CN115631823A - Similar case recommendation method and system - Google Patents
Similar case recommendation method and system Download PDFInfo
- Publication number
- CN115631823A CN115631823A CN202211077445.0A CN202211077445A CN115631823A CN 115631823 A CN115631823 A CN 115631823A CN 202211077445 A CN202211077445 A CN 202211077445A CN 115631823 A CN115631823 A CN 115631823A
- Authority
- CN
- China
- Prior art keywords
- symptom
- case
- cases
- similarity
- symptoms
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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
Abstract
The invention discloses a similar case recommendation method and a system, belonging to the technical field of clinical decision support systems, aiming at solving the technical problems that the realization difficulty of similar case recommendation by applying a deep learning algorithm and a natural language processing technology is extremely high, the feasibility is not high, and the adopted technical scheme is as follows: the method comprises the following specific steps: data preprocessing: extracting the symptoms and diagnosis information of the electronic medical record of the admission record of the case; processing the cases into standard symptoms and standard diagnosis lists by taking the cases as units, extracting treatment information when the cases are treated and storing the treatment information in a dictionary, and assisting in obtaining the similarity degree of the cases subsequently; wherein, the information of seeing a doctor includes age and department; acquiring symptom weight: extracting the weight of symptoms in disease diagnosis based on the symptoms of the knowledge map and the knowledge related to the disease; obtaining the similarity degree of cases: acquiring the similarity degree of other cases and the target case according to the symptoms and the diagnosis list of the target case and the fusion weight; and (5) recommending similar cases.
Description
Technical Field
The invention relates to the technical field of a Clinical Decision Support System (CDSS), in particular to a method and a System for recommending similar cases.
Background
Similar case recommendation belongs to a breakthrough of applying a recommendation system to intelligent medical treatment, namely, a case with the same illness and symptoms as a patient is recommended to a doctor so as to assist the doctor in clinical diagnosis and treatment.
At present, most of similar case recommendation methods are artificial intelligence algorithms based on deep learning, and case similarity is calculated by means of understanding semantics, calculating indexes such as cosine similarity and the like and using a method for calculating text similarity in a natural language processing technology.
However, the difficulty of implementing similar case recommendation by using a deep learning algorithm and a natural language processing technology is high, the interpretability of a deep learning model is not strong, a clinician is required to evaluate the similarity of sample cases to obtain a training sample, and errors of different evaluation opinions of different physicians are easy to occur, so that the reliability of the training sample is difficult to guarantee; the support of mass case data is a necessary basis for the accuracy of the model, data acquisition is very difficult for general enterprises and individuals, and the personal data with extremely high privacy, such as the electronic medical record, is particularly difficult to acquire and process; finally, the deep learning model has high requirements on hardware configuration such as a server, and further increases the implementation difficulty and the implementation cost.
In summary, the existing similar case recommendation has the problems that the implementation difficulty of implementing the similar case recommendation by using a deep learning algorithm and a natural language processing technology is very high, and the feasibility is not high.
Disclosure of Invention
The technical task of the invention is to provide a similar case recommendation method and system to solve the problems of high implementation difficulty and low feasibility of similar case recommendation based on deep learning and natural language processing technologies.
The technical task of the invention is realized in the following way, and the method for recommending similar cases specifically comprises the following steps:
data preprocessing: extracting symptoms and diagnosis information of the electronic medical record of the admission record of the case; processing the cases into standard symptoms and standard diagnosis lists by taking the cases as units, extracting treatment information when the cases are treated and storing the treatment information in a dictionary, and assisting in obtaining the similarity degree of the cases subsequently; wherein, the information of seeing a doctor includes age and department;
acquiring symptom weight: extracting the weight of the symptoms in disease diagnosis based on the symptoms of the knowledge map and the knowledge related to the disease;
acquiring the similarity degree of cases: acquiring the similarity degree of other cases and the target case according to the symptoms and the diagnosis list of the target case and the fusion weight;
similar case recommendation: according to the calculation results of symptoms, diagnosis and case similarity, a case list with the similarity from big to small is obtained by sorting in a reverse order, and cases with high similarity, same department and same age period (0-6 years, 6-18 years, 18-60 years and over 60 years) are preferentially recommended according to the age of the case and the clinic information.
