CN113571145A - Information matching method and system for health management decision - Google Patents
Information matching method and system for health management decision Download PDFInfo
- Publication number
- CN113571145A CN113571145A CN202110920691.7A CN202110920691A CN113571145A CN 113571145 A CN113571145 A CN 113571145A CN 202110920691 A CN202110920691 A CN 202110920691A CN 113571145 A CN113571145 A CN 113571145A
- Authority
- CN
- China
- Prior art keywords
- information
- main body
- health medical
- medical information
- health
- 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
- 230000036541 health Effects 0.000 title claims abstract description 272
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000004364 calculation method Methods 0.000 claims description 45
- 238000012545 processing Methods 0.000 claims description 41
- 238000013507 mapping Methods 0.000 claims description 32
- 230000008569 process Effects 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 14
- 239000013598 vector Substances 0.000 claims description 8
- 239000013604 expression vector Substances 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 239000003814 drug Substances 0.000 description 11
- 229940079593 drug Drugs 0.000 description 9
- 238000011282 treatment Methods 0.000 description 9
- 201000010099 disease Diseases 0.000 description 7
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 7
- 210000004072 lung Anatomy 0.000 description 6
- 239000000126 substance Substances 0.000 description 6
- YOHYSYJDKVYCJI-UHFFFAOYSA-N n-[3-[[6-[3-(trifluoromethyl)anilino]pyrimidin-4-yl]amino]phenyl]cyclopropanecarboxamide Chemical compound FC(F)(F)C1=CC=CC(NC=2N=CN=C(NC=3C=C(NC(=O)C4CC4)C=CC=3)C=2)=C1 YOHYSYJDKVYCJI-UHFFFAOYSA-N 0.000 description 5
- 206010008479 Chest Pain Diseases 0.000 description 4
- 239000002246 antineoplastic agent Substances 0.000 description 4
- 229940041181 antineoplastic drug Drugs 0.000 description 4
- 238000003759 clinical diagnosis Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 208000024891 symptom Diseases 0.000 description 4
- 206010011224 Cough Diseases 0.000 description 3
- 206010027476 Metastases Diseases 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 102000052116 epidermal growth factor receptor activity proteins Human genes 0.000 description 3
- 108700015053 epidermal growth factor receptor activity proteins Proteins 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 230000009401 metastasis Effects 0.000 description 3
- 238000012821 model calculation Methods 0.000 description 3
- 239000013589 supplement Substances 0.000 description 3
- 102000008096 B7-H1 Antigen Human genes 0.000 description 2
- 108010074708 B7-H1 Antigen Proteins 0.000 description 2
- 101000984753 Homo sapiens Serine/threonine-protein kinase B-raf Proteins 0.000 description 2
- 101100091501 Mus musculus Ros1 gene Proteins 0.000 description 2
- 206010028980 Neoplasm Diseases 0.000 description 2
- 206010036790 Productive cough Diseases 0.000 description 2
- 102100027103 Serine/threonine-protein kinase B-raf Human genes 0.000 description 2
- 208000009956 adenocarcinoma Diseases 0.000 description 2
- 210000000621 bronchi Anatomy 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 210000001165 lymph node Anatomy 0.000 description 2
- 208000002154 non-small cell lung carcinoma Diseases 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009469 supplementation Effects 0.000 description 2
- 102100033814 Alanine aminotransferase 2 Human genes 0.000 description 1
- 206010016326 Feeling cold Diseases 0.000 description 1
- 101000779415 Homo sapiens Alanine aminotransferase 2 Proteins 0.000 description 1
- 208000007433 Lymphatic Metastasis Diseases 0.000 description 1
- 208000002193 Pain Diseases 0.000 description 1
- 206010037660 Pyrexia Diseases 0.000 description 1
- 230000001919 adrenal effect Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 238000001574 biopsy Methods 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 210000003109 clavicle Anatomy 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 208000017574 dry cough Diseases 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002055 immunohistochemical effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000002085 irritant Substances 0.000 description 1
- 231100000021 irritant Toxicity 0.000 description 1
- 210000005015 mediastinal lymph node Anatomy 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 206010029410 night sweats Diseases 0.000 description 1
- 230000036565 night sweats Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 238000000554 physical therapy Methods 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000001959 radiotherapy Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 238000011269 treatment regimen Methods 0.000 description 1
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
- 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)
- Medical Informatics (AREA)
- Public Health (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The application discloses an information matching method and system for health management decision, which comprises the following steps: acquiring decision main body information; structuring the decision main body information to obtain structured main body information; coding the structured main body information to obtain a main body information coding result; calculating the correlation degree of the main body information coding result and the health medical information coding result of each health medical information; and extracting the health medical information corresponding to the health medical information coding result with the association degree exceeding a preset first association threshold value to obtain the target health medical information. According to the method and the device, the target health medical information with the relevance degree meeting the preset condition is screened out and provided for reference of medical care personnel, and the health medical information in different directions and types is pre-coded and used for matching, so that the matching dimensionality is increased, the limitations of single type of health management decision content support content, low decision support efficiency and the like are effectively improved, and the reliability and the accuracy of the provided reference information are ensured to be high.
Description
Technical Field
The invention relates to the field of intelligent medical treatment, in particular to an information matching method and system for health management decision.
Background
At present, diagnosis and treatment schemes are provided for patients mainly by means of self knowledge storage and personal experience of medical and health staff in the clinical diagnosis and treatment process, and the situation that the medical and health staff in China receive system education and training is greatly different, so that clinical diagnosis and treatment normative difference in different areas is large, medical service levels are uneven, and medical health requirements of the masses are difficult to meet.
However, in the prior art, decision support in a few single aspects such as triage advice, diagnosis reference, medication advice and the like can be provided only based on limited data such as diagnosis information, symptom signs and the like extracted from decision subject information, and the breadth and depth of the decision information are very limited, for example, the medication advice usually only includes limited information such as drug names, drug usage amounts and the like, information such as reasonable medication early warning, medication evidence interpretation, medication burden and the like is lacked, homogeneous amplification of different types of information cannot be realized through one technical path, complex decision requirements in a disease full-period management process cannot be supported quickly and efficiently from multi-dimensional fine-grained data, and diversified decision requirements in other health management decision directions besides clinical diagnosis and treatment cannot be covered.
On the other hand, in the prior art, in other health management related scenes except clinical diagnosis and treatment, multi-dimensional data support intelligent decision assistance is also lacked, for example, community medical care decision assistance for slow patient groups, nutritional health management assistance for various groups, hospital performance management and control, medical quality management assistance, and the like.
Therefore, a more comprehensive and effective information matching method for health management decision is needed, which can provide effective reference information for medical care personnel.
Disclosure of Invention
In view of the above, the present invention provides an information matching method and system for health management decision, which can provide effective reference information for medical care personnel. The specific scheme is as follows:
an information matching method for health management decisions, comprising:
acquiring decision main body information;
carrying out structuralization processing on the decision main body information to obtain structuralization main body information;
coding the structured main body information to obtain a main body information coding result for embodying the characteristics of the structured main body information;
calculating the correlation degree of the main body information coding result and the health medical information coding result of each health medical information;
extracting the health medical information corresponding to the health medical information coding result of which the correlation degree with the main body information coding result exceeds a preset first correlation threshold value to obtain target health medical information;
the health medical information coding result of each health medical information is obtained by coding each health medical information in advance.
Optionally, the process of encoding the structured main body information to obtain a main body information encoding result for embodying the characteristics of the structured main body information includes:
coding the structured main body information by using a preset information coding model to obtain a main body information coding result for embodying the characteristics of the structured main body information;
the information coding model is obtained by utilizing historical decision main body information for training in advance.
Optionally, the process of encoding the structured main body information by using a preset information encoding model to obtain a main body information encoding result for embodying the characteristics of the structured main body information includes:
calculating a representation vector of each feature in the structured main body information by using the information coding model;
and integrating the expression vectors of all the characteristics by using the information coding model to obtain the main body information coding result for embodying the characteristics of the structured main body information.
Optionally, the process of calculating the association degree between the main body information coding result and the health medical information coding result of each piece of health medical information includes:
calculating the association degree of the main body information coding result and the health medical information coding result of each health medical information by using a preset association degree calculation model;
the relevance calculation model is obtained by utilizing historical subject information codes and historical health medical information code results in advance for training.
Optionally, the method further includes:
and classifying the health medical information in advance based on preset information, and clustering and storing the health medical information into a health medical information base.
Optionally, the process of extracting the health medical information corresponding to the health medical information coding result of which the association degree with the main information coding result exceeds a preset first association threshold to obtain the target health medical information includes:
extracting the associated health medical information and the mapping health medical information of which the association degree with the encoding result of the associated health medical information exceeds a preset second association threshold from the health medical information base to obtain target health medical information comprising the associated health medical information and the mapping health medical information;
the related health medical information is health medical information corresponding to the health medical information coding result of which the degree of association with the main body information coding result exceeds a preset first association threshold value.
Optionally, the method further includes:
adding new information classification and adding new health medical information under any information classification.
Optionally, before the encoding processing is performed on the structured main body information, the method further includes:
performing query intention analysis on the structured main body information by using a preset intention analysis model to obtain a query intention corresponding to the structured main body information;
judging whether key features corresponding to the query intention are missing or not by utilizing a preset feature and intention mapping relation;
if the key features are missing, the key features missing in the decision main body information are obtained from a corresponding database or the key features missing in the decision main body information supplemented by a user are received, so that complete decision main body information is obtained;
taking the complete decision main body information as decision main body information again so as to carry out structuralization processing to obtain the structuralization main body information;
the intention analysis model is obtained by training with historical structured subject information in advance, and the mapping relation between the features and the intention records the corresponding relation between the query intention and the corresponding key features.
Optionally, the method includes:
randomly removing features in the historical decision main body information;
counting error probabilities between each query intention and the missing features, which cause the calculation errors of the relevance calculation model;
and determining the corresponding relation between each query intention and the key characteristics according to the error probability, and generating the characteristic-intention mapping relation.
The invention also discloses an information matching system for health management decision, which comprises:
the main body information acquisition module is used for acquiring decision main body information;
the structuralization processing module is used for structuralizing the decision main body information to obtain structuralization main body information;
the coding processing module is used for coding the structured main body information to obtain a main body information coding result for embodying the characteristics of the structured main body information;
the relevancy calculation module is used for calculating the relevancy between the main body information coding result and the health medical information coding result of each piece of health medical information;
the medical information extraction module is used for extracting the health medical information corresponding to the health medical information coding result of which the correlation degree with the main body information coding result exceeds a preset first correlation threshold value to obtain target health medical information;
the health medical information coding result of each health medical information is obtained by coding each health medical information in advance.
In the invention, the information matching method for health management decision comprises the following steps: acquiring decision main body information; carrying out structuralization processing on the decision main body information to obtain structuralization main body information; coding the structured main body information to obtain a main body information coding result for embodying the characteristics of the structured main body information; calculating the correlation degree of the main body information coding result and the health medical information coding result of each health medical information; extracting the health medical information corresponding to the health medical information coding result of which the correlation degree with the main body information coding result exceeds a preset first correlation threshold value to obtain target health medical information; the health medical information coding result of each health medical information is obtained by coding each health medical information in advance.
