obtain a typical disease co-occurrence mode;

and receiving the auxiliary operation of the main examining physician, and adjusting the typical disease co-occurrence mode to obtain a unified diagnosis library.

3. A method for recommending unified diagnosis according to claim 1 or 2, wherein the method for obtaining the primary/secondary typical disease diagnosis code set comprises the following steps:

in cluster C_kSelecting the people with similarity greater than the similarity threshold value with the detected person as the core physical examination people;

selecting the diagnostic codes of the subjects with the frequency greater than the frequency threshold value from the core physical examination population to form a typical disease diagnostic code set;

computing cluster C_kAverage order of each typical disease diagnostic code in (1);

according to said average order, for the cluster C_kReordering the extracted typical disease diagnostic code set to obtain cluster C_kThe major disease diagnosis type and the minor disease diagnosis type.

4. A unified diagnosis recommendation method according to claim 1 or 2, characterized in that said similarity measure comprises information amount measure of disease diagnosis codes, similarity measure between disease diagnosis codes and similarity measure between disease diagnosis code sets.

5. A recommendation device for unified diagnosis, characterized in that the device comprises:

a similarity measurement module for measuring the similarity according to the detected resultThe diagnostic code of the subject, determining the diagnostic code and cluster C_kSimilarity measure results of the primary/secondary canonical disease diagnosis coding set in (1); the cluster C_kIs K clusters divided according to the similarity between the diagnostic information of N subjects;

the matching detection module is used for selecting the cluster to which the similarity value greater than the preset threshold value belongs and acquiring a disease co-occurrence mode corresponding to each cluster from a unified diagnosis library; and

and the recommending module is used for performing visual analysis on the ICD body structure by combining the acquired disease co-occurrence mode with the diagnosis codes, identifying the nearest father node according to the semantic relation among the diagnosis codes under the same branch, obtaining the disease type and reordering to obtain the unified diagnosis of the examinee.

6. The apparatus for recommending unified diagnosis according to claim 5, further comprising a unified diagnosis library building module, said unified diagnosis library building module comprising:

the typical disease co-occurrence mode acquisition module is used for displaying the main/secondary typical disease diagnosis code sets in the ICD body structure, identifying a nearest father node according to the semantic relation between the disease diagnosis codes under the same branch to obtain a new disease type, and reordering the new disease types to obtain a typical disease co-occurrence mode;

and the unified diagnosis library generation module is used for receiving the auxiliary operation of the main examining physician and adjusting the typical disease co-occurrence mode to obtain a unified diagnosis library.

7. A recommendation device for unified diagnosis according to claim 5 or 6, characterized in that it further comprises a primary/secondary canonical disease diagnosis code set acquisition module, said primary/secondary canonical disease diagnosis code set acquisition module comprises:

a core physical examination population determination module for clustering C_kSelecting the people with similarity greater than the similarity threshold value with the detected person as the core physical examination people;

the typical disease diagnosis code set generation module is used for selecting diagnosis codes of the subjects with the occurrence frequency larger than a frequency threshold value from the core physical examination population to form a typical disease diagnosis code set;

average order calculation module for calculating cluster C_kAverage order of each typical disease diagnostic code in (1);

an obtaining module, configured to pair the clusters C according to the average order_kReordering the extracted typical disease diagnostic code set to obtain cluster C_kThe major disease diagnosis type and the minor disease diagnosis type.

8. A unified diagnosis recommendation device according to claim 5 or 6 characterized in that it further comprises similarity measure module, said similarity measure module comprises information measure module of disease diagnosis codes, similarity measure module between disease diagnosis codes and similarity measure module between disease diagnosis code sets.

9. An electronic device, comprising:

a processing unit;

a storage unit for storing processing unit executable instructions;

wherein the processing unit is configured to: performing the method of any one of claims 1 to 4.

10. A storage medium having computer-readable program instructions stored therein, which when executed by a processing unit, implement the method of any one of claims 1 to 4.

Images