Preferably, the data preprocessing further comprises normalizing the symptoms and diagnosis information of all cases in the case bank, and the details are as follows:
taking the case number as an index, and acquiring a symptom list and a diagnosis list of all cases in a case library in a cache;
according to the mapping relation table of the original words and the normalized words in the knowledge graph, obtaining a normalized word list of case symptoms and diagnosis by matching the names of the original words and the entity types of the original words and storing the normalized words in a cache; wherein, the original word name comprises a symptom name and/or a diagnosis name; the primitive word entity types include symptoms and/or diagnoses.
Preferably, the symptom weights are obtained as follows:
acquiring data related to symptoms and diseases in a knowledge map, extracting all relevant knowledge of which the knowledge type (initial entity type-relation-termination entity type) is disease-related symptom-symptom, and storing the knowledge in a symptom and disease knowledge table;
acquiring the disease number N of diseases related to any symptom i by extracting the symptom-disease related knowledge in the knowledge map i And knowledge baseNumber of diseases N, thereby calculating symptom weights.
Preferably, the number N of diseases related to any symptom i is obtained by extracting the related knowledge of symptom-diseases in the knowledge map i And the total number of diseases N in the knowledge base, thereby calculating the symptom weight as follows:
performing de-duplication counting on the normalized words of the initial entities to obtain the total number N of diseases in the knowledge base;
carrying out duplication removal extraction on the termination entity normalized words in the symptom and disease knowledge table to obtain a symptom list of a knowledge base;
according to a symptom list of a knowledge base, sequentially extracting an initial entity normalization word set corresponding to each symptom i, and calculating the number of initial entity normalization words to obtain the number N of diseases related to the diseases of each symptom i i ；
Associating a number of diseases N to a disease according to a list of symptoms of a knowledge base and each symptom i i And obtaining a symptom-weight dictionary, wherein the formula is as follows:
wherein, W i Symptom coefficients as the degree of case similarity; N-N i Indicating the number of diseases that are not associated with the symptom; n represents the total number of diseases in the knowledge base.
More preferably, the degree of similarity between the obtained cases is specifically as follows:
matching the symptoms and the diagnosis lists of other cases in the case library by using the symptoms and the diagnosis lists of the target case to respectively obtain intersection sets of the symptoms and the diagnoses of the target case and the other cases;
acquiring symptom similarity according to an intersection set of symptoms of the target case and other cases;
acquiring diagnosis similarity according to the intersection set of the diagnoses of the target case and other cases;
carrying out weighted operation on the symptom similarity and the diagnosis similarity to obtain the case similarity;
the basic information of the case is stored in the cache in the form of a table with the case number, age, department, symptom similarity, diagnosis similarity, and case similarity as the header.
More preferably, the symptom similarity is specifically as follows:
let the symptom list of the compared cases be [ i 1 ,i 2 ,i 3 …i n ](ii) a The symptom intersection set of the compared case and the target case is [ i 1 ,i 2 ,i 3 …i m ](ii) a Symptom weight is W i Then, the calculation formula of the corresponding case and the target case is as follows:
wherein, S represents the degree of symptom similarity; w i1 +W i2 +W i3 +…+W im Accumulating the symptom weights representing the symptom intersection set to obtain an intersection score; n represents the number of elements in the symptom list of the compared cases;
the diagnostic similarity is specifically as follows:
let the list of diagnoses of the cases to be compared be [ z ] 1 ,z 2 ,z 3 …z k ](ii) a The set of diagnostic intersections of the compared case and the target case is [ z ] 1 ,z 2 ,z 3 …z j ](ii) a The calculation formula of the degree of similarity of the symptoms of the corresponding case and the target case is as follows:
wherein, S' represents the degree of symptom similarity; j represents the number of diagnostic intersection sets; k represents the number of elements in the diagnosis list of the compared cases;
the case similarity degree is obtained by averaging the symptom similarity degree S and the diagnosis similarity degree S', and the formula is as follows:
wherein F represents the degree of case similarity.