The decision-making subject information is subjected to structured processing and unified format, then coding processing is carried out to obtain a subject information coding result, the correlation degree calculation is carried out on the subject information and a large number of health medical information coding results subjected to pre-coding processing, health medical information corresponding to the health medical information coding results with the correlation degree exceeding a preset first correlation threshold is screened out to serve as target health medical information and is provided for medical care personnel to refer, and the matching dimensionality is increased by carrying out pre-coding processing on health medical information in different directions and types and using the health medical information for matching, so that the limitations of single type of health management decision-making content support content, small information quantity, narrow applicable scene, low decision-making support efficiency and the like can be effectively improved, and the reliability and high accuracy of the provided reference information are ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic flow chart of an information matching method for health management decision according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a decision-making subject information tag according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart illustrating another information matching method for health management decision according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of an overall framework of an information matching method for health management decision according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a health medical data label and medical insurance data corresponding to mapping health medical information B006361 according to an embodiment of the present invention;
fig. 6 is a schematic view of another health care data tag mapped to health care information X006259 and data of clinical application guidelines for novel anti-tumor drugs (2020 edition) by the ministry of health care institute of electrical and health care, according to an embodiment of the present invention;
FIG. 7 is a flowchart illustrating another information matching method for health management decision according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of an information matching system for health management decision according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention discloses an information matching method for health management decision, which is shown in figure 1 and comprises the following steps:
s11: and obtaining the decision main body information.
Specifically, when health management needs to be performed on a patient, decision subject information of the patient is needed, and the decision subject information includes information needed for describing the patient, wherein the decision subject information may include at least one of user basic information, outpatient medical record information, hospitalization medical record information, physical examination information, medical advice information, examination information, operation information, and traditional Chinese medicine physiotherapy information.
S12: and structuring the decision main body information to obtain structured main body information.
Specifically, for convenience of storage and subsequent calling of the decision-making subject information, the decision-making subject information is subjected to structured processing to obtain structured subject information with a unified format. The decision-making main body information is subjected to structuring processing, a large amount of structured main body information can be effectively stored for a long time, and meanwhile, due to the uniform format, the structured main body information is convenient to analyze, process, call and the like in the future.
Specifically, in the process of performing the structural processing on the decision main body information, feature extraction and simplification may be performed on the decision main body information, for example, a decision main body information tag is set, and a corresponding feature tag is extracted from the decision main body information according to a preset tag, so as to obtain the refined and simplified structural main body information.
For example, the decision body information includes: patient, female, 63 years old was admitted to the hospital for "cough with chest pain left for more than a month". The patient has no obvious reason before one month, cough appears, the main is irritant dry cough, a small amount of expectoration is accompanied by left chest pain, the left chest pain appears dull pain, the expectoration is obvious when the patient inhales, the radiation is not emitted to other parts, no obvious chest distress and short breath exist, no chilliness, fever and night sweat exist, attention is not given to the patient at first, and the symptom does not change well. Visit at my hospital 2 weeks ago, examine chest CT: the focus is frequently seen in both lungs, and the left upper lung focus is not excluded from tumors. Examine PET/CT show: double lung metastasis, left cervical lymph node, bilateral clavicle area lymph nodes, mediastinal lymph node, double glottic lymph node metastasis, brain metastasis, right adrenal metastasis. Examining an electronic bronchoscope left main bronchus and left upper lung lobe bronchoscope lung biopsy: the upper left bronchus has new creature, is crisp in texture and easy to bleed, and is biopsied and pathologically shown as follows: the detection of cancer cells indicates NSCLC adenocarcinoma; immunohistochemical results show: ALK-ventana (-), BRAF (-), EGFR (19) (0), EGFR (21) (0), Ki-67(+ 30%), PD-L1(TPS 10%), ROS-1 (-). And (3) diagnosis: left lung upper lobe bronchial adenocarcinoma, non-small cell lung carcinoma, cT4N3M1c (stage IVB), ALK (-), BRAF (-), EGFR (-), Ki-67(+ 30%), ROS-1(-), PD-L1(TPS 10%). After extracting the decision main body information tag in the decision main body information by using the tag extraction model, the decision main body information tag shown in fig. 2, including sex, age, symptom sign tag, disease name tag, case classification tag, disease stage tag, molecular classification tag, and the like, can be obtained.
The decision main body information label can comprise at least one of a main body basic information label, a symptom sign label, a disease classification label, a medication name label, a medicine usage amount label, an operation classification label, a radiotherapy scheme label, other treatment labels, a cost label and a scientific research label; wherein, the main basic information label can further comprise at least one of a sex label, an age label, a native label, an occupation label, a national label, a marital label, a regional label, a medical institution type label, a department label, a community label, a consumable label, an equipment label, etc.; the disease classification label may include at least one of an ICD encoding label, a disease name label, a pathology typing label, a molecular typing label, a prognosis typing label, a disease stage label, a treatment stage label, an optional treatment regimen classification label, and the like.
S13: and coding the structured main body information to obtain a main body information coding result for embodying the characteristics of the structured main body information.
Specifically, the characteristics recorded in the form of characters are difficult to retrieve, the retrieval efficiency is low, in order to match the health medical information associated with the structured main body information, the structured main body information is coded, the characteristics of the structured main body information are further extracted and highlighted through the coding, a main body information coding result capable of reflecting the characteristics of the structured main body information is obtained, and the characteristic matching through the coding result is quicker and more accurate.
S14: and calculating the correlation degree of the main body information coding result and the health medical information coding result of each health medical information.
Specifically, the main information coding result can be matched with the health medical information coding result of each health medical information after being obtained, and in the matching process, the association degree of the main information coding result and the health medical information coding result of each health medical information is calculated according to various coefficients such as the similarity and the weight of the coding results.
The health medical information coding result of each health medical information is obtained by coding each health medical information in advance.
Specifically, a large amount of health medical information is collected in advance, in order to enable the health medical information to be effectively matched with the main information coding result of the decision main information, the health medical information also needs to be subjected to structural processing to obtain structural medical information, then the structural medical information is also subjected to coding processing to obtain each health medical information coding result corresponding to each structural medical information, so that the subsequent correlation calculation can be carried out with the main body information coding result of the structured main body information, the health medical information may include various medical related data, such as clinical research data, pharmaceutical data, medical data, and the like, and the whole content that the decision subject information may relate to is covered as comprehensively as possible, so that it is ensured that any decision subject information can search the health medical information corresponding to the decision subject information, and it is ensured that corresponding information support can be provided for the user.
When the health medical information is subjected to structural processing, labeling processing can be also performed on the health medical information, a health medical information label is set, a corresponding characteristic label is extracted from the health medical information according to the preset label, and then the simplified structural medical information is obtained.
S15: and extracting the health medical information corresponding to the health medical information coding result of which the correlation degree with the main body information coding result exceeds a preset first correlation threshold value to obtain the target health medical information.
Specifically, after the correlation degrees of the main body information coding result and each health medical information coding result are compared, the health medical information coding result with the correlation degree exceeding a preset first correlation threshold is selected from the main body information coding results, the health medical information coding result is determined to be suitable for the main body, the other health medical information coding results with the low correlation degrees can be regarded as information irrelevant to the decision main body information or information already covered by the health medical information with the high correlation degree, and it can be understood that after the health medical information coding result exceeding the preset first correlation threshold is determined, the health medical information corresponding to each health medical information coding result can be used as target health medical information and provided to a user, such as medical staff, as reference information so as to provide information reference for patients.
Therefore, the embodiment of the invention performs structural processing on decision main body information to unify the format, performs coding processing to obtain a main body information coding result, performs correlation calculation on the main body information coding result and a large number of health medical information coding results subjected to pre-coding processing, screens out health medical information corresponding to the health medical information coding result with the correlation exceeding a preset first correlation threshold value as target health medical information, provides the target health medical information for reference for medical care personnel, performs pre-coding processing on health medical information in different directions and types and is used for matching, and increases matching dimensionality, so that the limitations of single type of health management decision content support content, less information quantity, narrow applicable scene, low decision support efficiency and the like can be effectively improved, and the reliability and high accuracy of the provided reference information are ensured.
The embodiment of the invention discloses a specific information matching method for health management decision, and compared with the previous embodiment, the embodiment further explains and optimizes the technical scheme. Referring to fig. 3 and 4, specifically:
s21: and classifying the health medical information in advance based on preset information, and clustering and storing the health medical information into a health medical information base.
Specifically, in order to effectively store the health medical information, a health medical information library may be established, and in order to further improve the retrieval efficiency, the health medical information may be stored in a cluster manner based on preset information classification, so that when the query direction of the subject information can be determined clearly, the health medical information corresponding to at least one information classification may be retrieved for use in subsequent association degree calculation.
S22: acquiring decision main body information;
s23: structuring the decision main body information to obtain structured main body information;
s24: and coding the structured main body information by using a preset information coding model to obtain a main body information coding result for embodying the characteristics of the structured main body information.
Specifically, the information coding model is obtained by training in advance by using the historical decision-making subject information, and the model can be used for coding the structured subject information more quickly and accurately.
Further, S24 performs encoding processing on the structured main body information by using a preset information encoding model, and the obtained main body information encoding result for embodying the characteristics of the structured main body information may specifically include S241 and S242; wherein the content of the first and second substances,
s241: calculating the expression vector of each feature in the structured main body information by using the information coding model;
s242: and integrating the expression vectors of the features by using the information coding model to obtain a main body information coding result for embodying the characteristics of the structured main body information.
Specifically, the information coding model may first obtain a representation vector of each feature of the decision main body information by using a word embedding technique (including, but not limited to, word2vec and glove) or a pre-training model (including, but not limited to, bert (bidirectional Encoder responses from transforms), XLNet, GPT2, CNN (Convolutional Neural Networks), RNN (Recurrent Neural Networks), LSTM (Long Short-term memory Networks), and gru (gate recovery unit)), and then integrate (for example, concatenate or average, etc.) the feature representation vectors related to the decision main body information, where the integrated result is the information coding result. It is understood that the health care information encoding result is obtained in the same manner.
S25: calculating the association degree of the main body information coding result and the health medical information coding result of each health medical information by using a preset association degree calculation model;
specifically, the calculation speed and accuracy can be effectively improved by establishing a relevancy calculation model by using a natural language technology, wherein the relevancy calculation model is obtained by utilizing the historical subject information coding and the historical health medical information coding result in a training mode in advance.
S26: and extracting the associated health medical information and the mapping health medical information of which the association degree with the encoding result of the associated health medical information exceeds a preset second association threshold from the health medical information base to obtain the target health medical information comprising the associated health medical information and the mapping health medical information.
Specifically, the health medical information is clustered and stored in the health medical information base, and the associated health medical information and the mapped health medical information construct an efficient mapping relation in advance, so that while the associated health medical information is extracted from the health medical information base, the mapped health medical information with the association degree exceeding a preset second association threshold value with the encoding result of the associated health medical information can be extracted according to the information classification and the mapping relation and used as auxiliary information, and the auxiliary information and the information are used as target health medical information and provided for medical workers to make reference, and the richness of reference content is increased.
For example, as shown in fig. 5 and 6, after the labeling processing and the information coding result comparison between "other information-medical insurance data B006361" and "other information-national health committee" clinical application guideline for novel anti-tumor drug "(2020 version)" X006259 "show that the degree of association is 97%, which exceeds the preset threshold value 95%, the information X006259 can be used as the mapping health information of the information B006361 in the category of" other information-national health committee "clinical application guideline for novel anti-tumor drug" (2020 version) ". As shown in fig. 2, after being subjected to labeling processing and information encoding, the structured main information is compared with an information encoding result of the information B006361, and the degree of association is 93%, which exceeds a preset threshold value by 90%. Therefore, "other information-medical insurance data B006361" is used as the associated data in the decision target data of this time, and "other information-national health committee" guidance principle for clinical application of novel antitumor drugs (2020 version) "X006259" is used as the mapping data in the decision target data of this time, and is output together. Therefore, the health medical information is clustered and stored in the health medical information base, an efficient mapping relation is established by associating the health medical information with the mapping health medical information, various associated health medical information and mapping health information of the health medical information under a required application scene can be rapidly acquired, and multi-dimensional information reference is provided. The related health medical information is health medical information corresponding to a health medical information coding result of which the degree of association with the main body information coding result exceeds a preset first association threshold value.
The health medical information coding result of each health medical information is obtained by coding each health medical information in advance.
S27: adding new information classification and adding new health medical information under any information classification.
Specifically, the information classification and the data stored in a certain information classification can be continuously expanded in the health medical information base so as to adapt to the medical health decision query requirement in a new direction.
The embodiment of the invention also discloses a specific information matching method for health management decision, which is shown in fig. 7 specifically as follows:
s31: acquiring decision main body information;
s32: structuring the decision main body information to obtain structured main body information;
s33: and performing query intention analysis on the structured main body information by using a preset intention analysis model to obtain a query intention corresponding to the structured main body information.
Specifically, in order to improve the analysis speed, avoid querying all health medical information in the health medical information base, and also avoid the situation of poor matching effect caused by lack of key information in the decision main body information, after the structured main body information is obtained, a preset intention analysis model is used for performing the inquiry intention analysis on the structured main body information, so that the inquiry direction of the decision main body information, namely the inquiry intention corresponding to the structured main body information, can be obtained.
The intention analysis model is obtained by training historical structured subject information in advance, and the corresponding relation between the query intention and the corresponding key characteristic is recorded in the characteristic-intention mapping relation.
It will be appreciated that the analysis of query intent may be implemented using natural language processing techniques, or may be specified by direct user input when entering decision-making externalization information.
S34: and judging whether the key features corresponding to the query intention are missing or not by utilizing the preset mapping relation between the features and the intention.
Specifically, a mapping relation, that is, a preset mapping relation between the feature and the intention, is established in advance for the query intention and the key feature, so that the key feature required for achieving the query effect of the query intention can be determined according to the query intention, and thus whether the corresponding key feature is missing in the structured main body information can be judged.
Specifically, the structured main body information may include a feature tag, and according to a tag of a key feature corresponding to each intention recorded in the mapping relationship between the feature and the intention, it is determined whether the structured main body information includes a tag of a key feature corresponding to the current query intention, and if not, it may be determined that the key feature is missing.
S35: if the key features are missing, the key features missing in the decision main body information are obtained from the corresponding database or the key features missing in the decision main body information supplemented by the user are received, and the complete decision main body information is obtained;
s36: and taking the complete decision main body information as the decision main body information again so as to carry out structuring processing to obtain the structured main body information.
Specifically, if the key features corresponding to the query intention are missing, the key features missing from the decision subject information may be automatically obtained from the corresponding database, for example, the missing key features are obtained from the database including the personal information of the subject corresponding to the decision subject information, after the missing key features are supplemented, the supplemented complete decision subject information is re-obtained as new decision subject information, S32 is re-executed to perform the structuring processing to obtain the structured subject information, so that the following steps S33 to S39 are executed to re-perform the information matching of the health management decision, thereby effectively avoiding the situation that the reference information obtained by the final calculation is wrong and the judgment of the user is affected, and of course, the user may also actively supplement, after the key features corresponding to the query intention are missing, only the reminding information is generated to remind the decision subject information that the key features are missing, and the user can supplement the information by himself, and the complete decision main body information can be obtained after the key features missing in the decision main body information supplemented by the user are received.
Further, in the case of automatically supplementing the key features missing in the decision main body information, prompt information may be generated as to whether to allow automatic supplementation, and after the user inputs an instruction to allow automatic supplementation, the key features missing in the decision main body information are automatically supplemented.
Specifically, of course, if the key features corresponding to the query intention are not missing, the step S37 is executed to perform the next matching process.
S37: and coding the structured main body information to obtain a main body information coding result for embodying the characteristics of the structured main body information.
S38: calculating the correlation degree of the main body information coding result and the health medical information coding result of each health medical information;
s39: and extracting the health medical information corresponding to the health medical information coding result of which the correlation degree with the main body information coding result exceeds a preset first correlation threshold value to obtain the target health medical information.
The health medical information coding result of each health medical information is obtained by coding each health medical information in advance.
Further, the generation process of the feature-to-intention mapping relationship in S34 may specifically include S341 to S343; wherein the content of the first and second substances,
s341: and randomly removing the features in the historical decision main body information.
Specifically, in order to obtain the influence of each feature on different query intentions, part of features are randomly removed from the historical decision-making main body information, so that when the relevance calculation is performed on the historical decision-making main body information with the missing features, the relevance calculation is wrong due to the missing features, and the currently missing features are key features of correct health medical information which cannot be associated due to the missing features.
For example, n features in the historical decision-making subject information are randomly removed, wherein n is equal to [1, the maximum feature number of the current decision-making subject information).
S342: and counting the error probability between each query intention and the missing feature, which causes the calculation error of the relevance calculation model.
In particular, each historical decision-making subject information may support a variety of query intents, such as rehabilitation query intents, treatment query intents, or drug information query intents, therefore, the new random features are continuously removed by using the historical decision main body information, so that the error probability of the calculation error of the correlation calculation model caused by each query intention and missing features which can be covered by the historical decision main body information can be counted, of course, the query intentions which can be covered by one piece of historical decision main body information are limited, and the samples are limited, so that the random features can be removed by continuously utilizing new historical decision subject information, and counting the error probability of the correlation calculation model calculation error caused between each query intention and the missing feature, and finally counting the error probability of the correlation calculation model calculation error caused between each query intention and the missing feature with a wide enough coverage range.
Specifically, due to the absence of the features, the relevance calculation model cannot calculate the health medical information corresponding to the absent features, so that the key features of the query intentions can be estimated by counting the error probability of the relevance calculation model calculation error caused by each query intention and the absent features.
For example, due to the absence of the feature a, the error probability of the relevance calculation model calculating the calculation error of the health care information corresponding to the historical decision-making subject information and the query intention B is improved from 10% to 80%.
S343: and determining the corresponding relation between each query intention and the key characteristics according to the error probability, and generating a characteristic-intention mapping relation.
Specifically, the degree of closeness of the relationship between each feature and each query intention can be inferred according to the error probability of the calculation error of the relevance degree calculation model caused by each query intention and the missing feature, the error probability of the calculation error of one or more query intention relevance degree calculation models caused by the missing of one feature is greatly improved, if the error probability exceeds the preset error probability, the probability indicates that the feature is the key feature of the query intention, for example, if the error probability of the calculation error of the relevance degree calculation model for calculating the historical decision main information and the healthy medical information corresponding to the query intention B is improved from 10% to 80% due to the missing of the feature A, if the error probability exceeds the preset error probability 50%, the feature A seriously affects the calculation accuracy of the relevance degree calculation model for the healthy medical information in the direction of the query intention B, therefore, the feature A is a key feature for the query intention B, and the corresponding relation between each query intention and the key feature is repeatedly counted in such a way, so that a feature-intention mapping relation covering each query intention and the key feature can be generated.
Specifically, the judgment between the key features and the query intention can be judged according to an association formula;
the correlation formula is:
in the formula, the feature M is composed of n features in S341,
representing features M being key features of the query intention KProbability.
Correspondingly, the embodiment of the present invention further discloses an information matching system for health management decision, as shown in fig. 8, the system includes:
a main body information obtaining module 11, configured to obtain decision main body information;
the structuralization processing module 12 is configured to perform structuralization processing on the decision main body information to obtain structuralization main body information;
the encoding processing module 13 is configured to perform encoding processing on the structured main body information to obtain a main body information encoding result for embodying the characteristics of the structured main body information;
the association degree calculation module 14 is used for calculating the association degree between the main body information coding result and the health medical information coding result of each health medical information;
the medical information extraction module 15 is configured to extract the health medical information corresponding to the health medical information coding result of which the association degree with the main information coding result exceeds a preset first association threshold value, so as to obtain target health medical information;
the health medical information coding result of each health medical information is obtained by coding each health medical information in advance.
Therefore, the embodiment of the invention performs structural processing on decision main body information to unify the format, performs coding processing to obtain a main body information coding result, performs correlation calculation on the main body information coding result and a large number of health medical information coding results subjected to pre-coding processing, screens out health medical information corresponding to the health medical information coding result with the correlation exceeding a preset first correlation threshold value as target health medical information, provides the target health medical information for reference for medical care personnel, performs pre-coding processing on health medical information in different directions and types and is used for matching, and increases matching dimensionality, so that the limitations of single type of health management decision content support content, less information quantity, narrow applicable scene, low decision support efficiency and the like can be effectively improved, and the reliability and high accuracy of the provided reference information are ensured.
Specifically, the encoding processing module 13 is specifically configured to perform encoding processing on the structured main body information by using a preset information encoding model to obtain a main body information encoding result for embodying the characteristics of the structured main body information;
the information coding model is obtained by utilizing historical decision main body information for training in advance.
Specifically, the encoding processing module 13 includes a feature vector calculating unit and a vector integrating unit; wherein the content of the first and second substances,
the characteristic vector calculation unit is used for calculating the expression vector of each characteristic in the structured main body information by using the information coding model;
and the vector integration unit is used for integrating the expression vectors of the characteristics by using the information coding model to obtain a main body information coding result for reflecting the characteristics of the structured main body information.
Specifically, the association degree calculating module 14 is specifically configured to calculate, by using a preset association degree calculating model, an association degree between the subject information encoding result and the health medical information encoding result of each piece of health medical information;
the relevance calculation model is obtained by utilizing historical subject information codes and historical health medical information code results in advance for training.
Specifically, the system can further comprise a cluster storage module; wherein the content of the first and second substances,
and the clustering storage module is used for classifying the health medical information in advance based on preset information and clustering and storing the health medical information into the health medical information base.
Specifically, the medical information extraction module 15 is specifically configured to extract the associated health medical information and the mapped health medical information whose association with the associated health medical information coding result exceeds a preset second association threshold from the health medical information library, so as to obtain target health medical information including the associated health medical information and the mapped health medical information;
the related health medical information is health medical information corresponding to a health medical information coding result of which the degree of association with the main body information coding result exceeds a preset first association threshold value.
Specifically, the system also comprises a medical information adding module; wherein the content of the first and second substances,
and the medical information adding module is used for adding new information classification and adding the health medical information under any information classification.
Specifically, the system can further comprise an intention query module, a feature missing judgment module and a feature missing reminding module; wherein the content of the first and second substances,
the intention query module is used for performing query intention analysis on the structured main body information by using a preset intention analysis model to obtain a query intention corresponding to the structured main body information;
the characteristic missing judging module is used for judging whether the key characteristic corresponding to the query intention is missing or not by utilizing the preset characteristic and intention mapping relation;
the characteristic supplement module is used for acquiring the missing key characteristics in the decision main body information from a corresponding database to obtain complete decision main body information if the characteristic missing judgment module judges that the key characteristics are missing;
the matching calling module is used for calling the main body information acquisition module to obtain the complete decision main body information as new decision main body information again;
the intention analysis model is obtained by training historical structured subject information in advance, and the corresponding relation between the query intention and the corresponding key characteristic is recorded in the characteristic-intention mapping relation.
Specifically, the feature missing determination module includes:
the characteristic removing unit is used for randomly removing the characteristics in the historical decision main body information;
counting error probability which causes the calculation error of the correlation calculation model between each query intention and the missing feature;
and determining the corresponding relation between each query intention and the key characteristics according to the error probability, and generating a characteristic-intention mapping relation.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The technical content provided by the present invention is described in detail above, and the principle and the implementation of the present invention are explained in this document by applying specific examples, and the above description of the examples is only used to help understanding the method of the present invention and the core idea thereof; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. An information matching method for health management decisions, comprising:
acquiring decision main body information;
carrying out structuralization processing on the decision main body information to obtain structuralization main body information;
coding the structured main body information to obtain a main body information coding result for embodying the characteristics of the structured main body information;
calculating the correlation degree of the main body information coding result and the health medical information coding result of each health medical information;
extracting the health medical information corresponding to the health medical information coding result of which the correlation degree with the main body information coding result exceeds a preset first correlation threshold value to obtain target health medical information;
the health medical information coding result of each health medical information is obtained by coding each health medical information in advance.
2. The information matching method for health management decision as claimed in claim 1, wherein the process of encoding the structured subject information to obtain a subject information encoding result for embodying the characteristics of the structured subject information comprises:
coding the structured main body information by using a preset information coding model to obtain a main body information coding result for embodying the characteristics of the structured main body information;
the information coding model is obtained by utilizing historical decision main body information for training in advance.
3. The information matching method for health management decision making according to claim 2, wherein the process of encoding the structured subject information by using a preset information encoding model to obtain a subject information encoding result for embodying the features of the structured subject information comprises:
calculating a representation vector of each feature in the structured main body information by using the information coding model;
and integrating the expression vectors of all the characteristics by using the information coding model to obtain the main body information coding result for embodying the characteristics of the structured main body information.
4. The information matching method for health management decision as claimed in claim 2, wherein the process of calculating the correlation degree between the body information coding result and the health medical information coding result of each health medical information comprises:
calculating the association degree of the main body information coding result and the health medical information coding result of each health medical information by using a preset association degree calculation model;
the relevance calculation model is obtained by utilizing historical subject information codes and historical health medical information code results in advance for training.
5. The information matching method for health management decisions of claim 1, further comprising:
and classifying the health medical information in advance based on preset information, and clustering and storing the health medical information into a health medical information base.
6. The information matching method for health management decision making according to claim 5, wherein the process of extracting the health medical information corresponding to the health medical information coding result whose association degree with the main body information coding result exceeds a preset first association threshold to obtain the target health medical information comprises:
extracting the associated health medical information and the mapping health medical information of which the association degree with the encoding result of the associated health medical information exceeds a preset second association threshold from the health medical information base to obtain target health medical information comprising the associated health medical information and the mapping health medical information;
the related health medical information is health medical information corresponding to the health medical information coding result of which the degree of association with the main body information coding result exceeds a preset first association threshold value.
7. The information matching method for health management decisions of claim 5, further comprising:
adding new information classification and adding new health medical information under any information classification.
8. The information matching method for health management decision according to any of claims 1 to 7, wherein before the encoding process of the structured subject information, the method further comprises:
performing query intention analysis on the structured main body information by using a preset intention analysis model to obtain a query intention corresponding to the structured main body information;
judging whether key features corresponding to the query intention are missing or not by utilizing a preset feature and intention mapping relation;
if the key features are missing, the key features missing in the decision main body information are obtained from a corresponding database or the key features missing in the decision main body information supplemented by a user are received, so that complete decision main body information is obtained;
taking the complete decision main body information as decision main body information again so as to carry out structuralization processing to obtain the structuralization main body information;
the intention analysis model is obtained by training with historical structured subject information in advance, and the mapping relation between the features and the intention records the corresponding relation between the query intention and the corresponding key features.
9. The information matching method for health management decision as claimed in claim 8, wherein the generation process of the feature-to-intention mapping relation comprises:
randomly removing features in the historical decision main body information;
counting error probabilities between each query intention and the missing features, which cause the calculation errors of the relevance calculation model;
and determining the corresponding relation between each query intention and the key characteristics according to the error probability, and generating the characteristic-intention mapping relation.
10. An information matching system for health management decisions, comprising:
the main body information acquisition module is used for acquiring decision main body information;
the structuralization processing module is used for structuralizing the decision main body information to obtain structuralization main body information;
the coding processing module is used for coding the structured main body information to obtain a main body information coding result for embodying the characteristics of the structured main body information;
the relevancy calculation module is used for calculating the relevancy between the main body information coding result and the health medical information coding result of each piece of health medical information;
the medical information extraction module is used for extracting the health medical information corresponding to the health medical information coding result of which the correlation degree with the main body information coding result exceeds a preset first correlation threshold value to obtain target health medical information;
the health medical information coding result of each health medical information is obtained by coding each health medical information in advance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110920691.7A CN113571145A (en) | 2021-08-11 | 2021-08-11 | Information matching method and system for health management decision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110920691.7A CN113571145A (en) | 2021-08-11 | 2021-08-11 | Information matching method and system for health management decision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113571145A true CN113571145A (en) | 2021-10-29 |
Family
ID=78171266
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110920691.7A Pending CN113571145A (en) | 2021-08-11 | 2021-08-11 | Information matching method and system for health management decision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113571145A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030167189A1 (en) * | 2000-05-15 | 2003-09-04 | Alan G Gorman | System and method of drug disease matching |
| US20140172899A1 (en) * | 2012-12-14 | 2014-06-19 | Microsoft Corporation | Probability-based state modification for query dialogues |
| CN110619962A (en) * | 2019-08-26 | 2019-12-27 | 上海同筵信息科技有限公司 | Doctor-patient shared network medical service system |
| CN112069298A (en) * | 2020-07-31 | 2020-12-11 | 杭州远传新业科技有限公司 | Human-computer interaction method, device and medium based on semantic web and intention recognition |
| CN112151187A (en) * | 2020-10-15 | 2020-12-29 | 腾讯科技（深圳）有限公司 | Information query method and device, computer equipment and storage medium |
| CN112395392A (en) * | 2020-11-27 | 2021-02-23 | 浪潮云信息技术股份公司 | Intention identification method and device and readable storage medium |
| CN112863630A (en) * | 2021-01-20 | 2021-05-28 | 中国科学院自动化研究所 | Personalized accurate medical question-answering system based on data and knowledge |
| CN112992367A (en) * | 2021-03-23 | 2021-06-18 | 崔剑虹 | Smart medical interaction method based on big data and smart medical cloud computing system |
| CN113160954A (en) * | 2021-04-07 | 2021-07-23 | 泰康保险集团股份有限公司 | Medical resource allocation method and device, storage medium and electronic equipment |
-
2021
- 2021-08-11 CN CN202110920691.7A patent/CN113571145A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030167189A1 (en) * | 2000-05-15 | 2003-09-04 | Alan G Gorman | System and method of drug disease matching |
| US20140172899A1 (en) * | 2012-12-14 | 2014-06-19 | Microsoft Corporation | Probability-based state modification for query dialogues |
| CN110619962A (en) * | 2019-08-26 | 2019-12-27 | 上海同筵信息科技有限公司 | Doctor-patient shared network medical service system |
| CN112069298A (en) * | 2020-07-31 | 2020-12-11 | 杭州远传新业科技有限公司 | Human-computer interaction method, device and medium based on semantic web and intention recognition |
| CN112151187A (en) * | 2020-10-15 | 2020-12-29 | 腾讯科技（深圳）有限公司 | Information query method and device, computer equipment and storage medium |
| CN112395392A (en) * | 2020-11-27 | 2021-02-23 | 浪潮云信息技术股份公司 | Intention identification method and device and readable storage medium |
| CN112863630A (en) * | 2021-01-20 | 2021-05-28 | 中国科学院自动化研究所 | Personalized accurate medical question-answering system based on data and knowledge |
| CN112992367A (en) * | 2021-03-23 | 2021-06-18 | 崔剑虹 | Smart medical interaction method based on big data and smart medical cloud computing system |
| CN113160954A (en) * | 2021-04-07 | 2021-07-23 | 泰康保险集团股份有限公司 | Medical resource allocation method and device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111316281B (en) | Semantic classification method and system for numerical data in natural language context based on machine learning | |
| CN112863630A (en) | Personalized accurate medical question-answering system based on data and knowledge | |
| US11915127B2 (en) | Prediction of healthcare outcomes and recommendation of interventions using deep learning | |
| CN108461110B (en) | Medical information processing method, device and equipment | |
| CN107680661B (en) | System and method for estimating medical resource demand | |
| US20170147753A1 (en) | Method for searching for similar case of multi-dimensional health data and apparatus for the same | |
| CN111710429A (en) | Information pushing method and device, computer equipment and storage medium | |
| Xu et al. | Intelligent emotion detection method based on deep learning in medical and health data | |
| CN109074858A (en) | There is no hospital's matching in the health care data library for going identification of obvious standard identifier | |
| CN115579104A (en) | Artificial intelligence-based liver cancer full-course digital management method and system | |
| CN113724815A (en) | Information pushing method and device based on decision grouping model | |
| Qiu et al. | Egocentric image captioning for privacy-preserved passive dietary intake monitoring | |
| Shankar et al. | A novel discriminant feature selection–based mutual information extraction from MR brain images for Alzheimer's stages detection and prediction | |
| JP2015018462A (en) | Medical chart system and medical chart search method | |
| Leng et al. | Bi-level artificial intelligence model for risk classification of acute respiratory diseases based on Chinese clinical data | |
| CN110752027A (en) | Electronic medical record data pushing method and device, computer equipment and storage medium | |
| CN113241198B (en) | User data processing method, device, equipment and storage medium | |
| Mohapatra et al. | Automated invasive cervical cancer disease detection at early stage through deep learning | |
| CN113571145A (en) | Information matching method and system for health management decision | |
| CN113724878B (en) | Medical risk information pushing method and device based on machine learning | |
| EP3891761A1 (en) | Integrated system and method for personalized stratification and prediction of neurodegenerative disease | |
| Aulia et al. | Optimization of the Electronic Nose Sensor Array for Asthma Detection Based on Genetic Algorithm | |
| Ray et al. | Physical activity classification with dynamic discriminative methods | |
| Rout et al. | Identification of similar gastrointestinal images through content based image retrieval system based on analytical hierarchical process | |
| US20210133627A1 (en) | Methods and systems for confirming an advisory interaction with an artificial intelligence platform |
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 |