CN112700861B - Accompanying symptom interaction method and accompanying symptom interaction system - Google Patents
Accompanying symptom interaction method and accompanying symptom interaction system Download PDFInfo
- Publication number
- CN112700861B CN112700861B CN202011563391.XA CN202011563391A CN112700861B CN 112700861 B CN112700861 B CN 112700861B CN 202011563391 A CN202011563391 A CN 202011563391A CN 112700861 B CN112700861 B CN 112700861B
- Authority
- CN
- China
- Prior art keywords
- symptom
- candidate
- accompanying
- symptoms
- chief complaint
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/042—Backward inferencing
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- Primary Health Care (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Epidemiology (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Software Systems (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention provides a concomitant symptom interaction method and system, and belongs to the technical field of intelligent medical treatment. The method comprises the following steps: s1) forming a knowledge base according to the existing medical knowledge data; s2) acquiring a candidate accompanying symptom complete set of the chief complaint symptoms from the knowledge base based on the acquired patient chief complaint symptoms, and acquiring a co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom in the candidate accompanying symptom complete set; s3) obtaining a accompanied symptom candidate set according to a preset rule; s4) correcting the co-occurrence weight scores according to a preset rule, and sorting according to the co-occurrence weight score scores to obtain a sorted accompanied symptom candidate set; s5) carrying out the pushing flow of the ordered accompanied symptom candidate set according to a preset rule. The scheme of the invention ensures that the symptom acquisition interaction process is more orderly, credible and efficient, directly obtains the optimal information helpful for diagnosis, avoids invalid consultation, reduces the number of interaction rounds, and simultaneously improves the participation degree and the satisfaction degree of users.
Description
Technical Field
The invention relates to the technical field of intelligent medical treatment, in particular to an accompanying symptom interaction method and an accompanying symptom interaction system.
Background
With the increasing maturity of artificial intelligence technology, intelligent inquiry technology has also been developed. In the process of collecting patient information, how to accurately deduce and list candidate accompanying symptoms of a patient according to the effective primary symptoms or chief complaints of the patient for the patient to select is an important means for embodying intellectualization. Most inquiry interactive system can carry out symptom inquiry one by one at consultation patient symptom in-process, and the number of rounds of letting the patient answer is more, and the operation is wasted time and energy, causes relatively poor experience sense and participation sense for the patient, also reduces system information collection efficiency simultaneously, increases the collection cost. In addition, the inquiry sequence is disordered, repeated consultation, invalid consultation, low-level question, paradoxical question and the like are frequently generated, the collected information has little value or little diagnostic significance to diagnosis, the workload of subsequent treatment and screening of doctors is increased, and diagnosis of the doctors is possibly misled and even the health of patients is endangered. Aiming at the problems of excessive inquiry rounds and low information arrangement accuracy of the conventional human-computer interaction inquiry system, which cause low user feeling, a new accompanying symptom interaction method needs to be created.
Disclosure of Invention
The embodiment of the invention aims to provide a concomitant symptom interaction method and a concomitant symptom interaction system, so as to at least solve the problems of excessive inquiry rounds and low information arrangement accuracy of the conventional man-machine interaction inquiry system, which cause low user feeling.
In order to achieve the above object, a first aspect of the present invention provides a concomitant symptom interaction method applied to an interactive inquiry system, the method including: s1) forming a knowledge base according to the existing medical knowledge data; s2) based on the obtained patient chief complaint symptoms, obtaining a candidate accompanying symptom complete set of the chief complaint symptoms from the knowledge base, and obtaining a co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom in the candidate accompanying symptom complete set; s3) filtering the candidate accompanied symptom complete set according to a preset rule to obtain an accompanied symptom candidate set; s4) correcting the co-occurrence weight scores of the candidate accompanying symptoms and the chief complaint symptoms in the accompanying symptom candidate set according to a preset rule, and sorting the candidate accompanying symptoms according to the corrected co-occurrence weight score scores to obtain a sorted accompanying symptom candidate set; s5) carrying out the pushing process of the ordered accompanied symptom candidate set according to a preset rule, judging the state of the pushing process in real time until the state of the pushing process reaches a preset standard, and determining to finish the acquisition of the accompanied symptom information of the patient.
Optionally, in step S1), the existing medical knowledge data at least includes: the clinical diagnosis relationship data or medical literature is used for obtaining common and rare relationships between diagnosis diseases and symptoms, relationships between diseases and people, relationships between diseases and related medical histories, relationships between diseases and physical signs and relationships between diseases and examination.
Optionally, the obtaining of the co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom in the candidate accompanying symptom complete set includes: respectively setting a relation weight score of each candidate accompanying symptom in the candidate accompanying symptom total set and a relation weight score of the chief complaint symptom according to the knowledge base; respectively calculating the co-occurrence weight score of each candidate accompanying symptom and the main complaint symptom according to the relationship weight score of each candidate accompanying symptom and the relationship weight score of the main complaint symptom; the calculation formula is as follows:
Vrm＝prik+prim
wherein M ∈ M ═ 1, 2.. said, M is said candidate set of accompanying symptoms; prik(ii) a relationship weight score for the complaint symptom; prim(ii) a relationship weight score for the mth candidate concomitant symptom; vrmAnd (3) scoring the co-occurrence weight of the m candidate accompanying symptoms and the chief complaint symptoms.
Optionally, in step S3), the filtering the candidate complete set of accompanying symptoms according to a preset rule to obtain a candidate set of accompanying symptoms includes: acquiring patient image information, and filtering candidate accompanying symptoms which are not in accordance with actual symptoms of patients in the candidate accompanying symptom complete set according to the patient image information to obtain a candidate accompanying symptom intermediate set; and comparing the co-occurrence weight scores of the candidate accompanying symptoms in the candidate accompanying symptom intermediate set and the chief complaint symptoms with a preset score threshold, filtering the candidate accompanying symptoms of which the co-occurrence weight scores of the candidate accompanying symptoms and the chief complaint symptoms are smaller than the preset score threshold, and taking the filtered candidate accompanying symptom intermediate set as the accompanying symptom candidate set.
Optionally, in step S4), the modifying the co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom in the accompanying symptom candidate set according to a preset rule includes: calculating TF-IDF values of candidate concomitant symptoms in the concomitant symptom candidate set respectively; updating each candidate accompanying symptom relation weight score according to the TF-IDF value of each candidate accompanying symptom and the relation weight score of each candidate accompanying symptom; the calculation rule is as follows:
prii,new＝tfidfi×prii
wherein I belongs to I ═ I (1, 2...., I), I is a candidate set of accompanying symptoms; tfidfiA TF-IDF value for the ith candidate co-symptom in said set of co-symptom candidates;priia relationship weight score for the ith candidate concomitant symptom in the set of concomitant symptom candidates; prib,newUpdating the value of the relationship weight score for the type b candidate accompanying symptom in the accompanying symptom candidate set; modifying the co-occurrence weight scores of the candidate accompanying symptoms and the main complaint symptoms in the accompanying symptom candidate set according to the updated value of the candidate accompanying symptom relation weight score; the correction relation is as follows:
Vri＝prik+prii,new
wherein VriAnd (4) scoring the co-occurrence weight of the corrected i-th candidate accompanying symptom and the chief complaint symptom.
Optionally, the separately calculating the TF-IDF value of each candidate concomitant symptom in the concomitant symptom candidate set includes: calculating TF values and IDF values of the candidate concomitant symptoms respectively, and then calculating TF-IDF values of the candidate concomitant symptoms according to the TF values and the IDF values of the candidate concomitant symptoms; wherein, the calculation rule of the TF-IDF value is as follows:
tfidfi＝tfi,j×idfi
wherein J ∈ J ═ 1, 2.... J, J is a department set; tf isi,jTF values in the jth department for the ith candidate concomitant symptom; idfiThe IDF value of the ith candidate concomitant symptom.
Optionally, tf isi,jThe calculation rule of (1) is:
wherein K belongs to K ═ 1, 2.. times, K, and K is all symptom sets appearing in department j; sigmaknk,jIs the sum of the number of occurrences of all symptoms in department j; n isi,j(ii) the number of occurrences of candidate concomitant symptoms in department j; the idfiThe calculation rule of (1) is:
wherein | D | is the total number of departments; i ∈ J | { J: i ∈ J } | is the total number of departments presenting with the candidate accompanying symptom in the ith.
In optional step S5), the performing, according to a preset rule, a push procedure of the sorted accompanied symptom candidate set, and determining a state of the push procedure in real time until the state of the push procedure reaches a preset standard includes: s51) sequentially determining N candidate concomitant symptoms in the concomitant symptom candidate set; s52) the determined N candidate accompanying symptoms are sequentially pushed to a patient end for selection by a user, and response information corresponding to the user is recovered; s53) sequentially verifying whether the response information matches a preset disease hypothesis: if not, the steps S51-S53 are repeatedly executed from the first candidate accompanying symptom that has not been verified as answer information until all candidate accompanying symptoms in the accompanying symptom candidate set are verified.
A second aspect of the present invention provides a companion symptom interactive system, comprising: the acquisition unit is used for acquiring patient chief complaint symptoms and patient portrait information; the processing unit is used for forming a knowledge base according to existing medical knowledge data, acquiring a candidate accompanying symptom complete set of the chief complaint symptoms from the knowledge base based on the patient chief complaint symptoms, and acquiring a co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom in the candidate accompanying symptom complete set; the candidate accompanied symptom complete set is further used for filtering the candidate accompanied symptom complete set according to a preset rule to obtain an accompanied symptom candidate set; the symptom candidate set is also used for correcting the co-occurrence weight scores of the candidate accompanying symptoms and the chief complaint symptoms in the accompanying symptom candidate set according to a preset rule, and sorting the candidate accompanying symptoms according to the corrected co-occurrence weight score values to obtain a sorted accompanying symptom candidate set; the storage unit is used for storing the knowledge base; the pushing unit is used for pushing the sorted accompanied symptom candidate set according to a preset rule; the processing unit is further used for judging the state of the pushing flow in real time until the state of the pushing flow reaches a preset standard, and determining that the acquisition of the accompanying symptom information of the patient is completed.
In another aspect, the present invention provides a computer-readable storage medium having instructions stored thereon, which when executed on a computer, cause the computer to perform the above-described companion symptom interaction method.
Through the technical scheme, in order to accurately acquire the information of the patient, from the perspective of differential diagnosis, the concept of reverse reasoning is adopted based on the symptom co-occurrence principle in the diagnosis data of expert knowledge, the most possible related symptoms of the diagnosis result are orderly listed, and are displayed to the patient to be selected according to the probability sequence, so that more symptom information of the patient can be mined more quickly and accurately (the patient can give other symptom information of the patient more quickly and accurately), and reasonable information is provided for subsequent differential diagnosis and generation of medical records. The scheme of the invention ensures that the symptom acquisition interaction process is more orderly, credible and efficient, directly obtains the optimal information helpful for diagnosis, avoids invalid consultation, reduces the number of interaction rounds, and simultaneously improves the participation degree and the satisfaction degree of users.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention without limiting the embodiments of the invention. In the drawings:
FIG. 1 is a flow chart of the steps of a concomitant symptom interaction method provided by one embodiment of the present invention;
FIG. 2 is a flowchart of the steps provided by one embodiment of the present invention to acquire a concomitant symptom candidate set;
FIG. 3 is a flowchart of the steps provided by one embodiment of the present invention to rank candidate accompanying symptoms;
FIG. 4 is a flowchart illustrating the steps of a TF-IDF value calculation method for each candidate concomitant symptom according to one embodiment of the present invention;
fig. 5 is a system configuration diagram of a companion symptom interactive system according to an embodiment of the present invention.