A similar case recommendation system, the system comprising a data layer and a data processing layer; the data layer comprises a case management module, and the data processing layer comprises a preprocessing module, a symptom weight acquisition module, a case similarity acquisition module and a similar case recommendation module;
the case management module is used for maintaining basic information of cases, structured large-section texts of electronic medical records and symptom and diagnosis information;
the preprocessing module is used for carrying out normalization processing on symptoms and diagnosis information, carrying out mapping from original words to normalized words on all cases in a case library including the symptoms and the diagnosis information of a target case according to an original word and normalized word mapping table of a knowledge map, obtaining standard symptoms and standard diagnosis information of the case and storing the standard symptoms and the standard diagnosis information into a list form; the normalization processing is to obtain the normalization word corresponding to the symptom and diagnosis according to the mapping relation table of the original word and the normalization word of the knowledge graph;
the symptom weight acquisition module is used for acquiring symptom weights based on the knowledge type in the knowledge map as the relevant knowledge of the disease-relevant symptom-symptom and forming a symptom-weight dictionary;
the case similarity obtaining module is used for calculating the symptom similarity degree, the diagnosis similarity degree and the case similarity degree of all other cases by taking the symptoms and the diagnoses of the target case as comparison standards based on the transmitted symptom and diagnosis lists and the symptom weight dictionary of all cases in the case library, and storing the symptom similarity degree, the diagnosis similarity degree and the case similarity degree in a dictionary form;
the similar case recommending module is used for obtaining a case list with the similarity from large to small by sequencing in a reverse order according to the symptoms, diagnosis and calculation results of the case similarity, and preferentially recommending cases with high similarity, same department and same age (0-6 years, 6-18 years, 18-60 years and over 60 years) by combining the age of the cases and the clinic information.
Preferably, the working process of the symptom weight obtaining module is as follows:
(1) Acquiring data related to symptoms and diseases in the knowledge map, and extracting all relevant knowledge of which the knowledge type (initial entity type-relation-termination entity type) is disease-related symptom-symptom;
(2) The initial entity of all relevant knowledge of the disease-relevant symptom-symptom is subjected to de-duplication counting, and the total number N of the diseases in the knowledge base is calculated;
(3) Performing de-duplication extraction on terminating entity words of all relevant knowledge of disease-relevant symptom-symptom to obtain a symptom list of a knowledge base;
(4) Sequentially extracting initial entity normalization word sets corresponding to each symptom i according to the symptom lists of the knowledge base, and obtaining the disease number N of the diseases related to each symptom i by calculating the number of the initial entity normalization words i ；
(5) And for the symptom list of the knowledge base, sequentially obtaining the number N of diseases associated with each symptom according to the obtained symptoms i And the acquired total number of diseases N, resulting in a symptom-weight dictionary, the formula is as follows:
wherein, W i Symptom coefficients as the degree of case similarity; N-N i Indicating the number of diseases that are not associated with the symptom; n represents the total number of diseases in the knowledge base.
An electronic device, comprising: a memory and at least one processor;
wherein the memory has stored thereon a computer program;
the at least one processor executes the computer program stored by the memory such that the at least one processor performs the similar case recommendation method as described above.
A computer-readable storage medium having stored therein a computer program executable by a processor to implement the similar case recommendation method as described above.
The similar case recommendation method and the similar case recommendation system have the following advantages:
the method calculates the similarity of the cases by normalizing the symptoms of the cases and the diagnosis information and fusing the weight of the symptoms, so that the recommendation accuracy of the similar cases is high and the recommendation is easy to realize;
the method overcomes the defect of high implementation difficulty in making similar case recommendation based on deep learning and natural language processing technology, performs reasoning operation of similar degree by comparing symptoms of cases with diagnosis information, introduces the concept of symptom weight by related knowledge of symptoms and diseases in a medical knowledge map, and integrates the concept into calculation of case similarity degree, and aims to improve the proportion of rare symptoms in similarity prediction and improve the calculation accuracy of case similarity degree while being easy to implement;
the invention can recommend case information similar to the target case based on the symptoms and diagnosis information extracted from the electronic medical record of the target case and the weight of the symptoms in disease diagnosis, and assists clinical diagnosis and clinical decision;
and (IV) considering that the similarity degree of the symptoms of the cases is not only related to the number of the symptoms on the matching but also related to the rareness degree of the symptoms on the matching, the invention adds the concept of symptom weight to improve the accuracy of calculating the similarity degree of the symptoms in the next step.
Drawings
The invention is further described below with reference to the accompanying drawings.
Fig. 1 is a flow chart of a similar case recommendation method.
Detailed Description
The similar case recommendation method and system of the present invention will be described in detail below with reference to the drawings and specific embodiments of the specification.
Example 1:
the embodiment provides a similar case recommendation method, which specifically comprises the following steps:
s1, normalizing the symptoms and diagnosis information of all cases in a case library to respectively obtain a standard symptom list and a standard diagnosis list of each case;
s2, acquiring data related to symptoms and diseases in the knowledge graph to calculate symptom weight;
s3, according to the matching condition of the standard symptom list of each case and the standard symptom list of the target case, combining symptom weights and calculating the similarity degree of the symptoms; calculating the diagnosis similarity degree according to the matching condition of the diagnosis set of each case and the target case diagnosis list; averaging the similarity degree of the symptoms and the similarity degree of diagnosis to obtain the similarity degree of each case and the target case;
and S4, reordering the 5 cases with the highest similarity according to the ages and department information of the cases, and advancing the ranking of the cases in the same department and age period with the target case in the 5 cases with the highest similarity to obtain a final similar case recommendation result.
In this embodiment, the normalization process of the symptoms and diagnosis information of all cases in the case library in step S1 is specifically as follows:
s101, acquiring a symptom list and a diagnosis list of all cases in a case library in a cache by taking the case number as an index;
s102, obtaining a case symptom and diagnosed normalization word list by matching an original word name (symptom name/diagnosis name) and an original word entity type (symptom/diagnosis) according to an original word and normalization word mapping relation table in the knowledge map, and storing the case symptom and diagnosed normalization word list in a cache; the mapping table structure of the original words and the normalization words of the knowledge graph is as follows:
| serial number | Name of field | Meaning of a field |
| 1 | original_word | Primitive word |
| 2 | entity_type_id | Entity type id |
| 3 | normalize_term_id | Entry word id |
| 4 | normalize_term_cn | Chinese name of word |
| 5 | status | Knowledge state |
| 6 | update_time | Update time |
The symptom weight obtained in step S2 of this embodiment is specifically as follows:
s201, acquiring data related to symptoms and diseases in a knowledge map, extracting all related knowledge of which the knowledge type (initial entity type-relation-termination entity type) is disease-related symptom-symptom, and storing the knowledge in a symptom and disease knowledge table; as shown in the following table:
s202, through knowledge mappingExtracting related knowledge of symptom-disease, and obtaining disease number N of disease associated with any symptom i i And the total number of diseases N in the knowledge base, thereby calculating a symptom weight.
In step S202 of this embodiment, the number N of diseases associated with any symptom i is obtained by extracting the symptom-disease related knowledge in the knowledge graph i And the total number of diseases N in the knowledge base, thereby calculating the symptom weight as follows:
s20201, performing de-duplication counting on the initial words of the initial entities to obtain the total disease number N of the knowledge base;
s20202, carrying out duplication removal and extraction on the termination entity normalized words in the symptom and disease knowledge table to obtain a symptom list of a knowledge base;
s20203, sequentially extracting an initial entity normalization word set corresponding to each symptom i according to the symptom list of the knowledge base, and calculating the number of initial entity normalization words to obtain the number N of diseases related to the diseases of each symptom i i ；
S20204, list of symptoms according to knowledge base and number of diseases N associated with disease per symptom i i And obtaining a symptom-weight dictionary, wherein the formula is as follows:
wherein, W i Symptom coefficients as the degree of case similarity; N-N i Indicating the number of diseases that are not associated with the symptom; n represents the total number of diseases in the knowledge base.
The degree of similarity between the acquired cases in step S3 in this embodiment is specifically as follows:
s301, matching symptoms and diagnosis lists of other cases in the case library by using the symptoms and diagnosis lists of the target case to obtain intersection sets of the symptoms and diagnosis of the target case and the other cases respectively;
s302, acquiring symptom similarity according to an intersection set of symptoms of the target case and other cases;
s303, acquiring diagnosis similarity according to the intersection set of the diagnoses of the target case and other cases;
s304, carrying out weighted operation on the symptom similarity and the diagnosis similarity to obtain case similarity;
s305, combining the basic information of the case, and storing the basic information in a cache in a table form with the case number, age, department, symptom similarity, diagnosis similarity and case similarity as a header.
The symptom similarity in step S302 of this embodiment is specifically as follows:
let the symptom list of the compared cases be [ i 1 ,i 2 ,i 3 …i n ](ii) a The symptom intersection set of the compared case and the target case is [ i 1 ,i 2 ,i 3 …i m ](ii) a Symptom weight is W i Then, the calculation formula of the corresponding case and the target case is as follows:
wherein S represents the degree of symptom similarity; w i1 +W i2 +W i3 +…+W im Accumulating the symptom weights representing the symptom intersection set to obtain an intersection score; n represents the number of symptom list elements of the compared cases;
the diagnosis similarity in step S303 in this embodiment is specifically as follows:
let the list of diagnoses of the cases to be compared be [ z ] 1 ,z 2 ,z 3 …z k ](ii) a The set of diagnostic intersections of the compared case and the target case is [ z ] 1 ,z 2 ,z 3 …z j ](ii) a The calculation formula of the degree of similarity of the symptoms of the corresponding case and the target case is as follows:
wherein, S' represents the degree of symptom similarity; j represents the number of diagnostic intersection sets; k represents the number of elements in the diagnosis list of the compared cases;
the case similarity degree in step S304 of the present embodiment is obtained by averaging the symptom similarity degree S and the diagnosis similarity degree S', and the formula is as follows:
wherein F represents the degree of case similarity.
According to the front-end requirement, the generated tables can be sorted according to the symptom similarity degree, the diagnosis similarity degree and the case similarity degree respectively, records of the top 5 in ranking under different sorting conditions are obtained respectively, the records of the top 5 are reordered according to the age and the department, the cases in the same department and the same age period in the records of the top 5 are ranked in advance, and the final similar case recommendation result can be obtained.
Example 2:
the embodiment provides a similar case recommendation system, which comprises a data layer and a data processing layer; the data layer comprises a case management module, and the data processing layer comprises a preprocessing module, a symptom weight obtaining module, a case similarity obtaining module and a similar case recommending module;
the case management module is used for maintaining basic information of cases, structured large-section texts of electronic medical records and symptom and diagnosis information;
the preprocessing module is used for carrying out normalization processing on symptoms and diagnosis information, carrying out mapping from original words to normalized words on all cases in a case library according to original words and normalized word mapping tables of the knowledge map, wherein the symptoms and the diagnosis information of a target case are obtained from the cases in the case library, and the standard symptoms and the standard diagnosis information of the cases are obtained and stored in a list form; the normalization processing is to obtain the normalization word corresponding to the symptom and diagnosis according to the mapping relation table of the original word and the normalization word of the knowledge graph;
the symptom weight acquisition module is used for acquiring symptom weights based on the knowledge type in the knowledge graph as the related knowledge of the disease-related symptoms-symptoms and forming a symptom-weight dictionary;
the case similarity obtaining module is used for calculating the symptom similarity degree, the diagnosis similarity degree and the case similarity degree of all other cases by taking the symptoms and the diagnoses of the target case as comparison standards based on the transmitted symptom and diagnosis lists and the symptom weight dictionary of all cases in the case library, and storing the symptom similarity degree, the diagnosis similarity degree and the case similarity degree in a dictionary form;
the similar case recommending module is used for obtaining a case list with the similarity from large to small by sequencing in a reverse order according to the symptoms, diagnosis and calculation results of the case similarity, and preferentially recommending cases with high similarity, same department and same age (0-6 years, 6-18 years, 18-60 years and over 60 years) by combining the age of the cases and the clinic information.
The working process of the symptom weight obtaining module in this embodiment is specifically as follows:
(1) Acquiring data related to symptoms and diseases in the knowledge map, and extracting all relevant knowledge of which the knowledge type (initial entity type-relation-termination entity type) is disease-related symptom-symptom;
(2) The initial entity of all relevant knowledge of the disease-relevant symptom-symptom is normalized and counted to obtain the total number N of the diseases in the knowledge base;
(3) Performing de-duplication extraction on terminating entity words of all relevant knowledge of disease-relevant symptom-symptom to obtain a symptom list of a knowledge base;
(4) Sequentially extracting initial entity normalization word sets corresponding to each symptom i according to symptom lists of the knowledge base, and obtaining the disease number N of diseases related to each symptom i by calculating the number of the initial entity normalization words i ；
(5) And for a symptom list of the knowledge base, sequentially obtaining the number N of diseases related to each symptom according to the obtained number i And the total number of acquired diseases N, resulting in a symptom-weight dictionary, the formula being as follows:
wherein, W i Symptom coefficients as the degree of case similarity; N-N i Indicating the number of diseases that are not associated with the symptom; n represents the total number of diseases in the knowledge base.
Example 3:
the present embodiment also provides an electronic device, including: a memory and a processor;
wherein the memory stores computer execution instructions;
the processor executes the computer-executable instructions stored by the memory to cause the processor to perform a similar case recommendation method in any of the embodiments of the present invention.
The processor may be a Central Processing Unit (CPU), but may also be other general purpose processors, digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like. The processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may be used to store computer programs and/or modules, and the processor may implement various functions of the electronic device by executing or executing the computer programs and/or modules stored in the memory and invoking data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the terminal, and the like. In addition, the memory may include high speed random access memory, and may include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a memory card only (SMC), a Secure Digital (SD) card, a flash memory card, at least one disk storage period, a flash memory device, or other volatile solid state memory device.
Example 4:
the embodiment of the invention also provides a computer-readable storage medium, wherein a plurality of instructions are stored, and the instructions are loaded by the processor, so that the processor executes the similar case recommendation method in any embodiment of the invention. Specifically, a system or an apparatus equipped with a storage medium on which software program codes that realize the functions of any of the embodiments described above are stored may be provided, and a computer (or a CPU or MPU) of the system or the apparatus is caused to read out and execute the program codes stored in the storage medium.
In this case, the program code itself read from the storage medium can realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.
Examples of the storage medium for supplying the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (e.g., CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RYM, DVD-RW, DVD + RW), a magnetic tape, a nonvolatile memory card, and a ROM. Alternatively, the program code may be downloaded from a server computer via a communications network.
Further, it should be clear that the functions of any one of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform a part or all of the actual operations based on instructions of the program code.
Further, it is to be understood that the program code read out from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion unit connected to the computer, and then causes a CPU or the like mounted on the expansion board or the expansion unit to perform part or all of the actual operations based on instructions of the program code, thereby realizing the functions of any of the above-described embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A similar case recommendation method is characterized by comprising the following specific steps:
data preprocessing: extracting the symptoms and diagnosis information of the electronic medical record of the admission record of the case; processing the cases into standard symptoms and standard diagnosis lists by taking the cases as units, extracting treatment information when the cases are treated and storing the treatment information in a dictionary, and assisting in obtaining the similarity degree of the cases subsequently; wherein, the information of seeing a doctor includes age and department;
acquiring symptom weight: extracting the weight of the symptoms in disease diagnosis based on the symptoms of the knowledge map and the knowledge related to the disease;
acquiring the similarity degree of cases: acquiring the similarity degree of other cases and the target case according to the symptoms and the diagnosis list of the target case and the fusion weight;
similar case recommendation: according to the calculation results of symptoms, diagnosis and case similarity, a case list with the similarity from large to small is obtained by sequencing in a reverse order, and cases with high similarity, same department and same age are preferentially recommended according to the age of the cases and the clinic information.
2. The method of recommending similar cases according to claim 1, wherein the preprocessing of the data further comprises normalizing the symptoms and diagnostic information of all cases in the case bank as follows:
taking the case number as an index, and acquiring a symptom list and a diagnosis list of all cases in a case library in a cache;
according to the mapping relation table of the original words and the home words in the knowledge map, the home word list of case symptoms and diagnosis is obtained by matching the names of the original words and the entity types of the original words and stored in a cache; wherein, the original word name comprises a symptom name and/or a diagnosis name; the primitive word entity types include symptoms and/or diagnoses.
3. The similar case recommendation method according to claim 1, wherein the symptom weights are obtained as follows:
acquiring data related to symptoms and diseases in a knowledge map, extracting all related knowledge of which the knowledge type (initial entity type-relation-termination entity type) is disease-related symptom-symptom, and storing the knowledge in a symptom and disease knowledge table;
obtaining the disease number N of the disease associated with any symptom i by extracting the related knowledge of symptom-disease in the knowledge map i And the total number of diseases N in the knowledge base, thereby calculating a symptom weight.
4. The method for recommending similar cases as claimed in claim 3, wherein the number N of diseases associated with any symptom i is obtained by extracting the symptom-disease related knowledge from the knowledge-graph i And the total number of diseases N in the knowledge base, thereby calculating the symptom weight as follows:
performing de-duplication counting on the normalized words of the initial entities to obtain the total number N of diseases in the knowledge base;
carrying out duplication removal extraction on the termination entity normalized words in the symptom and disease knowledge table to obtain a symptom list of a knowledge base;
according to a symptom list of a knowledge base, sequentially extracting an initial entity normalization word set corresponding to each symptom i, and calculating the number of initial entity normalization words to obtain the number N of diseases related to the diseases of each symptom i i ；
Associating a number of diseases N to a disease according to a list of symptoms of a knowledge base and each symptom i i And obtaining a symptom-weight dictionary, wherein the formula is as follows:
wherein, W i Symptom coefficients as the degree of case similarity; N-N i Indicating the number of diseases that are not associated with the symptom; n represents the total number of diseases in the knowledge base.
5. The similar case recommendation method according to any one of claims 1-3, wherein the degree of similarity between cases is obtained as follows:
matching the symptoms and the diagnosis lists of other cases in the case library by using the symptoms and the diagnosis lists of the target case to respectively obtain intersection sets of the symptoms and the diagnoses of the target case and the other cases;
acquiring symptom similarity according to an intersection set of symptoms of the target case and other cases;
acquiring diagnosis similarity according to the intersection set of the diagnoses of the target case and other cases;
carrying out weighted operation on the symptom similarity and the diagnosis similarity to obtain the case similarity;
the basic information of the case is combined, and the basic information is stored in a cache in a table form with the case number, age, department, symptom similarity degree, diagnosis similarity degree and case similarity degree as a header.
6. The similar case recommendation method according to claim 5, wherein the symptom similarity is specifically as follows:
let the symptom list of the compared cases be [ i 1 ,i 2 ,i 3 …i n ](ii) a The set of symptom intersections of the compared case and the target case is [ i 1 ,i 2 ,i 3 …i m ](ii) a Symptom weight is W i Then, the calculation formula of the corresponding case and the target case is as follows:
wherein S represents the degree of symptom similarity; w i1 +W i2 +W i3 +…+W im Accumulating the symptom weights representing the symptom intersection set to obtain an intersection score; n represents the number of elements in the symptom list of the compared cases;
the diagnostic similarity is specifically as follows:
let the list of diagnoses of the cases to be compared be [ z ] 1 ,z 2 ,z 3 …z k ](ii) a The set of diagnostic intersections of the compared case and the target case is [ z ] 1 ,z 2 ,z 3 …z j ](ii) a Then the corresponding caseThe degree of similarity to symptoms of the target case is calculated as follows:
wherein, S' represents the degree of symptom similarity; j represents the number of diagnostic intersection sets; k represents the number of elements in the diagnosis list of the compared cases;
the case similarity is obtained by averaging the symptom similarity S and the diagnosis similarity S', and is expressed as follows:
wherein F represents the degree of case similarity.
7. A similar case recommendation system, comprising a data layer and a data processing layer; the data layer comprises a case management module, and the data processing layer comprises a preprocessing module, a symptom weight acquisition module, a case similarity acquisition module and a similar case recommendation module;
the case management module is used for maintaining basic information of cases, structured large-section texts of electronic medical records and symptom and diagnosis information;
the preprocessing module is used for carrying out normalization processing on symptoms and diagnosis information, carrying out mapping from original words to normalized words on all cases in a case library including the symptoms and the diagnosis information of a target case according to an original word and normalized word mapping table of a knowledge map, obtaining standard symptoms and standard diagnosis information of the case and storing the standard symptoms and the standard diagnosis information into a list form;
the symptom weight acquisition module is used for acquiring symptom weights based on the knowledge type in the knowledge graph as the related knowledge of the disease-related symptoms-symptoms and forming a symptom-weight dictionary;
the case similarity obtaining module is used for calculating the symptom similarity degree, the diagnosis similarity degree and the case similarity degree of all other cases by taking the symptoms and the diagnoses of the target case as comparison standards based on the transmitted symptom and diagnosis lists and the symptom weight dictionary of all cases in the case library, and storing the symptom similarity degree, the diagnosis similarity degree and the case similarity degree in a dictionary form;
the similar case recommending module is used for obtaining a case list with the similarity degree from large to small by sequencing in a reverse order according to the symptoms, diagnosis and calculation results of the case similarity degree, and preferentially recommending cases with high similarity degree, same department and same age period by combining the age of the cases and the clinic information.
8. The similar case recommendation system according to claim 7, wherein the symptom weight obtaining module operates as follows:
(1) Acquiring data related to symptoms and diseases in the knowledge map, and extracting all related knowledge of which the knowledge type (initial entity type-relation-termination entity type) is disease-related symptom-symptom;
(2) The initial entity of all relevant knowledge of the disease-relevant symptom-symptom is normalized and counted to obtain the total number N of the diseases in the knowledge base;
(3) Carrying out de-duplication extraction on termination entity normalized words of all relevant knowledge of disease-relevant symptom-symptom to obtain a symptom list of a knowledge base;
(4) Sequentially extracting initial entity normalization word sets corresponding to each symptom i according to the symptom lists of the knowledge base, and obtaining the disease number N of the diseases related to each symptom i by calculating the number of the initial entity normalization words i ；
(5) And for the symptom list of the knowledge base, sequentially obtaining the number N of diseases associated with each symptom according to the obtained symptoms i And the acquired total number of diseases N, resulting in a symptom-weight dictionary, the formula is as follows:
wherein, W i Symptom coefficients as the degree of case similarity; N-N i Indicating the number of diseases that are not associated with the symptom;n represents the total number of diseases in the knowledge base.
9. An electronic device, comprising: a memory and at least one processor;
wherein the memory has stored thereon a computer program;
the at least one processor executing the memory-stored computer program causes the at least one processor to perform the similar case recommendation method of any of claims 1-6.
10. A computer-readable storage medium, in which a computer program is stored which is executable by a processor to implement the method of similar case recommendation according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211077445.0A CN115631823A (en) | 2022-09-05 | 2022-09-05 | Similar case recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211077445.0A CN115631823A (en) | 2022-09-05 | 2022-09-05 | Similar case recommendation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115631823A true CN115631823A (en) | 2023-01-20 |
Family
ID=84902344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211077445.0A Pending CN115631823A (en) | 2022-09-05 | 2022-09-05 | Similar case recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115631823A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116504354A (en) * | 2023-06-28 | 2023-07-28 | 合肥工业大学 | Intelligent service recommendation method and system based on intelligent medical treatment |
-
2022
- 2022-09-05 CN CN202211077445.0A patent/CN115631823A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116504354A (en) * | 2023-06-28 | 2023-07-28 | 合肥工业大学 | Intelligent service recommendation method and system based on intelligent medical treatment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112037880B (en) | Medication recommendation method, device, equipment and storage medium | |
| Fang et al. | Feature Selection Method Based on Class Discriminative Degree for Intelligent Medical Diagnosis. | |
| CN110069779B (en) | Symptom entity identification method of medical text and related device | |
| CN112183026B (en) | ICD (interface control document) encoding method and device, electronic device and storage medium | |
| CN113707297A (en) | Medical data processing method, device, equipment and storage medium | |
| CN110299209B (en) | Similar medical record searching method, device and equipment and readable storage medium | |
| CN111710420A (en) | Complication morbidity risk prediction method, system, terminal and storage medium based on electronic medical record big data | |
| CN112257422A (en) | Named entity normalization processing method and device, electronic equipment and storage medium | |
| US11604778B1 (en) | Taxonomic fingerprinting | |
| CN112885478A (en) | Medical document retrieval method, medical document retrieval device, electronic device, and storage medium | |
| CN113707303A (en) | Method, device, equipment and medium for solving medical problems based on knowledge graph | |
| CN109213871A (en) | Patient information knowledge mapping construction method, readable storage medium storing program for executing and terminal | |
| CN110752027B (en) | Electronic medical record data pushing method, device, computer equipment and storage medium | |
| WO2022222942A1 (en) | Method and apparatus for generating question and answer record, electronic device, and storage medium | |
| WO2022229964A1 (en) | Method of generating a diseases database, usage of the diseases database, and system therefor | |
| CN111785383A (en) | Data processing method and related equipment | |
| WO2022227171A1 (en) | Method and apparatus for extracting key information, electronic device, and medium | |
| CN115631823A (en) | Similar case recommendation method and system | |
| CN114420279A (en) | Medical resource recommendation method, device, equipment and storage medium | |
| CN110569349A (en) | Big data-based method, system, equipment and storage medium for pushing articles for education | |
| CN112635072A (en) | ICU (intensive care unit) similar case retrieval method and system based on similarity calculation and storage medium | |
| Wang et al. | Research on named entity recognition of doctor-patient question answering community based on bilstm-crf model | |
| CN113571196A (en) | Method and device for constructing medical training sample and method for retrieving medical text | |
| CN108831560B (en) | Method and device for determining medical data attribute data | |
| CN116578704A (en) | Text emotion classification method, device, equipment and computer readable medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |