CN113657550A - Patient marking method, device, equipment and storage medium based on hierarchical calculation - Google Patents
Patient marking method, device, equipment and storage medium based on hierarchical calculation Download PDFInfo
- Publication number
- CN113657550A CN113657550A CN202111017437.2A CN202111017437A CN113657550A CN 113657550 A CN113657550 A CN 113657550A CN 202111017437 A CN202111017437 A CN 202111017437A CN 113657550 A CN113657550 A CN 113657550A
- Authority
- CN
- China
- Prior art keywords
- patient
- risk
- historical
- target
- marked
- 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
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000004364 calculation method Methods 0.000 title abstract description 7
- 239000003814 drug Substances 0.000 claims abstract description 187
- 201000010099 disease Diseases 0.000 claims abstract description 83
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 83
- 238000003745 diagnosis Methods 0.000 claims abstract description 37
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 23
- 238000003058 natural language processing Methods 0.000 claims abstract description 21
- 229940079593 drug Drugs 0.000 claims description 126
- 238000001914 filtration Methods 0.000 claims description 17
- 238000002372 labelling Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 206010012601 diabetes mellitus Diseases 0.000 description 42
- 239000008280 blood Substances 0.000 description 6
- 210000004369 blood Anatomy 0.000 description 6
- 208000017169 kidney disease Diseases 0.000 description 6
- 102000017011 Glycated Hemoglobin A Human genes 0.000 description 4
- 208000018262 Peripheral vascular disease Diseases 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 108091005995 glycated hemoglobin Proteins 0.000 description 4
- 208000019622 heart disease Diseases 0.000 description 4
- 208000015210 hypertensive heart disease Diseases 0.000 description 4
- 208000026106 cerebrovascular disease Diseases 0.000 description 3
- 208000002249 Diabetes Complications Diseases 0.000 description 2
- 206010012655 Diabetic complications Diseases 0.000 description 2
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 description 2
- 208000031226 Hyperlipidaemia Diseases 0.000 description 2
- 206010055171 Hypertensive nephropathy Diseases 0.000 description 2
- 208000013016 Hypoglycemia Diseases 0.000 description 2
- 208000008589 Obesity Diseases 0.000 description 2
- 208000017442 Retinal disease Diseases 0.000 description 2
- 206010038923 Retinopathy Diseases 0.000 description 2
- 201000004239 Secondary hypertension Diseases 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 239000008103 glucose Substances 0.000 description 2
- 230000002218 hypoglycaemic effect Effects 0.000 description 2
- 235000020824 obesity Nutrition 0.000 description 2
- 239000000047 product Substances 0.000 description 2
- 229940077274 Alpha glucosidase inhibitor Drugs 0.000 description 1
- 206010003011 Appendicitis Diseases 0.000 description 1
- 229940123208 Biguanide Drugs 0.000 description 1
- 239000003888 alpha glucosidase inhibitor Substances 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 150000004283 biguanides Chemical class 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 229940029980 drug used in diabetes Drugs 0.000 description 1
- 229940124645 emergency medicine Drugs 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000005802 health problem Effects 0.000 description 1
- 230000002473 insulinotropic effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000000018 receptor agonist Substances 0.000 description 1
- 229940044601 receptor agonist Drugs 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000013517 stratification Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
-
- 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
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Medicinal Chemistry (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Chemical & Material Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The application discloses a patient marking method, a patient marking device, patient marking equipment and a patient marking storage medium based on hierarchical calculation, relates to the technical field of intelligent medical treatment, and aims to analyze historical hospitalizing times and historical hospitalizing record marks of target patients who miss high-risk diagnosis labels, high-risk medicine labels and medicine type thresholds of the target diseases, and by means of hierarchical recognition and multi-dimensional recognition, comprehensive risk recognition is guaranteed, and marking accuracy is improved. The method comprises the following steps: acquiring a high-risk diagnosis label, a high-risk medicine label and a medicine type threshold; identifying a target patient among patients to be marked for the target disease; counting the historical hospitalizing times of the target patient in a historical time period; and when the historical hospitalizing times are larger than or equal to the hospitalizing time threshold value, acquiring the historical hospitalizing record of the target patient, identifying the historical hospitalizing record by adopting a natural language processing algorithm, determining the project index of the target patient, and marking the target patient by utilizing the risk label corresponding to the project index.
Description
Technical Field
The present application relates to the field of smart medical technology, and in particular, to a patient marking method, apparatus, device and storage medium based on hierarchical computation.
Background
With the continuous development of intelligent medical technology, people are more and more concerned about self health problems, and many hospitals, institutions and the like can mark risks of patients, so that services such as disease treatment and product recommendation are provided for the patients. Taking diabetes as an example, the purpose of marking the diabetes is to distinguish the risk levels of different diabetes, and supplement a diabetes risk label, so that the platform can analyze the diabetes with different risk levels, and then provide services such as disease treatment and product recommendation for the patients with different risk levels.
In the related art, many intelligent medical platforms obtain the index threshold value of the high-risk diabetic in the medical guide when marking the diabetic patient, and set the preset condition for screening the diabetic patient according to the index threshold value. If the type of diabetes is type one, without regular visits, the blood glucose indicator is greater than a preset threshold, the diabetic is flagged as high risk, otherwise, as medium low risk.
In the process of implementing the invention, the inventor finds that the related art has at least the following technical problems:
when the platform distinguishes the risk level of the diabetes mellitus patients, only the known diabetes mellitus types and the blood glucose indexes are used for rough distinguishing, so that some diabetes mellitus patients can be in a high-risk state, but the diabetes mellitus patients are classified in low-risk patients due to the fact that high-risk conditions set for the diabetes mellitus medically are not met, the risk level identification is not comprehensive enough, and the marking accuracy is low.
Disclosure of Invention
In view of this, the present application provides a patient marking method, apparatus, device and storage medium based on hierarchical computation, and mainly aims to solve the problems of incomplete risk level identification and low marking accuracy at present.
According to a first aspect of the present application, there is provided a method of patient labeling based on hierarchical computation, the method comprising:
obtaining a high-risk diagnosis label, a high-risk drug label and a drug category threshold, wherein the high-risk diagnosis label and the high-risk drug label are obtained in medical guidelines and have an association relation with the target disease, the drug category threshold is obtained by calculating a category average of historical drug categories of a plurality of sample patients, and the plurality of sample patients are marked as high-risk;
determining a target patient among patients to be labeled for a target disease according to the high-risk diagnosis label, the high-risk drug label and the drug category threshold, wherein the target patient is a patient among the patients to be labeled who does not hit the high-risk diagnosis label, the high-risk drug label and the drug category threshold of the target disease;
counting the historical hospitalizing times of the target patient in a historical time period;
and when the historical hospitalizing times are larger than or equal to the hospitalizing time threshold value, acquiring the historical hospitalizing record of the target patient, identifying the historical hospitalizing record by adopting a natural language processing algorithm, determining the project index of the target patient, and marking the target patient by utilizing the risk label corresponding to the project index.
Optionally, the determining a target patient among the patients to be marked for the target disease according to the high risk diagnostic label, the high risk drug label and the drug category threshold comprises:
acquiring the patient to be marked of the target disease, filtering the patient to be marked with the high-risk diagnosis label associated with the disease information in the patient to be marked, and marking the filtered patient to be marked as high-risk;
filtering the patients to be marked with the drug information associated with the high-risk drug label in the filtered patients to be marked, and marking the filtered patients to be marked as high-risk;
counting historical medicine types of each patient to be marked in the current remaining patients to be marked, filtering the patient to be marked of which the historical medicine type reaches the medicine type threshold value, and marking the filtered patient to be marked as high-risk;
and taking the patient to be marked which is left after filtering as the target patient.
Optionally, the counting the historical drug categories of each patient to be marked in the patients to be marked currently remaining includes:
for each patient to be marked in the currently remaining patients to be marked, reading a plurality of pieces of used medicine information of the patient to be marked in a specified time period;
obtaining a plurality of preset drug types related to the target disease, comparing the plurality of pieces of used drug information with the plurality of preset drug types, and determining the number of the types of the preset drug types hit by the plurality of pieces of used drug information;
and taking the category number as the historical medicine category of the patient to be marked.
Optionally, after the counting the historical number of times of hospitalizations of the target patient within the historical time period, the method further comprises:
determining a plurality of sample patients, querying each sample patient in the plurality of sample patients for a risk label of the target disease associated with the sample patient, the risk label being high risk or medium low risk;
grouping the plurality of sample patients according to the risk labels, and dividing the sample patients with consistent risk labels into the same group to obtain two sample patient groups;
for each sample patient group in the two sample patient groups, counting the historical hospitalization times of each sample patient included in the sample patient group as sample hospitalization times;
calculating the average medical treatment times of each sample patient group based on the sample medical treatment times included in each sample patient group to obtain two average medical treatment times;
and processing the two average hospitalizing times by adopting a quartile method to obtain a median of the two average hospitalizing times, and taking the median as the threshold of the hospitalizing times.
Optionally, after the counting the historical number of times of hospitalizations of the target patient within the historical time period, the method further comprises:
when the historical number of visits is less than the number of visits threshold, marking the target patient as high risk.
Optionally, the obtaining a historical visit record of the target patient, identifying the historical visit record by using a natural language processing algorithm, and determining the project index of the target patient includes:
acquiring historical clinic records of the target patient, and dividing the historical clinic records according to time nodes indicated by preset time intervals to obtain at least two clinic record segments;
for each of the at least two visit record segments, performing character recognition on the visit record segment by adopting the natural language processing algorithm, and determining a plurality of project names included in the visit record segment;
determining a designated item name of a designated item, inquiring the occurrence times of the designated item name in the plurality of item names, and using the occurrence times as a segment index of the visit record segment;
respectively counting segment indexes of each of the at least two visit record segments to obtain at least two segment indexes of the at least two visit record segments;
using the at least two segment indicators as project indicators for the target patient.
Optionally, the marking the target patient by using the risk label corresponding to the item index includes:
when the project index comprises at least two segment indexes which are both larger than an index threshold value, marking the target patient by adopting a risk identification indicating medium and low risk;
when there are segment indicators of the at least two segment indicators included in the project indicator that are below the indicator threshold, the target patient is flagged with a risk indicator indicating a high risk.
According to a second aspect of the present application, there is provided a patient marking apparatus based on hierarchical computation, the apparatus comprising:
an obtaining module, configured to obtain a high-risk diagnostic label, a high-risk drug label, and a drug category threshold, where the high-risk diagnostic label and the high-risk drug label are obtained in a medical guideline and have an association relationship with the target disease, the drug category threshold is obtained by calculating a category average of historical drug categories of a plurality of sample patients, and the plurality of sample patients are labeled as high-risk;
a first determination module, configured to determine a target patient from among patients to be labeled for a target disease according to the high-risk diagnosis label, the high-risk drug label, and the drug category threshold, where the target patient is a patient from among the patients to be labeled who does not hit the high-risk diagnosis label, the high-risk drug label, and the drug category threshold for the target disease;
the statistical module is used for counting the historical hospitalizing times of the target patient in a historical time period;
and the marking module is used for acquiring the historical treatment record of the target patient when the historical treatment times are more than or equal to the treatment times threshold, identifying the historical treatment record by adopting a natural language processing algorithm, determining the project index of the target patient, and marking the target patient by utilizing the risk label corresponding to the project index.
Optionally, the first determining module is configured to obtain the patient to be labeled with the target disease, filter the patient to be labeled with the disease information associated with the high-risk diagnosis label in the patient to be labeled, and label the filtered patient to be labeled as a high risk; filtering the patients to be marked with the drug information associated with the high-risk drug label in the filtered patients to be marked, and marking the filtered patients to be marked as high-risk; counting historical medicine types of each patient to be marked in the current remaining patients to be marked, filtering the patient to be marked of which the historical medicine type reaches the medicine type threshold value, and marking the filtered patient to be marked as high-risk; and taking the patient to be marked which is left after filtering as the target patient.
Optionally, the first determining module is configured to, for each patient to be marked in the currently remaining patients to be marked, read a plurality of pieces of information about the used drugs of the patient to be marked in a specified time period; obtaining a plurality of preset drug types related to the target disease, comparing the plurality of pieces of used drug information with the plurality of preset drug types, and determining the number of the types of the preset drug types hit by the plurality of pieces of used drug information; and taking the category number as the historical medicine category of the patient to be marked.
Optionally, the apparatus further comprises:
a second determination module for determining a plurality of sample patients, querying a risk label of the target disease associated with each of the plurality of sample patients, the risk label being high risk or medium low risk;
the grouping module is used for grouping the plurality of sample patients according to the risk labels, and dividing the sample patients with consistent risk labels into the same group to obtain two sample patient groups;
the statistic module is further used for counting the historical hospitalizing times of each sample patient included in the sample patient groups as the sample hospitalizing times for each sample patient group in the two sample patient groups;
the calculation module is used for calculating the average hospitalizing times of each sample patient group based on the sample hospitalizing times included in each sample patient group to obtain two average hospitalizing times;
and the processing module is used for processing the two average hospitalizing times by adopting a quartile method to obtain a median of the two average hospitalizing times, and taking the median as the hospitalizing time threshold.
Optionally, the tagging module is further configured to tag the target patient as high risk when the historical number of visits is less than the number of visits threshold.
Optionally, the marking module is configured to obtain a historical visit record of the target patient, and divide the historical visit record according to a time node indicated by a preset time interval to obtain at least two visit record segments; for each of the at least two visit record segments, performing character recognition on the visit record segment by adopting the natural language processing algorithm, and determining a plurality of project names included in the visit record segment; determining a designated item name of a designated item, inquiring the occurrence times of the designated item name in the plurality of item names, and using the occurrence times as a segment index of the visit record segment; respectively counting segment indexes of each of the at least two visit record segments to obtain at least two segment indexes of the at least two visit record segments; using the at least two segment indicators as project indicators for the target patient.
Optionally, the tagging module is configured to tag the target patient with a risk identifier indicating a medium-low risk when the item indicator includes at least two segment indicators that are both greater than an indicator threshold; when there are segment indicators of the at least two segment indicators included in the project indicator that are below the indicator threshold, the target patient is flagged with a risk indicator indicating a high risk.
According to a third aspect of the present application, there is provided a computer device comprising a memory storing a computer program and a processor implementing the steps of the method of any of the first aspects when the computer program is executed.
According to a fourth aspect of the present application, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of any of the first aspects described above.
By means of the technical scheme, the patient marking method, the patient marking device, the patient marking equipment and the storage medium based on the hierarchical calculation are provided, the historical hospitalizing times of the target patient in the historical time period are counted for the target patient with the high-risk diagnosis label, the high-risk medicine label and the medicine type threshold value of the target disease missing in the target disease, and when the historical hospitalizing times are more than or equal to the hospitalizing time threshold value, acquiring the historical hospitalizing record of the target patient, identifying the historical hospitalizing record by adopting a natural language processing algorithm, determining the project index of the target patient, marking the target patient by utilizing a risk label corresponding to the project index, and utilizing a layered identification mode, the risk identification of the patient is carried out in multiple dimensions such as the use of medicines, related indexes, the number of times of hospitalizing and the like, so that the comprehensive identification risk level is ensured, and the marking accuracy is improved.
The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the application. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a schematic flow chart illustrating a patient labeling method based on hierarchical computation according to an embodiment of the present application;
FIG. 2A is a flow chart of a patient labeling method based on hierarchical computation according to an embodiment of the present application;
FIG. 2B is a schematic flow chart illustrating a patient labeling method based on hierarchical computation according to an embodiment of the present application;
FIG. 3A is a schematic diagram illustrating a patient marking device based on hierarchical computation according to an embodiment of the present application;
FIG. 3B is a schematic diagram illustrating a patient marking device based on hierarchical computation according to an embodiment of the present application;
fig. 4 shows a schematic device structure diagram of a computer apparatus according to an embodiment of the present application.
Detailed Description
Exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The embodiment of the application provides a patient marking method based on hierarchical calculation, as shown in fig. 1, the method includes:
101. the method comprises the steps of obtaining a high-risk diagnosis label, a high-risk medicine label and a medicine type threshold value, wherein the high-risk diagnosis label and the high-risk medicine label are obtained in a medical guide and have an association relation with a target disease, the medicine type threshold value is obtained by calculating a type average value of historical medicine types of a plurality of sample patients, and the plurality of sample patients are marked as high-risk.
102. And determining a target patient from the patients to be marked of the target disease according to the high-risk diagnosis label, the high-risk medicine label and the medicine category threshold, wherein the target patient is a patient who does not hit the high-risk diagnosis label, the high-risk medicine label and the medicine category threshold of the target disease.
103. And counting the historical hospitalizing times of the target patient in the historical time period.
104. And when the historical hospitalizing times are larger than or equal to the hospitalizing time threshold value, acquiring the historical hospitalizing record of the target patient, identifying the historical hospitalizing record by adopting a natural language processing algorithm, determining the project index of the target patient, and marking the target patient by utilizing the risk label corresponding to the project index.
According to the method provided by the embodiment of the application, for the target patient who has missed the high-risk diagnosis label, the high-risk medicine label and the medicine type threshold of the target disease, the historical hospitalization times of the target patient in a historical time period are counted, when the historical hospitalization times is larger than or equal to the hospitalization time threshold, the historical hospitalization record of the target patient is obtained, the historical hospitalization record is identified by adopting a natural language processing algorithm, the project index of the target patient is determined, the target patient is marked by utilizing the risk label corresponding to the project index, and the risk identification of the patient is carried out in multiple dimensions such as medicine use, related index use, hospitalization times and the like by utilizing a layered identification mode, so that the comprehensive identification risk level is ensured, and the marking accuracy is improved.
The embodiment of the present application provides a patient marking method based on hierarchical computation, as shown in fig. 2A, the method includes:
201. the target patient is determined among the patients to be marked for the target disease.
Diabetes is not clearly distinguishable in risk stratification from other diseases, and clinically it can be classified into type one diabetes and type two diabetes. Type one diabetes patients are high risk groups for type two diabetes patients because in real life type one diabetes patients are more likely to develop disease and often are accompanied by high risk complications. Some current online platforms are generally judged according to whether a patient belongs to a type one diabetes patient when marking whether the patient belongs to a high risk of diabetes, and are marked as a high risk if the patient belongs to a type one diabetes and are marked as a low risk otherwise. However, the applicant recognizes that the way of marking the patients is rough, and there is a high possibility that some diabetic patients are already in a high risk state in the medicine, some indexes or medical situations, but are classified in the middle-low risk patients due to the fact that the judgment conditions for the diabetes are not met, so that the risk level identification is not comprehensive enough, and the marking accuracy is low.
Therefore, the method can be applied to an intelligent medical platform, so that the intelligent medical platform can count the historical hospitalizing times of a target patient in a historical time period for the target patient who has not hit a high-risk diagnosis label, a high-risk medicine label and a medicine type threshold value of the target disease, obtain the historical hospitalizing record of the target patient when the historical hospitalizing times is larger than or equal to the hospitalizing times threshold value, identify the historical hospitalizing record by adopting a natural language processing algorithm, determine a project index of the target patient, mark the target patient by using the risk label corresponding to the project index, and perform the risk identification of the patient in multiple dimensions such as medicine, related indexes, hospitalizing times and the like by using a layered identification mode, thereby ensuring the comprehensive identification risk level and improving the marking accuracy. The intelligent medical platform can be operated based on an independent server, and can also be operated based on a server providing basic cloud computing such as cloud service, a cloud database, cloud computing, a cloud function, cloud storage, Network service, cloud communication, middleware service, domain name service, security service, Content Delivery Network (CDN), big data and an artificial intelligence platform. Thus, medical data, such as personal health records, prescriptions, examination reports, etc., relating to the target patient can be downloaded on the server upon subsequent identification of the relevant information of the target patient.
The target patient is a patient who misses the high-risk diagnosis label, the high-risk drug label and the drug type threshold of the target disease in the patient to be labeled, that is, many high-risk patients can be identified by some existing identification rules, the identified patients obviously belonging to high risk can be directly labeled as high risk, and the patient who does not obviously hit the high risk condition is taken as the target patient to be subsequently identified in multiple dimensions. In order to identify the patients with high risks, the intelligent medical platform obtains high-risk diagnosis labels and high-risk drug labels which are associated with the target diseases in advance from a medical guideline, determines a plurality of sample patients marked as high risks, calculates the average value of the types of the historical drugs of the sample patients to obtain a drug type threshold, and then determines the target patients to perform subsequent multi-dimensional identification by using the high-risk diagnosis labels, the high-risk drug labels and the drug type threshold, wherein the process of determining the target patients in the patients to be marked of the target diseases is as follows:
firstly, a patient to be marked of a target disease is obtained, the patient to be marked with the disease information associated with the high-risk diagnosis label is filtered in the patient to be marked according to the high-risk diagnosis label, the high-risk medicine label and the medicine type threshold, and the filtered patient to be marked is marked as a high risk. Specifically, when identifying the disease information, two methods may be adopted, one method is to acquire patient data information of the patient to be marked, identify the patient data information in a text identification manner, extract the disease category in the patient data information as the target disease category, and use the patient to be marked associated with the high-risk diagnosis label of the target disease in the target disease category as the target patient. It should be noted that, because the patient to be marked may go to a hospital for a diagnosis due to other diseases, such as cold, appendicitis, etc., and the patient data information includes the disease category unrelated to the target disease, all the disease categories of the patient to be marked can be screened by using the disease category related to the target disease known in the medical guideline, and the disease category related to the target disease in the patient data information can be used as the target disease category. Alternatively, the patient to be labeled is identified in an ICD (International Classification of Diseases) disease code for the patient to be labeled as to whether the patient meets a high risk diagnostic label for the target disease, and upon determining that the patient meets the high risk diagnostic label, the patient to be labeled is filtered. For example, in the case of diabetes, patients to be labeled who meet high risk diagnosis labels such as type one diabetes, specific type diabetes, diabetic complications, secondary hypertension, cerebrovascular disease, heart disease, kidney disease, peripheral vascular disease, hypertensive heart disease, hypertensive renal disease, hypertensive heart and kidney disease, hypoglycemia, hyperlipidemia, obesity, etc. are identified in the ICD disease code, and are filtered, and further, since patients suffering from these diseases are already at high risk and do not need to be subsequently evaluated, the filtered patients to be labeled are directly labeled as high risk.
Then, the patients to be marked with the drug information associated with the high-risk drug label are filtered from the filtered patients to be marked, and the filtered patients to be marked are marked as high-risk. Specifically, the medicine information used for treating the target disease in the patient data information may be extracted in a text recognition manner, and specifically includes the medicine type filled in the medical order corresponding to the target disease and the medicine type purchased by the medical insurance card in a pharmacy or a hospital. In addition, in the actual operation process, the extracted medicine type can be judged by combining with the medical guideline, and the medicine which is described in the medical guideline and can treat the target disease is reserved as the medicine information. In this way, the patient to be marked with the high-risk drug label associated in the drug information is directly filtered as the high-risk patient. For example, continuing with the example of diabetes, patients with diabetes are identified as being at risk to be marked with diabetes drugs, cerebrovascular drugs, heart disease drugs, kidney disease drugs, peripheral vascular disease drugs, retinopathy drugs, diabetes emergency drugs, and these patients are filtered.
Then, considering that the patients using the multiple drugs related to the target disease are actually more serious and should be marked as high risk, it is necessary to count the historical drug categories of each patient to be marked in the current remaining patients to be marked, filter the patients to be marked whose historical drug categories reach the drug category threshold, and mark the filtered patients to be marked as high risk. Taking any one of the remaining patients to be marked as an example, the process of counting the historical drug categories is as follows: first, for each patient to be marked in the remaining patients to be marked currently, a plurality of pieces of medication information of the patient to be marked in a specified time period are read, such as statistics of the medication information of the patient to be marked in the past year. Then, a plurality of preset drug types related to the target disease are obtained, the plurality of pieces of used drug information are compared with the plurality of preset drug types, the number of the types of the preset drug types hit by the plurality of pieces of used drug information is determined, and the number of the types is used as the historical drug types of the patient to be marked. That is, the number of the drug types counted is used as the historical drug type by inquiring how many drugs related to the target disease have been used by the patient to be labeled in the past year. After the historical drug categories are determined, the patients to be marked whose historical drug categories reach the drug category threshold may be filtered, and the filtered patients to be marked may be marked as high risk. Continuing with the example of diabetes, the pre-set drug classes for diabetes medication are insulinotropic agents, alpha-glucosidase inhibitors, biguanides, GLP receptor agonists, etc., such that if the above-mentioned 3 classes are hit in the historical drug classes of the patient to be labeled, and the threshold of the drug class is also 3, the patient to be labeled is labeled as high risk. It should be noted that the drug category threshold may be set according to an index threshold described in a medical guideline, or a minimum value of the number of the drug categories used by the diabetic in the historical claims information may be used as the drug category threshold.
Through the process, the patient to be marked which obviously belongs to high risk is filtered and directly marked as high risk, and the patient to be marked which is left after filtering is the patient which needs to be subjected to multi-dimensional risk assessment subsequently, so that the patient to be marked which is left after filtering is taken as the target patient.
202. Counting the historical hospitalizing times of the target patient in the historical time period, and executing the following step 203 when the historical hospitalizing times are greater than or equal to a hospitalizing time threshold value; when the historical number of visits is less than the number of visits threshold, the following step 204 is performed.
In the embodiment of the application, considering that the risk of the patient who does not have regular medical treatment is actually higher than that of the patient who has regular medical treatment, the intelligent medical platform can count the historical medical treatment times of the target patient in the historical time period, and determine whether the target patient is at a high risk according to the historical medical treatment times. When the historical hospitalizing times are evaluated, a standard for judging whether the historical hospitalizing times are enough, namely a hospitalizing time threshold value, needs to be set, so that the historical hospitalizing times and the hospitalizing time threshold value can be compared in the following process, and how to mark the target patient can be determined. The specific process for generating the medical treatment times threshold is as follows:
first, a plurality of sample patients are determined, and a risk label of a target disease associated with each of the plurality of sample patients is queried, wherein the risk label is high risk or medium low risk. Grouping a plurality of sample patients according to the risk labels, and dividing the sample patients with consistent risk labels into the same group to obtain two sample patient groups. Then, for each of the two sample patient groups, the historical hospitalization times of each sample patient included in the sample patient group are counted as sample hospitalization times, and the average hospitalization times of each sample patient group is calculated based on the sample hospitalization times included in each sample patient group, resulting in two average hospitalization times. And finally, processing the two average hospitalizing times by adopting a quartile method to obtain a median of the two average hospitalizing times, and taking the median as a hospitalizing time threshold. The above process is to group the sample patients according to high risk, medium and low risk, then count the average number of medical visits for both the high risk group and the medium and low risk group, and after obtaining two average number of medical visits, determine the median of the two average number of medical visits by using quartile, and take the median as the threshold of medical visits.
In this way, the historical hospitalizing times of the target patient in the historical time period are counted, and when the historical hospitalizing times is greater than or equal to the hospitalizing time threshold, it indicates that the target patient actually carries out hospitalizing regularly, the target patient cannot be directly marked as high risk, and further evaluation is required, so that the following step 203 is executed; when the historical hospitalization times are less than the hospitalization time threshold, it indicates that the target patient is not hospitalized regularly, there is a high risk, and it needs to be marked as a high risk, so the following step 204 is performed.
203. And when the historical hospitalizing times are larger than or equal to the hospitalizing time threshold value, acquiring the historical hospitalizing record of the target patient, identifying the historical hospitalizing record by adopting a natural language processing algorithm, determining the project index of the target patient, and marking the target patient by utilizing the risk label corresponding to the project index.
In the embodiment of the application, when the historical hospitalization times are greater than or equal to the hospitalization time threshold, it indicates that the target patient is actually hospitalized regularly, and the target patient cannot be directly marked as high risk, and further evaluation is required. Considering that some patients only perform some main item examinations related to the target disease in the historical process and do not count up the regular medical visits, in the continuous evaluation process, the historical medical visits of the target patients need to be acquired, and the historical medical visits are divided according to time nodes indicated by preset time intervals to obtain at least two medical visit record segments. For example, the medical records of the target patient in the past two years are acquired, and if the preset time interval is one year, the medical records of each year in the past two years are obtained by splitting. And then, for each of the at least two visit record segments, performing character recognition on the visit record segment by adopting a natural language processing algorithm, and determining a plurality of project names included in the visit record segment. The natural language processing algorithm may be an NLP (Neuro-linear Programming) algorithm, all the project names are acquired as samples, the NLP algorithm is used to train the samples to obtain a language identification model, and then the visit record segments are input to the language identification model, so that a plurality of project names in the visit record segments can be directly output, or the project names are directly marked with yellow in the visit record segments. Then, a specified item name of the specified item is determined, the number of occurrences of the specified item name in the plurality of item names is queried, and the number of occurrences is used as a segment index of the visit record segment. For example, when the item names "blood sugar" and "glycated hemoglobin (HBA 1C)" are specified for a diabetic patient, the number of occurrences of "blood sugar" and "glycated hemoglobin (HBA 1C)" in the visit record fragment is counted as a fragment index, that is, the item name glycated hemoglobin (HBA1C) or blood sugar count is counted once.
By repeatedly executing the above process, the segment indexes of each of the at least two visit record segments can be respectively counted to obtain at least two segment indexes of the at least two visit record segments, and the at least two segment indexes are used as project indexes of the target patient. After the project index of the target patient is determined, when the project index comprises at least two segment indexes which are both larger than the index threshold value, the target patient is indicated to be in fact focused on the examination of the relevant target disease, the risk is low, and therefore, the target patient is marked by adopting a risk identifier indicating medium and low risk. And when the item index comprises at least two segment indexes with segment indexes lower than the index threshold value, the target patient is not focused on daily examination, the risk of the target patient is higher, and therefore, the target patient is marked by using a risk identifier indicating high risk.
204. When the historical number of visits is less than the number of visits threshold, the target patient is flagged as high risk.
In the embodiment of the present application, when the historical number of times of medical visits is less than the threshold number of times of medical visits, it indicates that the target patient does not have a regular medical visit, there is a higher risk, and it needs to be marked as a high risk, and therefore, the target patient is marked as a high risk.
In summary, taking diabetes as an example, the technical process of the present application is as follows:
referring to fig. 2B, among patients to be labeled with diabetes, patients to be labeled who meet high risk diagnostic markers such as type one diabetes, type specific diabetes, diabetic complications, secondary hypertension, cerebrovascular disease, heart disease, kidney disease, peripheral vascular disease, hypertensive heart disease, hypertensive renal disease, hypertensive heart and kidney disease, hypoglycemia, hyperlipidemia, obesity, and the like are filtered. Then, the patients to be marked with diabetes medicine, cerebrovascular disease medicine, heart disease medicine, kidney disease medicine, peripheral vascular disease medicine, retinopathy medicine and diabetes emergency medicine are filtered. And finally, filtering the patients to be marked with the used historical drug types reaching the drug type threshold value, and taking the rest patients to be marked as target patients. Next, the historical times of medical visits of the target patient in the historical time period are counted, and when the historical times of medical visits is greater than or equal to the medical visit threshold value, the target patient belongs to regular medical visits, so the number of times that the target patient examines blood sugar and glycated hemoglobin (HBA1C) per year is continuously calculated, and when the number of times is greater than or equal to 1, the target patient is marked as medium-low risk, and when the number of times is equal to 0, the target patient is marked as high-risk. And when the historical hospitalization times are less than the hospitalization time threshold, the target patient belongs to irregular hospitalization, and the target patient is marked as high risk.
According to the method provided by the embodiment of the application, for the target patient who has missed the high-risk diagnosis label, the high-risk medicine label and the medicine type threshold of the target disease, the historical hospitalization times of the target patient in a historical time period are counted, when the historical hospitalization times is larger than or equal to the hospitalization time threshold, the historical hospitalization record of the target patient is obtained, the historical hospitalization record is identified by adopting a natural language processing algorithm, the project index of the target patient is determined, the target patient is marked by utilizing the risk label corresponding to the project index, and the risk identification of the patient is carried out in multiple dimensions such as medicine use, related index use, hospitalization times and the like by utilizing a layered identification mode, so that the comprehensive identification risk level is ensured, and the marking accuracy is improved.
Further, as a specific implementation of the method shown in fig. 1, the present application provides a patient marking apparatus based on hierarchical computation, as shown in fig. 3A, the apparatus includes: an acquisition module 301, a first determination module 302, a statistics module 303, and a labeling module 304.
The obtaining module 301 is configured to obtain a high-risk diagnostic label, a high-risk drug label, and a drug category threshold, where the high-risk diagnostic label and the high-risk drug label are obtained in a medical guideline and have an association relationship with the target disease, the drug category threshold is obtained by calculating a category average of historical drug categories of a plurality of sample patients, and the plurality of sample patients are labeled as high-risk;
the first determining module 302 is configured to determine a target patient among patients to be labeled with a target disease according to the high-risk diagnosis label, the high-risk drug label, and the drug category threshold, where the target patient is a patient among the patients to be labeled who does not hit the high-risk diagnosis label, the high-risk drug label, and the drug category threshold of the target disease;
the statistic module 303 is configured to count the historical hospitalizing times of the target patient within a historical time period;
the marking module 304 is configured to, when the historical medical treatment times are greater than or equal to the medical treatment time threshold, obtain a historical medical treatment record of the target patient, identify the historical medical treatment record by using a natural language processing algorithm, determine a project index of the target patient, and mark the target patient by using a risk label corresponding to the project index.
In a specific application scenario, the first determining module 302 is configured to obtain the patient to be marked with the target disease, filter, in the patient to be marked, the patient to be marked with the high-risk diagnosis label associated with disease information, and mark the filtered patient to be marked as a high risk; filtering the patients to be marked with the drug information associated with the high-risk drug label in the filtered patients to be marked, and marking the filtered patients to be marked as high-risk; counting historical medicine types of each patient to be marked in the current remaining patients to be marked, filtering the patient to be marked of which the historical medicine type reaches the medicine type threshold value, and marking the filtered patient to be marked as high-risk; and taking the patient to be marked which is left after filtering as the target patient.
In a specific application scenario, the first determining module 302 is configured to, for each patient to be marked in the currently remaining patients to be marked, read a plurality of pieces of used medicine information of the patient to be marked within a specified time period; obtaining a plurality of preset drug types related to the target disease, comparing the plurality of pieces of used drug information with the plurality of preset drug types, and determining the number of the types of the preset drug types hit by the plurality of pieces of used drug information; and taking the category number as the historical medicine category of the patient to be marked.
In a specific application scenario, as shown in fig. 3B, the apparatus further includes: a second determination module 305, a grouping module 306, a calculation module 307 and a processing module 308.
The second determining module 305 is configured to determine a plurality of sample patients, query a risk label of the target disease associated with each of the plurality of sample patients, wherein the risk label is a high risk or a medium or low risk;
the grouping module 306 is configured to group the plurality of sample patients according to the risk labels, and divide the sample patients with the consistent risk labels into the same group to obtain two sample patient groups;
the statistic module 303 is further configured to, for each sample patient group of the two sample patient groups, count the historical hospitalization times of each sample patient included in the sample patient group as the sample hospitalization times;
the calculating module 307 is configured to calculate an average medical treatment time of each sample patient group based on the sample medical treatment times included in each sample patient group, so as to obtain two average medical treatment times;
the processing module 308 is configured to process the two average medical treatment times by using a quartile method to obtain a median of the two average medical treatment times, and use the median as the medical treatment time threshold.
In a specific application scenario, the labeling module 304 is further configured to label the target patient as high risk when the historical number of times of medical visits is less than the number of times of medical visits threshold.
In a specific application scenario, the marking module 304 is configured to obtain a historical visit record of the target patient, and divide the historical visit record according to a time node indicated by a preset time interval to obtain at least two visit record segments; for each of the at least two visit record segments, performing character recognition on the visit record segment by adopting the natural language processing algorithm, and determining a plurality of project names included in the visit record segment; determining a designated item name of a designated item, inquiring the occurrence times of the designated item name in the plurality of item names, and using the occurrence times as a segment index of the visit record segment; respectively counting segment indexes of each of the at least two visit record segments to obtain at least two segment indexes of the at least two visit record segments; using the at least two segment indicators as project indicators for the target patient.
In a specific application scenario, the labeling module 304 is configured to label the target patient with a risk identifier indicating a medium-low risk when the project index includes at least two segment indexes that are both greater than an index threshold; when there are segment indicators of the at least two segment indicators included in the project indicator that are below the indicator threshold, the target patient is flagged with a risk indicator indicating a high risk.
The device provided by the embodiment of the application, to the target patient who does not hit the high risk diagnosis label, the high risk medicine label and the medicine type threshold of the target disease in the target disease, the historical hospitalization times of the target patient in the historical time period is counted, when the historical hospitalization times is larger than or equal to the hospitalization time threshold, the historical hospitalization record of the target patient is obtained, the historical hospitalization record is identified by adopting a natural language processing algorithm, the project index of the target patient is determined, the target patient is marked by utilizing the risk label corresponding to the project index, the risk identification of the patient is carried out in multiple dimensions such as medicine use, related index and hospitalization times by utilizing a layered identification mode, the comprehensive identification risk level is ensured, and the marking accuracy is improved.
It should be noted that, for other corresponding descriptions of the functional units related to the patient marking device based on hierarchical computation provided in the embodiment of the present application, reference may be made to the corresponding descriptions in fig. 1 and fig. 2A to 2B, which are not described herein again.
In an exemplary embodiment, referring to fig. 4, there is further provided a device, which includes a bus, a processor, a memory, and a communication interface, and may further include an input/output interface and a display device, wherein the functional units may communicate with each other through the bus. The memory stores a computer program and the processor executes the program stored in the memory to perform the patient tagging method based on hierarchical computation in the above embodiments.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the hierarchical computation based patient labeling method.
Through the above description of the embodiments, those skilled in the art will clearly understand that the present application can be implemented by hardware, and also by software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.), and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the implementation scenarios of the present application.
Those skilled in the art will appreciate that the figures are merely schematic representations of one preferred implementation scenario and that the blocks or flow diagrams in the figures are not necessarily required to practice the present application.
Those skilled in the art will appreciate that the modules in the devices in the implementation scenario may be distributed in the devices in the implementation scenario according to the description of the implementation scenario, or may be located in one or more devices different from the present implementation scenario with corresponding changes. The modules of the implementation scenario may be combined into one module, or may be further split into a plurality of sub-modules.
The above application serial numbers are for description purposes only and do not represent the superiority or inferiority of the implementation scenarios.
The above disclosure is only a few specific implementation scenarios of the present application, but the present application is not limited thereto, and any variations that can be made by those skilled in the art are intended to fall within the scope of the present application.
Claims (10)
1. A method for patient labeling based on hierarchical computation, comprising:
obtaining a high-risk diagnosis label, a high-risk drug label and a drug category threshold, wherein the high-risk diagnosis label and the high-risk drug label are obtained in medical guidelines and have an association relation with the target disease, the drug category threshold is obtained by calculating a category average of historical drug categories of a plurality of sample patients, and the plurality of sample patients are marked as high-risk;
determining a target patient among patients to be labeled for a target disease according to the high-risk diagnosis label, the high-risk drug label and the drug category threshold, wherein the target patient is a patient among the patients to be labeled who does not hit the high-risk diagnosis label, the high-risk drug label and the drug category threshold of the target disease;
counting the historical hospitalizing times of the target patient in a historical time period;
and when the historical hospitalizing times are larger than or equal to the hospitalizing time threshold value, acquiring the historical hospitalizing record of the target patient, identifying the historical hospitalizing record by adopting a natural language processing algorithm, determining the project index of the target patient, and marking the target patient by utilizing the risk label corresponding to the project index.
2. The method of claim 1, wherein the identifying a target patient among the patients to be labeled for a target disease based on the high risk diagnostic label, the high risk drug label, and the drug category threshold comprises:
acquiring the patient to be marked of the target disease, filtering the patient to be marked with the high-risk diagnosis label associated with the disease information in the patient to be marked, and marking the filtered patient to be marked as high-risk;
filtering the patients to be marked with the drug information associated with the high-risk drug label in the filtered patients to be marked, and marking the filtered patients to be marked as high-risk;
counting historical medicine types of each patient to be marked in the current remaining patients to be marked, filtering the patient to be marked of which the historical medicine type reaches the medicine type threshold value, and marking the filtered patient to be marked as high-risk;
and taking the patient to be marked which is left after filtering as the target patient.
3. The method according to claim 2, wherein the counting each patient to be marked historical drug category in the currently remaining patients to be marked comprises:
for each patient to be marked in the currently remaining patients to be marked, reading a plurality of pieces of used medicine information of the patient to be marked in a specified time period;
obtaining a plurality of preset drug types related to the target disease, comparing the plurality of pieces of used drug information with the plurality of preset drug types, and determining the number of the types of the preset drug types hit by the plurality of pieces of used drug information;
and taking the category number as the historical medicine category of the patient to be marked.
4. The method of claim 1, wherein after the statistics of the historical number of visits by the target patient over the historical time period, the method further comprises:
determining a plurality of sample patients, querying each sample patient in the plurality of sample patients for a risk label of the target disease associated with the sample patient, the risk label being high risk or medium low risk;
grouping the plurality of sample patients according to the risk labels, and dividing the sample patients with consistent risk labels into the same group to obtain two sample patient groups;
for each sample patient group in the two sample patient groups, counting the historical hospitalization times of each sample patient included in the sample patient group as sample hospitalization times;
calculating the average medical treatment times of each sample patient group based on the sample medical treatment times included in each sample patient group to obtain two average medical treatment times;
and processing the two average hospitalizing times by adopting a quartile method to obtain a median of the two average hospitalizing times, and taking the median as the threshold of the hospitalizing times.
5. The method of claim 1, wherein after the statistics of the historical number of visits by the target patient over the historical time period, the method further comprises:
when the historical number of visits is less than the number of visits threshold, marking the target patient as high risk.
6. The method of claim 1, wherein the obtaining a historical encounter record of the target patient, identifying the historical encounter record using a natural language processing algorithm, and determining the project index of the target patient comprises:
acquiring historical clinic records of the target patient, and dividing the historical clinic records according to time nodes indicated by preset time intervals to obtain at least two clinic record segments;
for each of the at least two visit record segments, performing character recognition on the visit record segment by adopting the natural language processing algorithm, and determining a plurality of project names included in the visit record segment;
determining a designated item name of a designated item, inquiring the occurrence times of the designated item name in the plurality of item names, and using the occurrence times as a segment index of the visit record segment;
respectively counting segment indexes of each of the at least two visit record segments to obtain at least two segment indexes of the at least two visit record segments;
using the at least two segment indicators as project indicators for the target patient.
7. The method of claim 6, wherein the marking the target patient with the risk label corresponding to the item indicator comprises:
when the project index comprises at least two segment indexes which are both larger than an index threshold value, marking the target patient by adopting a risk identification indicating medium and low risk;
when there are segment indicators of the at least two segment indicators included in the project indicator that are below the indicator threshold, the target patient is flagged with a risk indicator indicating a high risk.
8. A patient marking device based on hierarchical computation, comprising:
an obtaining module, configured to obtain a high-risk diagnostic label, a high-risk drug label, and a drug category threshold, where the high-risk diagnostic label and the high-risk drug label are obtained in a medical guideline and have an association relationship with the target disease, the drug category threshold is obtained by calculating a category average of historical drug categories of a plurality of sample patients, and the plurality of sample patients are labeled as high-risk;
a first determination module, configured to determine a target patient from among patients to be labeled for a target disease according to the high-risk diagnosis label, the high-risk drug label, and the drug category threshold, where the target patient is a patient from among the patients to be labeled who does not hit the high-risk diagnosis label, the high-risk drug label, and the drug category threshold for the target disease;
the statistical module is used for counting the historical hospitalizing times of the target patient in a historical time period;
and the marking module is used for acquiring the historical treatment record of the target patient when the historical treatment times are more than or equal to the treatment times threshold, identifying the historical treatment record by adopting a natural language processing algorithm, determining the project index of the target patient, and marking the target patient by utilizing the risk label corresponding to the project index.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111017437.2A CN113657550A (en) | 2021-08-31 | 2021-08-31 | Patient marking method, device, equipment and storage medium based on hierarchical calculation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111017437.2A CN113657550A (en) | 2021-08-31 | 2021-08-31 | Patient marking method, device, equipment and storage medium based on hierarchical calculation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113657550A true CN113657550A (en) | 2021-11-16 |
Family
ID=78482644
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111017437.2A Pending CN113657550A (en) | 2021-08-31 | 2021-08-31 | Patient marking method, device, equipment and storage medium based on hierarchical calculation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113657550A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114334141A (en) * | 2022-03-09 | 2022-04-12 | 首都医科大学附属北京友谊医院 | Early warning method and device for cancer risk caused by X-ray examination |
| CN116072298A (en) * | 2023-04-06 | 2023-05-05 | 之江实验室 | Disease prediction system based on hierarchical marker distribution learning |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020120471A1 (en) * | 2000-08-30 | 2002-08-29 | Healtheheart, Inc. | Patient analysis and research system and associated methods |
| US20020123906A1 (en) * | 2000-12-29 | 2002-09-05 | Goetzke Gary A. | Chronic pain patient risk stratification system |
| TW200731148A (en) * | 2006-02-13 | 2007-08-16 | Yi-Cheng Yu | Health management system with circular risk references |
| KR20140141520A (en) * | 2013-05-30 | 2014-12-10 | 가톨릭대학교 산학협력단 | Method and system for predicting high risk based on patient safety |
| CN107358047A (en) * | 2017-07-13 | 2017-11-17 | 刘峰 | Diabetic assesses and management system |
| CN109166105A (en) * | 2018-08-01 | 2019-01-08 | 中国人民解放军南京军区南京总医院 | The malignancy of tumor risk stratification assistant diagnosis system of artificial intelligence medical image |
| CN109949936A (en) * | 2019-03-13 | 2019-06-28 | 成都数联易康科技有限公司 | A kind of Risk Forecast Method of being hospitalized again based on deep learning mixed model |
| CN110211690A (en) * | 2019-04-19 | 2019-09-06 | 平安科技（深圳）有限公司 | Disease risks prediction technique, device, computer equipment and computer storage medium |
| CN110993103A (en) * | 2019-11-28 | 2020-04-10 | 阳光人寿保险股份有限公司 | Method for establishing disease risk prediction model and method for recommending disease insurance product |
| CN111554365A (en) * | 2019-03-20 | 2020-08-18 | 华中科技大学同济医学院附属协和医院 | Chronic disease comprehensive service platform |
| CN111933237A (en) * | 2020-08-31 | 2020-11-13 | 平安国际智慧城市科技股份有限公司 | Chronic disease processing method and device and related equipment |
| CN111986809A (en) * | 2020-08-10 | 2020-11-24 | 武汉麦咚健康科技有限公司 | Hypertension drug recommendation system and method for doctor-assisted judgment |
| CN112036749A (en) * | 2020-08-31 | 2020-12-04 | 平安医疗健康管理股份有限公司 | Method and device for identifying risk user based on medical data and computer equipment |
| CN112132624A (en) * | 2020-09-27 | 2020-12-25 | 平安医疗健康管理股份有限公司 | Medical claims data prediction system |
-
2021
- 2021-08-31 CN CN202111017437.2A patent/CN113657550A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020120471A1 (en) * | 2000-08-30 | 2002-08-29 | Healtheheart, Inc. | Patient analysis and research system and associated methods |
| US20020123906A1 (en) * | 2000-12-29 | 2002-09-05 | Goetzke Gary A. | Chronic pain patient risk stratification system |
| TW200731148A (en) * | 2006-02-13 | 2007-08-16 | Yi-Cheng Yu | Health management system with circular risk references |
| KR20140141520A (en) * | 2013-05-30 | 2014-12-10 | 가톨릭대학교 산학협력단 | Method and system for predicting high risk based on patient safety |
| CN107358047A (en) * | 2017-07-13 | 2017-11-17 | 刘峰 | Diabetic assesses and management system |
| CN109166105A (en) * | 2018-08-01 | 2019-01-08 | 中国人民解放军南京军区南京总医院 | The malignancy of tumor risk stratification assistant diagnosis system of artificial intelligence medical image |
| CN109949936A (en) * | 2019-03-13 | 2019-06-28 | 成都数联易康科技有限公司 | A kind of Risk Forecast Method of being hospitalized again based on deep learning mixed model |
| CN111554365A (en) * | 2019-03-20 | 2020-08-18 | 华中科技大学同济医学院附属协和医院 | Chronic disease comprehensive service platform |
| CN110211690A (en) * | 2019-04-19 | 2019-09-06 | 平安科技（深圳）有限公司 | Disease risks prediction technique, device, computer equipment and computer storage medium |
| CN110993103A (en) * | 2019-11-28 | 2020-04-10 | 阳光人寿保险股份有限公司 | Method for establishing disease risk prediction model and method for recommending disease insurance product |
| CN111986809A (en) * | 2020-08-10 | 2020-11-24 | 武汉麦咚健康科技有限公司 | Hypertension drug recommendation system and method for doctor-assisted judgment |
| CN111933237A (en) * | 2020-08-31 | 2020-11-13 | 平安国际智慧城市科技股份有限公司 | Chronic disease processing method and device and related equipment |
| CN112036749A (en) * | 2020-08-31 | 2020-12-04 | 平安医疗健康管理股份有限公司 | Method and device for identifying risk user based on medical data and computer equipment |
| CN112132624A (en) * | 2020-09-27 | 2020-12-25 | 平安医疗健康管理股份有限公司 | Medical claims data prediction system |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114334141A (en) * | 2022-03-09 | 2022-04-12 | 首都医科大学附属北京友谊医院 | Early warning method and device for cancer risk caused by X-ray examination |
| CN116072298A (en) * | 2023-04-06 | 2023-05-05 | 之江实验室 | Disease prediction system based on hierarchical marker distribution learning |
| CN116072298B (en) * | 2023-04-06 | 2023-08-15 | 之江实验室 | Disease prediction system based on hierarchical marker distribution learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200258608A1 (en) | Medical database and system | |
| US11410760B2 (en) | Medical evaluation system and method for use therewith | |
| Murff et al. | Electronically screening discharge summaries for adverse medical events | |
| CN113657550A (en) | Patient marking method, device, equipment and storage medium based on hierarchical calculation | |
| US20220051114A1 (en) | Inference process visualization system for medical scans | |
| CN103370629A (en) | Clinical quality analytics system | |
| CN115762769A (en) | Intelligent risk early warning system after ERCP (effective Range planning) operation | |
| CN115775635A (en) | Medicine risk identification method and device based on deep learning model and terminal equipment | |
| US20140244295A1 (en) | Clinical quality analytics system with recursive, time sensitive event-based protocol matching | |
| CN115239945A (en) | Image annotation reliability prediction method and device, electronic equipment and storage medium | |
| CN109509550B (en) | Treatment information processing method, device, equipment and medium based on data analysis | |
| KR102407987B1 (en) | Methdo and apparatus for building bio data hub | |
| Ko et al. | Real‐Time Remote Patient Monitoring and Alarming System for Noncommunicable Lifestyle Diseases | |
| CN107480295A (en) | The modification method of user data | |
| CN113643776A (en) | Patient marking method, device, equipment and storage medium based on hierarchical calculation | |
| Song et al. | Un-apriori: A novel association rule mining algorithm for unstructured EMRs | |
| CN111710431A (en) | Method, device, equipment and storage medium for identifying synonymous diagnosis name | |
| CN115545955B (en) | Method and device for detecting abnormal data in medical archive data and electronic equipment | |
| Dang et al. | Sequence-based measure for assessing drug-side effect causal relation from electronic medical records | |
| US20240006067A1 (en) | System by which patients receiving treatment and at risk for iatrogenic cytokine release syndrome are safely monitored | |
| CN118692622A (en) | Big data-based target area crowd health data processing method and device | |
| CN112687367A (en) | Medical record grouping method, device and equipment based on dynamic disease condition and storage medium | |
| CN114240146A (en) | Diagnosis and treatment evaluation method and device and electronic equipment | |
| CN116245098A (en) | Injury monitoring method based on NLP technology | |
| Degli Esposti et al. | and Sandra Garcıa-Blasco |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220525 Address after: 518000 China Aviation Center 2901, No. 1018, Huafu Road, Huahang community, Huaqiang North Street, Futian District, Shenzhen, Guangdong Province Applicant after: Shenzhen Ping An medical and Health Technology Service Co.,Ltd. Address before: Room 12G, Area H, 666 Beijing East Road, Huangpu District, Shanghai 200001 Applicant before: PING AN MEDICAL AND HEALTHCARE MANAGEMENT Co.,Ltd. |