Description of the reference numerals
10-an acquisition unit; 20-a processing unit; 30-a storage unit; 40-push unit.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present invention, are given by way of illustration and explanation only, not limitation.
Fig. 5 is a system configuration diagram of a companion symptom interactive system according to an embodiment of the present invention. As shown in fig. 5, the embodiment of the present invention provides a concomitant symptom interaction system, including: the acquisition unit 10 is used for acquiring patient chief complaint symptoms and patient portrait information; the processing unit 20 is used for forming a knowledge base according to the existing medical knowledge data, acquiring a candidate accompanying symptom complete set of the chief complaint symptoms according to the knowledge base, and acquiring a co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom; the candidate accompanied symptom complete set is further used for filtering the candidate accompanied symptom complete set according to a preset rule to obtain an accompanied symptom candidate set; the system is also used for correcting the co-occurrence weight scores of the candidate accompanying symptoms and the main complaint symptoms in the accompanying symptom candidate set according to a preset rule, and sorting the candidate accompanying symptoms according to the size sequence of the scores to obtain a sorted accompanying symptom candidate set; a storage unit 30 for storing the knowledge base; a pushing unit 40, for the user to sequentially push the candidate accompanying symptoms in the ordered accompanying symptom candidate set for the user to select, and to recover the response information of the user; the processing unit 20 is further configured to analyze whether the pushing process state meets a preset standard in real time.
Fig. 1 is a flowchart of a method for accompanying symptom interaction according to an embodiment of the present invention. As shown in fig. 1, embodiments of the present invention provide a concomitant symptom interaction method that lists multiple symptoms in a single interaction for patient selection may be an effective and rapid way to gather patient information. Moreover, in order to accurately acquire patient information, from the perspective of differential diagnosis, on the basis of the symptom co-occurrence principle in diagnosis data of expert knowledge, the idea of reverse reasoning is adopted, the most likely related symptoms to the diagnosis result are orderly listed, and are displayed to the patient to be selected according to the probability sequence, so that more symptom information of the patient can be mined more quickly and accurately, and reasonable information is provided for subsequent differential diagnosis and generation of medical records. Firstly, the system intelligently judges a possible candidate accompanying symptom complete set based on knowledge in a knowledge base according to the patient primary symptoms or chief complaints obtained by initial consultation; secondly, filtering candidate accompanying symptoms according to the acquired symptoms and other portrait information of the patient, for example, removing the symptoms which do not accord with sex characteristics, the symptoms which do not accord with age characteristics and the symptoms which are inconsistent or repeated with the known symptoms; thirdly, acquiring candidate accompanying symptom probability ordering by adopting the co-occurrence relation and other attribute relations of all symptoms under the diseases in the knowledge base; and finally, judging conditions, if the conditions meet preset threshold conditions, finishing the information acquisition of the symptom link and entering other subsequent interactive links, and otherwise, performing the next round of accompanying consultation again. Specifically, the method comprises the following steps:
step S10: and forming a knowledge base according to the existing medical knowledge data.
Specifically, in order to accurately acquire symptom information that may co-occur under a certain disease, a large amount of medical knowledge sample acquisition is required. Through the medical knowledge, under the condition that a certain symptom is found, the conventional co-occurrence symptom is judged to narrow the disease range, and the diagnosis accuracy and efficiency of doctors are improved. Preferably, clinical diagnostically relevant data acquisition is performed on the one hand, and in order to make the acquired information more accurate and authoritative, such information collation is preferably performed by various famous physicians or medical professionals. The medical knowledge information acquired by the medical experts is more practical and has important reference significance corresponding to the aim of actual patients. On the other hand, medical knowledge acquisition is performed through medical documents, namely, the knowledge acquisition is performed through the existing modes of medical web pages, medical data, medical academic papers and the like. By the mode, on the basis of experience data of medical experts which are older than months, a better academic theory basis is provided for medical knowledge through advanced medical documents, and the empirical knowledge and the advanced theoretical knowledge are integrated, so that the medical knowledge obtained through arrangement is more authoritative. The processing unit 20 sorts and classifies the acquired information to generate a corresponding medical knowledge database, and stores the medical knowledge database in the storage unit 30, so as to facilitate later extraction and use. The method and the core idea are that the related symptoms are quickly extracted and confirmed through a certain symptom or a certain disease, so the collated knowledge base at least comprises the following information: the common rare relationship between the diagnosed disease and symptoms, the relationship between the disease and the population, the relationship between the disease and the related medical history, the relationship between the disease and the physical signs, and the relationship between the disease and the examination. By using these relational data, it is possible to quickly extract possible accompanying symptom information by relating a certain symptom or a certain disease, i.e., a search keyword.
In one possible embodiment, the diseases, the attribute parameters corresponding to the diseases and the accompanying symptoms are arranged into a complete schema rule, taking kawasaki disease as an example, the schema rule about kawasaki disease is shown in table one:
TABLE 1 Kawasaki disease schema rule
Step S20: acquiring patient chief complaint symptoms, acquiring a candidate accompanying symptom complete set of the chief complaint symptoms according to the knowledge base, and acquiring a co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom.
Specifically, the collecting unit 10 collects the chief complaint symptoms of the user, then transmits the chief complaint symptoms to the processing unit 20, and the processing unit 20 searches the knowledge base by using the chief complaint symptoms as search keywords to obtain possible diseases of the chief complaint symptoms and corresponding possible accompanying symptoms. For example, if the patient's chief complaint is fever and rash, and it is found that the diseases with these two symptoms may be many diseases such as kawasaki disease, viral infection, bacterial infection, chicken pox, scarlet fever, etc., but it is difficult to diagnose the patient only by these two typical symptoms, the processing unit 20 extracts the schema rules of the possible diseases, and then extracts other symptoms of each disease, and several of these other accompanying symptoms constitute a candidate accompanying symptom set.
After the candidate complete set of accompanying symptoms is obtained, the probability that each candidate complete set and the chief complaint symptom appear together needs to be judged, that is, the more practical the accompanying symptoms appearing together with the chief complaint symptom are, the smaller the scope of the disease can be. For example, in kawasaki disease, conjunctival congestion is a symptom with a relatively obvious characteristic in addition to fever and rash, and the symptom is not present in other diseases at all, so if information on whether the patient has the symptom or not can be obtained, a large number of suspicious diseases can be filtered, and the disease range can be further narrowed. In clinical medical diagnosis, a plurality of diseases exist in a department, one disease has symptom expressions with different degrees, relationship weight scores are set for all symptoms under a certain disease according to the symptom expression degrees under the diseases, the rule is set to convert the very common, rare and negative relationships in the schema rule into specific scores, the preset score is preferably 1-10, wherein the very common score is the highest, and the negative score is 0. The chief complaint symptoms are scored by the same rule for relationship weight, for example, the relationship weight score of fever and rash under one disease scope of kawasaki disease is set to 10 points, and the relationship weight score of lip bleeding under one disease scope of kawasaki disease is set to 5 points. Then, calculating the co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom according to the relationship weight score of the chief complaint symptom and the relationship weight score of each candidate accompanying symptom, wherein the calculation rule is as follows:
Vrm＝prik+prim
wherein M belongs to M ═ M (1, 2...., M), and M is a candidate accompanied symptom complete set; prikA relationship weight score for a chief complaint symptom; prim(ii) a relationship weight score for the mth candidate concomitant symptom; vrmAnd (4) scoring the co-occurrence weight of the m candidate accompanying symptoms and the chief complaint symptoms. That is, for the above example, when the lips bleedWhen fever and rash were the chief complaint symptoms as candidate accompanying symptoms, the weight score of the co-occurrence of the accompanying symptoms and the chief complaint symptoms was 15 points in the Kawasaki disease domain.
Step S30: and filtering the candidate complete set of accompanying symptoms according to a preset rule to obtain a candidate set of accompanying symptoms. Specifically, as shown in fig. 2, the method includes the following steps:
step S301: acquiring patient image information, and filtering the candidate accompanying symptoms in the candidate accompanying symptom complete set which do not accord with the actual candidate accompanying symptoms of the patient according to the patient image information to obtain an accompanying symptom candidate intermediate set.
Specifically, when generating the candidate accompanying symptom complete set, only the chief complaint symptom of the patient is considered, and other information of the patient is considered. However, certain diseases occur only in a portion of the population, such as sexually divided diseases and age-divided diseases. By the Kawasaki disease schema rule, people can know that the information of the easily-occurring people and the information of the people who are not likely to be caused by the disease are recorded in the schema rule of each disease. The acquisition unit 10 acquires image information of a patient, the image information of the patient including: the patient's age, sex, region, preference, family history, and past history. After acquiring the patient image information, the processing unit 20 performs disease-enabling and corresponding candidate accompanying symptom filtering according to the patient image information, for example, if the patient is a male, all gynecological diseases in the suspicious disease can be excluded, and the corresponding gynecological symptoms can also be excluded correspondingly. Besides filtering out accompanying symptoms which do not conform to the age and sex of the patient, the image information of the patient may also include confirmed symptom information, and in order to avoid the reduction of the experience of the patient caused by secondary confirmation, the symptom information which is inquired in the image information of the patient is preferably filtered out. After candidate accompanying symptoms which do not conform to the age, sex and inquiry of the patient are filtered out, the remaining candidate accompanying symptoms are combined into an accompanying symptom candidate intermediate set.
Step S302: respectively comparing the co-occurrence weight scores of the candidate accompanying symptoms and the chief complaint symptoms with a preset score threshold, filtering candidate accompanying symptoms of which the co-occurrence weight scores of the candidate accompanying symptoms and the chief complaint symptoms are smaller than the candidate accompanying symptoms corresponding to the preset score threshold, and obtaining an accompanying symptom candidate set.
Specifically, in step S20, a co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom is obtained, and a higher co-occurrence weight score of a candidate accompanying symptom and a chief complaint symptom indicates that the candidate accompanying symptom is more likely to be caused by the disease in the same manner as the chief complaint symptom in the current disease scope. Conversely, the smaller the co-occurrence weight score between the candidate accompanying symptom and the chief complaint symptom, the smaller the effort for confirming a disease by the candidate accompanying symptom. To reduce the number of data calculations, candidate concomitant symptoms with little effect on disease identification were excluded. A score threshold is preset, for example, 15 points, and when the co-occurrence weight score of a candidate accompanying symptom and a chief complaint symptom is less than 15 points, the candidate accompanying symptom is excluded from the candidate intermediate set of accompanying symptoms. And sequentially comparing the co-occurrence weight scores of the candidate accompanying symptoms and the chief complaint symptoms in the accompanying symptom candidate intermediate set with a preset score threshold, and excluding all candidate accompanying symptoms corresponding to the co-occurrence weight scores smaller than the preset score threshold to obtain an accompanying symptom candidate set.
Step S40: and correcting the co-occurrence weight scores of the candidate accompanying symptoms and the main complaint symptoms in the accompanying symptom candidate set according to a preset rule, and sequencing the candidate accompanying symptoms according to the value order to obtain a sequenced accompanying symptom candidate set.
Specifically, in order to prevent the co-occurrence score of common symptoms in a general department from being too high in a single department, a concept of the importance degree of the symptoms in the single department is introduced, and the importance degree of each symptom in the department is acquired. Specifically, as shown in fig. 3, the method includes the following steps:
step S401: and calculating TF-IDF values of the candidate concomitant symptoms in the concomitant symptom candidate intermediate set respectively.
Specifically, the intermediate set of accompanying symptom candidates contains all possible accompanying symptoms of a disease, but some symptoms are only commonly found in a small part of individuals, namely, some symptoms are not necessarily found in the disease, and the incidental symptoms are also commonly found, namely, the symptoms which accidentally cause a current disease and the common symptoms of the disease coexist. If the frequency of occurrence of a symptom corresponding to a certain disease is smaller, the probability that the corresponding symptom is an occasional co-occurrence symptom is higher. In order to achieve the above object, it is preferable that the TF-IDF value is introduced to determine the frequency of occurrence of each candidate accompanying symptom, and the smaller the TF-IDF value is, the greater the possibility that the corresponding accompanying symptom is an occasional co-occurring symptom. Specifically, as shown in fig. 4, the method includes the following steps:
step S4011: the TF value of each candidate accompanying symptom is calculated.
Specifically, in the TF-IDF value, TF indicates the frequency of occurrence of a term in a document, that is, the frequency of occurrence of a symptom in a certain department in the present invention, and indicates that the symptom is more likely to be a disease in the department as the frequency of occurrence of the symptom in the department is higher. Assuming that a hospital has J departments, a department set J is composed of all departments, wherein J ∈ J ═ 1, 2. The emerging symptoms within each department are then obtained, which all together make up the emerging symptom set K for that department, where K e K ═ 1, 2. Then assuming that a total of I candidate accompanying symptoms are obtained in the accompanying symptom candidate set, the accompanying symptom candidate set is denoted as I, where I ∈ I ═ 1, 2. Based on the above settings, TF values of each candidate accompanying symptom are calculated according to the following calculation rule:
therein, sigmaknk,jIs the sum of the number of occurrences of all symptoms in department j; n isi,jIs the number of times the ith candidate accompanying symptom appears in department j.
Step S4012: the IDF value of each candidate accompanying symptom is calculated.
Specifically, among TF-IDF values, the main idea of IDF is: if the documents containing a certain entry are fewer and the IDF is larger, the entry has good category distinguishing capability. The fewer departments with candidate accompanying symptoms are present in all departments, the easier it is for the accompanying symptoms to be diagnosed, i.e. ranging over only a small fraction of the departments. The IDF value of each candidate accompanying symptom is calculated by each set setting in step S3022, and the calculation rule is:
wherein | D | is the total number of departments; i ∈ J | { J: i ∈ J } | is the total number of departments presenting with the candidate accompanying symptom in the ith.
Step S4013: the TF-IDF value of each candidate accompanying symptom is calculated.
Specifically, among TF-IDF values of 0 to 1, the symptoms frequently appearing in general departments have lower TF-IDF values, whereas those frequently appearing in single departments and rarely appearing in other departments have higher TF-IDF values. The calculation rule of the TF-IDF value is as follows:
tfidfi＝tfi,j×idfi
step S402: and updating the relation weight score of each candidate accompanying symptom according to the TF-IDF value of each candidate accompanying symptom and the relation weight score of each candidate accompanying symptom.
Specifically, in order to avoid the co-occurrence score of common symptoms in general departments in a single department from being too high, the calculated candidate accompanying symptoms have too large effect, but the actual significance is not large, so that the diagnosis of the disease is deviated. Updating the candidate accompanying symptom relation weight score through the TF-IDF value of each candidate accompanying symptom, so that the action effect of each candidate accompanying symptom in diagnosis of each disease is closer to the reality. The relationship weight score update rule of each candidate accompanying symptom is as follows:
prii,new＝tfidfi×prii
wherein prib,newAn updated relationship weight score for the b-th candidate accompanying symptom in the accompanying symptom candidate set.
In step S403, a co-occurrence weight score between each candidate accompanying symptom and the chief complaint symptom in the accompanying symptom candidate set is performed according to the updated value of each candidate accompanying symptom relationship weight score.
Specifically, since the co-occurrence weight score between each candidate accompanying symptom and the chief complaint symptom is the overall effect of the relationship weight score between each candidate accompanying symptom and the relationship weight score between the chief complaint symptom, it is necessary to correct the co-occurrence weight score between each candidate accompanying symptom and the chief complaint symptom after updating each candidate accompanying symptom relationship weight score. The correction rule is as follows:
Vri＝prik+prii,new
wherein VriAnd (4) scoring the co-occurrence weight of the corrected i-th candidate accompanying symptom and the chief complaint symptom.
Step S404: and sorting the candidate accompanying symptoms according to the corrected co-occurrence weight score value magnitude order to obtain a sorted accompanying symptom candidate set.
Specifically, the higher the co-occurrence weight score of the candidate accompanying symptom and the main complaint symptom, the more the candidate accompanying symptom and the main complaint symptom can prove that the current suspected disease is the target disease under a certain disease scope. The co-occurrence weight scores of each candidate accompanying symptom and the chief complaint symptom are transversely compared, and the larger the value is, the more the patient should confirm the co-occurrence weight scores, and the faster the disease range can be narrowed. And (4) carrying out sequential arrangement of the candidate accompanying symptoms according to the numerical values, and then obtaining a sorted accompanying symptom candidate set.
Step S50: and carrying out a pushing process of the ordered accompanied symptom candidate set according to a preset rule, and analyzing the state of the pushing process in real time until the pushing process reaches a preset standard, thereby completing the acquisition of the accompanied symptom information of the patient.
Specifically, after obtaining the high-probability accompanying symptom candidate set corresponding to the chief complaint symptom, the processing unit 20 sends the sorted accompanying symptom candidate set to the pushing unit 40, and the pushing unit 40 selects a fixed number of candidate accompanying symptoms to be pushed to the patient end according to a preset rule for the patient to select and confirm, preferably, the number of candidate accompanying symptoms pushed each time is 4-6. For example, from the first of the ordered accompanying symptom candidate sets, the first 6 candidate accompanying symptoms are selected, and corresponding keyword labels are generated for the user to select. The pushing unit 40 obtains the response information of the user in real time and transmits the response information back to the processing unit 20, and the processing unit 20 gradually eliminates the suspected diseases according to the supplemented symptom information and the schema rules of the suspected diseases. And when a plurality of suspected diseases exist after the first suspected disease is eliminated, continuously starting from the 7 th candidate concomitant symptom in the sorted concomitant symptom candidate set, pushing the candidate concomitant symptoms of 7-13 for the user to select, then eliminating the suspected diseases again until only one suspected disease remains, judging the corresponding remaining diseases of the diseases suffered by the patient, and outputting an interactive result. And if the guess diseases can not be reduced to one after all the candidate accompanying symptoms in the ordered accompanying symptom candidate set are pushed, forming a collection of the rest guess diseases and pushing the collection to the doctor end for further confirmation and diagnosis of the doctor to complete the interactive process.
Embodiments of the present invention also provide a computer-readable storage medium having instructions stored thereon, which, when executed on a computer, cause the computer to perform the above-mentioned concomitant symptom interaction method.
Those skilled in the art will appreciate that all or part of the steps in the method for implementing the above embodiments may be implemented by a program, which is stored in a storage medium and includes several instructions to enable a single chip, a chip, or a processor (processor) to execute all or part of the steps in the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
While the embodiments of the present invention have been described in detail with reference to the accompanying drawings, the embodiments of the present invention are not limited to the details of the above embodiments, and various simple modifications can be made to the technical solution of the embodiments of the present invention within the technical idea of the embodiments of the present invention, and the simple modifications are within the scope of the embodiments of the present invention. It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. In order to avoid unnecessary repetition, the embodiments of the present invention will not be described separately for the various possible combinations.
In addition, any combination of the various embodiments of the present invention is also possible, and the same should be considered as disclosed in the embodiments of the present invention as long as it does not depart from the spirit of the embodiments of the present invention.
Claims (10)
1. An accompanying symptom interaction method applied to an interactive inquiry system, wherein the method comprises the following steps:
s1) forming a knowledge base according to the existing medical knowledge data;
s2) based on the obtained patient chief complaint symptoms, obtaining a candidate accompanying symptom complete set of the chief complaint symptoms from the knowledge base, and obtaining a co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom in the candidate accompanying symptom complete set;
s3) filtering the candidate accompanied symptom complete set according to a preset rule to obtain an accompanied symptom candidate set;
s4) correcting the co-occurrence weight scores of the candidate accompanying symptoms and the chief complaint symptoms in the accompanying symptom candidate set according to a preset rule, and sorting the candidate accompanying symptoms according to the corrected co-occurrence weight score scores to obtain a sorted accompanying symptom candidate set; wherein the content of the first and second substances,
the co-occurrence weight score correction is implemented based on the TF-IDF value of each candidate co-occurrence symptom, wherein,
the TF value of each candidate accompanying symptom is the frequency of each candidate accompanying symptom appearing in a preset department;
the calculation rule of the IDF value of each candidate accompanying symptom is as follows:
wherein | D | is the total number of departments;
i belongs to J, i is the total number of departments with the ith candidate accompanying symptoms;
i ∈ I ═ 1, 2...., I), I being the accompanying symptom candidate set;
j ∈ J ═ 1, 2.... J, J is a department set;
s5) carrying out the pushing process of the ordered accompanied symptom candidate set according to a preset rule, judging the state of the pushing process in real time until the state of the pushing process reaches a preset standard, and determining to finish the acquisition of the accompanied symptom information of the patient.
2. The concomitant symptom interaction method according to claim 1, wherein in step S1), the existing medical knowledge data at least includes:
the clinical diagnosis relationship data or medical literature is used for obtaining common and rare relationships between diagnosis diseases and symptoms, relationships between diseases and people, relationships between diseases and related medical histories, relationships between diseases and physical signs and relationships between diseases and examination.
3. The accompanying symptom interacting method according to claim 1, wherein the obtaining of the co-occurrence weight score between each candidate accompanying symptom in the candidate accompanying symptom ensemble and the chief complaint symptom in step S2) comprises:
respectively setting a relation weight score of each candidate accompanying symptom in the candidate accompanying symptom total set and a relation weight score of the chief complaint symptom according to the knowledge base; wherein the content of the first and second substances,
the relation weight score of each candidate accompanying symptom and the relation weight score of the chief complaint symptom are respectively obtained based on the occurrence probability of each candidate accompanying symptom and the chief complaint symptom under a preset disease, wherein the higher the occurrence probability of the candidate accompanying symptom and the chief complaint symptom under the preset disease is, the higher the corresponding relation weight score is;
respectively calculating the co-occurrence weight score of each candidate accompanying symptom and the main complaint symptom according to the relationship weight score of each candidate accompanying symptom and the relationship weight score of the main complaint symptom; the calculation formula is as follows:
Vrm＝prik+prim
wherein M ∈ M ═ 1, 2.. said, M is said candidate set of accompanying symptoms;
prik(ii) a relationship weight score for the complaint symptom;
prim(ii) a relationship weight score for the mth candidate concomitant symptom;
Vrmand (3) scoring the co-occurrence weight of the m candidate accompanying symptoms and the chief complaint symptoms.
4. The accompanying symptom interaction method of claim 1, wherein in step S3), the filtering the candidate complete set of accompanying symptoms according to a preset rule to obtain a candidate set of accompanying symptoms comprises:
acquiring patient image information, and filtering candidate accompanying symptoms which are not in accordance with actual symptoms of patients in the candidate accompanying symptom complete set according to the patient image information to obtain a candidate accompanying symptom intermediate set;
and comparing the co-occurrence weight scores of the candidate accompanying symptoms in the candidate accompanying symptom intermediate set and the chief complaint symptoms with a preset score threshold, filtering the candidate accompanying symptoms of which the co-occurrence weight scores of the candidate accompanying symptoms and the chief complaint symptoms are smaller than the preset score threshold, and taking the filtered candidate accompanying symptom intermediate set as the accompanying symptom candidate set.
5. The accompanying symptom interacting method of claim 3, wherein in step S4), the modifying the co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom in the accompanying symptom candidate set according to a preset rule includes:
calculating TF-IDF values of candidate concomitant symptoms in the concomitant symptom candidate set respectively;
updating each candidate accompanying symptom relation weight score according to the TF-IDF value of each candidate accompanying symptom and the relation weight score of each candidate accompanying symptom; the calculation rule is as follows:
prii,new＝tfidfi×prii
wherein I belongs to I ═ I (1, 2...., I), I is a candidate set of accompanying symptoms;
tfidfia TF-IDF value for the ith candidate co-symptom in said set of co-symptom candidates;
priia relationship weight score for the ith candidate concomitant symptom in the set of concomitant symptom candidates;
prib,newupdating the value of the relationship weight score for the type b candidate accompanying symptom in the accompanying symptom candidate set;
modifying the co-occurrence weight scores of the candidate accompanying symptoms and the main complaint symptoms in the accompanying symptom candidate set according to the updated value of the candidate accompanying symptom relation weight score; the correction relation is as follows:
Vri＝prik+prii,new
wherein VriAnd (4) scoring the co-occurrence weight of the corrected i-th candidate accompanying symptom and the chief complaint symptom.
6. The accompanying symptom interacting method according to claim 5, wherein the calculating the TF-IDF value of each candidate accompanying symptom in the accompanying symptom candidate set, respectively, includes:
calculating TF values and IDF values of the candidate concomitant symptoms respectively, and then calculating TF-IDF values of the candidate concomitant symptoms according to the TF values and the IDF values of the candidate concomitant symptoms; wherein, the calculation rule of the TF-IDF value is as follows:
tfidfi＝tfi,j×idfi
wherein J ∈ J ═ 1, 2.... J, J is a department set;
tfi,jTF values in the jth department for the ith candidate concomitant symptom;
idfithe IDF value of the ith candidate concomitant symptom.
7. The companion symptom interactive method of claim 6 wherein tf is the same as tfi,jThe calculation rule of (1) is:
wherein K belongs to K ═ 1, 2.. times, K, and K is all symptom sets appearing in department j;
∑knk,jis the sum of the number of occurrences of all symptoms in department j;
ni,jis the number of times the ith candidate accompanying symptom appears in department j.
8. The accompanying symptom interaction method according to claim 1, wherein in step S5), the performing a push procedure of the sorted accompanying symptom candidate sets according to a preset rule, and determining a state of the push procedure in real time until the state of the push procedure reaches a preset standard includes:
s51) sequentially determining N candidate concomitant symptoms in the concomitant symptom candidate set;
s52) the determined N candidate accompanying symptoms are sequentially pushed to a patient end for selection by a user, and response information corresponding to the user is recovered;
s53) sequentially verifying whether the response information matches a preset disease hypothesis:
if not, the steps S51-S53 are repeatedly executed from the first candidate accompanying symptom that has not been verified as answer information until all candidate accompanying symptoms in the accompanying symptom candidate set are verified.
9. A companion symptom interactive system, the system comprising:
the acquisition unit is used for acquiring patient chief complaint symptoms and patient portrait information;
the processing unit is used for forming a knowledge base according to existing medical knowledge data, acquiring a candidate accompanying symptom complete set of the chief complaint symptoms from the knowledge base based on the patient chief complaint symptoms, and acquiring a co-occurrence weight score of each candidate accompanying symptom and the chief complaint symptom in the candidate accompanying symptom complete set; the candidate accompanied symptom complete set is further used for filtering the candidate accompanied symptom complete set according to a preset rule to obtain an accompanied symptom candidate set; the symptom candidate set is also used for correcting the co-occurrence weight scores of the candidate accompanying symptoms and the chief complaint symptoms in the accompanying symptom candidate set according to a preset rule, and sorting the candidate accompanying symptoms according to the corrected co-occurrence weight score values to obtain a sorted accompanying symptom candidate set; wherein the content of the first and second substances,
the co-occurrence weight score correction is implemented based on the TF-IDF value of each candidate co-occurrence symptom, wherein,
the TF value of each candidate accompanying symptom is the frequency of each candidate accompanying symptom appearing in a preset department;
the calculation rule of the IDF value of each candidate accompanying symptom is as follows:
wherein | D | is the total number of departments;
i ∈ J | { J: i ∈ J } | is the total number of departments presenting with the ith candidate accompanying symptom
I ∈ I ═ 1, 2...., I), I being the accompanying symptom candidate set;
j ∈ J ═ 1, 2.... J, J is a department set;
the storage unit is used for storing the knowledge base;
the pushing unit is used for pushing the sorted accompanied symptom candidate set according to a preset rule;
the processing unit is further used for judging the state of the pushing flow in real time until the state of the pushing flow reaches a preset standard, and determining that the acquisition of the accompanying symptom information of the patient is completed.
10. A computer readable storage medium having stored thereon instructions which, when executed on a computer, cause the computer to perform the companion symptom interaction method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011563391.XA CN112700861B (en) | 2020-12-25 | 2020-12-25 | Accompanying symptom interaction method and accompanying symptom interaction system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011563391.XA CN112700861B (en) | 2020-12-25 | 2020-12-25 | Accompanying symptom interaction method and accompanying symptom interaction system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112700861A CN112700861A (en) | 2021-04-23 |
| CN112700861B true CN112700861B (en) | 2022-02-01 |
Family
ID=75510659
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011563391.XA Active CN112700861B (en) | 2020-12-25 | 2020-12-25 | Accompanying symptom interaction method and accompanying symptom interaction system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112700861B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116403735B (en) * | 2023-06-05 | 2023-08-11 | 山东志诚普惠健康科技有限公司 | Data interaction system and method of cloud health service platform |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451388A (en) * | 2016-05-31 | 2017-12-08 | 百度（美国）有限责任公司 | For the methods, devices and systems for automating medical diagnosis |
| CN108133752A (en) * | 2017-12-21 | 2018-06-08 | 新博卓畅技术（北京）有限公司 | A kind of optimization of medical symptom keyword extraction and recovery method and system based on TFIDF |
| CN109256206A (en) * | 2018-08-14 | 2019-01-22 | 平安医疗健康管理股份有限公司 | Interrogation data processing method, device, computer equipment and storage medium |
| CN111180081A (en) * | 2019-12-30 | 2020-05-19 | 众安信息技术服务有限公司 | Intelligent inquiry method and device |
| CN111599488A (en) * | 2020-05-19 | 2020-08-28 | 万达信息股份有限公司 | Intelligent inquiry implementing method, system and storage medium |
| CN111899884A (en) * | 2020-06-23 | 2020-11-06 | 北京左医科技有限公司 | Intelligent auxiliary inquiry method, device and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130268203A1 (en) * | 2012-04-09 | 2013-10-10 | Vincent Thekkethala Pyloth | System and method for disease diagnosis through iterative discovery of symptoms using matrix based correlation engine |
-
2020
- 2020-12-25 CN CN202011563391.XA patent/CN112700861B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451388A (en) * | 2016-05-31 | 2017-12-08 | 百度（美国）有限责任公司 | For the methods, devices and systems for automating medical diagnosis |
| CN108133752A (en) * | 2017-12-21 | 2018-06-08 | 新博卓畅技术（北京）有限公司 | A kind of optimization of medical symptom keyword extraction and recovery method and system based on TFIDF |
| CN109256206A (en) * | 2018-08-14 | 2019-01-22 | 平安医疗健康管理股份有限公司 | Interrogation data processing method, device, computer equipment and storage medium |
| CN111180081A (en) * | 2019-12-30 | 2020-05-19 | 众安信息技术服务有限公司 | Intelligent inquiry method and device |
| CN111599488A (en) * | 2020-05-19 | 2020-08-28 | 万达信息股份有限公司 | Intelligent inquiry implementing method, system and storage medium |
| CN111899884A (en) * | 2020-06-23 | 2020-11-06 | 北京左医科技有限公司 | Intelligent auxiliary inquiry method, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112700861A (en) | 2021-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112037880B (en) | Medication recommendation method, device, equipment and storage medium | |
| US11392856B2 (en) | Methods and systems for an artificial intelligence support network for behavior modification | |
| CN109299239B (en) | ES-based electronic medical record retrieval method | |
| US11461664B2 (en) | Methods and systems for an artificial intelligence alimentary professional support network for vibrant constitutional guidance | |
| CN112786194A (en) | Medical image diagnosis guide inspection system, method and equipment based on artificial intelligence | |
| CN114171176A (en) | Hospital triage data processing method and system based on Internet | |
| CN111813957A (en) | Medical diagnosis guiding method based on knowledge graph and readable storage medium | |
| US11275985B2 (en) | Artificial intelligence advisory systems and methods for providing health guidance | |
| CN112614565A (en) | Traditional Chinese medicine classic famous prescription intelligent recommendation method based on knowledge-graph technology | |
| US11581094B2 (en) | Methods and systems for generating a descriptor trail using artificial intelligence | |
| CN115497616B (en) | Method, system, equipment and storage medium for auxiliary decision-making of infectious diseases | |
| US20210397996A1 (en) | Methods and systems for classification using expert data | |
| Travers et al. | Using nurses’ natural language entries to build a concept-oriented terminology for patients’ chief complaints in the emergency department | |
| CN116910172B (en) | Follow-up table generation method and system based on artificial intelligence | |
| CN114416967A (en) | Method, device and equipment for intelligently recommending doctors and storage medium | |
| CN113851208A (en) | Medical examination recommendation system and method based on explicit topic allocation technology | |
| CN115798733A (en) | Intelligent auxiliary reasoning system and method for orphan disease | |
| CN112700861B (en) | Accompanying symptom interaction method and accompanying symptom interaction system | |
| CN111415760A (en) | Doctor recommendation method, system, computer equipment and storage medium | |
| CN110610766A (en) | Apparatus and storage medium for deriving probability of disease based on symptom feature weight | |
| Plaza et al. | Retrieval of similar electronic health records using UMLS concept graphs | |
| JP4729444B2 (en) | Health guidance support system | |
| CN115631823A (en) | Similar case recommendation method and system | |
| JP2023175142A (en) | Disease evaluation index calculation system, method, and program | |
| US11810669B2 (en) | Methods and systems for generating a descriptor trail using artificial intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |